Transcript
Page 1: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Summary of LTTP Emulsion and Foamed Bitumen Treated Sections

Technical Memorandum

Final

March, 2007

PREPARED FOR:

Gauteng Department of Public Transport, Roads and Works Directorate: Design Private Bag X3 Lynn East, 0039

SABITA Postnet Suite 56 Private Bag X21 Howard Place 7450

PREPARED BY:

Modelling and Analysis Systems CC. PO Box 882 Cullinan 1000 (www.modsys1.com)

Authors:

F Long

F Jooste

Technical Memorandum CSIR/BE/IE/ER/2007/0006/B

Page 2: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Acknowledgements

- I -

ACKNOWLEDGEMENTS

The sponsors of this project; Gauteng Department of Transport, Roads and Public Works (represented for this study by Elzbieta Sadzik) and SABITA (represented by Trevor Distin and Piet Myburgh) are gratefully acknowledged. The project has been managed by Les Sampson of the Asphalt Academy. Successful completion of this project could only be realised by enlisting the help of many people who willingly contributed their time, effort and resources. The following people are specifically acknowledged for their valuable input to the project (by organisation and alphabetical order). • ASCH Consulting Engineers: Shaheen Nackerdien • Africon: Pablo Balmaceda • Arcus Gibb: Nichol van der Walt • BKS: Nick Kong and Ofal Motlankene • Cayad: Heuppeschenne Lekay • City of Cape Town: Mark Bondietti • CSIR: Adrian Bergh, Louw du Plessis, Dave Jones, Wynand Steyn and Hechter Theyse • Freetrans: Karl Arnold and Deirdre Elsmere • Gautrans: Alison Burger, Enrico Fletcher, Elzbieta Sadzik and Valerie van Staden • Goba Moahloli Keeve Steyn: John Hodgson, Mike Kelly, Brian Spottiswoode (retired), Andre

Ungerer and Tom van Zyl • HHO: Andrew Laatz • Jeffares and Green: Paul Olivier • Kwazulu-Natal: Pat Dorkin, Rob Lindsay, Mike Proudfoot • Kwezi-V3: Trevor Morgan • Loudon International/Siyenza: Dave Collings, Barry Nothard and Erik Uppink • Madan Singh and Associates: Ridwaan Ghany • MicroZone: Johan Smit • Ndodana Consulting Engineers: Henk Diedericks and Martinus Wilken • Norwegian Public Works Department: Charles Overby • SANRAL: Rob Damhuis, Nuno Gomez, Ron Harmse, Louw Kannemeyer, Werner Lategan, Renaldo

Lorio, Dennis Rossmann, Corne Roux, Fanie van Aardt, Kobus van der Walt and Mias Wiese • SCIP: Ben Botes • SNA: Johan Opperman and Rudi van Huyssteen • Stewart Scott International: Jaco Liebenberg • TRAC: Gawie Jordaan • Tshepega: Gerrit Jordaan and PW de Bruin • University of Pretoria: Emile Horak • UWP: Mike White, Adrian Skea • Vela-VKE: Mynhardt Augustyn, Derek Burger, Joe Grobler, Simon Kotze, Arthur Taute and John de

Wet • WCPA: Sydney Crocker, Mervyn Henderson and Andre van der Gryp • WSP: Andries Bester, Hugh Thompson, Mike Hughes (independent), Hennie Landman, Chris Roux

Page 3: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Table of Contents

- I -

TABLE OF CONTENTS

ACKNOWLEDGEMENTS..............................................................................................................................I 1 BACKGROUND AND INTRODUCTION ............................................................................................1 1.1 Project Objectives ..........................................................................................................................1 1.2 Scope and Context of This Document ...........................................................................................1 1.3 Report Structure.............................................................................................................................2 2 METHOD OF DATA COLLECTION ...................................................................................................3 2.1 Identifying and Collecting Data for BSM Sections .........................................................................3 2.2 Emulsion and Foamed Bitumen Sections Identified ......................................................................3 3 SELECTION AND PRÉCIS OF LTPP SECTIONS INVESTIGATED.................................................9 3.1 Selection of Sections to Investigate ...............................................................................................9

3.1.1 Identification of Deficiencies in Collected Data....................................................................9 3.2 LTPP Sections investigated .........................................................................................................10 3.3 Précis of LTPP Sections ..............................................................................................................11

3.3.1 D2388 (Cullinan) ................................................................................................................11 3.3.2 MR27 (near Stellenbosch) .................................................................................................11 3.3.3 N1 Section 1 (Kraaifontein)................................................................................................12 3.3.4 N1 Section 13 (Springfontein and Trompsburg) ................................................................12 3.3.5 N1 Section 14 (Springfontein and Trompsburg) ................................................................12 3.3.6 N2 Section 16 (Kwelera, East London)..............................................................................13 3.3.7 N2 Section 20 (Mount Frere to Mount Ayliff) .....................................................................13 3.3.8 N3 Section 4 (near Mooi River) .........................................................................................13 3.3.9 N3 Section 12, Northbound (Modderfontein to Buccleugh)...............................................14 3.3.10 N3 Section 12, Southbound (Modderfontein to Buccleugh) ..............................................14 3.3.11 N4 Section 1 (Scientia interchange to Pienaars River) .....................................................14 3.3.12 N4 Section 5X (Wonderfontein to Crossroads, KM 20 to 25)............................................14 3.3.13 N4 Section 5X (Wonderfontein to Crossroads, KM 27.6 to 30.6)......................................15 3.3.14 N7 Section 7 (near Kammieskroon)...................................................................................15 3.3.15 N12 Section 19 (near Daveyton, Experimental Section 1) ................................................15 3.3.16 N12 Section 19 (near Daveyton, Experimental Section 2) ................................................16 3.3.17 P23/1 (Kroonstad to Steynsrus).........................................................................................16 3.3.18 Road 1386 (Moloto Rd), Stabilised Subbase ....................................................................16 3.3.19 Road 1386 (Moloto Rd), Unstabilised Subbase ................................................................17 3.3.20 MR504 (near Shongweni)..................................................................................................17 3.3.21 N11 Section 8 (Ermelo to Hendrina)..................................................................................17 3.3.22 P24/1 (near Vereeniging)...................................................................................................18 3.3.23 P243/1 (near Vereeniging).................................................................................................18 3.3.24 R22-4 (MR439) (near Phelandaba) ...................................................................................18 3.3.25 Same-Himo (Tanzania)......................................................................................................18 3.3.26 TR16-3 (R27) (Nieuwoudtville to Calvinia) ........................................................................19 3.3.27 Summary of Key Information from LTPP Summaries........................................................19

4 IDENTIFICATION OF TRENDS........................................................................................................22 5 SUMMARY AND RECOMMENDATIONS ........................................................................................26 5.1 Summary......................................................................................................................................26 5.2 Recommendations .......................................................................................................................26 6 REFERENCES..................................................................................................................................27 APPENDIX A: PROJECT WORK PROPOSAL.........................................................................28 APPENDIX B: LTPP SUMMARIES ...........................................................................................29

Page 4: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Lists of Table and Figures

- II -

LIST OF FIGURES

Figure 1: Visual Summary of LTPP Sections (Foamed Bitumen Sections)..................................... 24 Figure 2: Visual Summary of LTPP Sections (Emulsion Sections).................................................. 25

LIST OF TABLES

Table 1: List of Emulsion Roads ........................................................................................................5 Table 2: List of Foamed Bitumen Roads ...........................................................................................8 Table 3: Pavement Structures for MR504 Sections ....................................................................... 17 Table 4: Summary of LTPP Sections Information used in Structural Design Method

Development..................................................................................................................... 20

Page 5: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Background and Introduction

- 1 -

1 BACKGROUND AND INTRODUCTION

The stabilization of materials using emulsion and foamed bitumen materials is an accepted method of road construction in South Africa, particularly for rehabilitation projects involving recycling of the existing pavement structure. With the increased focus on conservation of natural resources, the use of these materials as a method of pavement rehabilitation is expected to increase.

In the last few years a need for updated guidelines for the design and use of emulsion and foamed bitumen treated materials has been identified by the broader industry. Because of the many similarities between foamed and emulsified bitumen treated materials it was decided that it would be appropriate to publish one guideline document for Bitumen Stabilized Materials (BSM) and that this guideline would cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document. The project was designed to address two key deficiencies in existing guidelines: (a) structural pavement design; and (b) mix design.

In view of the considerable scope of the project, it was planned in several phases, starting with an Inception Study (Phase 1) completed during 2006. The second phase of the project is currently in progress and is scheduled for completion in 2008. This technical memorandum forms part of the second phase of the structural pavement design component of the study. The work proposal for the structural design component is included in Appendix A.

1.1 PROJECT OBJECTIVES

The objectives of this project, as outlined in the project proposal, are as follows:

• To expand the Long Term Pavement Performance (LTPP) database compiled during the Inception Study to a sufficient level to enable the development of classification based design guidelines.

• Develop a classification based design method.

• Peer review of the recommended design guidelines.

1.2 SCOPE AND CONTEXT OF THIS DOCUMENT

This memorandum describes the expansion of the Long Term Pavement Performance (LTPP) database compiled during the Inception Study. This involved:

• The collection and summary of data on additional LTPP sections;

• Updating of the data collected for the LTPP sections included in the Inception Study, and

• Review of the summary compiled by one or more persons involved in the design, construction or maintenance of the section.

Note that no additional data collection or analysis was done on HVS test sections, and therefore these sections are not covered in this memorandum. Details on the HVS test sections can be found in the Inception Study report (Long and Jooste, 2006).

The data collected in this component of the project, and the HVS data collected in the Inception Study, were used to develop the structural design method for BSM pavements. This development work is described by Jooste and Long (2007) in “A Knowledge Based Structural Design Method for Pavements Incorporating Bitumen Stabilised Materials.” In conjunction with this work, a material classification system for pavement materials was developed and is described in Jooste, Long and Hefer (2007), “A Method for Consistent Classification of Materials for Pavement Rehabilitation and Design.”

Page 6: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Background and Introduction

- 2 -

1.3 REPORT STRUCTURE

• Section 2 of this memorandum outlines the method of data collection used to identify potential emulsion and foamed bitumen test sections and lists the sections identified.

• Section 3 presents the sections included in the database and a précis of each section. The detailed LTPP summaries are included in Appendix B.

• Section 4 identifies and discusses some trends in the data from the LTPP sections studied.

This report is classified as a Technical Memorandum, as defined by the documentation guidelines of the Gauteng Department of Public Transport, Roads and Works. As such, the main purpose of the document is to record technical processes and data, and not to present a formal, finished document for general distribution.

Page 7: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 3 -

2 METHOD OF DATA COLLECTION

In this section, the method for identifying road sections to evaluate containing emulsion and foamed bitumen is discussed. The sections identified are listed and ranked according the type and quantity of data available for that section. Finally, some of the difficulties of identifying the sections and collecting the relevant data are discussed.

2.1 IDENTIFYING AND COLLECTING DATA FOR BSM SECTIONS

The method of collecting all available data on foam and emulsion roads involved a four step process, as outlined below. 1. Obtain a list of all potential foam and emulsion roads.

This was a fourfold exercise. Firstly, the road owners were contacted to request a list of all the foam and emulsion roads in their network. The following road owners were contacted: SANRAL, WCPA, KZN-Dot, Freetrans and Eastern Cape DOT. Secondly, a 1996 CSIR report titled “Database of Emulsion-Treated Base (ETB) pavements in South Africa” was consulted (Wright et al, 1991). This report contained a list of ETB pavements in South Africa. Thirdly, responses from a questionnaire sent to SABITA members were collated. This included a lengthy list of emulsion roads compiled by Mr Hugh Thompson of WSP and Mr Adrian Bergh of CSIR (previously from MBS, now WSP). Finally, the consultants contacted as part of the data collection process were asked for information on additional sections. These actions resulted in a list of potential roads for this study. Using this procedure, a list of potential foam and emulsion roads was compiled in the Inception Study. In Phase 2, this list was updated, modified and expanded.

2. Contact road agencies to request available information on foam and emulsion roads.

Using the list compiled in Step 1, the Road Owners (this time including the City of Cape Town) were again contacted and the following information was requested for the roads on the list: - Design reports - Construction reports - As-built and tender documents - Materials - Behaviour and performance measurements, including the current condition - Maintenance and rehabilitation history - Traffic - Contact persons

3. Contact Consultants to request available information on foam and emulsion roads.

Using the list compiled in Step 1, relevant consultants were contacted and the same information as listed above was requested for the roads on the list.

4. Compile list of roads and prioritize sections.

Using the information retrieved, a master list of all projects was compiled. A total of 134 roads were identified. Of these, 109 were emulsion roads and 25 were foam roads. There are significantly less foam roads because the technology is newer and hence there are fewer roads that have been treated with foam than with emulsion.

2.2 EMULSION AND FOAMED BITUMEN SECTIONS IDENTIFIED

The complete lists of emulsion and foamed sections identified for potential study are shown in Table 1 and Table 2, respectively. The following information is included: Road name, section description, information source, region, information rating, year constructed and pavement structure. These lists have been updated since the Inception Study, and where possible the pavement structure has been included. The sections investigated as part of the Inception Study are highlighted in yellow, and because these sections were already investigated, the pavement structure is not included.

Page 8: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 4 -

The roads on the list that were thought to be of use were prioritized according to the kind and amount of information available. The roads were ranked 1 to 4, where 1 indicates the road had good information available, and 4 indicates little available information or advice was received or that the section was not worth investigating.

Not all information was available for all the sections, therefore some cells are blank. There are also some entries noted with a “?”, which denotes this information needed to be checked or obtained if that road was to be investigated further.

The additional sections studied as part of Phase 2 of the project are highlighted in the tables in green. The method of selecting these sections is described in Section 3.

Page 9: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 5 -

Table 1: List of Emulsion Roads Road Name

Section Description Region Info Rating

Constr. Pavement structure

Km 17.9–18.81 & 19.4-19.634: 25 AC, 150 ETB (2% cem, 1% bit), G6, G7

1386 Several short sections between Km 14.65 and 27.62. Roodeplaat Dam to Moloto Section

Gauteng 1 2001

Km 14.65-17.9 & 19.735-20.6: 25 AC, 150 ETB (2% cem, 1% bit), 150 C3 (varying cem contents), G6, G7

2229 N3 near Balfour Gautrans? 3 no pavement structure A74 Petrus Steyn Free State 3 1992 13.2 Seal, 25 AC, 150 BT2 A160 Harrismith Free State 1 1982 40 AG, 100 BT1, 150 C4, 150 G7 Artesia to Dibete

Km 0 - 25.5 (maybe not all be useable)

Botswana 1 No ETB in rehab options??

Candle Berry Drive

Nahoon, East London Eastern Cape

2 single seal, 125 ETB (3% res bit), in situ natural gravel

Cape Town AZ Berman CCT 3 Cape Town Bonteheuwel Ave CCT 3 Cape Town Gunners Circle Phase 1 CCT 3 Cape Town Morgenster Rd CCT 3 Cape Town Ou Kaapse Weg CCT 2001/20

03/2004 40 AC, 250 ETB (G4-G5) (2% cem, 2.2% emul), 150 gravel subbase, in situ sandy and rocky subgrade

Cape Town Old Nooiensfontein Rd CCT 2001 25 AC, 150 ETB (2% cem, 2% emul), 100 gravel subbase, in situ sandy and clayey subgrade

Cape Town Rhodes Drive CCT 3 1999 40 AC, 200 ETB (AC+G6 laterite) (2% cem, 3% emul), 400 sandstone boulder packed layer, in situ weathered granite subgrade

Cape Town Signal Hill Rd CCT 3 Cape Town Silverstream Rd CCT 3 Cape Town Spine Road CCT 3 Cape Town Vanguard Drive CCT 3 Cape Town Victoria Rd Phase 1&2 CCT 3 D2388 Between Road 25 and Onverwacht

(3.5 km), near Cullinan N/A 1

D540 Derdepoort

D540 near Derdepoort. Existing gravel road upgraded to paved road.

? 3 2002

Esikhawent near Richard's Bay 4 ~1977 crushed stone base in some areas and mixture of crushed waste and natural gravel in others

Forest-Haven Dr

Bus route in Durban. KZN 3 1988 110 AC, 150 ETB

Gaberone - Francistown

2 km Botswana 1996? cement & emulsion stabilised calcrete

Heilbron to Memel

Unknown road code. Freetrans road from Heilbron to Memel. Large ETB patches done.

? 3 2002

K151 Rigel Avenue

N1 to Hans Strydom. ? 1 2004

K27- Two sections between Km 47 and 49. Midrand to Olifantsfontein

N/A 2 ?

MR27 Stellenbosch to Somerset West W. Cape 1 1988 MR37 Panorama/Plattekloof, Cape town Western

Cape 2 1983? 40 recycled premix, 100 ETB (0.8%

res bit), 100 CTB, 200 laterite, cape sand

M43 (Barry Marais Rd)

Boksburg to Vosloorus Boksburg 3 1986 25 AC bit rubber, 20 AC, 150 ETB

N1-1 N1 Section 1 Kraaifontein to Paarl, Km 31.1 to 54.1 Slow Lane

Western Cape

1 1984

N1-13 Trompsburgh to Springfontein a.k.a Wurasoord to Trompsburgh (according to H.Thompson)

Free State 1 1980

N1-14 Trompsburgh to Springfontein a.k.a Wurasoord to Trompsburgh (according to H.Thompson)

Free State 1 1980

Page 10: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 6 -

Road Name

Section Description Region Info Rating

Constr. Pavement structure

N1-19, Km 3.8 - 6.4

Kromelmboog River to Vaal, Southbound

Freestate/Gauteng

3 2002 40 AC, 210 G1 to C4, 1.5% cem, 2% 60% emuls

N1-29X Northbound, Km 0 to 28.3 (1.8-10.7 & 22.5-28.3 ETB part)

Northern/Limpopo

3 Nov-02 S2, 25 AC, 280 ETB (1.8% res bit, 2% cem), 150 G5, 150 G7

N1-29 Km 28.3 to Km 70.0, Northern Jun-03 13.2 seal, 30 AC, 250/125 ETB (C3, 2.5% cem, 1% bit), C4 or nothing, 150 G6, 150 G10

N2 -6 Eastbound, Km 12.98 to 15.54 Southern? 3 2001 6.7 S1, 6.7 S2, 150 ETB, 150 G5, 250 G7,

N2-11E? Heidelberg to Dekriet ?? W. Cape 3 1971 60 AC, 150 BS, 150 G5, 300 G7 Km 22-37.6 end of Freeway. HVS test (Jordaan) near Km 30.8

EC 1 ? N2-16

Km 38.5 to 38.7 and 37.6 to 37.8, Northbound, Slow Lane. HVS test near Km 30.8

EC 1 1982

N2-20 Km 30 to 39, Thabankulu to Ingceweleni/Mzintlava River

Southern 1 1999/2000

35 AC, 180 ETB (1.75/2% cem, 2.5% emuls) dolerite and sandstone, 120/200 G4, 150 C4, 150 G7, G9

Northbound, Km 60.1 to 106.76 ? 3 1990 40 AC, 150 BES, 150 C4 (1974), 125 G8

Kokstad to Harding. Trial sections with DISR at Km 44.52 to 45.583

KZN/EC 3 ~1994 40AS, 60 BC, 200 GEMS (5% emul, 2% PBFC), 400 G5, G9

N2-21

Staffords Post to Transkei Border KZN 3 1965 40 AS, 60 BC, 200 GEMS (5% emulsion), G10 SG

N2-25S Isipingo to Mount Edgecombe, KM 30.728 - 30.73

38 AS, 100 - 15- BS, 300 C3/150 C3+150 C4, 250 G8 / 150 G8/ 150 C3

N3 Harrismith to Warden KZN 2 N3-4 Kildare to Mooi River. Test sections

from Nottingham Road Interchange KZN 1

N3-11 Villiers to Heidelberg ? 3 N3-12 Johannesburg SANRAL? 1 mid

1980s 40 mm semi-gap, crushed stone ETB, 208 mm ETB (C2 + surfacing) (1/5% emulsion, 0.9% res bind)

N4-1 Pretoria to Pienaars River Northern 1 ? N4-2 East and Westbound, Km 0 to 30.6

(or 55??). Pienaars River to Bronkhorstspruit

Northern 1 2003 115 - 180 ETB (1.5% cem, 30 litres emul/m3, 1.5% net bit) (G4+G2)

N4-5X Eastbound Km 27.4 to 54.2 1 1997 N6-4 Km 2 to 22 QT to Baily Station EC 3 N6-6 Smithfield to Rouxville 3 N6-8 Northbound, Km 0 to 19,

Reddersburg to R702 intersection Free State 3 1996 150 BES, 300 G5, 300 G10

N7-7 Springbok, Kammieskroon (LTPP data from Km 47 to 47.2)

Northern Cape

1 1987

N7-8 Vioolsdrift 3 N11-3 Northbound, Km 13.5 to 19 EC 3 1970 13.2 S2, 6.7 S2, 13.2 S1, 40AG, 100

BEM, 150 G5, 150 G7 N11-8 Southbound 33.6 to 38 Km 38 to

45.9. Ermelo/Hendrina vicinity. Mpumulanga

1 2003/2004

20 UTFC, 175 etb BT1 (1% cem, 2.4% bit emul/1.5% net bit)

N12 Springs to Ogies 3 N12-14 Northbound, Km 26.1 to 63.16 3 1958 13 S1, 4 S59 S1, 100 BES, 100 C2,

150 G7 N12-19 ETB section on N12 for which LTPP

data is available and associated HVS testing

Northern 1 ?

P1-1 Pretoria North to Moloto 3 P4-2 Balfour Gautrans? 3 No pavement structure available P5-1 from P101-1 for 5 Km, starting east

of rail bridge Gautrans? 3

P5/1 Springs to Delmas Gauteng? 2 1981 13.2 Seal, 19 AC, 100 ETB, 150 C3 subbase

P6-2 Benoni - Heidelberg Gautrans? 3 1996 150 G2 ETB (3% emul), 150 stab subbase, chert SG

P6-2??? Tierfontein Free State 3 1992? SL1, 25AC, 150 BT2, 150 C4 or 150 C3, 150 G6/G5 or 150 G7

Page 11: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 7 -

Road Name

Section Description Region Info Rating

Constr. Pavement structure

P9-2 Petrus Steyn Free State 1 1990 ST1, 10 AC, 150 BT2, 150 G6 P9-3 Petrus Steyn. Suspect this is same

as previous two. HVS testing was conducted at ~ km 38.3 to Km 38.5

Free State 1 1993 Design: 25Ac + S1, 125 G2/BT1 (2% lime, 20litres/m3), 150 C3/C4, 150 C4/15- G6

P9-4 Sasolburg to Heilbron 3 1994 P11-1 Viljoensdrift Free State 2 1984 40 AC, 100 BT1, 300 C3, 150 G5,

150 G6 P12/1 Sasolburg to Koppies 3 ? P16-1 Swartspruit to Manhaarrand 2 1993 13.2 Seal, 25 AC, 150 BT2, 150 C4,

150 G5 Magaliesburg Gautrans? 3 1980? 25 AC, 100 ETB, C3 subbase Tarlton to Magalies River 3 ?

P16-1

Vrede Free State 3 ? P18-8 Bethlehem Free State 3 1992 13.2 seal, 25 AC, 150 BT2, 150 C4,

150 G6, 150 G7 P18-9 Bethlehem Free State 3 1993 13.2 seal, 25 AC, 150 BT2, 150 C3,

150 G6, 150 G8 P18-11 Vrede Free State 3 1993 13.2 seal, 25 AC, 150 BT2, 150 G5 P21-2 & 3 Lindley to Bethlehem or Steynsrust 3 ? P23/1 Kroonstad Free

State? 2 1992 13 seal, 25 AC, 150 BT2/C3, 300

G5, 150 G7 P36/1 Heilbron to Kroonstad 3 ? P36-3 Heilbron Free State 3 ? 13 seal, 25 AC, 150 BT2 P41/1, FS Frankfort to Villiers 3 1990 8T13, 20 AC, 150 BT2, 150 C3, 150

G6 P41-2, FS Tweeling to Frankfort. (Km 0 – 7) Free State 2 No pavement structure details in

design report. 8 T13, 20 AC, 150 BT2, 150 C3, 150 G6

P41-3, FS Tweeling to Reitz (Km 0 – 27.06) Free State 2

1989/1990/1992

? No pavement structure details in design report. 13.2 S, 20 AC, 150 BT2, 150 G5, 150 G6

P57-3 Bethlehem Free State 3 1992? 8 T13, 25 AC, 150 BT2, 150 C4, 150 G6

P59-1 Bloemfontein to Dealsville Free State 2 ? P59-2 Boshof Free State 1991 20 AC, 150 BT2, 150 C4, 150 G6 P62-1 Bultfontein to Wesselsbron, Km

49.85 - 51.2 Free State ~1991 13 seal, 20 surfacing, Reworked to

prescribed depth, stabilised base with emulsion

P62-2 Losdoring Free State 3 1991? 15 ST2, 25 AC, 150 BT2, 150 G5 P64-2 Reitz Free State 3 1994? 150 BT2, 150 C4, 150 G7, 150 G9 P66-1 Wepener Free State 1 1989 double seal, 175 ETB (0.9% res bit),

in situ decomposed dolerite P73/1 Johannesburg South 4 P88-1 Potchefstroom Gautrans? 3 no pavement structure P103 Hans Strijdom, JHB Gautrans 3 P167-1 Vanderbijlpark to Heidelberg 2 1981 13.2 seal, 25 AC, 125 ETB slag, lime

stab subbase R61-4 Eastbound Km 0 to 61.4 3 1992 13.2 S1, 150 BES, 150 C3, 250 G7 R510-1 Northbound, Km 2.8 to 22.95 3 1966 seals, 150 BES, 125 C3, 150 G7 S66 300 m section Free State 3 1992? 25 AC, 150 BT2, 150 C4, 150 G6 TR1/5 Namibia Namibian 3 TR13/3 Britstown, 50 km W. Cape 1 1993? double seal, 150 ETB, 150 cem stab

(2.5% cem), 150 G5 TR18/1 Queenstown E. Cape 3 1985? 13.7 Double seal, 150 ETB (1.5%

em), 100 G5 TR74/1 Elliot E. Cape 3 1984? 13.7 Double seal, 150 ETB (2-2.5%

em), 100 G5 Unknown Bloemfontein to Dealsville 3 Unknown Tweespruit to Excelcior Free State 3 N/A Kyalami Race Track 4 N/A Main Roads, Maseru 4

Page 12: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Method of Data Collection

- 8 -

Table 2: List of Foamed Bitumen Roads Road Name

Section Description Region Info Rating

Constr. Pavement structure

Athens to Corinth Highway

Athens to Corinth Highway in Greece Greece 2 post 2001

40 AC surfacing, 50 AC base, 200-300 mm foam, in situ support

D67 Heidelberg ? 3 ?

MR25/63 Wartburg, KZN Kwazulu-Natal

3 2003

MR439 (R22-4)

MR439 (R22-4) from Mseleni to Phelandaba. 250 mm Foamed Bitumen Base, Km 82.7 to 116.27

Kwazulu-Natal

1 2003 S6, 250 FBT (1-2% lime, 3-4% foam), 300 G7/G9, G10 selected fill)

MR504 P504 near Shongweni in KZN Kwazulu-Natal

1 1995

N3-2 Nchanga to Cato Ridge 1994

N6-4 Km 2 to 22 QT to Baily Station E. Cape 3 ?

N7 (TR11/1)

Western Cape. NB Slow KM 5.6 - 11.6, SB Slow 5.8 to 18.3

Western Cape

1 2002 novachip, 35 AC, 250 foam (1% cem, 2.3% foam) (200 G2 + 50 G5), G5 subbase,

Ermelo/Hendrina vicinity, Km 33.4 to Km 38

Northern 1 2003 200 FTB (1% cem 32.5, 1.5% bit) N11-8

Ermelo/Hendrina vicinity, Km 38 to Km 45.9

Northern 1 2003 180 FTB (1.5% cem 42.5, 1.5% bit)

Newlands West Drive

Newlands West Drive between Sooklall and Newcentre Drives, 0.98 km

Kwazulu-Natal

2 2003 40 AC, 125 foam RAP overlay, 200 mm existing asphalt or in situ stabilised C4, existing crushed rock, existing G9 subgrade

P17-5 Bushbuck Ridge to Hazyview Km 41.2 to 67.4

Northern 2 1999/2000

novachip, 0-150 foam weathered granite (1.5% cem, 1.5% bit),

P17-5 / R40 /S59?

Klaserie to Bushbuck Ridge, Km 0 - 40.7

Northern 2 2001 6.7 Seal, 13.2 seal, 160 - 250 foam weathered granite (1.5% cem, 1.5% bit),

P24/1 & P88/1

Near Vereeniging, Km 65.455 to 69.930 & Km 6.2 to 11.192

Gauteng 1999 bitumen rubber?, 150 FTB (2% cem, 1.5/1.7% foam)

P25 Heidelberg to Meyerton ? 3 ?

P41/2 Gauteng

Vereeniging/Benoni district. Bitty milling, Km 9.318 to 38.88. between P243/1 and P4/2

Gauteng 2 2000/2001

13.2 mm seal, 20 ACM, 280/200 in situ recycling

P52-3 Bethal/Kriel Mpumalanga

3 2005

P185/1 Standerton to Kinross Mpumalanga

3 ?

P243/1 Several sections between Km 12.1 and 25.6. HVS test section

Gauteng 1 2000 13.2 seal, 20 ACM, 250 foam (2% cem, 1.8% foam)

P423-0 / MR 423

P423-0 Bruyns Hill To Nagle Dam Kwazulu-Natal

2 1996

P466-2 P466-2 near Sodwana Kwazulu-Natal

3 1994 cohesionless sand

Same-Himo Pilot Project

Same-Himo Pilot Project, Tanzania Tanzania 1 1992

TR16/3 (R27)

Nieuwoudtville to Calvinia (21.4 km) Northern cape?

1 2003/2004

Trunk road T1

Zambia Zambia 3 Jun-03 50 AC, 150/175 foam, 150 G5, 300 G6, G8 subgrade

Unknown Letsitele to Gravellotte Northern 3 ?

Page 13: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 9 -

3 SELECTION AND PRÉCIS OF LTPP SECTIONS INVESTIGATED

3.1 SELECTION OF SECTIONS TO INVESTIGATE

In the Inception Study, several roads were studied in detail and summaries of the sections were documented. The selection of roads for further study was based on the availability of construction, behaviour, performance and traffic data, and on the age of the section. For Phase 2 of the study, the database was expanded using the same criteria, but it was also attempted to fill deficiencies in the sections studied in the Inception Study (and discussed in the next section). Some of the sections studied were also included as a specific request by client body representatives.

3.1.1 Identification of Deficiencies in Collected Data

An objective of the Inception Study was to identify major gaps in the available information and make recommendations for additional data collection. As a result, the following deficiencies in the collected data were identified (Long and Jooste, 2006):

• None of the LTPP sections investigated have failed, and only one HVS section failed before water was forcibly added. It was therefore impossible to investigate the mode of failure for foam and emulsion pavements. It was therefore recommended that examples of failed emulsion and foam pavements are identified and investigated.

• There was a lack of foamed bitumen sections in the list of pavements identified and studied. This is because the technology is much newer than emulsion technology and has not been as widely accepted in the South African pavement industry. It was thus recommended that additional foamed bitumen sections be identified and investigated.

• The majority of emulsion pavements investigated had a stabilized support, and there was therefore a need to gather more information on their performance on unstabilized support. Many rehabilitation projects using deep in situ recycling (DISR) recycle the existing base and subbase layer into a thick base layer (300 to 350 mm) without reworking the support. This type of rehabilitation option is becoming increasing popular and current mechanistic designs show pavements with this structure to have a limited capacity. It was therefore important to identify and investigate more pavements with the structure typical of DISR projects to establish their structural capacity.

• Since most of the emulsion sections had a stabilised support, it is difficult to identify whether the good performance experienced is because of the good support or due to the addition of emulsion to the base layer. It was therefore recommended to study emulsion sections with an unstablilised support.

• The majority of foam pavements investigated had a natural gravel support. Additional pavements where the foam layer was placed on a stabilized subbase were required.

• None of the sections (HVS and LTPP) had a poor subgrade and therefore there was no indication of the performance of foam and emulsion pavements on poor subgrades.

• Only one ETB section was built on a new road, and all the foamed bitumen sections were constructed during rehabilitation of an existing pavement. There was therefore a need to identify and investigate both foam and emulsion pavements that were constructed as part of a new road.

Where possible, selections of the additional sections to study further was based on mitigating the deficiencies listed above.

Page 14: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 10 -

3.2 LTPP SECTIONS INVESTIGATED

Based on the selection criteria discussed in the sections above, the following sections were studied in detail in either the Inception Study or Phase 2 of the project:

Emulsion

• D2388 (Cullinan) • MR27 (near Stellenbosch) • N1 Section 1 (Kraaifontein) • N1 Section 13 (Springfontein and Trompsburg) • N1 Section 14 (Springfontein and Trompsburg) • N12 Section 19 (Experimental Section 1) (near Daveyton) • N12 Section 19 (Experimental Section 2) (near Daveyton) • N2 Section 16 (Kwelera, East London) • N2 Section 20 (Mount Frere to Mount Ayliff) • N3 Section 4 (near Mooi River) • N4 Section 1 (Scientia interchange to Pienaars River) • N4 Section 5X (KM 20 to 25) (Wonderfontein to Crossroads) • N4 Section 5X (KM 27.6 to 30.6) (Wonderfontein to Crossroads) • N7 Section 7 (near Kammieskroon) • MR27 (near Stellenbosch) • P23/1 (Kroonstad to Steynsrus) • Road 1386, Moloto Rd (section with stabilised subbase) (near Pretoria) • Road 1386, Moloto Rd (section with unstabilised subbase) (near Pretoria)

Foamed bitumen

• MR504 (1) (near Shongweni) • MR504 (2) (near Shongweni) • MR504 (3) (near Shongweni) • N11-8 (Ermelo to Hendrina) • P24/1 (near Vereeniging) • P243/1 (near Vereeniging) • R22-4 (MR439) (near Phelendaba) • Same-Himo (1) (Tanzania) • Same-Himo (2) (Tanzania) • Same-Himo (3) (Tanzania) • TR16-3 (R27) (Nieuwoudtville to Calvinia)

For each section, a detailed summary was compiled, which summarizes all the available data and provides some observations on each section. The summaries compiled as part of the Inception Study were updated with the latest information available, specifically the current condition where possible. All the summaries were also reviewed by personnel involved in the design, construction and maintenance of the sections. The summaries are included in Appendix B and are organized into the following sections:

• Header information: road type, road code, construction year, limits considered (kilometres), project description, available documentation

• Area and climate • Construction history • Behaviour and material quality indicators • Performance indicators and treatment history • Traffic loading indicators • Factual observations • Observations based on estimates or interpretations • Reviewer(s)

The sections selected for investigation are often shorter subsections of a larger construction section. The shorter sections were selected to tie into the available documentation and to ensure the section had the

Page 15: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 11 -

same pavement structure and treatments. Some sections are as short as 200 meters, as originally these lengths were included in the CSIR/SABITA LTPP study (Wright et al, 1991).

Under “Traffic Loading Indicators”, all available traffic data are listed and then used to determine an estimate of the cumulative traffic. This estimate is very approximate, and was based on rough visual fits to the data. The aim was to be able to classify the cumulative traffic according to available information.

Different sources of information for the same section often provided conflicting information. In these cases, attempts were made to ascertain which reference was correct by checking with people involved in the project. It this was not possible, an assumption was made as to which information was most likely to be correct and in these cases this is clearly stated.

The summary for the 3 Same-Himo sections were not reviewed. The reviewer was approached several times and committed to completing the review but failed to do so.

3.3 PRÉCIS OF LTPP SECTIONS

This section contains a brief précis of the LTPP sections studied. The detailed summaries are included in Appendix B. The emulsion roads are listed first, followed by the foam roads.

3.3.1 D2388 (Cullinan)

This road is situated between R25 and Onverwacht, near Cullinan in Gauteng. The area has a moderate climate. In 1997 the emulsion treated base layer was constructed using labour-intensive methods with 1% cement and 1% residual binder. HVS testing was done on this road from 1996 to 2000 and is discussed in the Inception Study report (Long and Jooste, 2006).

The reconstructed pavement structure is as follows:

• 8 mm slurry seal surfacing; • 100 mm emulsion treated base (1% cement and 1.67% emulsion/1% residual binder), natural

gravel; • 150 cement stabilised natural gravel (3% cement), and • 300 mm G6 selected layer.

Double seals were applied to the road after 6 and 7 years of service.

The riding quality on the section indicates that the road is in a poor condition, and therefore for the purpose of the structural design method development the road is considered to be in a warning/failed state.

An estimated 0.2 to 0.5 million equivalent standard axles (MESA) have been carried in 8 years of service.

3.3.2 MR27 (near Stellenbosch)

This road is situated between Somerset West and Stellenbosch in the Western Cape. The climate in the area is moderate. The existing cement treated base (CTB) pavement was rehabilitated with emulsion in 1988. The rehabilitated pavement structure is as follows:

• 13.2 mm and 6.7 mm double seal; • 40 mm continuously graded asphalt; • 100 mm emulsion treated base (1.67% emulsion/1% residual binder); • 100 mm cement stabilised natural gravel; • 150 mm laterite selected layer, and • In situ sand.

After 14 years a double seal (13 mm and 7 mm) with a polymer modified binder was placed.

Page 16: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 12 -

All the behaviour and performance indicators (rutting, riding quality, visual condition and deflections) indicate that the road is in a good condition.

The cumulative traffic in 18 years of service (1988 to 2005) is estimated as 3.3 to 6.8 MESA.

3.3.3 N1 Section 1 (Kraaifontein)

This road is near Kraaifontein in the Western Cape, which has a moderate climate. The road was rehabilitated in 1984 by recycling the cement treated base with emulsion. Part of the original cement treated base was retained as a subbase. The rehabilitated pavement structure is as follows:

• 40 mm semi-gap graded asphalt with rolled in chips; • 40 mm continuously graded asphalt; • 100 mm emulsion treated base (1% residual binder); • 100 mm cement stabilized subbase; • 150 mm G5 selected layer, and • 250 mm G8 upper subgrade and selected layer.

After 10 years of service, the asphalt was milled off and replaced. The behaviour and performance data indicate the road is in a sound condition.

The road has carried an estimated 11.5 to 15.8 MESA in 21 years of service.

3.3.4 N1 Section 13 (Springfontein and Trompsburg)

This section is between Springfontein and Trompsburg in the Free State, and the area has a dry climate. Between 1978 and 1980 the road was rehabilitated by recycling the cement stabilised crushed stone base layer (including recrushing) and recompacting with the addition of emulsion and cement. The lower part of the original cemented base was left as a subbase. The pavement structure is as follows:

• 13 S1 seal; • 25 mm asphalt; • 150 mm emulsion treated base (1% cement and 0.7% bitumen); • 100 mm natural gravel stabilised with cement and 0.6 to 0.7% binder, and • 150 mm G6 selected layer.

After 11 years a diluted emulsion surface treatment was applied, and after 16 years the road was patched and sealed.

After 25 years of service the rut depths and riding quality are at acceptable levels. After 26 years of service the condition of the road was reported as satisfactory to poor and hence the road is considered as being in a warning condition.

The road has carried an estimated 10 to 13 MESA from 1980 to 2005 (25 years).

3.3.5 N1 Section 14 (Springfontein and Trompsburg)

This section is between Springfontein and Trompsburg in the Free State, and the area has a dry climate. Between 1978 and 1980 the road was rehabilitated by recycling the cement stabilised crushed stone base layer (including recrushing) and recompacting with the addition of emulsion and cement. The pavement structure is as follows:

• 13 mm S1 seal; • 25 mm asphalt; • 150 mm emulsion treated base (1% cement and 0.7% bitumen); • 100 mm decomposed dolerite, and • 300 mm G6 selected layer.

After 11 years a diluted emulsion surface treatment was applied, and after 16 years the road was patched and sealed.

Page 17: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 13 -

After 25 years of service the rut depths and riding quality are at acceptable levels. After 26 years of service the condition of the road was reported as satisfactory to poor and hence the road is considered as being in a warning condition. The road has carried an estimated 10 to 13 MESA from 1980 to 2005 (25 years).

3.3.6 N2 Section 16 (Kwelera, East London)

This road is situated near East London in the Eastern Cape in a wet climatic region. The road was reconstructed with an ETB in 1980 and 1981 by reworking the existing cement stabilised base as a subbase and importing and emulsion treating a new base layer. HVS testing was done on this section and is discussed in the Inception Report (Long and Jooste, 2006). Based on all the available information, the pavement structure is assumed to be as follows:

• 40 mm gap-graded asphalt; • 140 mm emulsion treated sandstone base (1% cement and 1% residual bitumen); • 130 mm stabilised subbase (1.5% lime and 1.5% slagment), and • Clayey subgrade.

After 10 and 25 years of service seals were placed on the road. Due to the seals, the condition of the road is sound.

The road has carried approximately 2.4 to 3.2 MESA from 1980 to 2006 (26 years).

3.3.7 N2 Section 20 (Mount Frere to Mount Ayliff)

This section is in the Eastern Cape between Mount Frere and Mount Ayliff in a wet to moderate climate. The road was rehabilitated in 1999 and 2000 using deep in situ recycling and emulsion treatment. The rehabilitated pavement is as follows:

• 35 mm asphalt surfacing with rolled in chips; • 180 mm cement and emulsion treated base (1.75% cement, 1.5% residual bitumen); • 120 mm of original base (C4 in equivalent granular state); • 150 mm original subbase (C4 in equivalent granular state); • 150 mm original selected subgrade (G7), and • Original subgrade (G8 to G10).

No treatments have been applied to the road in the 6 years of service. Rutting and roughness after 6 years of service is good, and the road is therefore in a sound condition.

The road has carried an estimated 1.1 to 2.4 MESA in 6 years of service.

3.3.8 N3 Section 4 (near Mooi River)

This section is on the N3 Toll Road near Mooi River in the Durban direction. The area has a wet climate.

The road was rehabilitated in 1988 by recycling the existing surfacing, cement stabilized base and subbase layers into an emulsion base. The rehabilitated pavement structure is as follows:

• 40 mm semi-gap graded asphalt; • 200 mm emulsion treated base (1.2% residual binder); • 150 mm cemented crushed stone subbase, and • In situ subgrade.

No treatments have been applied to the road since the rehabilitation, and the available behaviour and performance data, particularly the rutting and visual condition, indicate the road is in a warning condition.

The road has carried an estimated 9.2 to 21 MESA in 17 years.

Page 18: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 14 -

3.3.9 N3 Section 12, Northbound (Modderfontein to Buccleugh)

This section is part of the ring road surrounding Johannesburg. The area has a moderate climate. The road was widened and rehabilitated in 1986 and 1987. In the slow lane the existing asphalt and cement treated crushed stone base was recycled with emulsion into a 100 mm layer. Thereafter the entire road was overlaid with a 100 mm ETB base. The pavement structure is assumed to be as follows:

• 40 mm semi-gap graded asphalt; • 200 mm ETB, recycled crushed stone (1.5% emulsion / 0.9% residual, unknown cement content); • 100 mm C3; • 100 mm C4; • 200 mm G8, and • 300 mm G9.

After 12 years a 40 mm asphalt overlay was placed. Based on the available behaviour and performance data (deflections, rutting and roughness) the section is in a sound condition after 20 years of service.

The road has carried an estimated 9.7 to 17.5 MESA in 20 years of service.

3.3.10 N3 Section 12, Southbound (Modderfontein to Buccleugh)

This section is part of the ring road surrounding Johannesburg, Gauteng. The area has a moderate climate. The road was widened and rehabilitated in 1986/1987. In the slow lane the existing asphalt and cement treated crushed stone base was recycled with emulsion. Thereafter the entire road was overlaid with an ETB base. Based on various sources of information, the pavement structure is assumed to be as follows:

• 40 mm semi-gap graded asphalt; • 360 mm ETB, recycled crushed stone (1.5% emulsion / 0.9% residual, unknown cement content); • 150 mm C4, and • G7 subgrade.

After 12 years a 40 mm asphalt overlay was placed. Based on the available behaviour and performance data (deflections, rutting and roughness) the section is in a sound condition after 20 years of service.

The road has carried an estimated 9.7 to 17.5 MESA in 20 years of service.

3.3.11 N4 Section 1 (Scientia interchange to Pienaars River)

The road is situated just east of Pretoria on the N4 in an area with a moderate climate. The road was rehabilitated in 1997. The rehabilitation involved recycling the existing cemented layers and treating with emulsion. The recycled pavement structure is as follows:

• 30 mm continuously graded asphalt and bitumen rubber single seal; • 170 mm emulsion treated base (2% cement, 1% net bitumen); • 130 mm cemented natural gravel subbase (C4), and • 500 mm selected layers and upper subgrade (G6 to G7).

No treatments have been applied to the road since the rehabilitation. After 9 years of service, the road is in a sound condition.

The cumulative traffic from 1997 to 2005 is estimated as 2.8 to 4.4 MESA.

3.3.12 N4 Section 5X (Wonderfontein to Crossroads, KM 20 to 25)

This N4 Section is situated between Wonderfontein and Crossroads in Mpumulanga and has a moderate to wet climate. The road was rehabilitated in 1996. In this section of the contract, the base layer was stabilised with cement and emulsion, and the majority of the subbase was also stabilised with cement and emulsion. The reconstructed pavement is assumed to be as follows:

Page 19: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 15 -

• 13.2 mm single seal; • 25 mm continuously graded asphalt; • 150 mm emulsion and cement stabilised crushed stone (0.4% binder and 1% cement), and • 150 C3 subbase.

No maintenance or rehabilitation occurred in 9 years of service. After 10 years of service, the road is in a good condition based on rutting, roughness and visual condition assessments.

The road has carried an estimated 1.5 to 5 MESA in 8 years of service.

3.3.13 N4 Section 5X (Wonderfontein to Crossroads, KM 27.6 to 30.6)

This section is situated on a climbing lane on the N4 between Wonderfontein and Crossroads in Mpumulanga and has a moderate to wet climate. The road was rehabilitated in 1996. In this section of the contract, the base layer was stabilised with cement and emulsion, and the subbase was stabilised with cement. The reconstructed pavement is assumed to be as follows:

• 13.2 mm single seal; • 25 mm continuously graded asphalt; • 150 mm emulsion and cement stabilised natural gravel (0.5% binder and 1 or 2% cement); • 150 mm C3 upper subbase (1-3% cement); • 150 mm C3 lower subbase (3-4% cement), and • 150 mm G5 or G6 selected layer or prepared subgrade.

No maintenance or rehabilitation has occurred in 10 years of service. After 8 years of service, the road is in a good condition. Although the condition after 8 years of service was reported as good, after 10 years of service the road in a poor condition and needs to be rehabilitated. The section is therefore considered to have failed.

The road has carried an estimated 1.5 to 5 MESA in 8 years of service.

3.3.14 N7 Section 7 (near Kammieskroon)

This section is near Kammieskroon in the North Western Cape and is situated in a dry climate. The emulsion treated pavement was constructed as part of the rehabilitation of an existing pavement in 1986 and 1987. The reconstructed pavement is assumed to be as follows:

• Seal; • 150 mm ETB (2% emulsion and 1% cement) continuously graded asphalt; • 50 mm cement stabilised subbase (remainder of the original CTB); • 150 mm natural gravel selected layer or lower subbase, and • 250 mm selected subgrade.

A 9.5 mm seal was placed after 8 years of service and a fog spray after 18 years of service. Based on rutting, riding quality and the visual condition, the road can be considered to be in a sound condition.

The cumulative traffic carried in 18 years of service is estimated to be 1 to 1.4 MESA.

3.3.15 N12 Section 19 (near Daveyton, Experimental Section 1)

This experimental section is on the N12 near Daveyton in Gauteng and was part of a long term pavement performance (LTPP) monitoring experiment. The area has a moderate climate. The section was originally constructed with cement stabilised layers but premature failure resulted in rehabilitation with emulsion in 1974. The recycled pavement structure is as follows:

• 13.2 mm seal; • 50 mm continuously graded asphalt; • 120 mm emulsion treated base (1% net bitumen); • 80 mm crushed stone subbase; • 140 mm cement stabilised gravel subbase, and

Page 20: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 16 -

• 300 to 350 selected subgrade.

A fog spray was applied after 17 years of service, and after 26 years of service patching and crack sealing was done. Visual observations indicate that the condition of the road is deteriorating rapidly, and the road is therefore considered to be in a warning state.

The estimated traffic carried in the 30 years of service is 16.4 to 26.8 MESA for 30 years.

3.3.16 N12 Section 19 (near Daveyton, Experimental Section 2)

This experimental section is on the N12 near Daveyton in Gauteng and was part of a long term pavement performance (LTPP) monitoring experiment. The area has a moderate climate. The section was originally constructed with cement stabilised layers but premature failure resulted in rehabilitation with emulsion in 1974. HVS testing has recently been completed on this section of the N12 (Long and Jooste, 2006). The recycled pavement structure is as follows:

• 13.2 mm seal; • 50 mm continuously graded asphalt; • 135 mm emulsion treated base (1% net bitumen); • 130 mm cement stabilised gravel subbase, and • 300 to 350 selected subgrade.

A fog spray was applied after 17 years of service, and after 26 years of service patching and crack sealing occurred. Visual observations indicate that the condition of the road is deteriorating rapidly, and the section is therefore considered to be in a warning state.

The estimated traffic carried in the 30 years of service is 16.4 to 26.8 MESA for 30 years.

3.3.17 P23/1 (Kroonstad to Steynsrus)

This section is in the Free State Province between Kroonstad and Steynsrus in a moderate climatic area. The pavement was rehabilitated in 1992 by reworking the base and adding lime and emulsion. The pavement structure is as follows:

• 20 mm asphalt; • 150 mm emulsion treated base (1% emulsion and 1% cement), and • 150 mm G6 subbase.

No treatments were applied to the road in 13 years of service. After 14 years of service, the road was in a fair to poor condition with high rutting. The road will need to be repaired soon, and it is therefore considered to have failed.

The road is estimated to have carried approximately 0.5 and 1.3 MESA in 13 years of service.

3.3.18 Road 1386 (Moloto Rd), Stabilised Subbase

The road is situated north-east of Pretoria and runs towards Moloto. The area has a moderate climate. The road was rehabilitated in 2001 by reworking the existing base layer and treating with cement to form a new subbase, and importing new material and treating with emulsion and cement for a new base. The pavement structure is as follows:

• 13.2 mm seal with polymer modified binder; • 25 mm continuously graded asphalt; • 150 mm emulsion treated base (G5) with 2% cement and 1% bitumen; • 150 mm cement treated base (G7) with 2% cement; • 350 mm G8 selected layer, and • G5 to G10 subgrade.

The visual condition index and roughness measurements indicate that the road is deteriorating fairly rapidly, and is therefore considered to be in a warning condition.

Page 21: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 17 -

The road has carried an estimated 1.9 to 4.8 MESA after 5 years of service.

3.3.19 Road 1386 (Moloto Rd), Unstabilised Subbase

The road is situated north-east of Pretoria and runs towards Moloto. The area has a moderate climate. The road was rehabilitated in 2001 by reworking the existing base layer and treating with cement and emulsion. The pavement structure is as follows:

• 13.2 mm seal with polymer modified binder; • 25 mm continuously graded asphalt; • 150 mm emulsion treated base (G5-G6) with 2% cement and 1% bitumen; • 300 mm G5-G7 subbase, and • G5 to G6 subgrade.

The visual condition index, roughness and rutting measurements indicate that the road is in a sound condition.

The road has carried an estimated 1.9 to 4.8 MESA after 5 years of service.

3.3.20 MR504 (near Shongweni)

This road is near Shongweni in Kwazulu-Natal in an area with a wet climate. The road was rehabilitated using foamed bitumen in 1995. Three sections from this road were investigated in this study. The pavement structures are shown in Table 3.

Table 3: Pavement Structures for MR504 Sections Layer Section A Section B Section C Surfacing None Slurry or fog spray Slurry Foam base 160 mm RAP 125 mm natural gravel 180 mm natural gravel

Subbase 150 mm G6 natural gravel (imported) 150 mm reworked in situ G6 natural gravel

Subgrade Reworked in situ material In situ material

Section A received a single seal and Sections B and C received 30 mm of asphalt after 3 years of service. In the 12 years of service, the behaviour and performance indicators show that the road is deteriorating, and is therefore considered to be in a warning state.

The road has carried an estimated 0.7 to 2 MESA in 12 years.

3.3.21 N11 Section 8 (Ermelo to Hendrina)

This section is located between Hendrina and Ermelo in Mpumulanga and has a moderate climate. The road was rehabilitated in 2003 and 2004 using deep in situ recycling and treatment with foamed bitumen and cement. The original pavement had a cemented base and subbase. The pavement structure is as follows:

• 20 mm UTFC surfacing; • 180 mm foamed bitumen base with natural gravel (1.5% foamed bitumen and 1% cement); • 250 mm G6 to G8 subbase, and • G8 subgrade.

After 3 years of service, the deflections on the section are relatively high, but the road is reported to be in a sound condition.

The section has carried an estimated 0.6 to 1.1 MESA in 3 years of service.

Page 22: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 18 -

3.3.22 P24/1 (near Vereeniging)

This road is situated near Vereeniging in Gauteng and has a moderate climate. The original crushed stone base and cemented subbase was rehabilitated in 1999 using foamed bitumen. The rehabilitated pavement structure is as follows:

• 30 mm bitumen rubber surfacing; • 150 mm foamed bitumen base (1.5% foamed bitumen and 2% cement); • 210 mm cemented natural gravel subbase; • 330 mm selected layer (clayey sand), and • Silty sand subgrade.

No treatments have been applied to the road in 7 years of service and road is in a sound condition.

The road has carried an estimated 2 to 4.5 MESA in 6 years.

3.3.23 P243/1 (near Vereeniging)

This section is near Vereeniging and the Vaal Dam in Gauteng. The area has a moderate climate. The road was rehabilitated in 2000 by deep in situ recycling the original natural gravel cement stabilised base and subbase and treating with foamed bitumen and cement. A short section was also recycled and treated with emulsion and cement, but that section is not considered in this project. HVS testing was performed on both the foam and emulsion sections (Long and Jooste, 2006). The rehabilitated pavement structure is as follows:

• 13.2 mm single seal; • 20 mm HMA; • 250 mm foamed bitumen base (1.8% foamed bitumen and 2% cement); • 150 mm subbase (G8), and • G8 to G10 subgrade.

After 6 years of service, the road has not received any treatments and based on the rutting, riding quality and visual condition can be considered to be in a sound condition.

The road has carried an estimated 0.17 to 0.53 MESA after 6 years of service.

3.3.24 R22-4 (MR439) (near Phelandaba)

This section is situated between Mseleni and Phelandaba in Kwazulu-Natal. The area has a wet climate. The original road was a gravel track, which was upgraded using foamed bitumen treatment in 2002/2003. The resulting pavement structure is as follows:

• 6 mm coarse slurry seal; • 250 mm foamed bitumen base (3.5/4% foamed bitumen and 2% lime), and • G7 to G9 subgrade (sand).

A double seal was applied to the road after 4 years of service. From deflection measurements and visual observations, the road can be considered to be in a sound condition since the application of the seal.

An estimated 0.3 to 2.2 MESA have been applied in 3 years of service.

3.3.25 Same-Himo (Tanzania)

This road was a pilot project in Tanzania in an area with a wet climate. The road was reconstructed in 1992 using foamed bitumen treatment. The original pavement had a cement stabilised base. Three sections were studied from this pilot project. The pavement structure is as follows:

• 19 mm and 12 mm chip seal; • Section 1: 150 mm foamed bitumen with natural gravel (premixed);

Section 2: 150 mm foamed bitumen with natural gravel with some imported material (premixed); Section 3: 175 mm foamed bitumen treated natural gravel (in situ recycled), and

Page 23: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 19 -

• 100 mm subbase (G6).

The pavement has received some crack sealing, but no other treatments in the 15 years of service. The road is reportedly in a sound condition.

The road carried between 1.4 and 2.1 MESA after 8 years of service, but after 11 years of service it is estimated to have carried close to 4 MESA.

3.3.26 TR16-3 (R27) (Nieuwoudtville to Calvinia)

This section is between Nieuwoudtville and Calvinia in the Northern Cape. The area has a dry climate. The road was rehabilitated in 2003/2004 by recycling the natural gravel base and subbase and treating with foamed bitumen and cement. The pavement structure is as follows:

• 13.2 mm and 6.7 mm modified binder double seal; • 200 mm foamed bitumen base (1.5% cement, 2.5% foamed bitumen); • 175 mm natural gravel subbase (G5), and • Natural gravel subgrade (G5).

No treatments have been applied to the road in the 2 years of service. The behaviour and performance indicators show that the condition of the road is deteriorating fast and is therefore considered to be in a warning condition.

The road has carried an estimated 0.06 to 0.61 MESA.

3.3.27 Summary of Key Information from LTPP Summaries

A summary of the key information from the LTPP summaries is shown in Table 4. The data shown includes the climate, pavement structure, traffic, age, maintenance and condition. The pavements with foamed bitumen treated layers are highlighted. In Table 4, a question mark (?) indicates an assumed value, based on available evidence. These data were used to develop and calibrate a structural design method for bituminous stabilised materials (Jooste and Long, 2007), and the same table is also shown in the reference. As part of the structural design development, certain assumptions regarding the pavement materials were made and justified, but because of this there may be slight differences in Table 4 and the equivalent table given in Jooste and Long (2007). The surfacing thickness shown in the table may differ slightly from what is contained in the LTPP summaries because a thickness of 5 mm was assigned to a seal in the development of the structural design method.

In the proposed structural design method, there is no differentiation between foam and emulsion treated layers, hence in the table such layers are labelled BSM1 or BSM2 for bituminous stabilised material class 1 or 2. More information on the material classes is contained in Jooste and Long (2007).

Page 24: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 20 -

Table 4: Summary of LTPP Sections Information used in Structural Design Method Development

Project Climate Surface Base Subbase Selected Subgrade Traffic (mesa)

Age (years) Maintenance Condition

D2388, Km 2.1 to 2.3 (Cullinan) Mod Slurry 100 BSM2 150 C4 300 G6 G7 (?) 0.2 to 0.5 9

Double seals placed after 6 and 7 years in

service

Warning / Failed

MR27, Km 23.8 to 24.8, Slow Lanes (Stellenbosch) Mod 45 HMA

plus seal 100 BSM1 100 C3 150 G7 (?) G8 (?) 4 to 9 19 Double seal with PMB after 14 years. Good

N1 Section 1, Km 34.2 to 34.4 North and Km 34.8 to 35.0 South (Kraaifontein)

Mod 80 HMA 100 BSM1 100 C3 150 G5 G8 12 to 16 21 HMA surface replaced after 10 years Sound

N1 Section 13, Km 45.4 to 48.6 (Springfontein to Trompsburg)

Dry 30 HMA plus Seal 150 BSM1 100 BSM2 300 G6 G7 10 to 13 25

Fog spray at 11 years, base patching and seal

after 16 years Warning

N1 Section 14, Km 6.0 to 19.0 South (Springfontein to Trompsburg)

Dry 30 HMA plus Seal 150 BSM1 100 G4 300 G6 G7 10 to 13 25

Fog spray at 11 years, base patching and seal

after 16 years Warning

N2 Section 16, Km 38.5 to 38.7 and 37.6 to 37.8, North (Kwelera)

Wet 40 HMA 140 BSM2 130 C4 300 G7 (?) G9 (?) 2 to 3 26 Seal placed at 10 and 24 years in service Sound

N2 Section 20, Km 31.7 to 37.1 (Tabankulu to Mzintlava)

Wet 30 HMA 180 BSM1 270 G6 150 G7 G10 1 to 2 7 None to date Sound

N3 Section 4, Km 38.7 to 38.9 East (Mooi River) Wet 40 HMA 200 BSM1 150 C3 150 G7 (?) G8 9 to 21 17 None reported Warning

N3 Section 12, Km 34.5 to 36.0 North (Modderfontein to Buccleugh)

Mod 40 HMA 200 BSM1 200 C3 150 G9 G9 10 to 18 20 40 mm Overlay and

surface repairs in some areas after 12 years

Sound

N3 Section 12, Km 36.3 to 40.8 South (Modderfontein to Buccleugh)

Mod 40 HMA 360 BSM1 150 C4 150 G7 G8 (?) 10 to 18 20 40 mm Overlay and

surface repairs in some areas after 12 years

Sound

N4 Section 1, Km 19.3 to 25.6 (Scientia to Pienaars River)

Mod 30 HMA 170 BSM1 130 C4 300 G6 G7 3 to 4 9 None Sound

N4 Section 5X, Km 20.0 to 25.0 East (Wonderfontein to Crossroads)

Wet 30 HMA 150 BSM1 150 C3 150 G7 (?) G9 (?) 1.5 to 5 9 None Sound

N4 Section 5X, Km 27.6 to 29.0 East (Wonderfontein to Crossroads)

Wet 30 HMA 150 BSM2 300 C3 150 G6 G9 (?) 1.5 to 5 9 None Failed

Page 25: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Selection and Précis of LTPP Sections Investigated

- 21 -

Project Climate Surface Base Subbase Selected Subgrade Traffic (mesa)

Age (years) Maintenance Condition

N7 Section 7, Km 47.0 to 47.2 (Garies to Okiep) Dry Seal 150 BSM1 50 C3 150 G6 G8 1 to 1.5 18

Single seal after 8 years, Fog spray after

18 years Sound

N12 Section 19, Km 27.17 to Km 27.32 Slow Lane East, Experimental Section 2 (Daveyton)

Mod 55 HMA 135 BSM1 130 C3 220 G5 G6 16 to 27 30

Fog spray after 17 years, crack sealing and patching started after 26

years in service

Warning

N12 Section 19, Km 27.32 to Km 27.47 Slow Lane East, Experimental Section 1 (Daveyton)

Mod 55 HMA 120 BSM1 120 G3 80 C3 G6 16 to 27 30

Fog spray after 17 years, crack sealing and patching started after 26

years in service

Warning

P23-1, Km 3.0 to 4.0 (Kroonstad to Steynsrus) Mod 35 HMA 150 BSM2 150 G6 150 G8 G8 (?) 0.5 to 1.3 13 None Failed

Road 1386, Stabilised subbase, Km 14.7 to 17.9 and 19.7 to 20.7 (Moloto)

Mod 30 HMA 150 BSM2 150 C4 350 G8 G10 1.9 to 4.8 5 None Warning

Road 1386, Unstabilised subbase, Km 17.9 to 18.8 and Km 19.4 to 19.7 (Moloto)

Mod 30 HMA 150 BSM2 300 G7 - G6 1.9 to 4.8 5 None Sound

MR504, Km 3.0 to 3.7, Section A (Shongweni) Wet Surface

Seal 175 BSM1 150 G6 150 G10 G10 0.7 to 2 12 Initially un-surfaced; 19 mm Seal placed

after 3 years Warning

MR504, Km 3.0 to 3.7, Sections B and C (Shongweni)

Wet 30 HMA 175 BSM2 150 G6 150 G10 G10 0.7 to 2 12 Initially un-surfaced;

30 mm asphalt placed after 3 years

Warning

N11-8, Km 44.4 to 45.9, North (Ermelo to Hendrina) Mod 20 UTFC 180 BSM2 250 G8 - G8 0.6 to 1.1 3 None Sound

P24-1, Km 65.9 to 66.6 West (Vereeniging) Mod 30 HMA

(B/R) 150 BSM1 210 C4 300 G7 (?) G9 (?) 2 to 4.5 7 None Sound

P243/1, Km 14.4 to 21.14 (Vereeniging) Mod 25 HMA 250 BSM2 150 G8 - G10 0.2 to 0.5 6 None Sound

R22-4 (MR439), Km 0 to 31.2 (Phelandaba) Wet Slurry 250 BSM2 - - G9 0.3 to 2.2 4 Double seal after 4

years Sound

Same-Himo Road (Tanzania) Wet Surface

Seal 175/150 BSM2 100 G6 150 G7 (?) G8 1.4 to 4 11 None Sound

TR16-3, Km 7,4 to 14.4 and 25.0 to 30.0 (Nieuwoudtville to Calvinia)

Dry Double Seal 200 BSM2 175 G5 - G5 0.1 to 0.6 2 None Warning

Page 26: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Identification of Trends

- 22 -

4 IDENTIFICATION OF TRENDS

With such a large selection of LTPP data, it is possible to identify trends in the data and to make some observations about the relative performance of foam and emulsion roads. In Figure 1 and Figure 2, a visual summary of the LTPP sections is shown for the foam and emulsion sections, respectively. From the summaries presented earlier and the figures, the following trends are observed:

Condition

• Three of the 19 emulsion sections (16%) are considered to have failed, 6 of 19 (32%) are in a warning condition and 10 of 19 (53%) are in a sound condition. Generally, but not exclusively, the sections in a warning condition have carried more traffic than those in a sound condition. None of the foam sections have failed, although the foam sections are, on the whole, significantly younger than the emulsion sections. 4 of 11 foam sections (36%) are in a warning condition and 7 of 11 (64%) are in a sound condition.

Base thickness

• The BSM base thicknesses range from 100 to 360 mm. Of the emulsion sections, only 4 out of the 19 sections (21%) had a base or base and subbase with a combined thickness equal to or greater than 200 mm. Of the foam sections, only 3 of the 11 sections (27%) had a base thickness equal to or greater than 200 mm. The foam sections with the thicker base layers are all 6 years old or less. With the increase in deep in situ recycling, it is thought that the base thicknesses experienced in future pavements will exceed 200 mm more often than was experienced in the past.

BSM parent material

• In 12 of the 19 emulsion sections (64%), the parent material for the emulsion layer was cemented crushed stone, 5 of the 19 sections (26%) originally had a natural gravel base layer, 2 sections (11%) originally had crushed stone base layers and 1 section was originally cemented natural gravel. Of the foam sections, 9 out of 11 sections (82%) had natural gravel in the base layer, 1 section had RAP and 1 had a crushed stone base layer. The base layers where the parent material was a cemented crushed stone can be expected to perform better than layers with natural gravel. This means that it is likely that the emulsion sections will show better performance than the foam sections.

Support layers

• In virtually all cases the rehabilitation was performed on a road that had carried traffic for several years already. It can therefore be safely assumed that in almost all cases, the subbase and selected layers had been well compacted under traffic. The only sections that were newly constructed were Road D2388 near Cullinan, both N4-5X sections and R22-4 (MR439) on which new subbases were constructed.

• Fourteen of the 19 emulsion sections (74%) have a cemented subbase, whereas only 1 out of 11 foam sections (9%) have a cemented subbase. It is suspected that the presence of cemented subbases may be a key factor in the performance of the sections which have accommodated medium to heavy traffic. With a higher proportion of emulsion pavements having cemented subbases than foam, it is expected that the emulsion sections will show superior performance.

• Only 3 out of 19 emulsion pavements (16%) have a gravel subbase, whereas for the foam sections, 10 out of 11 (91%) have a gravel subbase. One emulsion section has, and no foam sections have, a crushed stone subbase.

• Sixteen out of 19 of emulsion sections (84%) and 8 of 11 of foam sections (73%) have subbase thicknesses 150 mm or less or no subbase at all. This includes sections that accommodated in excess of 10 MESA. The TRH4 catalogue (1996) allows no CTB or crushed stone bases on cement treated subbases with thicknesses below 200 mm (at times 300 mm) for traffic in the range of 3 to 10 MESA. Thus, compared to conventional design methods, significant savings can potentially be realized for BSM pavements if the LTPP data are incorporated in the design method.

Page 27: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Identification of Trends

- 23 -

Surfacing

• 17 out of 19 emulsion sections (89%) have an asphalt surfacing with or without a seal and only 2 sections (11%) have a single or slurry seal. However, only 3 out of 11 foam sections (27%) have an asphalt surfacing and 8 (73%) have seals or no surfacing. The presence of an asphalt layer probably contributes to improved performance and therefore it is expected that the emulsion sections will perform better than the foam sections.

Traffic accommodated

• Several of the sections evaluated have accumulated traffic in excess of 10 mesa, which exceeds that typically recommended for BSM pavements in current catalogues (Asphalt Academy, 2002; SABITA, 1993 and 1999). In TRH4, only pavements with granular and hot mix asphalt bases are recommended for traffic classes exceeding 10 MESA. The BSM LTPP data show that these catalogues may be conservative with respect to pavements that incorporate BSM layers.

Page 28: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Identification of Trends

- 24 -

3 6 9 12 15 18 21 24

MR 504 (1) No surfacing160 FTB150 Natural gravel (G6)12 Recycled asphalt (RAP)

P24/1 730 Surfacing150 FTB 210 CTB300 Natural gravel

Crushed stone

Section Pavement structure*

BSM Layer Parent Material

MESA Accommodated to Date

Slurry/fog seal125 FTB150 Natural gravel (G6)12 Natural gravelMR 504 (2)

Slurry/fog seal180 ETB150 Natural gravel (G6) 12 Natural gravelMR 504 (3)

Same-Himo (1) 11Double seal150 FTB100 mm natural gravel (G6)

Natural gravel

11Double seal150 FTB 100 mm natural gravel (G6)Same-Himo (2) Natural gravel

11Double seal175 FTB 100 mm natural gravel (G6)

Same-Himo (3) Natural gravel

AgeYears

Cemented subbase

Natural gravel subbase

BSM subbase

Crushed stone subbase

LEGEND:

Emulsion sections

Foamed bitumen sections

Sound ConditionWarning ConditionFailed Condition

Subbase Type BSM Parent Material

P243/1 625 HMA250 FTB 150 Natural gravel (G8) Natural gravel

Natural gravel

R22-4 (MR439) 4Slurry250 FTB Natural gravel and sand Natural gravel

N11-8 420 UTFC180 FTB 250 Natural gravel (G8)Natural gravel

Natural gravel

TR16-3 2Double seal200 FTB 175 Natural gravel (G5)Natural gravel

Natural gravel

Recycled asphalt (RAP)

Natural gravel

Cemented crushed stoneCemented natural gravel

Crushed stone

3 6 9 12 15 18 21 24

MR 504 (1) No surfacing160 FTB150 Natural gravel (G6)12 Recycled asphalt (RAP)

P24/1 730 Surfacing150 FTB 210 CTB300 Natural gravel

Crushed stone

Section Pavement structure*

BSM Layer Parent Material

MESA Accommodated to Date

Slurry/fog seal125 FTB150 Natural gravel (G6)12 Natural gravelMR 504 (2)

Slurry/fog seal180 ETB150 Natural gravel (G6) 12 Natural gravelMR 504 (3)

Same-Himo (1) 11Double seal150 FTB100 mm natural gravel (G6)

Natural gravel

11Double seal150 FTB 100 mm natural gravel (G6)Same-Himo (2) Natural gravel

11Double seal175 FTB 100 mm natural gravel (G6)

Same-Himo (3) Natural gravel

AgeYears

Cemented subbase

Natural gravel subbase

BSM subbase

Crushed stone subbase

LEGEND:

Emulsion sections

Foamed bitumen sections

Sound ConditionWarning ConditionFailed Condition

Sound ConditionWarning ConditionFailed Condition

Subbase Type BSM Parent Material

P243/1 625 HMA250 FTB 150 Natural gravel (G8) Natural gravel

Natural gravel

R22-4 (MR439) 4Slurry250 FTB Natural gravel and sand Natural gravel

N11-8 420 UTFC180 FTB 250 Natural gravel (G8)Natural gravel

Natural gravel

TR16-3 2Double seal200 FTB 175 Natural gravel (G5)Natural gravel

Natural gravel

Recycled asphalt (RAP)

Natural gravel

Cemented crushed stoneCemented natural gravel

Crushed stone

Figure 1: Visual Summary of LTPP Sections (Foamed Bitumen Sections)

Page 29: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Identification of Trends

- 25 -

N12-19 (1)50 HMA + Seal120 ETB80 Crushed stone140 Cemented natural gravel

N12-19 (2)50 HMA + Seal135 ETB130 CTB300 Natural gravel

N1-1325 HMA + Seal150 ETB 150 ETB300 Natural gravel (G6)

N2-16

30

30

25

25

3 6 9 12 15 18 21 24Cemented crushed stone

Cemented crushed stone

Cemented crushed stone

N1-1 2180 Surfacing100 ETB 100 CTBNatural gravel (G8)

Cemented crushed stone

140 Surfacing140 ETB 130 CTBNatural gravel (G6)

Cemented crushed stone & natural gravel

N3-4 1740 Surfacing200 ETB 150 CTB

Cemented crushed stone

19Seal150 ETB 50 CTB150 Natural gravel

N7-7 Cemented crushed stone

P23/1 1320 HMA150 ETB 150 Natural gravel (G6) Crushed stone

D2388 9Slurry100 ETB 150 CTB300 Natural gravel (G6)

Natural gravel

N4-5X (20-25) 9

9N4-5X (27-30)

25 AC + Seal150 ETB 150 CTB

Crushed stone

25 AC + Seal150 ETB 300 CTBNatural gravel (G5 or G6)

Natural gravel

N4-1 930 Surfacing170 ETB 130 CTB

Cemented crushed stone

Section Age Pavement BSM Parent Material

MESA Accommodated to Date

MR27 1940 HMA + Seal100 ETB 100 CTB150 Natural gravel

Cemented crushed stone

N1-14 25 Cemented crushed stone

25 HMA + Seal150 ETB 100 Natural gravel (G4)300 Natural gravel (G6)

N3-12 (North) 20Cemented crushed stone

40 Surfacing200 ETB 200 CTBNatural gravel (G8/G9)

N3-12 (South) 20Cemented crushed stone

40 Surfacing360 ETB 150 CTBNatural gravel (G7)

N2-20 635 HMA180 ETB 270 Natural gravel (G7)Natural gravel

Cemented natural gravel

1386 (SB) 530 HMA150 ETB 150 CTBNatural gravel

Natural gravel

1386 (SB) 530 HMA150 ETB 300 Natural gravel (G7)Natural gravel

Natural gravel

N12-19 (1)50 HMA + Seal120 ETB80 Crushed stone140 Cemented natural gravel

N12-19 (2)50 HMA + Seal135 ETB130 CTB300 Natural gravel

N1-1325 HMA + Seal150 ETB 150 ETB300 Natural gravel (G6)

N2-16

30

30

25

25

3 6 9 12 15 18 21 24Cemented crushed stone

Cemented crushed stone

Cemented crushed stone

N1-1 2180 Surfacing100 ETB 100 CTBNatural gravel (G8)

Cemented crushed stone

140 Surfacing140 ETB 130 CTBNatural gravel (G6)

Cemented crushed stone & natural gravel

N3-4 1740 Surfacing200 ETB 150 CTB

Cemented crushed stone

19Seal150 ETB 50 CTB150 Natural gravel

N7-7 Cemented crushed stone

P23/1 1320 HMA150 ETB 150 Natural gravel (G6) Crushed stone

D2388 9Slurry100 ETB 150 CTB300 Natural gravel (G6)

Natural gravel

N4-5X (20-25) 9

9N4-5X (27-30)

25 AC + Seal150 ETB 150 CTB

Crushed stone

25 AC + Seal150 ETB 300 CTBNatural gravel (G5 or G6)

Natural gravel

N4-1 930 Surfacing170 ETB 130 CTB

Cemented crushed stone

Section Age Pavement BSM Parent Material

MESA Accommodated to Date

MR27 1940 HMA + Seal100 ETB 100 CTB150 Natural gravel

Cemented crushed stone

N1-14 25 Cemented crushed stone

25 HMA + Seal150 ETB 100 Natural gravel (G4)300 Natural gravel (G6)

N3-12 (North) 20Cemented crushed stone

40 Surfacing200 ETB 200 CTBNatural gravel (G8/G9)

N3-12 (South) 20Cemented crushed stone

40 Surfacing360 ETB 150 CTBNatural gravel (G7)

N2-20 635 HMA180 ETB 270 Natural gravel (G7)Natural gravel

Cemented natural gravel

1386 (SB) 530 HMA150 ETB 150 CTBNatural gravel

Natural gravel

1386 (SB) 530 HMA150 ETB 300 Natural gravel (G7)Natural gravel

Natural gravel

Figure 2: Visual Summary of LTPP Sections (Emulsion Sections)

Page 30: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Summary and Recommendations

- 26 -

5 SUMMARY AND RECOMMENDATIONS

5.1 SUMMARY

Part of Phase 2 of the project for development and calibration of a structural design procedure for bituminous stabilised materials involved the collection of long-term pavement performance data for foam and emulsion pavements. This memorandum describes the method of selecting potential sections for more detailed analysis, the associated data collection and the actual sections studied. These LTPP data were used to develop and validate the structural design method, which is described by Jooste and Long (2007).

A total of 134 potential LTPP sections were identified, of which 109 are emulsion roads and 25 are foam roads. There are significantly less foam roads because the technology is newer and therefore there are fewer sections available. From these 134 roads, 30 sections from 19 sites were selected for further investigation. Of the 30 sections, 19 are emulsion sections and 11 are foam. The selection of the sections to study was determined by the available information, age, traffic carried and pavement structure. Some of the sections were studied as part of the Inception Study (Long and Jooste, 2006), and from this deficiencies in the data were identified. In Phase 2 of the project, the selection of additional sections to study also attempted to mitigate the deficiencies identified in the Inception Study.

LTPP summaries were drawn up for each section and include information on the road location, climate, year of construction, available documentation, construction history, pavement structure, behaviour and performance indicators, maintenance and traffic carried. The summaries were also all sent for review by engineers involved in the design, construction or on-going maintenance of the sections. The detailed summaries are included in Appendix B, and a précis of each section is given in Section 3.

Using the LTPP data collected, some trends in the data were identified and were discussed in Section 4.

5.2 RECOMMENDATIONS

The LTPP sections studied in this project are all existing roads that continue to accommodate traffic. Many of the sections are still in a sound or warning condition and therefore the traffic carried on the sections to date does not represent the traffic to failure. It is therefore recommended that future maintenance, rehabilitation, behaviour, performance and traffic data continue to be collected for all the sections. This will allow the confidence frontiers of performance, developed in the structural design method, to be updated with the relevant data as the sections age and reach failure.

In order to be able to continually improve the reliability and confidence in the structural design method, it is recommended that consultants and road authorities are encouraged to collect the appropriate data on existing or new BSM pavements. This aspect is planned for the additional work required for Phase 2 of this project, as detailed in the work proposal.

Page 31: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

References

- 27 -

6 REFERENCES

ASPHALT ACADEMY, Interim Technical Guideline: The Design and Use of Foamed Bitumen Treated Materials, TG2, September 2002.

JOOSTE, Long and Hefer. 2007. A Method for Consistent Classification of Materials for Pavement Rehabilitation and Design. Technical Memorandum. Modelling and Analysis Systems, Cullinan, South Africa. SABITA/Gauteng Department of Public Transport, Roads and Works, Pretoria, 2007. (GDPTRW report number: CSIR/BE/IE/ER/2007/0005/B).

JOOSTE, Long and Hefer. 2007. A Knowledge Based Structural Design Method for Pavements Incorporating Bitumen Stabilised Materials. Technical Memorandum. Modelling and Analysis Systems, Cullinan, South Africa. SABITA/Gauteng Department of Public Transport, Roads and Works, Pretoria, 2007. (GDPTRW report number: CSIR/BE/IE/ER/2007/0004/B).

LONG and Jooste. 2006. Updating Bituminous Stabilized Materials Guidelines: Structural Design Inception Study. Technical Memorandum. Modelling and Analysis Systems, Cullinan, South Africa. SABITA/Gauteng Department of Public Transport, Roads and Works, Pretoria, 2006.

SABITA, 1993. GEMS - The Design and Use of Granular Emulsion Mixes, Manual 14, 1993.

SABITA, 1999. ETB: The Design and Use of Emulsion-treated Bases, Manual 21, 1999.

Technical Recommendations for Highways, Structural Design of Flexible Pavements for Interurban and Rural Roads, TRH4, Draft, 1996.

WRIGHT, B.G., et al, Long Term Performance of Emulsion Treated Base Pavements, PR88/014, RDAC, CSIR, Pretoria, 1991.

Page 32: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Appendix A: Project Work Proposal

APPENDIX A: PROJECT WORK PROPOSAL

Page 33: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

1

Project Proposal Number: PP/2005/09/d Version: 1.0

Updating Bituminous Stabilized Materials Guidelines, Phase 2: Development and Calibration of Structural Design Procedure Submitted by: Dr Fenella Long, MAS Name Company Contact details Dr Fenella Long MAS (083) 399 0090, [email protected] Dr Wynand Steyn Transportek, CSIR (012) 841 2634, [email protected]

1. TERMS OF REFERENCE This proposal forms part of a larger study proposal, entitled “Compilation of a Bituminous Stabilized Materials Guideline Document for Foamed and Emulsified Bitumen Treated Materials” (Proposal Number PP/2005/09). As such, the tasks and methodology defined in this proposal comprise a sub-task of a larger project and should be viewed as such. The background, project objectives, expected benefits and implementation plans are described in detail in proposal PP/2005/09, and will not be restated here. Briefly, the project pertains to the updating of the TG2 and TG(X) guidelines, which are intended to guide and assist practitioners in the selection and design of pavements and materials that incorporate bituminous stabilization consisting of bitumen emulsion or foamed bitumen. The objectives of the larger project are to improve or redesign the modules relating to the mix design and structural design, to ensure that the guidelines reflect the latest best practice as well as all available research and field performance data. The overriding objective is to compile a complete guideline document incorporating the mix and structural design and construction guidelines. The first phase of the project was an inception study with two components, mix and structural design. An objective of the inception study was to plan further testing and development activities needed for the thorough revision of the TG2 and TG(X) guidelines. It was recommended that in Phase 2 of the project the mix design and structural design guidelines be developed and reviewed, and in Phase 3 the guideline document be compiled. This proposal pertains to the structural design component of Phase 2 of the larger project, which comprises the development of the structural design guidelines. The development of the mix design guidelines is detailed in a companion proposal (PP/2006/03/c). The two components of Phase 2 should be executed in parallel.

2. PROJECT OBJECTIVES This structural design guideline development project has three objectives:

1) To expand the LTPP database compiled during the Inception Study to a sufficient level to enable the development of classification based design guidelines.

2) Develop and populate a classification based design matrix.

3) Peer review of the recommended design guidelines.

Page 34: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

2

3. METHODOLOGY The methodology proposed in the following paragraphs is in accordance with the framework for a structural design method as documented in the Structural Design Inception Study Reports: • Long F.M., Jooste F.J. and Jenkins K.J., Updating Bituminous Stabilized Materials Guidelines:

Inception Study, Summary Report, February 2006, DRAFT. • Long F.M., and Jooste F.J., Updating Bituminous Stabilized Materials Guidelines: Structural Design

Inception Study, Technical Memorandum, February 2006, DRAFT. This development of the structural design guidelines for a Bituminous Materials Guideline will comprise of the following tasks: Task 1: Update LTPP Summaries with Recently Available Information At the time of the inception study some information on the LTPP sections studied was not available (e.g. the SANRAL 2005 network survey and the current condition). In this task, this information will be accessed and the LTPP summaries updated. It may not be possible to obtain actual performance data on the current condition, however, and in those cases the clients or relevant engineers will be contacted to obtain a subjective opinion of the current condition. Task 2: Peer Review of LTPP and HVS summaries Since the LTPP and HVS summaries developed in the Inception Study will form the empirical base for the structural design methodology, it is essential that these summaries be accurate and appropriate. This task therefore involves a peer review of the LTPP and HVS summaries developed in the Inception Study. The project engineers, resident engineers and clients originally involved in the design and construction of each section will be contacted to:

• Validate and/or correct the information contained in the LTPP summaries, and • Expand the summaries to include practical hints and tips that ensured successful construction.

The peer review will also be performed on any additional LTPP summaries compiled in Task 5. Task 3: Develop Material Classification Method This task involves developing a robust, appropriate method for a classification based design method. The following specific subtasks are involved:

• Study and select model for determining the appropriate material class from several indicators. • Develop and test the selected model. • Identify and contact practitioners, and hold workshop to select classification parameters and

ranges. This task will be executed in conjunction with the mix design project team, as the mix design output will form important inputs into the material classification system. However, the theoretical framework of the method can be developed prior to the availability of the recommended mix design method with the associated tests and classification limits. The method will then be refined in Task 7 when the mix design results are available and any additional LTPP data have been collected. Task 4: Develop Structural Design Matrix/Method In this task the design matrix will be populated with the available data. This requires:

• An in-depth study of the available LTPP and HVS data. • Formulation of the structure of the design matrix. • Assignment of the pavement structures to the design matrix. • Interpolation of new structures using mechanistic principles for unpopulated areas within the

design matrix. Task 5: Expand LTPP Database In this task, the database of LTPP pavements developed in the Inception Study will be expanded to address the deficiencies identified in the Inception Study and from the previous task. At this stage it

Page 35: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

3

is envisaged that an additional 10 to 16 sections will need to be included, however this will be finalised only after completion of Task 4. The data gathered in this task will include the following, where applicable:

• Identification and assessment of failed pavements; • Additional foamed bitumen pavements; • Pavements with natural gravel subbases (particularly with thick recycled bases); • Pavements on poor quality subgrades; • Foam pavements constructed on stabilised subbases, and • BSM pavements constructed as part of a newly constructed road.

The data will be obtained by three means. Firstly, analysis of existing as-built records, available behaviour and performance data and traffic assessments in line with the methodology used to prepare LTPP summaries in the Inception Study. Secondly, some sections will require field measurements, as discussed in the next task. Thirdly, some sections, such as the sections on poor quality subgrades will be developed by interpolating between, or extrapolating from, existing structures using mechanistic principles. Task 6: Collect Additional Field Data This task involves the collection of field data from in-service pavements. This is only required for pavement types where no data are already available, and will be limited to the collection of essential data in an effort to limit the total project cost. Some or all of the following data will be collected:

• Testpits and material tests • Deflection measurements (FWD) • Visual condition • Rutting measurements • DCP tests • Roughness measurements • Core extraction

The data will be used to compile LTPP summaries for the sections in line with the summaries prepared in the Inception Study. Task 7: Refine Material Classification and Design Method The material classification and design method developed in Tasks 3 and 4 will be refined with the data collected in Tasks 5 and 6 and on inclusion of the recommended mix design method selected in the companion Phase 2 project. This task can therefore only begin when the mix design method is finalised. The populated design matrix will also be refined and validated through liaison with identified expert practitioners. Task 8: Simple Methodology for Designing Structures not Included in the Design Matrix (optional) In this task, the method used to design the non-validated structures in the design matrix will be expanded and documented. This will be done by formalising the method and inputs used and defining the inputs required. This method should only be used by experienced practitioners, and could be included in the final BSM guideline as an appendix. Note that this task involves the use of mechanistic principles, and not the mechanistic-empirical method. As such, transfer functions will not be used. This task is optional. Task 9: Strategy for On-going Population of the Design Matrix This task involves the development of a strategy to facilitate the on-going population of the design matrix as data on more pavements become available, or as updated data on pavements included in the design matrix become available. This involves determining a strategy for alerting practitioners

Page 36: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

4

and road authorities to the need for more data, and the compilation of a detailed list of both the required and desired data. Task 10: Documentation The project findings will be detailed in a technical memorandum, which will serve as a backup to the guidelines to be presented in the guideline document. The findings will also be summarized in a project summary report.

4. DELIVERABLES The deliverables for this project will consist of a technical memorandum, a project summary report and a detailed presentation to the project funders. The project documentation will contain the following information:

• Expanded LTPP summaries; • Description of method selected for material classification from several indicators; • Populated design matrix; • Method for designing structures not explicitly included in design matrix (optional), and • Strategy for obtaining new data on pavements as it becomes available. The project documentation will be structured according to the Gautrans Documentation Guidelines so that raw data and technical discussions are limited to the technical memorandum, while the project summary report will concisely summarize key observations and recommendations.

5. SKILLS DEVELOPMENT The skills development part of this project will involve the use and associated mentoring of a Gautrans technician or young engineer in the project work, particularly in Tasks 5 and 6 as outlined in the Methodology section. The skills imparted will be:

• Data retrieval • Analysis of retrieved data • Exposure to project documentation, such as as-built records, design reports and Pavement

Management Systems • Basic reporting The cost of employment of the mentee will be carried by Gautrans as part of their normal salary. The cost of the mentoring by the project team is included in the Project Cost section of this proposal.

6. PROJECT PLAN 6.1. Project Team/Personnel The project team will consist of Dr Fritz Jooste, Dr Fenella Long and Sanet Jooste of MAS with Prof. Kim Jenkins acting in an advisory role. Details of the project team are summarized below. Table 1. Project Team

Name Organisation Contact details Hourly Rate

Dr Fritz Jooste (FJ) MAS (082) 578 5628, [email protected] R 500 Dr Fenella Long (FL) MAS (083) 399 0090, [email protected] R 450 Sanet Jooste (SJ) MAS (072) 266 1726, [email protected] R 420 Professor Kim Jenkins (KJ) SANRAL Chair (082) 920 7859, [email protected] R 500

The development of the mix design guidelines will be championed by Professor Kim Jenkins. The output from the mix design tests are an input into the structural design guidelines, and therefore the mix design and structural design teams will work in tandem to ensure the mix and structural design guidelines are aligned.

Page 37: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

5

6.2. Project Costs Detailed costs are only provided for Phase 2: Structural Design Procedure Development. These costs are detailed in Table 2. The cost of Task 6 is a range, which depends on the number of field sites from which field data will be collected. The number of sections will be finalised after the completion of Task 4 and in consultation with the clients. The work will be done by Fritz Jooste, Fenella Long and Sanet Jooste. Two workshops with experienced practitioners will be held during the project. The costs shown include the time cost of the project team, but do not include the cost of holding the workshops. The Project Manager’s costs are not included in this proposal, but are listed in the master proposal. Table 2. Project Costs for Structural Design Procedure Development Task Type Task Detailed Action

Minimum - Maximum Minimum - Maximum1 Update LTPP summaries with recently available

information R

2 Peer review LTPP and HVS summaries R5 Expand LTPP database R6 Collect additional field data

(3 to 10 sites at R 50 000 per site) R 150 000 - 500 000

3 Develop material classification method R4 Develop structural design matrix/method R7 Refine material classificaiton and design method R8* Simple methodology for designing structures not

included in design matrix R

9 Strategy for on-going population of design matrix R10 Document R

Skills development mentoring RLiaison, meetings and presentation to Funders RDirect Costs (Travel, documentation, etc.) R

R 772 000 - 1 122 000R 108 080 - 157 080R 880 080 - 1 279 080

* This task is optional

12 00020 000

Value Added Tax (14 %)Total

40 000

10 00040 000

Cost

Development tasks

760 000

Sub Total

20 000

40 000

50 00080 000

200 000

50 000

Cost

Other

Reporting tasks

410 000 -

220 000

50 000

92 000

60 000

Data gathering tasks

6.3. Time Line The Gantt chart is shown in Figure 1. The project will take 12 months to complete, however the start of some tasks is dependant on the completion of previous tasks and the Mix Design study. The estimates shown in Figure 1 are based on sufficient information coming from Part 1 of the applicable Mix Design Tasks (see Phase 2 Mix Design Proposal). Should this not be the case, the completion of the tasks shown will be delayed.

Task Detailed Action1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

1 Update LTPP summaries with recently available information2 Peer review LTPP and HVS summaries3 Develop material classification method4 Develop structural design matrix/method5 Expand LTPP database *6 Collect additional field data **7 Refine material classification and design method ***8 Simple methodology for designing structures not included in design matrix9 Strategy for on-going population of design matrix

10 DocumentNotes

* Scope is subject to output of Tasks 3 and 4** Start and scope of task is subject to output of Task 5

*** Start and scope of taks is subject to output of Tasks 5 & 6 and the Mix Design study

Month

Figure 1. Gantt Chart

Page 38: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Appendix B: LTPP Summaries

APPENDIX B: LTPP SUMMARIES

Page 39: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 D2388

Road Type: ETB Road Code: Road 2388, Cullinan Constructed: 1997 (reconstruction) Limits Considered: Km 2.1 to 2.3 Project Description: Road 2388, Between Road 25 and Onverwacht, near Cullinan,

focussing on Km 2.1 to 2.3 Available Documentation: 1. Demecon Consultants, “Preliminary Design Report: Portion of Road 2388 between Road

25 and Onverwacht”, File Number 10/4/1/4-2388(1), 1996. 2. Demecon Consultants, “The Upgrading of a Portion of Road 2388 Between Road 25 and

Onverwacht Using Labour Intensive Methods”, Project Document: Tender No GT 155 B(R), 1996.

3. As-built records obtained from Gautrans. 4. LTPP Data and associated information obtained from CSIR. 5. Gautrans PMS. 6. Discussion with, and review by, Hechter Theyse, CSIR.

Area and Climate The road is situated in the Gauteng province, approximately 10 km north-east of Cullinan. The road is located on the Gauteng Highveld which experiences hot humid summers and cool dry winters. The area has a Weinert-n value of approximately 2.5, and the average annual rainfall is approximately 735 mm, with 85 per cent of rainfall occurring in the summer months [1].

Construction History This road originally consisted of a gravel road which provided access to properties in the area. The road was declared a public district road in 1975, and can therefore be assumed to have been at least 20 years old at the time of the upgrading in 1997. The road was upgraded in 1997, using the following basic construction in the area under consideration: • 8 mm slurry seal surfacing; • 100 mm emulsion treated base (1% cement and 1.67% emulsion); • 150 mm cement stabilized natural gravel (3 % cement), and • 300 mm G6 selected layer (constructed in two layers). The pavement layers up to and including the subbase were constructed using conventional construction methods. The emulsion treated base was constructed using labour intensive construction methods. The specified method of construction for the base was as follows [2]:

“During construction, the gravel base material was mixed dry with the cement in a concrete mixer. Water and emulsion was then mixed separately and added to the gravel in the mixer and mixed until a uniform mix was obtained. The mix was then transported in wheelbarrows and placed in longitudinal strips contained between steel shutters or adjacent pre-constructed strips. The base was constructed in two layers of equal thickness, and compacted using a vibrating pedestrian roller.”

The subbase was supposedly stabilised, however the curing procedure was poor. Field resilient modulus tests indicated that some of the sections were stabilized, but some were effectively unstabilised. The laboratory UCS tests showed high values indicative of stabilised materials, but these were measured in the laboratory and do not therefore reflect the field construction. [6] Key material indicators extracted from the as-built data for the section under consideration are summarized in Table 1.

Page 40: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 D2388

Table 1: Summary of As-Built Information (from samples taken within Km 2.1 and 2.3) [1]

Layer Property Specification Observations from As-Built Records Material Imported

Weathered Dolerite

G5 Quartzitic Sandstone stabilized with 1% Portland Cement and 1.5% bitumen emulsion (60%).

Thickness 100 mm Layer thickness for four samples varied from 100 to 104 mm.

Grading Not Shown A sieve analysis is shown for samples in the area, but could not be assessed as the sieve sizes are not stated.

PI < 6 PI of 3 is indicated for all samples. Compaction > 97% Varied from approximately 95% to 98%.

100 mm Imported Emulsion Treated Base (based on 4 samples in area)

UCS at 95% Mod AASHTO

> 500 kPa Varied 490 and 510 kPa.

Material In-situ sand G5 Quartzitic Sandstone stabilized with 3.0 per cent Portland cement.

Thickness 150 mm 147 mm reported. Grading Not Shown A sieve analysis is shown for samples in the area, but

could not be assessed as the sieve sizes are not stated.

PI Not Shown 3 reported. Compaction > 95% 99 per cent reported.

150 mm Imported Stabilized Subbase (1 sample only)

UCS at 95% Mod AASHTO

Not Shown 1195 kPa reported.

Material In-situ sand G5 Quartzitic Sandstone. Thickness 150 mm 148 mm reported. Grading Not Shown A sieve analysis is shown for samples in the area, but

could not be assessed as the sieve sizes are not stated.

PI Not Shown 6 reported. Compaction > 93% 98 per cent reported.

150 mm Imported Upper Selected Layer (1 sample only)

CBR at 93% Mod AASHTO

Not Shown 34 per cent reported.

Material In-situ sand G5 Quartzitic Sandstone and Ferricrete. Thickness 150 mm 150 mm reported. Grading Not Shown A sieve analysis is shown for samples in the area, but

could not be assessed as the sieve sizes are not stated.

PI Not Shown 10 reported. Compaction > 93% 95 per cent reported.

150 mm Imported Lower Selected Layer (1 sample only)

CBR at 90% Mod AASHTO

Not Shown 16 per cent reported.

Behaviour and Material Quality Indicators • FWD deflections recorded as part of the HVS LTPP programme are summarized in

Table 2.

Table 2. FWD deflections [4] Normalized Maximum Deflection (micron) Location Measurement

Date Average 90th Percentile Feb 2005 355 416 Outer

Wheel Path Jul 2005 292 366 Feb 2005 241 310 Inner Wheel

Path Jul 2005 184 282 Feb 2005 296 309 Lane

Centre Jul 2005 224 294

• As expected, the deflections recorded in the winter months are lower than those recorded towards the end of the summer. The deflections in the outer wheel path are greater than

Page 41: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 D2388

those measured in the inner wheel path, but there is no consistent difference between the deflections measured in the lane centre and the inner wheel path.

Performance Indicators • The documentation for the LTPP data collection states that a seal was placed over the

section. The Gautrans PMS states that a 13 mm and 6 mm double seal with a modified binder was placed in July 2003 (as-built information code, PMS), and another 13 mm and 6 mm double seal in January 2004 (design information code, PMS).

• Rut depths measured in February and July 2005 (8 years after opening to traffic) showed very similar trends with an average rut depth of approximately 4 mm and a 90th percentile rut depth of 7.8 mm.

• Visual assessments conducted in February 2005 showed only surfacing distress and some undulation. Surfacing distress was generally limited to bleeding and undulation. No cracking was reported, but it should be noted that the road had been resealed a few years before the assessment was performed.

• Riding quality measured in 2004 from Km 2 to 3 and recorded in the Gautrans PMS shows the 95th percentile HRI is 16, the 5th percentile is 3 and the average is 7.6. This indicates the road is in a very poor condition. Riding quality was already poor immediately after construction because of the labour-intensive construction technique that was used [6].

• Visual assessments from the Gautrans PMS [5] are summarized in Table 3 as a Visual Condition Index (VCI). This is for Km 0 to 3.51 which includes the section under consideration. The VCI is low in 2003, just prior to the seal placement. Since 2005 the road is not in a good condition.

Table 3. VCI from Gautrans PMS Year VCI 2001 55 2002 50 2003 33 2004 90 2005 69 2006 68

Traffic Loading Indicators • The preliminary design report [1] showed traffic measurements taken in 1995. The data

showed 135 vehicles per day, of which 30 per cent were heavy vehicles (i.e. 40 heavy vehicles per day). The estimated traffic growth was 3 per cent, and the 20 year design traffic was estimated to be between 0.2 and 0.8 million standard axles.

• Traffic counts conducted in February 2005 as part of the LTPP data collection efforts measured 184 E80s per day in the westbound direction, and 7 E80s per day in the eastbound direction.

• Traffic counts from the Gauteng PMS are reported as AADT and percent heavy vehicles. The E80s per lane were calculated as the percentage heavies of the AADT, multiplying by E80/heavy factors of 2 and 3. .

The available traffic data is summarized in Figure 1, which shows the available measured data at different periods, together with a lower and higher best estimate of the traffic increase. It will be noted from the figure that the data measured in 1995 is somewhat lower than the initial traffic assumed for the higher estimate. This was done to obtain realistic growth rates that would more or less link the initial traffic with the traffic measured in 2005. This is not an

Page 42: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 D2388

unrealistic assumption, since the 1995 data was measured on the existing gravel road, and the traffic is likely to have increased after the upgrading of the road. Comments recorded in the LTPP data state that most heavy vehicles were observed carrying sand and it is therefore possible that some vehicles are overloaded, particularly in the westbound direction. Using the data projected along the upper and lower estimate lines shown in Figure 1, the cumulative E80s on the road between 1997 and 2005 are estimated to be between 0.2 and 0.5 MESA. This estimate accords with the 20 year design traffic as projected at the time of the preliminary design.

0

50

100

150

200

250

300

350

400

1994 1996 1998 2000 2002 2004 2006Year of Measurement

E80s

Per

Day

LTPP Data, Feb 2005Lower EstimateHigher Estimate1995 Data, E80/Heavy = 2.01995 Data, E80/Heavy = 2.51995 Data, E80/Heavy = 3.0Gautrans, E80/Heavy = 2Gautrans, E80/Heavy = 3

Figure 1: Traffic Data Summary with Estimated Growth Scenarios

Factual Observations • The Gautrans PMS states that a 13 mm and 6 mm double seal with a modified binder

was placed in July 2003 (6 years of service), and another 13 mm and 6 mm double seal in January 2004 (7 years of service).

• Rut depths measured in February and July 2005 (8 years after opening to traffic) showed an average rut depth of approximately 4 mm.

• Riding quality measured in 2004 (7 years of service) averages an HRI of 7.6. This indicates the road is in a very poor condition; however the riding quality was poor immediately after construction.

Observations Based On Estimates or Interpretations • An estimated 0.2 to 0.5 million E80s have been carried in 8 years of service. • The road was in a poor condition, based on the VCI (2006, 9 years of service). Rutting in

2004 (after seal placement) is fair. • The highest average deflection measured in 2005 was less than 350 microns.

Reviewed by: • Hechter Theyse, Transportek, CSIR.

Page 43: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 MR27

Road Type: ETB Road Code: MR27 near Stellenbosch Constructed: 1988 (reconstruction) Limits Considered: Slow lane at approximately Km 9.8 to 10.0 (both directions), now

between Km 23.8 and 24.8 Project Description: MR27, Stellenbosch Available Documentation: 1. Wright, B.G., et al, “Long Term Performance of emulsion treated base pavements”,

PR88/014, RDAC, CSIR, Pretoria, 1991. 2. As-built for reconstruction (from PAWC, compiled by UWP) 3. Rose, D.A., and G.J. Jordaan, “Initial assessment for pavement rehabilitation”, conducted

in 1985 (MR27). 4. Discussions with, and review by, Mr Mike White of Uhlmann Witthaus & Prins (UWP),

who were the consultants for the construction and subsequent rehabilitation actions. 5. Steyn, WJvdM, “Summary of field and laboratory test results on selected ETB

pavements”, CR-98/073, CSIR, Pretoria, 1997. 6. Theyse, H.L., “Towards Guidelines for the Structural Design of Pavements with Emulsion-

treated Layers”, CR-97/087, CSIR, Pretoria, 1998. 7. Discussions with Mr Andre van der Gryp from WCPA. 8. WCPA PMS.

Area and Climate The road is situated near Stellenbosch in the Western Cape, in a moderate climate with a Weinert n value of 5, and an average annual rainfall of around 620 mm [1]. The section under consideration is 200 m long and situated roughly 10 km from Stellenbosch [1], at a sign which states “Stellenbosch 10 km”. These stations do not, however, coincide with the WCPA database. By correlating the km stations for Somerset West and Stellenbosch, the 200 meter section must be between the WCPA Km 23.8 and 24.8 [7].

Construction History This ETB pavement was constructed as a major rehabilitation of an existing road which was carried out in 1988. The original pavement was constructed between 1966 and 1971 and consisted of the following [3]:

• Wearing course consisting of a 30 mm surfacing and possibly a double seal; • 2 layers of cement treated base, each 100 mm thick; • 150 mm ferricrete gravel subbase; • 250 mm ferricrete selected subgrade. According to the recorded history [1] the pavement structure after the 1988 rehabilitation was thus as follows [1]:

• Double seal (13.2 mm and 6.7 mm using 3% SBR modified cationic emulsion binder); • 40 mm continuously graded asphalt; • 100 mm emulsion treated base; • 100 mm cement stabilized subbase; • 150 mm laterite selected layer; • In-situ sand. An examination of the as-built sheets for the ETB layer showed the following key aspects [2]:

• The base layer construction is described as “milled cement treated base recompacted with 1% net bitumen (60% anionic emulsion) followed by a 40 mm layer of 20% recycled asphalt surfacing and a 13.2 mm/6.7 mm double seal”;

• For samples taken in the area under consideration, the base material consisted of a coarse material with less than 20% of material passing the 0.425 mm sieve (average percentage passing 0.425 mm is roughly 18%);

• The percentage passing the 19 mm sieve is roughly between 85 and 96%;

Page 44: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 MR27

• The PI is described as SP in all instances; • Average layer thicknesses varied from 88 to 107 mm. A discussion with Mr Mike White of UWP suggests that the milled base material was premixed with water, into which the 60% emulsion was mixed. The base was thus compacted towards the wet side of optimum, and the material was initially very soft under the rollers. A 1997 investigation [5], during which cores were extracted from the base at two locations, inside and outside the wheel path (i.e. to provide four cores in total) showed a residual binder content of 0.5%.

Behaviour and Material Quality Indicators • RSD deflections were measured around 1989 over a 200 mm long LTPP section situated

around Km 10. These deflections showed a deflection of roughly 437 micron for the northbound lane, and 226 micron for the southbound lane. The 95th percentile values for the north and southbound lanes were 586 and 329 micron, respectively.

• During a 1997 investigation [5] it was found that intact cores could not be extracted for UCS testing. This suggests that the material was in a rather granular state.

• DCP results obtained in 1997 [5] show an average penetration rate of 1 mm/blow in the ETB base, with a standard deviation of 0.5 mm. This corresponds to a 95th percentile that is roughly less than 2 mm/blow. The DCP investigation showed that 8 out of 9 DCP tests failed to penetrate at depths ranging from 140 to 211 mm. This depth roughly corresponds to the top of the cement stabilized subbase (i.e. lower base of the original pavement constructed around 1970).

• Moisture contents obtained from samples of the ETB base in 1997 [5] show an average moisture content of 5.2%, with a standard deviation of 0.7%.

• FWD deflections recorded in 1997 [5, 6] show that the deflection ranges roughly from 235 to 520 micron, with an average of around 390 micron.

• FWD peak deflections from the WCPA PMS [7] are shown in Table 1. The pavement was in a good condition after both 10 and 14 years of service, and the northbound carriageway has lower deflections than the southbound.

Table 1. FWD Deflections from WCPA PMS

Southbound Northbound 1998 2002 1998 2002

95th Percentile 314 398 438 287 5th Percentile 209 243 189 168 Average 270 317 308 227

Performance Indicators • In 2002, after 14 years of service, a 13.2 mm and 6.7 mm double seal with a polymer

modified binder was placed. [7] Before the reseal, approximately 0.2% of the slow lanes and the original fast lanes were patched in the base, and approximately 1.0% were surface patched.

• Rut depth measurements were taken around 1989 over a 200 mm long LTPP section situated around Km 10. The average rut for both directions was less than 4 mm, while the 95th percentile rut for both directions was less than 5 mm.

• The average rut depth measured in a 1997 study [5] was 5.5 mm, with a standard deviation of 1.4 mm. This corresponds to a 95th percentile rut of around 8 mm.

• Rutting is recorded in the WCPA PMS [8], as summarized in Table 2. It is likely that the 2002 Northbound data were recorded before the seal placement as the rutting decreases to minimal levels in 2006.

Page 45: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 MR27

Table 2. Rutting (mm) from WCPA PMS [8] Southbound Northbound

2001 2006 2002 2006 95th Percentile 6.0 4.0 9.1 4.0 5th Percentile 3.0 2.0 3.0 1.0 Average 4.2 3.2 5.7 2.3

• Riding quality measurements from the WCPA PMS are illustrated in Figure 1, where the

line is the 5th to 95th percentile, and the square the average. The 2006 data were collected every 10 metres whereas the other data were collected every 100 meters, which accounts for the larger variance in 2006. On the Southbound section, the IRI is relatively constant until 1998, and then reduces in 2001 only to increase again in 2006. On the Northbound section, the IRI increases, and then decreases in 2002, probably due to the placement of the seal. Except for 2006 on the Southbound direction, the IRI is in all cases close to or less than 3, indicating a fair to good roughness. On the Southbound direction in 2006, roughness is fair to poor.[8] There is no obvious reason for the increase in IRI and in general the pavement is still in a good condition. [7]

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006Year

IRI

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006Year

IRI

Southbound IRI Northbound IRI Figure 1. Riding Quality from WCPA PMS • The visual survey in 2006 showed little visual distress on the Northbound lane and on the

Southbound lane there was minor (not exceeding degree or extent 2) bleeding, transverse cracking and pumping.

• The VCI (pavement index) in 2006 ranged between 90 and 95%, which indicates a good condition. [8]

• According to Andre van der Gryp from WCPA, in June 2006 the section is in a good condition. [7]

Traffic Loading Indicators • A traffic analysis carried out in 1985, three years prior to the construction of the ETB

pavement. This analysis was based on data collected between 1970 and 1984, and showed the following key aspects and predictions [1,3]: o The heavy vehicle count for 1984 was estimated as being between 660 and 810

heavy vehicles per day for the slow and fast lanes combined. The heavy vehicle growth rate was estimated at 7%;

o A lane distribution factor of 0.9 was used for the slow lane; o An E80/heavy vehicle factor of 1.2 was assumed for the slow lane, with an expected

growth rate of 0 to 4% in this factor assumed. o The 1984 estimate of E80’s per day for the slow lane ranged from 356 to 438; o Based on the above assumptions, the 1984 estimate of the 20 year total equivalent

E80’s ranged from 7.3 to 9.0 million • An analysis of the traffic data performed in 1989 [1], estimated that the 20 year design

traffic (i.e. from 1988 to 2008) was roughly 11 million standard axles. • WCPA PMS traffic data is summarized in Table 3. A growth rate of 5.29% was calculated

[8].

Page 46: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 MR27

Table 3. Traffic data from WCPA [8]

E80s / day / slow lane** Year Heavy Vehicles* E80/heavy = 1.5 E80/heavy = 3 2000 571 407 814 1991 583 415 831 1986 635 452 905 1983 745 531 1062 1979 274 195 390

* assumed in two directions ** 0.95 lane distribution factor applied

• Using all the traffic estimates, and growth rates of 4 and 6%, the cumulative traffic in 18 years of service (1988 to 2005) is 3.3 to 6.8 million E80s and the 20 year design traffic (1988 to 2007) 4.1 to 8.8 million E80s. This is less than the estimate made in 1989 for the 20 year design traffic.

0

200

400

600

800

1000

1200

1400

1600

1975 1980 1985 1990 1995 2000 2005Year

E80s

per

day

per

slo

w la

ne

WCPA, E80/heavy 1.5WCPA, E80/heavy 3References 1 & 3Lower estimateUpper estimate

Figure 2. Traffic estimates

Factual Observations • In 2002, after 14 years of service, a double seal consisting of 13 mm and 7 mm seals with

a polymer modified binder was placed.[7]. • Rut depth measurements taken in 1989 averaged less than 4 mm, and 5.5 mm in 1997. • The average rut depth measured in a 1997 study [5] was 5.5 mm. • Rut depth measurements from 2001 on the Southbound lane averaged 4.2 mm, and from

2002 on the Northbound lane averaged 5.7 mm. • Rut depth measurements from 2006 on the Southbound lane averaged 3.2 mm, and on

the Northbound lane averaged 2.2 mm. • Riding quality (IRI) on the southbound section is close to or less than 2.5 (1992 to 2001)

but ranges between 2 and 4 in 2006. On the northbound section, the IRI is close to or less than 3 (1991 to 2006). Overall, the riding quality indicates a fair to good roughness.

• The VCI (pavement index) in 2006 ranged between 90 and 95%, which indicates a good condition.

Observations Based on Estimates or Interpretations • The cumulative traffic in 18 years of service (1988 to 2005) is estimated as 3.3 to 6.8

million E80s and the 20 year design traffic (1988 to 2007) 4.1 to 8.8 million E80s. • RSD peak deflections measured in 1989 averaged approximately 437 micron for the

Northbound lane, and 226 micron for the Southbound lane. • FWD deflections recorded in 1997 [5,6] show that the deflection averages 390 micron. • Average FWD peak deflections are 270 microns (Southbound, 1998), 317 microns

(Southbound 2002), 308 (Northbound 1998) and 227 (Southbound 2002) [7].

Page 47: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 MR27

• The road is in a good condition in June 2006 (4 years after seal application and 19 years after construction.)

Anecdotal Information: This was one of the first roads sealed with a modified binder. It lasted until the reseal in 2002. [4] The milled CTB was very coarse because the original CTB had strengths between 10 and 15 MPa, and the fines were cemented. The material was broken down in a milling drum (not a modern recycler) and the grading therefore depended on the fineness obtained. Approximately 10% of crusher dust was added to improve the grading, but this dust also did not have a large proportion passing the 0.075 mm sieve. [4] The contractor diluted the emulsion with the mixing water. The method was objected to by Mike White, however the specification did not cover the emulsion dilution and the contractor continued with his method. [4] Reviewed by: • Mr Mike White (UWP)

Page 48: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N1 Section 1

Road Type: ETB Road Code: N1 Section 1 near Kraaifontein Constructed: 1984 (reconstruction) Limits Considered: Km 34.2 to 34.4 Northbound, and Km 34.8 to 35, Southbound Project Description: N1 Section 1, Kraaifontein Available Documentation: 1. Wright, B.G., et al, “Long Term Performance of Emulsion Treated Base Pavements”,

PR88/014, RDAC, CSIR, Pretoria, 1991. 2. As-built for reconstruction (from LTPP Study) 3. Project history from SABITA seminar in 1984 (JA du Plessis) 4. Construction history and traffic data from the SANRAL PMS (from 1997). 5. Discussions with Mr. Mike White of Uhlmann Witthaus & Prins (UWP), who were the

consultants for the construction and subsequent rehabilitation actions. 6. Steyn, WJvdM, “Summary of field and laboratory test results on selected ETB

pavements”, CR-98/073, CSIR, Pretoria, 1997. 7. Theyse, H.L., “Towards Guidelines for the Structural Design of Pavements with Emulsion-

treated Layers”, CR-97/045, CSIR, Pretoria, 1998. 8. SANRAL PMS 9. Discussions with Heuppeschenne Lekay (Cayad), Maintenance Route Manager.

Area and Climate The road is situated near Kraaifontein in the Western Cape, in a moderate climate with a Weinert n value of 5, and an average annual rainfall of around 620 mm [1]. The section under consideration is situated between the Klapmuts and Paarl Interchanges [3].

Construction History This pavement was originally constructed between 1969 and 1972 [3], and consisted of a 50 mm surfacing over a 200 mm C2 cement treated base constructed in two 100 mm layers, with a 150 mm G5 subbase and 250 mm G8 selected layers. Problems were reported within a few years after construction, and by 1978, roughly 45% of the constructed CTB section was reported to show pumping and reflection cracks [3]. A heavy rehabilitation was performed in 1984, which converted the pavement structure to a emulsion treated base over a cemented subbase. The rehabilitation saw the recycling of the top 100 mm layer of the cement treated base, with the addition of 10 to 20 per cent crusher dust and 1 per cent bitumen by mass in the form of an emulsion [1]. According to the recorded history [1] and discussions with Mr. Mike White of UWP [5], the construction process involved the following: • The asphalt surfacing was milled off and stockpiled. • The C2 base was then milled 100 mm deep and the milled material was windrowed. • The remaining 100 mm of the old C2 base was then inspected and badly deteriorated

areas were removed (these comprised only approximately 15% of the slow lane area). • 10 to 20 per cent crusher dust was then mixed with the milled C2 base material. The

material was then premixed with water, after which a 60% emulsion was added to obtain a net bitumen content of 1%. Compaction took place below optimum moisture content.

• A 40 to 50 mm continuously graded premix was then added to bring the road surface level to the fast lane. This asphalt mix consisted of a 50/50 blend of milled RAP from the stockpile of milled asphalt and new aggregate and binder. Due to the high RAP content careful control was exercised over the mix grading.

• A 13.2 mm single bitumen-rubber seal was applied as a SAMI across the fast lanes which had extensive CTB shrinkage cracks.

• A 40 mm semi-gap graded asphalt was then overlaid over the slow and fast lanes. The pavement structure after the 1984 rehabilitation was thus as follows [1,4,5]: • 40 mm semi-gap graded asphalt with rolled in chips; • 40 mm continuously graded asphalt; • 100 mm emulsion treated base;

Page 49: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N1 Section 1

• 100 mm cement stabilized subbase; • 150 mm G5 selected layer; • 250 mm G8 upper subgrade and selected layer. ([8] states 250 mm G8 on Northbound

and 150 mm G7 on Southbound). The SANRAL PMS has no record on an ETB layer in these particular sections, however discussions with Mr. Mike White confirmed the ETB layer is present. Table 1: Summary of As-Built Information, Roughly Km 34 to 35, North and Southbound [2] Layer Property Specification Observations from As-Built Records

Thickness 100 Observed thicknesses vary from 90 to 140 mm, and on average are around 115 mm.

Material Type BT2 The source material was a cement treated Malmesbury Hornfels

Stabilizers 1% Net bitumen 1% Net bitumen was added of a 60% cationic stable grade emulsion.

100 mm Emulsion Treated Base

Grading/Fineness

Not shown Percentage passing the 0.425 sieve is generally below 20%, and generally around 15%. The grading modulus is generally above 2.4 and on average around 2.5. This suggests a fairly coarse material.

A 1997 investigation [6], during which cores were extracted from the base at two locations, inside and outside the wheel path (i.e. to provide four cores in total) showed a residual binder content of 0.8%.

Behaviour and Material Quality Indicators • Deflections (assumed to be measured with the Deflectograph) were measured in 1989,

roughly 5 years after the ETB construction. The average deflections for the north and southbound directions were 247 microns and 340 microns, with 95th percentile values of 365 microns and 499 microns, respectively. An assessment of the deflection bowl parameters, conducted as part of an earlier LTPP investigation of this pavement suggested that the pavement structure is fairly stiff and generally sound [1].

• FWD deflections recorded in 1997 [6,7] show that the deflection ranges roughly from 140 to 480 microns, with an average of around 315 microns. The 95th percentile is 425 microns.

• After 15 years of service (1999), FWD deflections measured by SANRAL are 379 and 501 microns on the northbound section and 350 and 440 microns on the southbound direction. After 21 years of service (2005), FWD deflections are 413 and 320 microns on the northbound section. No measurements from 2005 are available for the southbound direction.[8]

• DCP measurements taken in 1989 show a very low penetration rate in the base (generally below 1 mm/blow). The cement stabilized C2 layer could not be penetrated in most instances. The selected layers and subgrade at a depth of roughly 400 to 800 mm showed a high penetration rate and the in-situ CBR of these materials (as suggested by DCP measurements) seem to be below 10%.

• During a 1997 investigation [6] it was found that intact cores could not be extracted for UCS testing.

• DCP results obtained in 1997 [6] show an average penetration rate of 0.8 mm/blow in the ETB base, with a standard deviation of 0.3 mm. This corresponds to a 95th percentile that is roughly less than 2 mm/blow.

• The 1997 investigation [6] showed an average in-situ moisture content of 5.6%, with a standard deviation of 1%.

Performance Indicators and Treatment History • The rutting measured after 5 years in service (1989), showed an average rut of 10 mm

and 13 mm on the north and southbound sections, respectively, with a 95th percentile rut of roughly 15 and 19 mm, respectively [1]. Test pits were opened in some areas, and

Page 50: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N1 Section 1

showed that the rutting was confined to the asphalt layers, and did not originate from the ETB or the lower pavement layers. This was confirmed by Mr. Mike White of UWP [5].

• The average rut depth measured in a 1997 study [6] was 5.8 mm, with a standard deviation of 3.5 mm. This corresponds to a 95th percentile rut of around 12.8 mm. This observation appears to contradict the fact that the road was resurfaced in 1995/1996, but does however, tie in with the SANRAL measurements. Mike White disputes this rutting measurement because it was so soon after the overlay and because it does not agree with his observations of the road at the time.

• The SANRAL PMS rutting data are illustrated in Figure 1 [8]. High levels or rutting were measured in 1995, prior to the 1995 rehabilitation. The rutting was however, confirmed to all be in the semi-gap graded asphalt layer and not from the underlying layers. [5] From 1999 the rutting levels have remained fairly constant, although they are relatively high.

• Riding quality measurements taken in 1989 suggest a sound condition [5]. This is confirmed with SANRAL measurements taken from 1999 to 2002, shown in Figure 2.

• Owing to the rutting of the asphalt layers, the semi-gap grade asphalt was milled off the slow lane in 1995/6, and a new continuously graded asphalt surfacing with rolled in chips was constructed. The original semi-gap asphalt is still on the fast lane.

• According to Mr. Mike White, no other rehabilitation was performed on this road since 1995, which is confirmed by the SANRAL PMS. In 2006, the current condition of the road was reported as being good. However, reflection cracks from the cement stabilized subbase have appeared. These cracks generally appear as evenly spaced transverse cracks. These cracks pump during the rainy season. Mike White is of the opinion that the lower CTB layer has failed.

• The SANRAL visual surveys collected in 1995, 1996 and 1997 show the road is in a good condition with no structural problems.

• In July 2006, the northbound slow lane is experiencing some pumping in an area approximately 1.5 by 2 meters. There is also some crocodile cracking on about 25% of the section. The northbound fast lane is in a fairly good condition with a little cracking on about 5 to 10% of the section. The southbound slow lane has rutting of about 10 to 17 mm deep in about 15 meters of the 200 metre section. It is most likely to be rutting in the 1995/1996 overlay. The condition of the fast lane is satisfactory. Within 1 kilometre on either side of the specific section patching is required in some places. Overall, the condition is fair, but the road could deteriorate fairly rapidly if the rainfall is high in the upcoming rainy season. [5,9]

0

5

10

15

20

25

30

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003Year

Rut

dep

th (m

m)

0

1

2

3

4

5

1998 1999 2000 2001 2002 2003Year

IRI

Figure 1. Rut depths from SANRAL PMS Figure 2. IRI from SANRAL PMS

Traffic Loading Indicators Traffic data were obtained from the LTPP study report [1], and from the SANRAL PMS [8]. The traffic data are summarized in Table 2 and are illustrated in Figure 1. The estimated cumulative traffic in the slow (worst) lane from 1984 to 2005 is 11.5 to 15.8 million E80s, which are the areas under the lines shown in Figure 3. It is of interest to note that an analysis performed in 1985 [3] showed an estimated total number of cumulative axles of 11 million E80s for the period 1985 to 2005. This study however, probably used in E80 per heavy factor of 2, which is not representative of current and recent traffic levels.

Page 51: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N1 Section 1

Table 2. Traffic Data Summary for N1 Section 1 Northbound Slow Lane Southbound Slow Lane Year

Vehicles per day E80/lane/day Vehicles per day E80/lane/day 1985 [3] Not known 680 (800)1 1986 [1] Not Known 412 583 1995 [4] 24016 1443 (1698) 24016 1510 (1777) 1996 [8] 1946 2 1927 1997 [8] 2107 2160 1998 [8] 1644 1684 1999 [8] 2252 2057 2000 [8] 2875 2040 2001 [8] 2662 2644 2002 [8] 2623 3045 2003 [8] 2673 2645 2004 [8] 3112 3351

Note 1: A lane distribution factor of 0.85 was assumed. The value in brackets denote the estimated E80 per direction. Note 2: Data from SANRAL PMS for worst lane

0

500

1000

1500

2000

2500

3000

3500

4000

1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006Year

E80s

/day

/wor

st la

ne

North, SANRALSouth, SANRALNorth, historicalSouth, historical

Figure 3. Traffic Data An assessment made as part of the 1997 study conducted for SABITA [6], states that the traffic demand class for the pavement is ES100. This is based on an AADT of 15 900 and 9.9 per cent heavy vehicles. These values are not shown in Table 2 since it is not known to which direction the AADT refers to (or if it is for both directions).

Factual Observations • Since the rehabilitation in which the ETB base layer was constructed, the pavement has

been in service for 20 years without a need for structural strengthening or rehabilitation. • The asphalt surfacing was milled off and replaced after roughly 10 years in service. • Only one maintenance or rehabilitation action, which involved the surfacing layer only,

has been performed in a 20 year service period. • The rut depth in 2002 (18 years of service) is less than 10 mm. • The IRI in 2002 (18 years of service) is under 2.

Observations Based On Estimates or Interpretations • An estimated 11.5 to 16 million E80s have been carried in 21 years of service. • The road is still in a good condition, based on: deflections (peak deflection less than 500

microns after 15 years), IRI (less than 2 after 18 years), and Mike White’s comments. Although the rut depth is fairly high, the section is not in a terminal condition.

Anecdotal Comments The milled CTB was very coarse because the original CTB had strengths between 10 and 20 MPa, and the fines were cemented. The material was broken down in a milling drum (not a modern recycler) and the grading therefore depended on the fineness obtained.

Page 52: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N1 Section 1

Approximately 15-20% of crusher dust was added to improve the grading, but this dust also did not have a large proportion passing the 0.075 mm sieve. Some rutting had occurred on various isolated sections by 1999, particularly on the upgrades. The cause is suspected to be variations in the bitumen quality [5]. Several sections were removed and replaced during construction, and some disputed sections remained in place for monitoring. Semi-gap was a poor design choice but was based on the belief in those days that rolled-in chips required a semi-gap mix. The mix stability was further affected by a cost decision to source some of the fines from a local sand borrow pit on land belonging to the Department of Transport to save money. The particle shape of this wind blown sand was variable and although all the dynamic creep and Marshall stability check tests were within the specification, Mike White believes that this variable sand may have been the low rutting resistance of this mix as in a few places the mix has not rutted much, even on the slow lane sections which weren’t replaced in 1995/96.) Reviewed by: Mr Mike White, UWP Engineers

Page 53: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N1 Section 13

Road Type: ETB Road Code: N1 Section 13 Constructed: 1978 to 1980 Limits: Section 13, Km 45.394 to Km 48.560 (North and Southbound) Project Description: N1 Section 13 Springfontein and Trompsburg Available Documentation: 1. Mackintosh, Bergh and Sturgess, “Spesiale Instandhouding van Nasionale Roete 1

Seksies 13 & 14 Tussen Springfontein en Trompsburg” Ontwerpverslag, 1996. 2. As-built for original design (from WSP) 3. Steyn, WJvdM, “Tentative Classification of ETB Pavements in a design catalogue based

on available pavement data”, CSIR report CR-96/070, Pretoria, July 1997. (This report, however, provides no source for the data.)

4. SANRAL PMS. 5. Discussion with Ofal Motlankene (BKS), Route Manager. 6. Discussion with Corné Roux (SANRAL). 7. Discussion with Johan Opperman (SNA), previous Route Manager. 8. Discussion with and review by Mike Hughes (previously WSP).

Area and Climate The road is situated in the Free State province between Springfontein and Trompsburg, specifically between the two turnoffs to Springfontein. This area has a dry climate with a Weinert N value of 2 [1]. The average rainfall reported between 1984 and 1993 was roughly 460 mm per year [1].

Construction History The original pavement was constructed in 1969. The road was rehabilitated between 1978 and 1980 [1]. Before rehabilitation, the base layer consisted of a cement stabilized crushed stone. This full width of the layer was ripped, crushed off-site and then re-compacted with the addition of emulsion and cement. The in situ subbase was ripped and recompacted with the addition of emulsion. After rehabilitation, the generalized pavement structure consisted of [1,2,8]: • 25 mm AC + 13 S1; • 150 mm recovered, re-crushed cemented crushed stone, stabilized with 1% cement and

0.7% bitumen; • 100 mm decomposed dolerite, stabilized with cement and 0.6 to 0.7% binder; • Two 150 mm G6, weathered dolerite selected layers. The essential information obtained from the as-built sheets are summarized in Table 1. The UCS values clearly suggest the base layer had a high cohesion after construction. A discussion with the resident engineer on the project, Mr. Mike Hughes, confirmed that the base was treated with 24 l/m3 of emulsion and 1% cement. The exact variations and type of emulsion and binder are not shown on the as-built records, but the available information and conversation with the engineers involved suggests the structure as shown above is correct.

Page 54: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N1 Section 13

Table 1: Summary of As-Built Information [2] Layer Property Specification Observations from As-Builts

Stone Type Dolerite Spread Rate 0.01 Generally achieved 0.011 Binder 150/200 pen

13.2 mm Single Seal

Spray Rate Varies 1.3 to 1.9 Generally achieved Thickness 25 mm Average is 25 mm, but varies greatly between 14 mm

and 34 mm. % Binder 5 % Average is 5%, maximum is 5.6%. % Voids 4 to 6 % Largely in spec, but 38% of section shows values

greater than 6.0%. Stability 6 to 12 kN Generally above 10 kN.

25 mm Wearing Course+

Compaction 95% Almost all above spec, on average ± 96%. Material In-Situ Dolerite As-builts make no mention of emulsion, only cement

treated. Thickness 150 mm Varies greatly between 100 mm and 195 mm. On

average close to 150 mm. Grading Not Shown The grading modulus is almost always above 2.4 and

on average is around 2.6. The percentage passing the 0.425 mm sieve is always below 18, and on average around 15. This indicates a relatively coarse material meeting the specification for G1 to G3 material.

PI <= 6 Average PI = 5, with a maximum PI = 7. Compaction >= 88% ARD Generally achieved or exceeded, only 12% of areas

less than limit. Average close to 90%

150 mm In-situ treated crushed stone base

UCS Not Shown Only one result less than 1800 kPa, many greater than 1900 kPa. Not often measured.

Material In-situ decomposed dolerite

Only one 100 mm layer included in as-builts. No mention of cement or lime. Entire section of subbase reconstructed with 0.7% bitumen [8].

Thickness 100 mm All in spec, average is 115 mm and maximum is 130 mm.

Grading Not Shown GM > 1.5

The subbase grading modulus generally varies between 1.8 and 2.3. The percentage passing the 0.425 mm sieve is generally around 24%, but varies roughly between 15 and 33. This suggests a G4 or G5 material.

Bitumen Content

0.6 to 0.7 % Varies greatly between about 0.3 and 1.2%.

PI < 10 On average around 8, 3% above 10. Compaction > 95 Mod

AASHTO Generally in spec, on average around 97%, with 15% above 100%.

100 mm Bitumen Stabilized Subbase

CBR @ 95% Mod. AASHTO

> 70 Generally close to or above 100%, average is 122%.

Behaviour and Material Quality Indicators • On the whole construction section (N1-13 & N1-14), FWD deflections measured in

1994 (confirmed FWD by Hugh Thompson), showed that 94% of the deflections were below 700 micron, with all deflections being below 1000 micron. 84% of deflections were between 400 micron and 700 micron. The deflections are summarized in Table 2. The analysis of deflection bowl parameters [1] suggested that in the areas with a higher deflection the middle layer index was associated with lower traffic than the base layer index or the lower layer index. This suggested that the subbase and selected layers were the weaker zones in the pavement structure.

• FWD deflections measured in 1999 on the southbound direction [4] showed that 68% of the deflections were below 700 micron and none of the deflections were above 1000 micron. 50% of deflections were between 400 micron and 700 micron. The 95th percentile deflection was 880 microns and the average deflection 620 microns, which indicates a poor condition. The deflections are summarized in Table 2.

• FWD deflections measured in 2006 on the southbound direction [4], showed that 63% of the deflections were below 700 micron, with 3% of deflections above 1000 micron.

Page 55: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N1 Section 13

59% of deflections were between 400 micron and 700 micron. The 95th percentile deflection was 958 microns and the average deflection 659 microns, which still indicates a poor condition, however there has been an improvement since 1999. The improvement may be due to extensive patching done on the road. The deflections are summarized in Table 2.

Table 2: Summary of FWD deflections

19941 19992 20062 Deflections Both

directions Southbound Northbound Southbound

% < 700 94 68 75 63 % > 1000 0 0 3 3 % 400 – 700 84 50 72 59 Average 621 560 613 659 95th percentile 880 803 958

Note 1: Deflections across whole N1-13&14 construction sections. 2: Deflections across specific section under investigation

• A visual assessment of the pavement showed that the areas with high deflections (greater than 700 micron) were in a good condition, and a link between performance and deflection could apparently not be established [1].

• Based on the deflections measured in 1994, as well as the DCP data (presumed to be measured in 1994, according to DCP analysis sheets shown in [1]), the subgrade appeared to be of a good quality, with a backcalculated stiffness of 100 MPa or above. Most DCP derived CBR values are above 60%, with one isolated value of 28% [1].

• An examination of the 1994 DCP layer strength diagrams showed that the in situ shear strength of the selected layers and subgrade was often comparable to those of the base and subbase layers.

• One test pit was opened on this section in 1994 and shows that the base was generally classified as a G5 or G6. Mike Hughes commented that this classification is subjective because the in situ strength was not measured, and other roads with a G5 or G6 base may not perform as well as this section. The subbase and selected layers are not of a poorer quality, and appear to be G5 to G6 as well. The results from this test pit are similar to other test pits on the N1-13.

• The visual condition data from the SANRAL PMS, collected in 1995, 1996 and 1997 show that the degree and extent of crocodile cracking was less than 3 and no other type of cracking was detected. The rutting visual assessments showed no more than degree 3 and extent 3 rutting.

Performance Indicators and Treatment History • Based on rut depth data collected in 1994 (presumed from 1996 report [1]), 98% of

this project had rut depth values of less than 10 mm. 2% had rut depths between 10 and 20 mm.

• The rut depths from the SANRAL PMS are shown in Figure 2 [4]. The data are shown ranging from the 5th to the 95th percentile, with the average shown as the round marker. By 2002, 94% of the project had rut depths less than 10 mm and there was no rutting in excess of 20 mm [4]. This shows the road was in a good condition in 2002.

• The roughness (IRI) data from the SANRAL PMS [4] are shown in Figure 3, where the average is the round marker and the range is from the 5th to 95th percentile. The average IRI remained relatively constant from 1999 to 2002. Therefore, in terms of riding quality the road was still performing well in 2002.

• The road received a diluted emulsion in 1991 [1]. The rehabilitation design report compiled in 1996 recommended light rehabilitation over the full N1-13 and 14 construction section that consisted of the following:

Page 56: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N1 Section 13

Reconstruction of base and surfacing over approximately 2.3 km of the project length in the traffic lanes, and over approximately 4.1 km of the shoulder lanes. This comprises a total of roughly 20 per cent of the project length;

Diluted emulsion to improve adhesion of existing surfacing in certain areas; Fine slurry in areas with uneven surfacing; 13.2 mm bitumen rubber surfacing. The above-noted rehabilitation was carried out around 1996/1997. Based on

discussions with Mr Hugh Thompson, no other work has been performed on this section to date, and the condition of the road is good.

0

2

4

6

8

10

12

1995 1996 1997 1998 1999 2000 2001 2002Year

Rut

dep

th (m

m)

0

1

2

3

4

5

1999 2000 2001 2002Year

IRI

Figure 2. Rut depths from SANRAL PMS Figure 3. IRI from SANRAL PMS

• The SANRAL PMS shows that on the Section 13 segment a 13.2 mm S1 was placed in 1985 and again in 1995. However, Mike Hughes disputes this and is of the opinion that the section was patched and sealed in 1996/1997, which is in accordance with the previous bullet. [8]

• The current condition (July 2006, 26 years of service) of Section 13 is poor according to Ofal Motlankene [5], however Mike Hughes states that the specific section under consideration is in a satisfactory condition [8].

• The road is likely to be rehabilitated in the next 3 to 4 years. [6]

Traffic Loading Indicators • Traffic data provided by the (then) Department of Transport’s regional office in

Bloemfontein showed that, for the years 1990 to 1994 the percentage of heavy vehicles was between 30 and 40 per cent. A figure of roughly 23 per cent was reported for 1991, but the heavy vehicle percentages for 1990, 1993 and 1994 varied between 31 and 38 per cent [1].

• The E80 per heavy vehicle factor was estimated around 2.0 [1]. • The estimated E80 per lane per day was between 570 and 768 [1]. The figures vary

from year to year (and for Section 13 and 14), and no clear growth rate can be discerned in the reported data.

• Based on traffic counts performed at the time of the original construction in 1969, and 1978/1980 (when rehabilitation was performed), the traffic growth rate in this period was between 15.7 and 17.1%. Since 1981, the weighted average growth rate was 8.1% per year. [1].

• The growth in the E80’s per lane per day was estimated at 12.1% for the period 1969 to 1993, with a steady growth rate of 8% for the period between 1981 and 1993 [1].

• The 1996 rehabilitation report [1] estimated that, in the 20 year period between 1978 (the time of the emulsion treated base rehabilitation) and 1998, the road would have accommodated roughly between 4 and 5 MESA.

• The initial traffic in 1980 is estimated from the WSP traffic estimates [1], and extrapolated back using 3 growth rates. The initial E80s per lane per day are 300 (6%), 235 (8%), 185 (10%).

• SANRAL PMS records estimate the E80 per lane per day as [4]:

Page 57: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N1 Section 13

E80s per lane per day TRAFFIC

N1-13 N1-14 11/08/1995 1480 22/02/1996 1623 1654 25/04/1997 1625 08/08/1999 1877 01/01/2000 1850 01/01/2001 1902 01/01/2002 2010 01/01/2003 1994

0

500

1000

1500

2000

2500

78/01/01 83/06/24 88/12/14 94/06/06 99/11/27 05/05/19

E80

per d

ay

SANRAL data, N1-13SANRAL data, N1-14WSP ReportEstimated initial traffic (WSP)

01/07/2004 2316 2341.7 Figure 1. Traffic estimates

• The traffic estimates are shown in Figure 1. A range of total E80 traffic is estimated

from the area under the lines shown in Figure 1. The estimate for the total E80s from 1980 to 2005 is 10 to 13 million E80s per lane.

Factual Observations • The pavement was in service for roughly 11 years before a diluted emulsion surface

treatment was applied; • The measured rut depth after roughly 14 years in service was generally sound, with

98% of rut depths being below 10 mm. After 25 years in service the rut depth remained sound, with 94% of the rut depths being below 10 mm. After both 14 years and 25 years in service, there was no rutting exceeding 20 mm.

• The IRI remains consistently below 2.8 after 22 years in service, indicating a sound condition.

• Base repairs were recommended over approximately 20% of the project length after 15 years. The road was patched and sealed after 16 years of service.

• By 2005, the road has been in service for 25 years, and no reconstruction has been performed.

• The current condition of the road (2006, 26 years of service) is reported to be satisfactory to poor and is likely to rehabilitated in 3 to 4 years. [6].

Observations Based On Estimates or Interpretations • The estimated traffic between 1991 and 1994 was between 570 and 770 E80s per

day. The estimated growth rate in daily E80s was 8%. • The road has carried an estimated 10 to 13 million E80s in 25 years. • After 14 years in service, more than 84% of FWD deflections were between 400 and

700 micron and after 25 years, 59% (Southbound) and 72% (Northbound) of deflections were between 400 and 700 microns. This shows the pavement was deteriorating, although the performance is superior to the section on the N1-14.

• Based on deflection and DCP indicators, the subgrade on which this road is constructed is regarded as good.

• Based on the IRI and rutting, the road was performing well in 2002.

Reviewed by:

• Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 58: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N1 Section 14

Road Type: ETB Road Code: N1 Section 14 Constructed: 1978 to 1980 Limits: Southbound (Eastbound) Km 6.0 to Km 18.985 Project Description: N1 Section 14 between Springfontein and Trompsburg Available Documentation: 1. Mackintosh, Bergh and Sturgess, “Spesiale Instandhouding van Nasionale Roete 1

Seksies 13 & 14 Tussen Springfontein en Trompsburg” Ontwerpverslag, 1996. 2. As-built for original design (from WSP). 3. Steyn, WJvdM, “Tentative Classification of ETB Pavements in a design catalogue based

on available pavement data”, CSIR report CR-96/070, Pretoria, July 1997. (This report, however, provides no source for the data.)

4. SANRAL PMS. 5. Discussion with Ofal Motlankene (BKS), Route Manager. 6. Discussion with Corné Roux (SANRAL). 7. Discussion with Johan Opperman (SNA), previous Route Manager. 8. Discussion with and review by Mike Hughes (previously WSP).

Area and Climate The road is situated in the Free State province between Springfontein and Trompsburg. This area has a dry climate with a Weinert N value of 2. The average rainfall reported between 1984 and 1993 was roughly 460 mm per year [1].

Construction History The original pavement was constructed in 1969. The road was rehabilitated between 1978 and 1980 [1]. Before rehabilitation, the base layer consisted of a cement stabilized crushed stone. This layer was broken into “manageable” slabs by a pneumatic hammer on an excavator then re-compacted with the addition of emulsion and cement. [8] After rehabilitation, the generalized pavement structure consisted of [1]: • 25 mm AC + 13 S1; • 150 mm recovered, re-crushed cemented crushed stone, stabilized with 1% cement and

0.7% bitumen; • 150 mm weathered dolerite, stabilized with 2% lime and 0.7% binder; • Two 150 mm G6, weathered dolerite selected layers. However, the actual pavement structure on this section differs from the above structure, as shown in the as-built records and confirmed by Mike Hughes. The essential information obtained from the as-built sheets is summarized in Table 1. The UCS values clearly suggest the base layer had a high cohesion after construction. The exact variations and type of emulsion and binder are not shown on the as-built records. A discussion with the resident engineer on the project, Mike Hughes, confirmed that the base was treated with 24 l/m3 of stable grade anionic emulsion and 1% cement. On this section, a 600 mm strip of the old cement stabilised base was left intact on the outside edge of the pavement. [8].

Based on the as-built information, and in conversation with the engineers involved, the pavement structure on this section is assumed to be: • 25 mm AC + 13 S1; • 150 mm recovered, re-crushed cemented crushed stone, stabilized with 1% cement and

0.7% bitumen; • 100 mm decomposed dolerite, unstablized; • Two 150 mm G6, weathered dolerite selected layers.

Page 59: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N1 Section 14

Table 1: Summary of As-Built Information [2] Layer Property Specification Observations from As-Builts

Stone Type Dolerite Spread Rate 0.01 Generally achieved 0.011. Binder 150/200 pen

13.2 mm Single Seal

Spray Rate Varies 1.3 to 1.9 Generally achieved. Thickness 25 mm Average is 25 but varies greatly between 10 mm and

41 mm. % Binder 5 % 72% less than 5%, and on average close to 4.8%. % Voids 4 to 6 % Less than 50% in spec, with 32% less than 4.0%. Stability 6 to 12 kN 70% above 10 kN, all greater than 8.5 kN.

25 mm Wearing Course

Compaction 95% BRD 86% above spec, on average ± 97%. Material In-Situ Dolerite As-builts make no mention of emulsion, only cement

treated. Thickness 150 mm Varies greatly between 110 mm and 190 mm. On

average 145 mm. Grading Not Shown The grading modulus is almost always above 2.1 and

on average around 2.5. The percentage passing the 0.425 mm sieve is almost always below 30, and on average around 15. This indicates a relatively coarse material meeting the specification for G1 to G3 materials.

PI <= 6 Generally SP or NP. Compaction >= 88% ARD Average is 91%, but 21% is less than 88%.

150 mm In-situ treated crushed stone base

UCS Not Shown Not measured on this section. Material In-situ

decomposed dolerite

Only one 100 mm layer included in as-builts.

Thickness 100 mm Mostly in spec, average is 116 mm. Many around 130 to 150 mm.

Grading GM > 1.5 The subbase grading modulus generally varies between 1.5 and 2.4. The percentage passing the 0.425 mm sieve is generally around 29%, but varies roughly between 15 and 46. This suggests a G4 or G5 material.

PI < 10 Generally below 8, on average around 4. Compaction > 95 Mod

AASHTO Generally in spec, on average around 99, with 30% above 100%.

100 mm Unstabilized Subbase

CBR @ 95% Mod. AASHTO

> 70 All in spec, average is 141%.

Behaviour and Material Quality Indicators • On the whole construction section (N1-13 & N1-14), FWD deflections measured in 1994

(confirmed FWD by Hugh Thompson), showed that 94% of the deflections were below 700 micron, with all deflections being below 1000 micron, and 84% of deflections were between 400 micron and 700 micron. The deflections are summarized in Table 2. The analysis of deflection bowl parameters [1] suggested that in the areas with a higher deflection the middle layer index was associated with lower traffic than the base layer index or the lower layer index. This suggested that the subbase and selected layers were the weaker zones in the pavement structure.

• FWD deflections measured in 1999 on the southbound direction [4], showed that 14% of the deflections were below 700 micron, with 18% of deflections above 1000 micron. 12% of deflections were between 400 micron and 700 micron. The 95th percentile deflection was 1126 microns and the average deflection 859 microns, which indicates a poor condition. The deflections are summarized in Table 2.

• FWD deflections measured in 2006 on the southbound direction [4], showed that 15% of the deflections were below 700 micron, with 12% of deflections above 1000 micron. 14% of deflections were between 400 micron and 700 micron. The 95th percentile deflection was 1083 microns and the average deflection 829 microns, which still indicates a poor condition, however there has been little change from 1999 to 2004. The deflections are summarized in Table 2.

Page 60: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N1 Section 14

Table 2: Summary of FWD deflections 19941 19992 20062 Deflections Both

directions Southbound Southbound

% < 700 94 14 15 % > 1000 0 18 12 % 400 – 700 84 12 14 Average 859 829 95th percentile 1126 1083

Note 1: Deflections across whole N1-13&14 construction sections. 2: Deflections across specific section under investigation • A visual assessment of the pavement showed that the areas with high deflections (greater

than 700 micron) were in a good condition, and a link between performance and deflection could apparently not be established [1].

• Based on the deflections measured in 1994, as well as the DCP data (presumed to be measured in 1994, according to DCP analysis sheets shown in [1]), the subgrade appeared to be of a good quality, with a backcalculated stiffness of 100 MPa or above. One isolated DCP test suggests a subgrade CBR of around 6%, but most values are above 10% and many apparent in situ CBR values are above 100% (based on the DCP penetration rates) [1].

• An examination of the 1994 DCP layer strength diagrams showed that the in situ shear strength of the selected layers and subgrade was often comparable to those of the base and subbase layers.

• Six trial pits opened in 1994 show that the base was generally classified as a G5 or G6 (one G4). Mike Hughes commented that this classification is subjective because the in situ strength was not measured, and other roads with a G5 or G6 base may not perform as well as this section. The subbase and selected layers are not of a poorer quality, and appear to be G5 to G6 as well (one G4 class for the subbase was noted).

• The visual condition data from the SANRAL PMS, collected in 1995, 1996 and 1997 show that the degree and extent of all cracking metrics were less than 1. The rutting visual assessment showed 1 km of degree 2, extent 4 rutting, and the remaining sections had less than degree 3 and extent 3 rutting. In 1995 there was some degree 4, extent 5 bleeding and flushing, but this reduced to degree 2, extent 4 by 1997. The condition improved from 1995 to 1996.

Performance Indicators and Treatment History • Based on rut depth data collected in 1994 (presumed from 1996 report [1]), 88% of this

project had rut depth values of less than 10 mm. 9% had rut depths between 10 and 20 mm and 3% had rut depths in excess of 20 mm.

• The rut depths from the SANRAL PMS are shown in Figure 2 [4]. The data are shown ranging from the 5th to the 95th percentile, with the average shown as the square marker. By 2002, 78% of the project had rut depths less than 10 mm and no rut depths exceeded 20 mm [4]. This shows the road was in a fairly good condition in 2002. The roughness (IRI) data from the SANRAL PMS [4] are shown in Figure 3, where the average is the square marker and the range is from the 5th to 95th percentile. The average IRI remained relatively constant from 1999 to 2002. In terms of roughness, the road is still performing well in 2002.

• The road received a diluted emulsion in 1991 [1]. The rehabilitation design report compiled in 1996 recommended light rehabilitation that consisted of the following:

Reconstruction of base and surfacing over approximately 2.3 km of the project length in the traffic lanes, and over approximately 4.1 km of the shoulder lanes. This comprises a total of roughly 20 per cent of the project length;

Diluted emulsion to improve adhesion of existing surfacing in certain areas; Fine slurry in areas with uneven surfacing; 13.2 mm bitumen rubber surfacing; The above-noted rehabilitation was carried out around 1996/1997. Based on

discussions with Mr. Hugh Thompson, no other work has been performed on this section to date, and the condition of the road is good.

Page 61: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N1 Section 14

0

2

4

6

8

10

12

14

1995 1996 1997 1998 1999 2000 2001 2002Year

Rut

dep

th (m

m)

0

1

2

3

4

5

1999 2000 2001 2002Year

IRI

Figure 2. Rut depths from SANRAL PMS Figure 3. IRI from SANRAL PMS • The SANRAL PMS shows that on the Section 13 segment a 13.2 mm S1 was placed in

1985. A 6.7 mm S5 and 19 mm S1 were placed in 1992, and a 13.2 mm S1 was placed in 1995. However, Mike Hughes disputes this and is of the opinion that the section was patched and sealed in 1996/1997, which is in accordance with the previous bullet. [8]

• The current condition (July 2006, 26 years of service) is satisfactory although there are a few potholes. [5]

• The road is likely to be rehabilitated in the next 3 to 4 years. [6]

Traffic Loading Indicators • Traffic data provided by the (then) Department of Transport’s regional office in

Bloemfontein showed that, for the years 1990 to 1994, the percentage of heavy vehicles was between 30 and 40 per cent. A figure of roughly 23 per cent was reported for 1991, but the heavy vehicle percentages for 1990, 1993 and 1994 varied between 31 and 38 per cent [1].

• The E80 per heavy vehicle factor was estimated around 2.0 [1]. • The estimated E80 per lane per day was between 570 and 768 [1]. The figures vary from

year to year (and for Section 13 and 14), and no clear growth rate can be discerned in the reported data.

• Based on traffic counts performed at the time of the original construction in 1969, and 1980/1981 (when rehabilitation was performed), the traffic growth rate in this period was between 15.7 and 17.1%. Since 1981, the weighted average growth rate was 8.1% per year. [1].

• The growth in the E80’s per lane per day was estimated at 12.1% for the period 1969 to 1993, with a steady growth rate of 8% for the period between 1981 and 1993 [1].

• The 1996 rehabilitation report [1] estimated that, in the 20 year period between 1978 (the time of the emulsion treated base rehabilitation) and 1998, the road would have accommodated roughly between 4 and 5 MESA.

• The initial traffic in 1980 is estimated from the WSP traffic estimates [1], and extrapolated back using 3 growth rates. The initial E80s per lane per day is 300 (6%), 235 (8%), 185 (10%).

• SANRAL PMS records estimate the E80 per lane per day as shown in the table [4]: • The traffic estimates are shown in Figure 1. A range of total E80 traffic is estimated from

the area under the lines shown in Figure 1. The estimate for the total E80s from 1980 to 2005 is 10 to 13 million E80s.

Page 62: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N1 Section 14

E80s per lane per day TRAFFIC

N1-13 N1-14 11/08/1995 1480 22/02/1996 1623 1654 25/04/1997 1625 08/08/1999 1877 01/01/2000 1850 01/01/2001 1902 01/01/2002 2010 01/01/2003 1994

0

500

1000

1500

2000

2500

78/01/01 83/06/24 88/12/14 94/06/06 99/11/27 05/05/19

E80

per d

ay

SANRAL data, N1-13SANRAL data, N1-14WSP ReportEstimated initial traffic (WSP)

01/07/2004 2316 2341.7 Figure 1. Traffic estimates

Factual Observations • The pavement was in service for roughly 11 years before a diluted emulsion surface

treatment was applied; • The measured rut depth on the whole construction section after roughly 14 years in

service was generally sound, with 88% of rut depths being below 10 mm. On the specific section under consideration, after 25 years in service the rut depth remained sound, with 78% of the rut depths being below 10 mm.

• After roughly 14 years in service 3 per cent of the whole construction project length showed rutting in excess of 20 mm. After 25 years in service no rutting exceeded 20 mm on the specific segment under consideration.

• The IRI remains consistently below 3 after 22 years in service, indicating a sound condition.

• Base repairs were recommended over approximately 20% of the project length after 15 years. The road was patched and sealed after 16 years of service.

• By 2005, the road has been in service for 25 years, and no reconstruction has been performed.

• The current condition of the road (2006, 26 years of service) is reported to be satisfactory, albeit with some potholes and is likely to rehabilitated in 3 to 4 years.

Observations Based On Estimates or Interpretations • The estimated traffic between 1991 and 1994 was between 570 and 770 E80s per day.

The estimated growth rate in daily E80s was 8%. • The road has carried an estimated 10 to 13 million E80s in 25 years. • After 14 years in service, more than 80% of FWD deflections were between 400 and 700

micron and after 26 years, 14% of deflections were between 400 and 700 microns. This shows the pavement was deteriorating.

• Based on deflection and DCP indicators, the subgrade on which this road is constructed is regarded as good.

• Based on the IRI and rutting, the road was performing well in 2002.

Anecdotal Information According to Johan Opperman, the 600 mm cemented edge that was left during the rehabilitation could have constributed to the rutting experienced on the wider N1 Section 13 and 14 section. Where rutting occurred, the pavement was milled out to a depth of 150 mm, and replaced with emulsion. These repairs were done around May 2002. Andre Holder and Dennis Rossmann were involved in the investigation, but the exact locations of such repairs are unknown.

Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 63: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 6 N2 Section 16

Road Type: ETB Road Code: N2 Section 16 Constructed: 1981 (reconstruction) Limits Considered: Km 38.5 to 38.7 (was Km 32.1 to 32.3) and Km 37.6 to 37.8 (was

Km 31.2 to 31.4) of northbound (eastbound) slow lane Project Description: N2-16 Kwelera, East London Available Documentation: 1. Wright, B.G., et al, “Long Term Performance of emulsion treated base pavements”,

PR88/014, RDAC, CSIR, Pretoria, 1991. 2. Partial as-built records contained in the project file for project PR88/014. 3. SANRAL PMS. 4. Steyn, WJvdM, “Summary of field and laboratory test results on selected ETB

pavements”, CR-98/073, CSIR, Pretoria, 1997. 5. Theyse, H.L., “Towards Guidelines for the Structural Design of Pavements with Emulsion-

treated Layers”, CR-97/045, CSIR, Pretoria, 1998. 6. Discussions and email communications with Mr Mynhardt Augustyn and Mr Derek Burger

of Vela VKE (respectively assistant resident engineer and contract engineer for some of the rehabilitation performed on this road).

7. Discussions with Dr G. Jordaan who was involved in Heavy Vehicle Simulator (HVS) testing on this section around 1988.

8. Jordaan, GJ and CJ Nienaber, “HVS Testing on a Rehabilitated Pavement Containing A Cold Mix Recycled Layer – National Route N2 near Kwelera, East London”, CSIR Report I/FP/5/88, May 1988.

9. Discussions with Mr Renaldo Lorio, SANRAL. 10. Discussions with, and review by, Dr Gerrit Jordaan, Tshepega. 11. As-built records. 12. Madan Singh and Associates, “Reseal: Contract No NRA N.002-160-2003/2 for National

Route 2 Section 16: Gqunubi river (Km 27.40) to Kei Mouth/Blue Water I/S (Km 50.20)”, April 2004.

13. Review by Mr Nicol van der Walt, previously National Roads Project Engineer.

Area and Climate The road is situated near East London in the Eastern Cape. The region is classified as wet, with an approximate Weinert n-value of 2, and an average annual rainfall of around 928 mm [1]. The kilometre stations on the road were moved by 6.4 km in the direction of King Williams Town in the late 1980s. [9]

Construction History Several maintenance and rehabilitation actions have been performed on Section N2-16. While many of these actions involved emulsion treatment of the base, the exact location of the treatment is not always known. The rehabilitation design was done in-house by engineers of National Roads (NTC at the time) and directly documented for contract which was supervised by VKE. Therefore, no detailed rehabilitation reports are available for this road. The construction history reported here focuses on a section of road for which a relatively certain construction history could be determined. This was done using the available documentation [1], the partial as-built records [2], as-built records [11], the SANRAL PMS [3] and communications with relevant consultants and researchers [6,7]. The construction history for the sections was as follows: • A reconstruction was performed from 1979 to 1981. This action involved the cement and

lime stabilization of the original CTB base, which was severely failed in places and parts of the selected layers/subgrade where necessary, and overlay with a new base using emulsion treatment (140 mm thick). This treatment was performed continuously on all lanes of both directions in the areas between Km 33.272 and 40.76 (was Km 26.872 and 36.360) [11]. The HVS testing of 1988 was located on the eastbound slow lane at Km 37.2 (was Km 30.8) [8].

• According to Nicol van der Walt [13], the section under consideration originally had had a G2 base on a G6 to G8 subbase.

Page 64: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 6 N2 Section 16

• The stabilizer was bulk distributed and the emulsion was applied by a bitumen distributor followed by a water browser to dilute the emulsion. The material was blade mixed with a grader [13].

• During the reconstruction in 1980 to 1981, a 40 mm gap-graded asphalt was placed as a surfacing [2,3,11].

• In 1987, major rehabilitation was performed on the N2-16, but this did not involve the sections noted above [2,3,6];

• A double seal with bitumen rubber was placed on the above-noted sections in 1990 [3,6]. Specific material information extracted from the as-built records are shown in Table 1. Table 1: Summary of As-Built Information for Km 37.6 to 38.7 [11]1 Layer Property Specification Observations from As-Builts

Material 26% 13 mm Quartzite crusher run, 43% 13 mm quartzite, 31% Igoda Sand, 1% lime.

Thickness 40 mm Average = 41 mm. Binder 5.5% (60/70 pen) Average = 5.6%. Voids 4.0 % Average = 4.3%, ranges from 3.6 – 5.2%. Stability Not specified Average = 10.7%.

40 mm Premix

Compaction 97% Average = 98.4%, minimum 97.5%. Material Quartzitic sandstone. Thickness 140 mm Average = 147.5 mm, only one location less than

spec. Grading P0.425 = 11-24%

P0.075 = 5-12% GM not specified

P0.425: Average = 15.7%, all in spec. P0.075: Average = 7.7%, all in spec. GM = 0.48 to 0.59, average = 0.54.

ACV 30 - 35 Average = 27.4, only 1/3 in spec. Bitumen content

1% residual bitumen by mass

No specific values given.

Cement content

1% cement by mass

No specific values given.

PI < 6 Maximum PI = 4.0. Compaction Rel comp > 86

AMD not specified

Relative compaction: Minimum = 86.4%, average = 87.9%. AMD: Minimum = 2464 kg/m3, Average = 2510 kg/m3.

140 mm Emulsion stabilised crushed stone base

UCS Not specified Minimum = 2.6 MPa, Average = 3.4 MPa. Material In situ decomposed dolorite plus basecourse plus

RAP. Thickness 125 mm Average = 132 mm, all in spec. Grading P0.075 < 25% P0.425: Average = 30.8%.

P0.075: Average = 13.8%, all in spec. GM = 0.88 to 1.16, average = 0.88.

Stabiliser content

1.5% lime, 1.5% slagment

No specific values given.

PI < 10 All S.P. Compaction Rel. comp > 95%

Relative compaction: Minimum = 94.3%, average = 96.5%. MDD: Minimum = 2125 kg/m3, Avg. = 2186kg/m3.

OMC Not specified OMC = 7.0 to 9.3, average = 8.0.

125 mm subbase, stabilised with lime / slagment

UCS > 1.5 MPa Average = 4.4 MPa, all in above spec. In situ material stabilised with lime / slagment when necessary

No selected subgrade in section under consideration

Note 1: The kilometres shown are the current kilometre markings, the as-builts were compiled using the kilometres at the time of construction and therefore the original kilometres were used to extract the relevant data.

Test pit examinations performed as part of the 1988 HVS experiment showed that the the emulsion treated base was found to be around 100 mm thick, as opposed to 150 mm. The subbase was found to consist of lime and slagment stabilized material. This material consisted of RAP, base and shoulder material [8]. No sign of stabilization of the selected

Page 65: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 6 N2 Section 16

layers was detected. It should be noted that this test pit investigation was probably performed near the HVS test, which was situated on an area with low deflections (around 230 micron RSD deflection). It is thus likely that the pavement structure in other areas (with higher deflections), is similar to, or weaker than, the structure determined from the 1988 HVS test pit investigation. According to Gerrit Jordaan, the base layer thickness varied greatly, and the variation was corrected with asphalt, therefore the surfacing thickness also varies. [10] Using all of the above information, and relying mostly on the as-built information for the specific sections under investigation, the pavement structure assumed for the purpose of this evaluation is as follows:

• 40 mm Gap-Graded Asphalt • 140 mm Quartzitic sandstone base, stabilized with 1% cement and 1% residual bitumen

by mass • 130 mm decomposed dolerite subbase, stabilized with 1.5% lime and 1.5% slagment • Clayey subgrade A 1997 investigation [4], during which cores were extracted from the base at two locations, inside and outside the wheel path (i.e. to provide four cores in total) showed a residual binder content of 0.5%.

Behaviour and Material Quality Indicators • DCP measurements taken around 1989 [1] show a relatively consistent pattern with a low

penetration rate observed in the ETB (estimated CBR of 100% or greater), while the stabilized subbase could not be penetrated in most of the tests.

• DCP measurements recorded around 1997 [4] show that the emulsion treated base could not be penetrated.

• In a 1997 investigation [4], it was found that intact cores could not be extracted from the emulsion treated base.

• The 1997 investigation [4] showed an average in situ moisture content of 3.9%, with a standard deviation of 0.6%.

• Deflection measurements (confirmed RSD by Andrew Laatz) taken in 1989 [1], show that the average deflection for the two sections was around 750 micron (Km 38.5 to 38.7) and 280 micron (Km 37.6 to 37.8). A small part of the 200 m test section from Km 38.5 to 38.7 showed deflections of more than 1 mm which affected the average deflection significantly. The 95th percentile deflections for the two sections were around 1200 and 670 micron.

• FWD deflections recorded in 1997 [4,5] on the section from Km 37.6 to 37.8 show that the deflection ranges roughly from 50 micron to more than 1 mm. The pattern of deflections observed was near identical to that shown in the 1989 survey, which show a low deflection (less than 300 micron) on the first 100 m of the LTPP section, and a high deflection (generally above 600 micron) over the second 100 m of the LTPP section.

• The deflections suggest that two uniform subsections (roughly 100 m long) are located within this LTPP section from Km 37.6 to 37.8. The first part has low deflections (generally below 300 micron), while the second part has significantly higher deflections (generally above 600 micron). Backcalculations performed as part of the study in 1997 [4,5] show clearly that the stiffness of the base and subbase is lower in the subsection with higher deflections. The drop in stiffness in the weaker subsection seems to be more pronounced for the subbase. Specifically, in the stronger subsection, the subbase appears to be in a stabilized, stiff state (stiffnesses above 500 MPa), whereas in the weaker subsection the subbase stiffnesses are below 200 MPa.

• FWD deflections measured every 200 metres in 1999, from the SANRAL PMS [3] confirm the high deflection (1.2 mm) on the second 100 m of the section at Km 37.6 to 37.8, and the low deflection at the beginning of the section (<150 microns). On the Km 38.5 to 38.7 section, the only deflection measurement is 600 microns, which is higher than the measurements just outside the section. The average maximum deflection is 373 microns on the approximately 2 km around and including these sections.

Page 66: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 6 N2 Section 16

• Similarly, FWD deflections measured in 2005 from the SANRAL PMS [3] show a single deflection of 991 microns from Km 37.6 to 37.8, and 498 and 216 microns on Km 38.5 to 38.7. The average maximum deflection on the 2 km section around these sections is 368 microns.

• Test pits were opened in August 2002, and the results for Km 38.6 on the northbound lane are shown in Table 2. In general, the base was classified as a G2 to G5, with a grading that fits the G2 envelope and was described as “a slightly moist dark brown very dense intact sandy gravel imported dolerite base.” The subbase layer is slightly moist yellow brown very dense intact sandy gravel imported slightly stabilised and unstabilised”, G5-G6 when unstabilised and C3 to C4 when stabilised, with a thickness of 100 to 275 mm. The subgrade is “imported sandy gravel material of intact weathered dolerite with occasional cobbles whilst dense intact silty sand was also found.” The subgrade quality varies from G6 to G10. [12]

Table 2. Results from test pit opened in 2002 at Km 38.6 (northbound) [12]

Depth Grading Modulus

Plasticity Index

MDD OMC Description

52 - 220 2.47 7 Dolerite 220 - 340 2.00 4 Dolerite 340 - 450 1.66 12 2214 8.7 Weathered sandstone 450 - 600 1.92 12 Sandstone with cobbles

600 - 1100 0.54 19 1779 14.5 Weathered dolerite

Performance Indicators and Treatment History • Rut depth measurements were taken around 1989 [1] and show that the 95th percentile

rut for the both sections was below 8 mm, with an average of around 4 mm. • The rut depth measurements of 1989 [1] are relatively uniform over the 200 m test

sections, and show no correlation with the deflections, which exhibit a clearly increasing pattern from the one side to the other side of the section.

• Rutting measurements from 1996 to 2002 from the SANRAL database are shown in Figure 1. The SANRAL measurements are determined every 100 metres, therefore there are only 2 data points per section. The data shown are for 2 km including the sections. The line represents the 5th to 95th percentile. After 22 years of service the rut depths are less than 10 mm, which shows the road is in a good condition.

• The average rut depth measured in a 1997 study [4] was 16.4 mm on the Section from Km 37.6 to 37.8, with a standard deviation of 4 mm. This corresponds to a 95th percentile rut of more than 20 mm.

• Rut depths measured (with a 3 metre straightedge) in April 2004 showed degree 5 rutting over the sections under consideration. The average and 90th percentile rut depths on the northbound lane are 9.5 mm and 11.5 mm (Km 35.2 to 37.6) and 21.6 mm and 57 mm (Km 37.8 to 41.2). It is not clear whether the high rutting values are actually on the sections highlighted in this summary. [12]

• Riding quality measurements taken around 1989 [1] show that the road was in a sound to warning condition.

• Riding quality measurements from 1999 to 2002 from the SANRAL database are shown in Figure 2. The SANRAL measurements are determined every 100 metres, therefore there are only 2 data points per section. The data shown are for 2 km including the sections. The line represents the 5th to 95th percentile. After 22 years in service the IRI are less than 3 (except for 2000), indicating the pavement is still in a good condition.

• HRI measured in 1999 is generally below 3 [12]. • Visual assessments performed in 1989 [1] show that the road is in a sound condition with

regards to deformation, but in a warning condition with respect to cracking. Observed cracks consisted mainly of transverse and longitudinal cracks, and some pumping was also noted (it is believed that the section was sealed shortly after this assessment was made [1,6]).

• Visual surveys from the SANRAL PMS in 1995, 1996 and 1997 show the pavement was in a good condition after 17 years in service, except for degree 2 bleeding and flushing over the whole section.

Page 67: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 6 N2 Section 16

• Visual assessments carried out in April 2004 showed degree 3 extent 5 bleeding/flushing over the whole rehabilitation section, some minor cracking and pumping on the specific sections of focus in this summary. A patch was observed at Km 37.6, but it is not clear whether this is on the northbound or southbound directions. It was also observed that gravel shoulders have eroded thus making shoulder levels lower and leaving the pavement prone to edge breaks. [12]

• The road was sealed in 2005 [9]. To apply the seal, a tack was placed first, then 19 mm stones, then 6.7 mm stones on top of the 19 mm stones (without bitumen addition), then a modified binder was applied followed by another application of 6.7 mm stones.

0

5

10

15

20

1995 1996 1997 1998 1999 2000 2001 2002Year

Rut

dep

th in

left

lane

(mm

)

Average, Km 37 to 39Average, Km 38.5 to 38.7Average, Km 37.6 to 37.8

0.0

1.0

2.0

3.0

4.0

5.0

1999 2000 2001 2002Year

IRI i

n le

ft la

ne

Average, Km 37 to 39Average, Km 38.5 to 38.7Average, Km 37.6 to 37.8

Figure 1. Rut depths from SANRAL PMS Figure 2. IRI from SANRAL PMS

Traffic Loading Indicators The 1989 study [1] of this section’s history showed that in 1980 the number of equivalent vehicle units was 2114 and the E80s per lane per day were approximately 150. In 1985 the evus were 2758 (assumed to be for both directions). This implies a growth rate of 5.5% [1]. In the 1989 study [1], a calculation assumed a 50/50 directional split, with an E80 per heavy factor of 1.7, and a growth in traffic and heavy vehicles of 6%. Using these assumptions, the E80s per lane per day for 1985 was estimated as 782. The cumulative traffic between 1982 (roughly the date of construction) and 2002 (a 20 year design period), was 8.9 million E80’s. Table 3 shows the traffic data from the SANRAL PMS and the SANRAL CTO books as the counted number of E80’s per day per worst lane in the Easterly direction N2-16. These values, and the counted E80s per day from [1], are significantly below the calculated values from [1]. Because the SANRAL data are collected over several years, and are more recent, the high values from [1] are ignored. Using the data shown in the figure in Table 3, the growth rate during the years measured by SANRAL is calculated as 5.3%. The initial traffic was extrapolated back using growth rates of 4 and 8%. The cumulative traffic is estimated between 2.4 and 3.3 million E80s to 2005 (25 years of service), which is the area below the dashed and solid lines, respectively.

Page 68: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 6 of 6 N2 Section 16

Table 3. Traffic from SANRAL PMS and CTO books (E80s per worst lane)

Year SANRAL PMS

CTO Books

1988 203 1989 276 1990 295 1991 533 1993 265 1996 324.3 1997 330.1 1999 391.4 2000 352.8 2001 367.8 2002 450.9 449 2003 441.9 2004 442.5 485

0

100

200

300

400

500

1980 1985 1990 1995 2000 2005Year

E80s

per

lane

per

day

SANRALHistorical [1]Estimated initial trafficCTO data

Factual Observations • A double seal was placed in 1990, after 10 years of service. In 2005 (25 years of

service), another seal was placed. • After 22 years of service the rutting is less than 10 mm. • After 24 years of service the rutting is close to or greater than 10 mm, with high values

(average exceeding 20 mm) experienced on Section 38.5 to 38.7. These values were however measured with a straightedge.

• After 22 years of service the riding quality IRI is less than 3.

Observations Based on Estimates or Interpretations • The road has carried an estimated 2.4 to 3.3 million E80s in the slow lane after 25 years

of service. • The visuals and DCP data indicate the road is in a good condition in 1997, after 17 years

of service. • The visuals conducted in 2004, 24 years of service, show some problems on the road, but

no significant failures. • The FWD data show the sections are not uniform, and on Km 37.6 to 37.8 high

deflections are apparent. • Due to the placement of the seal in 2005, the current (2006, 26 years of service) condition

of the road is good.

Anecdotal Information from Reviews • Gerrit Jordaan recommends ETB layers are covered with a modified binder asphalt layer,

not just a seal. A seal is too susceptible to punching and pumping. • PW de Bruin recommends the use of lime rather than cement, it seems to result in fewer

problems. • The traffic estimates seem low for this section. • The addition of the bitumen and cement was to enhance constructability and

waterproofing of the base. The additional strength obtained is a bonus and shows what tremendous benefits are obtained with small percentages of bitumen and cement [13].

Reviewed by: • Gerrit Jordaan, Tshepega • Nicol van der Walt, Arcus Gibb (previously National Roads)

Page 69: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N2 Section 20

Road Type: ETB Road Code: N2 Section 20 Constructed: 1999/2000 (rehabilitation) Limits Considered: Km 31.7 to 37.1 of east- and westbound lanes Project Description: N2-20 Tabankulu to Mzintlava River Available Documentation: 1. Hawkins Hawkins and Osborn, and Fongoqa Skade Toyi and Associates, “Repairs and

Overlay of National Route 2: Section 20, Between Tabankulu (Km 30.0) and Mzintlava River (Km 39.0), and Safely Improvements to Short Sections between Km 24.0 and Km 61.0”, Detail Assessment and Design Report, June 1999.

2. Hawkins Hawkins and Osborn, and Fongoqa Skade Toyi and Associates, “Repairs and Overlay of National Route 2: Section 20, Between Tabankulu (Km 30.0) and Mzintlava River (Km 39.0), and Safely Improvements to Short Sections between Km 24.0 and Km 61.0”, As Built Material Test Results, October 1999 to April 2000.

3. SANRAL PMS. 4. Discussion with Trevor Morgan, Kwezi-V3, Route Manager. 5. CTO Traffic Data. 6. Review by Andrew Laatz, HHO Africa.

Area and Climate The road is situated between Mount Frere and Mount Ayliff in the Eastern Cape. The specific section is between Tabankulu and the Mzintlava River. The region is in hilly terrain and experiences extreme cold and wet weather. With an average annual rainfall of around 710 mm (measured in nearby Kokstad) [1]. According to TRH4, the Weinert N-value is approximately 2, which is the boundary between the wet and moderate regions.

Construction History • The original pavement was constructed in 1960 and had a 6.1 metre cross-section. • The road was widened in 1979 to a surfacing width of 9.8 metres. The widening

essentially created a paved shoulder and only 0.6 metres of the widening is included in the travelling lane. An assessment in 1987 showed an extensive longitudinal crack (degree 3) over most of the section length where the widening had occurred. Some crocodile cracking was observed progressing from the longitudinal crack. [1]

• In 1988, due to severe bleeding and loss of skid resistance, the section was resealed with either a 13 mm or 19 mm seal.

• In 1993, due to continued bleeding and loss of skid resistance, Km 31.7 to Km 37.1 was resealed with a 19 mm and 9.5 mm double seal.

• The pavement structure before rehabilitation was as follows [1,2]: o 19 and 9.5 mm double reseal (1993) o 19 mm reseal (1988) o 50 mm asphalt surfacing (1979) o 50 mm asphalt levelling course (1979) o 200 mm weakly stabilised sandstone or dolerite base (1960) o 150 mm weakly stabilized highly weathered dolerite subbase (1960) o Selected materials o In situ subgrade of weathered shale

• In the rehabilitation, the existing asphalt surfacing and seal layers were deep in situ recycled with part of the existing granular base layer. Prior to the rehabilitation, minimal base layer failures were evident. The rehabilitated pavement is as follows [1,2]:

o 35 mm asphalt surfacing with rolled in chips o 180 mm cold in-situ recycled cement and emulsion treated base (1.75%

cement, 2.5% emulsion (1.5% residual bitumen)) o 120 mm of original base (G4 in equivalent granular state) o 150 mm original subbase (C4 in equivalent granular state) o 150 mm original selected subgrade (G7) o Original subgrade (G8 to G10)

Page 70: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N2 Section 20

Specific material information extracted from the detail assessment and the as-built records are shown in Table 1. In the detail assessment, data are extracted from 4 trial pits, two of which are just outside the section limits. However, the base and subbase construction was the same for these two trial pits therefore they are included to increase the reliability of the data. The trial pits and the rehabilitation, did not extend into the selected layer or subgrade and therefore no material properties for these underlying layers are available. [1, 2]

Behaviour and Material Quality Indicators • Before the rehabilitation, Deflectograph deflections measured in 1990 indicated the

pavement was very flexible (TRH12 criteria). The 25th to 75th percentile deflections were 0.63 to 0.86 mm. [1]

• Before the rehabilitation, DCPs measured at trial pit locations just outside the section showed that the pavement support is generally good and although the previously stabilised materials were observed to be in an equivalent granular state, the layers were difficult to penetrate. [1]

• FWD measurements from the SANRAL PMS are shown in Figure 1 for 1999, 2001 and 2005, where the lines indicate the 5th to 95th percentile and the marker the average. The data were measured on the eastbound lane only. The 1999 data may have been measured prior to the rehabilitation. The 2001 deflections are surprisingly high for a new pavement, however the pavement temperature at the time of the measurements was higher than for the measurements in other years. In 2005, 95 per cent of the deflections are less than 500 microns, which is reasonable for a pavement such as this one. [3]

0

100

200

300

400

500

600

700

800

900

1999 2001 2002 2003 2004 2005Year

Peak

def

lect

ion

(mic

rons

)

Figure 1. FWD measurements [3]

Page 71: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N2 Section 20

Table 1: Summary of As-Built Information for ~Km 31.7 to 37.1 [1] Layer Property Specification Observations from As-Builts

Material 13.2 Medium Continuously Graded Asphalt Surfacing – Roadmac surfacing and precoated chips

Thickness 35 mm Average = 38 mm, Range 30 to 49.5 mm, 22% out of spec.

Precoated chips Spread rate average = 3.21 kg/m2, Range 2.9 to 4.4 kg/m2 Sieve (mm) 19 4.75 0.425 0.075 Average 100 65 24 6 Minimum 100 55 18 5 Maximum 100 75 29 6

Grading

Standard deviation in segments is 0 to 12%. Binder content 4.9% Average = 4.4%, Range 1.3 to 4.9%

Standard deviation in segments is 0 to 0.74% Almost all results less than spec

Penetration1 60/70 Average = 65.5 kN, Range 56 to 77 kN 50 % of measurements out of spec.

Marshall stability

Average = 13.4 kN, Range 11.7 to 15.7 kN Standard deviation in segments is 0 to 1.48 kN

Marshall flow Average = 2.32 mm, Range 2 to 2.6 mm Standard deviation in segments is 0 to 0.6 mm

Tack application

0.6 l/m2 (no standard deviation given, suspect not measured value

Asphalt surfacing

Compaction 93% of TMRD Average = 93.6%, Range 91.4 to 96.6% Standard deviation in segments is 0 to 1.68% 22% out of spec.

Material Stabilised weathered sandstone / crushed sandstone / unstabilised or stabilised decomposed dolerite

Thickness 90 to 205 mm Grading Two trial pits meet G1-G3, two are finer than G4

grading envelope. Grading modulus = 1.88 to 2.32

PI 6 to 7 CBR @ 98% Mod Average = 88%, Range = 73 to 101%

Base, prior to rehabilitation

TRH14 class G4 to G5 Material Asphalt, sandstone and dolerite Emulsion 2.5% Bitumen emulsion Cement 1.75 % Cem I Thickness 180 mm Average = 183 mm, Range = 178 to 187.

Standard deviation in segments is 1.2 to 26 mm Sieve (mm) 19 4.75 0.425 0.075 Average 93 61 27 5 Minimum 90 53 21 4 Maximum 98 68 30 6.5

Grading

Standard deviation in segments is 0 to 19%. UCS @ 100% Mod Average = 1.6 MPa, Range = 1.4 to 1.8 MPa Reference density

Average = 1967 kg/m3, Range = 1825 to 2050 kg/m3

Reference OMC

Average = 8.8%, Range = 6.4 to 11.2 %

Base, after rehabilitation

Compaction 102 % Average = 102.4%, Range = 101.6 to 103.2% Standard deviation in segments is 0.5 to 1.8 % 11% below specification

Material Unstabilised or stabilised decomposed dolerite / mudstone gravel

Thickness 90 to 295 mm Grading Finer than G4 PI 3 to 5 CBR @ 93% Mod Average = 37%, Range = 35 to 38%

Subbase, prior to rehabilitation

TRH 14 class G5-G6 Note 1: Calculated on all bitumen samples, not just from those used on section.

Page 72: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N2 Section 20

Performance Indicators and Treatment History • Before the rehabilitation, the visual condition of the section shows severe bleeding and

loss of seal over the full length and width. [1] • Before the rehabilitation, rutting averaged 4.9 mm in both 1990 and 1999 (measured with

the deflectograph). [1] • Before the rehabilitation, a visual assessment in 1997 indicated only the binder condition

and surface failure were greater than degree 3 or extent 3. There was no cracking distress recorded. [3]

• There have been no treatments on the road since the rehabilitation. [3,4] • Rutting data from the SANRAL PMS are shown in Figure 2, where the lines indicate the

5th to 95th percentile and the marker the average. The data were measured on the eastbound lane only. The data show a reduction between 1999 and 2000, which is likely to indicate that the 1999 data were measured before the rehabilitation. Since the rehabilitation, the rutting has remained below 6 mm, which is reasonable for a relatively new pavement. [3]

• Roughness (IRI) data from the SANRAL PMS are shown in Figure 3, where the lines indicate the 5th to 95th percentile and the marker the average. The data were measured on the eastbound lane only. As with the rutting data, a reduction in the roughness is observed between 1999 and 2000, which is likely to indicate that the 1999 data were measured before the rehabilitation. Since the rehabilitation, the roughness has remained below 3, which is reasonable for a relatively new pavement. [3]

• Informal visual assessments by Andrew Laatz in 2002 indicated no distress, whereas in 2005 a noticeable amount of degree 2 cracking (longitudinal and crocodile) was observed, mainly in the West bound outer wheel tracks over a section of approximately 1 km [6].

• The current condition of the road (2007) is reasonable with no clearly observable rutting or cracking. [4]

0

1

2

3

4

5

6

7

8

9

10

11

12

1999 2000 2001 2002Year

Rut

ting

(Lef

t whe

el p

ath,

mm

)

0

1

2

3

4

5

1999 2000 2001 2002Year

Rou

ghne

ss (L

eft w

heel

pat

h IR

I)

Figure 2. Rut depths from SANRAL PMS Figure 3. IRI from SANRAL PMS

Traffic Loading Indicators • Traffic data was reported in the Detail Assessment report from three different sources,

and is given in Table 2. • Traffic data was also obtained from the SANRAL Traffic Count Yearbooks, and is also

given in Table 2. • The figure in Table 2 graphically illustrates the traffic growth. The traffic carried from the

time of rehabilitation to the beginning of 2007 is estimated as the shaded areas under the two lines. The estimated traffic carried is 1.1 to 2.4 million E80s in each direction.

Page 73: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N2 Section 20

Table 2. Traffic data, Average E80s per Day (Northbound/Southbound) Year Assessment

report CTO

1990 358 1991 507/1200a 1994 285/321b 1997 599/554c 1998 599/5541 2000 672/6251

2001 602/6021 540/4802

2002 615/5681

2003 622/6181

568/5192

2004 977/14011

830/9283

2005 653/5891 472/4923

0

300

600

900

1200

1500

1989 1991 1993 1995 1997 1999 2001 2003 2005 2007Year

E80s

per

lane

per

day

Northbound [1]Southbound [1]Northbound [5]Southbound [5]

a. HHO count b. CSIR count c. MIKROS count

1. CTO Station 409, Kokstad South, N2-21 Km 3.4 2. CTO Station 767, Mount Frere, N2-19 Km 90.0 3. CTO Station 3020, Mount Frere, N2-19 Km 89.4

Factual Observations • Since the 1999/2000 rehabilitation, no treatments have been applied to the road in the 6

years of service. • In the 6 years of service, rutting has remained below 6 mm, indicating good performance. • Similarly, in the 6 years of service, roughness has remained below an IRI of 3, indicating

good performance.

Observations Based on Estimates or Interpretations • The road has carried an estimated 1.1 to 2.4 million E80s in each direction after 6 years

of service. • In 2005 (5 years of service), 95 per cent of the deflections were less than 500 microns,

which is reasonable. • The current condition of the road is reasonable (6 years of service). Reviewed by: • Andrew Laatz, HHO Africa

Page 74: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 N3 Section 4

Road Type: ETB Road Code: N3-4 near Mooi River Constructed: 1988 Limits Considered: Km 38.7 to 38.9, Eastbound Slow Lane (in Durban direction, also

referred to as Southbound) Available Documentation: 1. Wright, B.G., et al, “Long term performance of emulsion treated base pavements”,

PR88/014, RDAC, CSIR, Pretoria, 1991. 2. Steyn, WJvdM, “Summary of field and laboratory test results on selected ETB

pavements”, CR-98/073, CSIR, Pretoria, 1997. 3. Theyse, H.L., “Towards guidelines for the structural design of pavements with emulsion-

treated Layers”, CR-97/045, CSIR, Pretoria, 1998. 4. Construction history and traffic data from the SANRAL PMS (from 1997). 5. Construction history, traffic, behaviour and performance data from SANRAL PMS (from

2005). 6. N3TC Data. 7. Discussions with Mike Bouwmeester, WSP Civil and Structural Engineers. 8. Discussions with Mike Hughes, previously WSP Civil and Structural Engineers. 9. Review by Brian Spottiswoode, previously Keeve Steyn.

Area and Climate The road is situated near Mooi River and close to the Nottingham Road intersection in Kwazulu Natal province. The road is located in a Wet region with a Weinert-n value of approximately 2 and an annual rainfall of around 870 m. The section of interest is located on a long straight section of road in a flat area on a shallow fill [1].

Construction History Based on information contained in the SANRAL PMS in 1997, the section was originally constructed in 1963, although 1966 is reflected in the 2005 PMS. Mike Hughes confirms the contruction took place in 1966 [8]. Rehabilitation of this road was conducted during 1988, and consisted of recycling of the existing surfacings, and cement stabilized base and subbase layers, in which these layers were milled and treated with bitumen emulsion [1,9]. The pavement structure after the rehabilitation in 1988 was as follows [1]:

• 40 mm semi-gap graded asphalt; • 200 mm emulsion treated recycled base layer; • 150 mm cemented crushed stone, • In-situ subgrade. Reference [1] states that the binder content of the ETB is 4% by mass, but the source data for this estimate is not known. Brian Spottiswoode stated that no more than 2% by mass of emulsion was used, i.e., 1.2% residual bitumen, and no additional cement was added to the existing C2 layer during the rehabilitation [9]. Cores extracted in 1997 [2] on the northbound carriageway showed a binder content of only 0.1%. This latter binder content estimate seems unlikely, since this section was one of the two (out of a total of eight ETB sections) for which an intact core could be extracted. The apparent binder content of 0.1% was the lowest of all the emulsion treated LTPP sections investigated at this time. However, photographs of the cores also show a dark colouring and a texture similar to asphalt treated material. The information extracted from the SANRAL PMS in 2005 does not show an ETB layer, and instead shows a 200 mm G2 layer. However, based on the other information reviewed, evidence suggests that this information is incorrect and the layer is an ETB. Ron Harmse, an RE on these sections, quoted a 3.5% binder content. It was however, not absolutely clear whether the section under consideration is the ETB section or whether it has been subjected to more recent construction. Mr Harmse confirmed that construction of the northbound and southbound sections were the same. No maintenance or rehabilitation was reported after the 1988 rehabilitation.

Page 75: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 N3 Section 4

Behaviour and Material Quality Indicators • A 1990 report of the condition of the section [1] reports Deflectograph deflections with a

mean of 166 micron and a 95th percentile of 238 micron; • Deflectograph deflections recorded in September, 1988 showed an average deflection of

190 micron [1]; • FWD deflections measured in 1997 on the north-bound carriageway♣ of the section under

consideration showed a maximum deflection with an average of around 150 mm, with all deflections being well below 200 micron.

• DCP measurements recorded around 1990 [1] showed a very low penetration rate (less than 1 mm/blow) in the emulsion treated base, while the subbase could not be penetrated with the DCP;

• Cores extracted from the emulsion treated base on the north bound carriageway in 1997 showed that intact cores could be extracted and tested at an average UCS of 3.1 MPa, with a standard deviation of 1.0 MPa [2];

• DCP tests conducted on the north bound carriageway at the section under consideration in 1997 showed an average penetration rate of 2.6 mm per blow;

• Moisture tests of extracted material on the north bound carriageway at the section under consideration in 1997 showed an average moisture content of 5.0 per cent;

• The SANRAL PMS [5] contains no FWD data for this section. • The deflection measured in 2004 at one place within the 200 metre section was 142

microns on the southbound lane, which is in agreement with the deflections on the 400 metres on either side of the section. [6]

Performance Indicators • Riding quality measurements recorded around 1990 [1] showed a pavement in a sound

riding quality condition. • The roughness in the left (slow lane) measured from 1999 to 2002 [5] is consistently less

than an IRI of 2.5 in the section of interest and for 400 metres on either side of the section. The roughness in 2002 is, however, higher than the preceding years. The data confirm that the road was in a good condition in 2002. [5]

• The roughness (left wheel path IRI) measured with a high-speed device averaged 1.88 in 2004 and 1.76 in 2005. The 95th percentile IRI was 2.63 and 2.45 in 2004 and 2005, respectively. [6]

• A 1990 report of the condition of the section [1] reports measured rut depth data with a mean of roughly 1 mm and a 95th percentile of roughly 3 mm.

• Rut depths measured in 1997 on the north-bound carriageway of the section under consideration showed an average rut of roughly 3 mm with a standard deviation of 1.3 mm.

• Rut depths in the section, and in the 400 metres on either side of the section are below 7 mm (1995, 1996 and 1999 to 2002). The rut depth increases with each year. The actual rut depths in the section increases from 3 mm in 1999 to 5.1 and 5.6 mm in 2002 [5].

• The rutting in the left wheel path measured with a high-speed device averaged 5 mm with a 95th percentile of 7.8 mm in 2004 and in 2005 averaged 9.0 mm with a 95th percentile of 12.8 mm. The rutting is therefore increasing. [6]

• The visual condition recorded in 1995 to 1997 show the pavement was in a good condition with the binder condition being degree 2 extent 5 in 1997 and surface failure degree 3 and extent 2. No other visual parameters are significant. [5]

♣ The earliest reports of measurements on this LTPP section [1] clearly show that the LTPP section is situated on the Southbound lanes, while the HVS test is on the Northbound lanes. Data recorded in the earlier measurements (around 1990) are explicitly shown as being recorded on the southbound lane [1]. The 1997 report by Steyn [2] does not state which lane the section is located, but laboratory result sheets all state the data is recorded on the northbound carriageway. The 1997 report by Theyse does not state in which lane measurements were taken, but since this report constantly refers to data collected in the study by Steyn [2], it can be assumed that the data was also collected in the north bound lane. Strictly speaking, the 1997 data can thus not be directly compared to the data measured in 1989 (as was done in the study by Theyse [3]), but the 1997 information is stated here for completeness. The two carriageways had the same construction history, therefore it can be assumed that the behaviour and performance history of the southbound lane (i.e. the lane of interest) would be roughly similar to or better than that of the northbound lane, since both sections are situated on the same subgrade and the southbound lane carries slightly less traffic.

Page 76: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 N3 Section 4

• The current condition of the section is good, with minimal rutting. There is some bleeding, which is from a seal defect. [7]

Traffic Loading Indicators Traffic data recorded in 1989 showed the following parameters:

• Total daily traffic (assumed to be both directions, all four lanes) was 8040 vehicles per day.

• The vehicle split was: 82% light vehicles; 6.7% medium heavy vehicles (2,3 and 4 axles); and 11.3 per cent heavy vehicles (5 or more axles).

• Design factors estimated at this time were: Expected growth in traffic and E80’s of 4.5%; 40% of traffic estimated to travel in the eastbound direction (direction of interest); 1.7 E80’s per vehicle assumed for medium heavy vehicles; 2.6 E80s per vehicle assumed for heavy vehicles.

Estimates of design traffic reported in 1990 [1] showed the following:

• E80’s per day estimated for the slow lane in 1989 were 1159. • Total cumulated traffic estimated for the 20 year period between 1988 and 2008 was

roughly 23 million E80’s. Traffic Data Recorded in 1993 [2] showed an AADT of 10 743 with 17% heavy vehicles. Using the 1989 data as base year information, this suggests roughly a 7 per cent growth in the AADT, and a 10 to 12 per cent growth in the percentage of heavy vehicles. The SANRAL PMS records for 1997 [4] show an AADT of 12 136 and an E80 per day of 1902, as measured in 1996. It is not stated whether this information is per direction and/or lane, but a comparison with earlier data suggest the AADT applies to all lanes for all directions, while the E80 per day applies to a design lane. The SANRAL PMS records for 2005 [5] show the traffic levels indicated in Table 1. These data are inconsistent, as illustrated in the figure. Using the data shown, which incorporates the data mentioned above, the initial traffic in 1988 was estimated using growth rates of 4 and 10%. The cumulative traffic to the end of 2005 (current state) and for a 20 year design life were calculated as the areas under the solid and dashed lines:

- 2005 (17 years of service): 9.2 to 21 million E80s. - 2008 (20 year design life): 12 to 21.9 million E80s.

Table 1. SANRAL PMS traffic records from 2005

Year E80s per worst lane per day

1995 1488.9

1996 2445.3

1997 1898.7

1998 1806.9

1999 863.3

2000 7375.8

2002 1361.2

2003 2639.9

2004 5869.7

0

1000

2000

3000

4000

5000

6000

7000

8000

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008Year

E80s

per

wor

st la

ne p

er d

ay

South/east boundHistoricalEstimated

Page 77: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 N3 Section 4

Factual Observations • Since the rehabilitation in which the ETB base layer was constructed in 1988, the

pavement has been in service for 17 years without a need for structural strengthening or rehabilitation.

• The rut depths in the section are 1 mm (average) in 1990, 3 mm in 1999 and 5.1 and 5.6 mm in 2002, 9 mm (average) in 2005 (17 years of service).

• The IRI in the slow lane in 2002 (15 years of service) is under 2.5 and in 2005 the 95th percentile is 2.5 (17 years of service).

• Deflectograph deflections averaged 190 microns in 1988 and 166 microns in 1990.

Observations Based On Estimates or Interpretations • An estimated 9.2 to 21 million E80s have been carried in 17 years of service. • The 20 design life is 12 to 23 million E80s. • The road is in a reasonable condition, based on: IRI (2005), rutting (2005), visual

condition (1997) and DCP (1990), and observations by Mike Bouwmeester. Reviewed by: • Mike Hughes (previously WSP) • Brian Spottiswoode (Project Engineer) • Hugh Thompson (WSP)

Page 78: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 6 N3 Section 12, Northbound

Road Type: ETB Road Code: N3 Section 12 Constructed: 1986/7 Limits: Northbound Km 34.45 to Km 36.0 Project Description: N3 Section 12 between Modderfontein and Buccleugh Available Documentation: 1. Discussion with Brian Spottiswoode, previously Keeve Steyn. 2. As-built drawing of pavement structures. 3. Keeve Steyn Incorporated and Goba Moahloli and Associates Inc., “National Route 3,

Section 12: Modderfontein Interchange to Buccleugh Interchange and National Route 1, Section 20: Rivonia to Buccleugh Interchange: Construction of an Asphalt Overlay and Repairs, Detailed Assessment Report”, October 1998.

4. Discussion with Pablo Balmaceda, Africon, previously Keeve Steyn. 5. SANRAL PMS. 6. CTO traffic records.

Area and Climate The road is situated in the Gauteng province and is part of the Johannesburg Ring Road. The specific section starts just north of the Modderfontein interchange and ends just south of the London Road interchange. This area has a summer rainfall and is classified as a moderate area which TRH4 classifies with a Weinert N value of 2 to 5. The average rainfall was 713 mm per year (1961 to 1990), with most of it occurring in the summer months. [3].

Construction History The original pavement was opened to traffic in 1972 and was rehabilitated in 1986 to 1987 [3]. As part of the rehabilitation, the road was widened from 2 to 3 lanes in each direction, with the extra lane created in the median. In the slow lane the existing asphalt and cement treated crushed stone base was recycled into an ETB. Thereafter the entire carriageway was overlaid with an ETB base followed by a semi-gap graded asphalt overlay. According to a drawing from the as-built records from the 1986/7 rehabilitation, the pavement structures are as follows [2]: Northbound Km 34.45 to 36.0, Slow lane • 40 mm semi-gap asphalt with 10 mm precoated chips; • 100 mm crushed stone treated with bitumen emulsion (BT1); • 100 mm recycled asphalt and crushed stone treated with bitumen emulsion (BT1); • 102 mm cement treated base, and • 102 mm stabilised gravel, and • 152 mm selected subgrade. According to Brian Spottiswoode, the precoated chips were 13 mm and were rolled in. The 100 mm subbase layer should be 150 mm. In the 1998 detailed assessment report, the existing pavement structure from 1986/7 is as follows [3]: Northbound Km 34.45 to 36.0, Slow lane • 40 mm AS; • 100 mm BT1; • 100 mm BT1; • 100 mm C3; • 100 mm C4, and • 150 mm G5 (legibility unclear).

Page 79: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 6 N3 Section 12, Northbound

According to Brian Spottiswoode, the 100 mm subbase layer should be 150 mm and the ETB layers contained 1.5% emulsion (0.9% residual binder) [1]. The SANRAL PMS records only reflect a 1972 pavement structure, and do not show the 1986/7 rehabilitation [5]. As part of the assessment in 1998, test pits were opened across the section and cores and asphalt blocks were obtained. The tests from the test pit materials are the only available source material properties. It must be noted that when the test pits were opened, the road was already 11 years old. Key results from the single pits on this section are summarized in Table 1. Table 1: Summary of Information from Test-pit at Km 35.79, Northbound Slow Lane [3] Layer Property Observations from As-Builts

Description Semi-gap graded with rolled in chips Thickness 40 mm Binder content 6.5% Penetration 20

P19 P9.5 P6.7 P4.75 P2.36 P0.150 P0.075 Grading 100% 72.3% 61.6% 56.7% 50.5% 16.4% 8.1%

Gap-graded asphalt

ITS 1320 kPa Description Emulsion treated and cement stabilised, very dense, sandy

gravel base. Thickness 200 mm MDD 2158 kg/m3 OMC 5 % CBR 90%: 10, 95%: 38, 98%: 79

ETB (2 layers combined)

DCP Penetration = refusal In situ CBR > 400 %

Description Cement stabilised sandy gravel subbase (UCS > 2000 kPa) Thickness 80 mm Approx TRH 14 class C3

Cement treated upper subbase DCP Penetration = refusal

In situ CBR > 400 % Description Slightly clayey, silty, coarse sand, medium dense Thickness 130 mm + 200 mm = 330 mm Approx TRH 14 class G8 Grading P0.075 = 28%

GM = 1.52 Plasticity Index 6 Density MDD = 2104 kg/m3

In situ density = 89.2% OMC 7.2% CBR 90%: 12, 95%: 40, 98%: 60

Lower subbase

DCP Penetration = 7.3 mm/blow In situ CBR = 32.8 %

Description Slightly clayey, slightly gravely, coarse sand (medium dense) Thickness 150 mm Approx TRH 14 class G9 Grading P0.075 = 30%

GM = 1.42 Plasticity Index Slightly plastic Density MDD = 2153 kg/m3

In situ density = 91.4% OMC 5.9% CBR 90%: 5.8, 95%: 32, 98%: 102

Selected layer

DCP Penetration = 8.6 mm/blow In situ CBR = 26.7 %

Description Slightly clayey, silty, coarse sand (loose) Thickness 150 mm Approx TRH 14 class G9

Subgrade

DCP Penetration = 11.5 mm/blow In situ CBR = 18.4 %

Page 80: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 6 N3 Section 12, Northbound

Based on the available information, the pavement structure on this Northbound section is assumed to be: • 40 mm semi-gap graded asphalt; • 200 mm ETB, recycled crushed stone (1.5% emulsion, 0.9% residual, unknown cement

content); • 100 mm C3; • 100 mm C4; • 200 mm G8, and • 300 mm G9.

Behaviour and Material Quality Indicators • In 1997, the FWD deflections (measured every 200 metres) averaged approximately 250

microns, with all values less than 300 except for one of approximately 470 microns [3]. • In 2005 on the slow lane the average FWD deflection was 188 microns, with a 95th

percentile of 279 microns. [5] This shows the road is in an excellent condition.

Performance Indicators and Treatment History • In 1998 the northbound section near the Marlboro Road Interchange (Km 39.0) was

showing some evidence of subsidence and longitudinal and transverse cracking of the asphalt surface. This area has a fairly high fill and runs approximately parallel to the Jukskei River. It was considered that the distress could be “the combined result of the presence of incompetent, clayey alluvial material and by the residual granite having a potential collapsing grain structure at this locality. These problems could have been further aggravated by insufficient compaction of the lower and selected fill layers of the road prism during the original construction”. [3]

• The SANRAL PMS indicates a 30 mm AO surfacing was placed in 1999 [5]. Pablo Balmaceda indicated a 40 mm bitumen rubber semi-open graded (BRASO) overlay was placed in 1999 [4].

• Visual inspections conducted in 1997 indicated little maintenance was carried out on the road since the 1986/7 rehabilitation. [3]

• In 1997, on the first third of the section, transverse, longitudinal and crocodile cracking as well as potholes and pumping were present in a warning condition. On the second third, transverse and longitudinal cracks were present in a fair condition. On the final third, transverse, longitudinal and crocodile cracking as well as potholes and pumping were present in short sections rated fair, warning and severe. [3]

• The visual condition from 1995 to 1997 from the SANRAL PMS shows the binder condition degree ranged from 2 to 4 to extent 5, pumping was degree and extent 2 and the cracking measures had a maximum degree and extent of 3.

• No rut data was collected as part of the 1998 assessment as no rutting was visible on the road. [3]

• Rutting was measured as part of the SANRAL network assessments, and is shown in Figure 1 where the bar represents the 5th to 95th percentiles and the marker the average. The rut depths decrease over time, however the actual rutting is relatively small and shows the road was in a good condition in all years measured. The data from 1996 do not follow the trend and seem to be incorrect.

• Roughness (IRI) was measured as part of the SANRAL network assessments, and is shown in Figure 2 where the bar represents the 5th to 95th percentiles and the marker the average. The roughness remains constant, and very low, from 1999 to 2002.

• The roughness in 1997 had an HRI of between 1 and 1.5. • The 1998 Detailed Assessment Report summarized the performance of the sections as

follows: “The section is structurally in a sound condition, as observed during the visual inspection. All distresses observed are surfacing related due to the ageing of the binder, low penetration grade asphalt binder used during the last repair action, low binder content, or possibly binder absorption by the aggregates (quartzitic). Transverse cracks were observed over the entire slow lane and partially on the other lanes as well as isolated crocodile cracking and pumping…. Pumping was found to be very localized in spite of the surfacing and fatigue cracking, which indicates that the ETB base is still performing

Page 81: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 6 N3 Section 12, Northbound

adequately.” A fog spray application, crack sealing and asphalt overlay with localised surfacing repairs were recommended for this section.

0

1

2

3

4

5

6

7

8

9

10

1995 1996 1997 1998 1999 2000 2001 2002Year

Left

whe

el p

ath

rut (

mm

)

0.0

1.0

2.0

3.0

4.0

5.0

1999 2000 2001 2002Year

Left

whe

el p

ath

IRI

Figure 1. Rut depths from SANRAL PMS Figure 2. IRI from SANRAL PMS

Traffic Loading Indicators • Traffic data provided in the 1998 Detailed Assessment Report is summarized in Table 2

and 3. The slow lane carries approximately 20% of the traffic volume in each direction. The average growth rate was determined as 8.1%. [3]

• The traffic data from the Comprehensive Traffic Observations Yearbooks is shown in Table 4.

• All the traffic data are shown in Figure 4. The minimum and maximum total traffic carried is estimated from the areas underneath the solid and dashed lines, respectively:

o 1986 to 1999: 6.8 to 11.8 million E80s o 1986 to 2006: 9.7 to 17.5 million E80s

Table 2: Historical Traffic Data [3] Date Source AADT

(both directions)

% Heavy Vehicles

E80’s per

heavy

Average daily E80s in worst

lane August 1991 CTO 339 (km 36.5) 69 759 4.2 1.2 1 678 November 1991 CTO 340 (km 40.8) 54 180 5.4 1.2 1 459 1997 NDOT PMS [3] 83 264 - - 2 100

Table 3: 1997 Traffic Counts [3]

Northbound Carriageway Southbound Carriageway Data type Slow lane

Middle lane

Fast lane

Total (all

lanes)

Total (all

lanes)

Fast lane

Middle lane

Slow lane

ADT 11 631 23 055 25 034 57 720 58 215 25 924 23 306 9 698 Average E80s / lane 2 678 771 5 3 454 3 393 2 795 2 596

E80s / heavy / lane 1.22 0.99 0.83 1.16 1.1 0.8 0.84 1.23

Page 82: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 6 N3 Section 12, Northbound

Table 4: CTO Traffic Data [6] Daily E80s in worst lane

Station 339 Station 340 Station 997 Year North South North South North South

E80s/heavy

1990 1679 1274 1.2 1993 1678 1459 1.2/1.6 1998 2677 2603 1595 2288 1.4 1999 1906 1848 1.8 2000 2437 1909 2379 2291 2 2001 2144 1769 2407 2267 2 2002 2447 2417 2 2003 2248 1937 2774 2571 1 2004 2009 1809 2441 2251 2537 2165 1

0

500

1000

1500

2000

2500

3000

3500

1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006Year

E80s

per

wor

st la

ne

CTO: Both lanesCTO: NorthCTO: South

Figure 3: Traffic Data

Factual Observations • The road was in service for 12 years before a 40 mm asphalt overlay was placed. • In 2006, the road has been in service for 19 years (7 years after the overlay). • The average FWD deflection after 10 years of service (1987 – 1997) was approximately

250 microns. The road was therefore in a reasonable condition prior to the overlay. • The average FWD deflection after 18 years of service (6 years after the overlay), the

average FWD deflection was 190 microns. The road is therefore still in a good condition. • After 10 years of service, the visual condition of the section showed some areas in warning

state and a few isolated areas in a severe state. • After both 10 and 15 years of service, rutting on the section was low. • After both 10 and 15 years of service, roughness on the section was good.

Observations Based On Estimates or Interpretations • The road was in a reasonable condition prior to the overlay (12 years of service), based on

the FWD deflections, visual observations, rutting and roughness measurements. • The road is in a good condition in 2005, 18 years of service, based on the FWD

deflections, visual observations, rutting and roughness measurements. • The estimated traffic carried is as follows:

1986 to 1999 (13 years of service): 6.8 to 11.8 million E80s 1986 to 2006 (20 years of service): 9.7 to 17.5 million E80s

Page 83: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 6 of 6 N3 Section 12, Northbound

Anecdotal Information Brian Spottiswoode: During the rehabilitation, the existing asphalt surfacing and G1 base were recycled in the slow lane only, adding about 1.5% emulsion (0.9% residual), then the entire road (and widening) were overlaid with a surface-enriched crushed stone ETB and a 40 mm semi-gap-graded asphalt. This surfacing became very brittle (because of the absorptive quartzitic aggregates that were used) and after several years peeled off in one section in the slow lane outer wheel path. The base however, remained intact (without potholing) for some months before the surfacing was replaced.

Reviewed by: • Tom van Zyl, Goba • Brian Spottiswoode, previously Keeve Steyn • Pablo Balmaceda, previously Keeve Steyn

Page 84: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N3 Section 12, Southbound

Road Type: ETB Road Code: N3 Section 12 Constructed: 1986/7 Limits: Southbound Km 36.3 to Km 40.84 Project Description: N3 Section 12 between Modderfontein and Buccleugh Available Documentation: 1. Discussion with Brian Spottiswoode, previously Keeve Steyn. 2. As-built drawing of pavement structures. 3. Keeve Steyn Incorporated and Goba Moahloli and Associates Inc., “National Route 3,

Section 12: Modderfontein Interchange to Buccleugh Interchange and National Route 1, Section 20: Rivonia to Buccleugh Interchange: Construction of an Asphalt Overlay and Repairs, Detailed Assessment Report”, October 1998.

4. Discussion with Pablo Balmaceda, Africon, previously Keeve Steyn. 5. SANRAL PMS. 6. CTO traffic records.

Area and Climate The road is situated in the Gauteng province and is part of the Johannesburg Ring Road. The specific section starts at the London Road interchange and ends between the Marlboro Road and Buccleugh interchanges. This area has a summer rainfall and is classified as a moderate area which TRH4 classifies with a Weinert N value of 2 to 5. The average rainfall was 713 mm per year (1961 to 1990), with most of it occurring in the summer months. [3].

Construction History The original pavement was opened to traffic in 1972 and was rehabilitated in 1986 to 1987 [3]. As part of the rehabilitation, the road was widened from 2 to 3 lanes in each direction, with the extra lane created in the median. In the slow lane the existing asphalt and cement treated crushed stone base was recycled into an ETB. Thereafter the entire carriageway was overlaid with an ETB base followed by a semi-gap graded asphalt overlay. According to a drawing from the as-built records from the 1986/7 rehabilitation, the pavement structures are as follows [2]: Southbound Km 36.3 to 40.84, Slow lane • 40 mm semi-gap asphalt with 10 mm precoated chips; • 150 mm crushed stone treated with bitumen emulsion; • 207 mm recycled asphalt and CTB treated with bitumen emulsion; • No information on selected layer other than showing presence of layer and possibility of

cement treatment, and • 152 mm selected subgrade. According to Brian Spottiswoode, the precoated chips were 13 mm and were rolled in. A 150 mm cement stabilised gravel subbase layer was included underneath the ETB layer and he has no recollection of any stabilization in the selected layer. In the 1998 detailed assessment report, the existing pavement structure from 1986/7 is as follows [3]: Southbound Km 36.3 to 40.84, Slow lane • 40 mm AS; • 150 mm BT1; • 207 mm BT1 recycled; • 150 C4, and • 150 G7.

Page 85: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N3 Section 12, Southbound

The SANRAL PMS records only reflect a 1972 pavement structure, and do not show the 1986/7 rehabilitation [5]. According to Brian Spottiswoode, the ETB layers contained 1.5% emulsion (0.9% residual binder) [1]. As part of the assessment in 1998, test pits were opened across the section and cores and asphalt blocks were obtained. The tests from the test pit materials are the only available source material properties. It must however be noted that when the test pits were opened, the road was already 11 years old. Key results from the 2 test pits opened and single location of coring and asphalt block extraction on this section are summarized in Table 1. Table 1: Summary of Information from Test-pits at Km 37.4 and Km 40.0 and Cores and Blocks from Km 40.5, Southbound Slow Lane [3] Layer Property Observations from As-Builts

Description Semi-gap graded with rolled in chips Thickness 40 mm Binder content Average = 6.3% (6.1%, 6.2%, 6.6%) Penetration Average = 27 (25, 28, 33)

P19 P9.5 P6.7 P4.75 P2.36 P0.150 P0.075 Grading 100% 68% 58% 54% 48% 15% 7%

Gap-graded asphalt

ITS Average = 1753 kPa (1253, 1190, 2787 kPa) Description Emulsion treated and cement stabilised, dense, gravel. Thickness Average = 370 mm (360, 170+210 mm) Density MDD: Average = 2138 kg/m3 (2152, 2124kg/m3)

In situ: 54.4% (one value reported) OMC Average = 5.6% kg/m3 (5.2, 6%) CBR Average 90%: 14, 95%: 34, 98%: 60

ETB (2 layers combined)

DCP Penetration = refusal and 1.6 mm/blow In situ CBR > 400 %

Description Cement stabilised, sand, very dense, UCS > 2 000 kPa Thickness Average = 160 mm (150, 170 mm) Approx TRH 14 class C3

Cement treated subbase

DCP Penetration = refusal and 1.9 mm/blow In situ CBR > 400 % or 180%

Description Sand, dense to loose Thickness Average = 305 mm (310, 150 + 150 mm) Approx TRH 14 class G7 Grading Average P0.075 = 23% (17, 26, 25%)

GM = 1.6 (1.67, 1.53, 1.61) Plasticity Index SP, 8, 12 Density MDD: Average = 2 045 kg/m3 (2 033, 2 105, 1 997 kg/m3)

In situ: 85.3% (64.2, 96.3, 95.5) OMC Average = 8.0% kg/m3 (7.9, 7.2, 8.9%) CBR Average 90%: 15, 95%: 38, 98%: 61

Selected layer

DCP Average penetration = 5.8 mm/blow (4.3, 3.8, 9.4) Average in situ CBR = 54.4 % (64.3, 75.2, 23.8)

Based on the available information, the pavement structure on this Southbound section is assumed to be: • 40 mm semi-gap graded asphalt; • 360 mm ETB, recycled crushed stone (1.5% emulsion (0.9% residual), unknown cement

content); • 150 mm C4, and • G7 subgrade.

Page 86: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N3 Section 12, Southbound

Behaviour and Material Quality Indicators • In 1997, the FWD deflections (measured every 200 metres) averaged approximately 290

microns, with a 90th percentile of approximately 380 microns. The majority of the higher deflections are on the latter part of the section [3].

• In 2005, the average FWD deflection was 164 microns, with a 95th percentile of 203 microns. [5] This shows the road is in a very good condition.

Performance Indicators and Treatment History • The SANRAL PMS indicates a 30 mm AO surfacing was placed in 1999 [5]. Pablo

Balmaceda indicated a 40 mm bitumen rubber semi-open graded (BRASO) overlay was placed in 1999 [4].

• Visual inspections conducted in 1997 indicated little maintenance was carried out on the road since the 1986/7 rehabilitation. [3]

• In 1997, from Km 36.3 to approximately Km 39.1, the road was in a good condition, with very few short sections highlighted in the visual condition survey. From Km 39.1 to 40.84, transverse and longitudinal cracking was present, mainly in a fair or warning condition with approximately 15% of the lane in a severe condition (in short sections). Some crocodile cracking, potholes and pumping was present in one 200 m section rated as severe.

• The visual condition from 1995 to 1997 from the SANRAL PMS shows the binder condition was degree 3/4, extent 5 in 1996, but degree and extent zero in 1997. Pumping is degree and extent 3 and the cracking measures had a maximum degree and extent of 3 except for a few short sections where the extent was rated 5.

• No rut data was collected as part of the 1998 assessment as no rutting was visible on the road. [3]

• Rutting was measured as part of the SANRAL network assessments, and is shown in Figure 1 where the bar represents the 5th to 95th percentiles and the marker the average. The rut depths decreased over time, however the actual rutting was relatively small and shows the road was in a good condition for all years measured. The data from 1996 do not follow the trend and seem to be incorrect.

• Roughness (IRI) was measured as part of the SANRAL network assessments, and is shown in Figure 2 where the bar represents the 5th to 95th percentiles and the marker the average. The roughness remains constant, and very low, from 1999 to 2002.

• The roughness in 1997 had an HRI below 2.5, with three isolated values between 2.5 and 4.

• The 1998 Detailed Assessment Report summarized the performance of the sections as follows: “Visually, the general condition of the pavement is fair, with isolated sections of the surfacing showing extensive cracking due to the ageing of the binder and the subsequent fatigue cracking caused by the traffic loading over a drier and brittle asphalt surfacing. At some locations, the surfacing cracks and transverse cracks have progressed into crocodile cracking. As no deformation is visible, it is deduced that the crocodile cracking is limited to the surfacing layer. Laboratory tests taken from Test Hole T5 revealed that the ETB layers are very dense and still in a good condition, the cement stabilized subbase is very stiff and the that the layers below the cement stabilised subbase have high moisture content.” Localized repairs over the areas with localised cracking and pumping, and some base repairs, followed by an asphalt overlay over the whole section was recommended.

Page 87: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N3 Section 12, Southbound

0

1

2

3

4

5

6

7

8

9

10

1995 1996 1997 1998 1999 2000 2001 2002Year

Left

whe

el p

ath

rut (

mm

)

0.0

1.0

2.0

3.0

4.0

5.0

1999 2000 2001 2002Year

Left

whe

el p

ath

IRI

Figure 1. Rut depths from SANRAL PMS Figure 2. IRI from SANRAL PMS

Traffic Loading Indicators • Traffic data provided in the 1998 Detailed Assessment Report is summarized in Table 2

and 3. The slow lane carries approximately 20% of the traffic volume in each direction. The average growth rate was determined as 8.1%. [3]

• The traffic data from the Comprehensive Traffic Observations Yearbooks is shown in Table 4.

• All the traffic data are shown in Figure 3. The minimum and maximum total traffic carried is estimated from the areas underneath the solid and dashed lines, respectively:

o 1986 to 1999: 6.8 to 11.8 million E80s o 1986 to 2006: 9.7 to 17.5 million E80s

Table 2: Historical Traffic Data [3] Date Source AADT

(both directions)

% Heavy Vehicles

E80’s per

heavy

Average daily E80s in worst

lane August 1991 CTO 339 (km 36.5) 69 759 4.2 1.2 1 678 November 1991 CTO 340 (km 40.8) 54 180 5.4 1.2 1 459 1997 NDOT PMS [3] 83 264 - - 2 100

Table 3: 1997 Traffic Counts [3]

Northbound Carriageway Southbound Carriageway Data type Slow lane

Middle lane

Fast lane

Total (all

lanes)

Total (all

lanes)

Fast lane

Middle lane

Slow lane

ADT 11 631 23 055 25 034 57 720 58 215 25 924 23 306 9 698 Average E80s / lane 2 678 771 5 3 454 3 393 2 795 2 596

E80s / heavy / lane 1.22 0.99 0.83 1.16 1.1 0.8 0.84 1.23

Table 4: CTO Traffic Data [6]

Daily E80s in worst lane Station 339 Station 340 Station 997 Year

North South North South North South E80s/heavy

1990 1679 1274 1.2 1993 1678 1459 1.2/1.6 1998 2677 2603 1595 2288 1.4 1999 1906 1848 1.8 2000 2437 1909 2379 2291 2 2001 2144 1769 2407 2267 2 2002 2447 2417 2 2003 2248 1937 2774 2571 1 2004 2009 1809 2441 2251 2537 2165 1

Page 88: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N3 Section 12, Southbound

0

500

1000

1500

2000

2500

3000

3500

1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006Year

E80s

per

wor

st la

neCTO: Both lanesCTO: NorthCTO: South

Figure 3: Traffic Data

Factual Observations • The road was in service for 12 years before a 40 mm asphalt overlay was placed. • In 2006, the road has been in service for 19 years (7 years after the overlay). • The average FWD deflection after 10 years of service (1987 – 1997) was approximately

290 microns. The road was therefore in an acceptable condition prior to the overlay. • The average FWD deflection after 18 years of service (6 years after the overlay) was 200

microns. The road was therefore still in a good condition. • After 10 years of service, the visual condition of the section showed some areas in warning

state and a few isolated areas in a severe state. • After both 10 and 15 years of service, rutting on the section was low. • After both 10 and 15 years of service, roughness on the section was good.

Observations Based On Estimates or Interpretations • The road was in a reasonable condition prior to the overlay (12 years of service), based on

the FWD deflections, visual observations, rutting and roughness measurements. • The road is in a good condition in 2005, 18 years of service, based on the FWD

deflections, visual observations, rutting and roughness measurements. • The estimated traffic carried is as follows:

o 1986 to 1999 (13 years of service): 6.8 to 11.8 million E80s o 1986 to 2006 (20 years of service): 9.7 to 17.5 million E80s

Anecdotal Information Brian Spottiswoode: During the rehabilitation, the existing asphalt surfacing and C2 base (208 mm total) were recycled in the slow lane only, adding about 1.5% emulsion (0.9% residual), then the entire road (and widening) was overlaid with a surface-enriched crushed stone ETB and a 40 mm semi-gap-graded asphalt. This surfacing became very brittle (because of the absorptive quartzitic aggregates that were used) and after several years peeled off in one section in the slow lane outer wheel path. The base however, remained intact (without potholing) for some months before the surfacing was replaced.

Reviewed by: • Tom van Zyl, Goba • Brian Spottiswoode, previously Keeve Steyn • Pablo Balmaceda, previously Keeve Steyn

Page 89: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 ` N4 Section 1

Road Type: ETB Road Code: N4 Section 1, Scientia I/C to Pienaars River Constructed: Feb 1999 Limits Considered: Km 19.28 to 25.6, Eastbound and Westbound Slow Lanes Available Documentation: 1. Mackintosh, Bergh and Sturgess, ““The Rehabilitation of National Route 4 Section 1 From

Scientia I/C to Pienaars River”, Design Report Volume 1, Report T853/1, Edenvale, July 1996.

2. Project Document from MBS. 3. As-built materials control sheets and design sheets from SANRAL and MBS. 4. SANRAL PMS 5. Discussion with, and review by, Mike Hughes, previously WSP. 6. Discussion with Nuno Gomez, SANRAL.

Area and Climate The road is situated just east of Pretoria in a region that TRH4 (1996) classifies as moderate, with a Weinert-n value of approximately 2.4 and an annual rainfall of 670 mm [1].

Construction History The design report of 1996 notes that the road had been in service for 27 years, which means the pavement was originally constructed in 1969. For the area under consideration, the as-built records suggest that the pavement structure before the ETB rehabilitation consisted of the following [1]:

• 25 mm AC (1969); 13.2 mm seal with crack sealing [5] (1972), 13.2 mm seal (1981) and 9.5 mm PMB seal (1990);

• 100 mm C1 class cemented base; • 100 mm C1 class cemented upper subbase; • 100 mm C3 class cemented lower subbase, and • 500 mm selected and subgrade layers, G6 to G7 class. Before rehabilitation, the area under consideration showed roughly 300 metres of severe distress, which appears to have consisted of severe cracking and pumping, with rocking of the cemented base and subbase slabs in some places [1]. No significant rutting or undulations were reported, with only 3% of measurements being in excess of 10 mm [1]. In the area under consideration, recycling and emulsion stabilization of the existing base was performed in 1997 [3]. SANRAL shows this rehabilitation to have occurred in 1999 [4], but this is for the next section of the N4 [5]. For this segment, the following material indicators were retrieved for the ETB layer from the as-built records [3]:

• Average layer thickness of the emulsion treated base was 170 mm (no standard deviation is provided);

• The source material is stated as 36% G1 material, imported from Ferro Crushers, with 64% in situ milled C3 material (note the disagreement between the as-builts [3] and design report [1]);

• For the eastbound direction, the treatment is shown as “1% OPC and 1% Nett Bitumen”. For the westbound direction, the treatment is shown as “1% OPC and 1% Emulsion”. The westbound direction should also read 1% bitumen [5].

• The percentages passing the 0.075 mm sieve are between 6 and 9%; • The percentages passing the 0.425 mm sieve are between 18 and 25%; • The percentages passing the 4.75 mm sieve are between 38 and 49%; • The percentages passing the 19 mm sieve are mostly between 77 and 89%; • The PI after treatment is noted as N.P. or S.P; • The reference density is indicated as roughly between 2470 and 2520 kg/m3, and • The design compaction is shown as 88% ARD. The materials control sheets show that

the compaction achieved was roughly between 86 and 90%, with an average of around 88%.

Page 90: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 ` N4 Section 1

The as-built sheets show that the design for the area under consideration included a 25 mm medium continuously graded asphalt, with a 13.2 mm bitumen rubber single seal. The pavement after rehabilitation can thus be assumed to be as follows:

• 30 mm surfacing, consisting of a continuously graded asphalt and a bitumen rubber single seal;

• 170 mm emulsion treated base, using 2% OPC and 1.0% net bitumen; • 130 mm cemented natural gravel subbase, possibly weakened in some areas, and • 500 mm selected layers and upper subgrade. Most likely G6 or G7 quality material.

Subgrade CBR (from laboratory) is generally above 15. The SANRAL PMS does not reflect rehabilitation of the base layer, and only shows the seal and surfacing in 1999 [4].

Behaviour and Material Quality Indicators • FWD deflections measured before the ETB rehabilitation were generally below 400

micron. For the area under consideration, only one 600 m section on the eastbound lane showed deflections above 400 micron;

• DCP tests conducted in the area under consideration before ETB rehabilitation showed that the base and/or upper subbase could generally not be penetrated to a depth of roughly 160 to 370 mm. Based on the DCP results, the lower subbase and upper subgrade layers appeared to be below the specifications [1].

• Test pit data collected in the area under consideration before ETB rehabilitation showed that the base generally did not conform to the specification for G4 material, owing to the coarseness of the material below the 4.75 mm sieve. Two trial pits in the eastbound direction showed that the material in the horizon between 260 to 370 mm (which should have been a cemented subbase), was classified as G7 [1]. The laboratory CBR values of the material in the selected layer and subgrade horizons were above 15%, and on average around 30%.

• FWD deflections measured in 1999 (2 years of service) show an average peak deflection of 240 microns and a 95th percentile of 345 microns on the Eastbound lane, and an average peak deflection of 220 microns and a 95th percentile of 340 microns on the Westbound lane.

• FWD deflections measured in 2005 (8 years of service) show an average peak deflection of 292 microns and a 95th percentile of 436 microns on the Eastbound lane, and an average peak deflection of 326 microns and a 95th percentile of 454 microns on the Westbound lane.

Performance Indicators • The roughness from the SANRAL PMS [4] in Figure 1 shows that for the years measured

the IRI in the left lane is low (less than 2.6), and the road is therefore in sound condition up to 5 years of service. The roughness remains constant from year to year.

• The rutting from the SANRAL PMS [4] in Figure 1 shows that the average rutting in the left lane is approximately 6 mm in 1999, 2000 and 2001 and reduces to approximately 5 mm in 2002. The 95th percentile rut is less than 8.2 mm or less after five years of service.

• No visual condition data are available in the SANRAL PMS after the 1997 rehabilitation. • In July 2006, the road is in a good condition. On one part of the N4-1, which may not be

part of this section, on the inner shoulder stones were coming loose and therefore a light fog spray was applied [6].

Page 91: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 ` N4 Section 1

0

1

2

3

4

5

1998 1999 2000 2001 2002 2003Year

IRI (

left)

Average WestboundAverage Eastbound5th to 95th percentile

0

1

2

3

4

5

6

7

8

9

10

11

12

1997 1998 1999 2000 2001 2002 2003Year

Rut

ting

(left)

(mm

)

Average, WestboundAverage, Eastbound5th to 95th percentile

Figure 1. SANRAL Roughness Data Figure 1. SANRAL Rutting Data

Traffic Loading Indicators • Table 1 summarizes the traffic estimates for the area under consideration, as determined

in 1996 [1]. At this time, the E80 per heavy vehicle was assessed as roughly 1.4. • The 1996 assessment estimated that, for various growth rate scenario’s, the 15 year

design traffic would be between 3 and 12 million E80’s (old E3 class) for the slow lane of both directions, except for an 8% growth rate, in which case the eastbound slow lane would have a design traffic of 12 to 50 million E80s (old E4 traffic class).

• The 1996 assessment estimated that, for various growth rate scenario’s, the 20 year design traffic would be between 12 and 50 million E80’s (old E4 class) for the eastbound slow lane, except for an 2% growth rate, in which case the eastbound slow lane would have a design traffic of 3 to 12 million E80s (old E3 traffic class). For the westbound slow lane, the 20 year design traffic was estimated as being between 3 to 8 million E80’s for all growth rate scenarios.

• The SANRAL PMS [4] traffic data are shown in Table 2. Some of the data were omitted because they were clearly erroneous. The traffic levels were estimated for 2011 (15 year design life) and 2017 (20 year design life) using growth rates of 4 and 8%. These estimates are shown in the figure in Table 2. The cumulative traffic was estimated as the areas under the dashed and solid lines in the figure:

o 1997 to 2005 (8 years): 2.8 to 4.4 million E80s o 1997 to 2011 (15 years): 5.3 to 8.4 million E80s o 1997 to 2017 (20 years): 8.1 to 15 million E80s

Table 1: Traffic Assessment Data from Rehabilitation Investigation of 1996 Eastbound Westbound Section Heavy Vehicles / Day

ADT (vpd)

E80s / slow lane / day*

Heavy Vehicles / Day

ADT (vpd) E80s / slow lane / day

Km 16.5 to 20.5 897 8880 1193 392 7318 521 Km 20.5 to 25.64 348 6215 463 348 6215 463

* Calculated using a lane distribution factor of 0.95 and E80s per heavy factor of 1.4. This factor is likely to have increased since 1996.

Page 92: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 ` N4 Section 1

Table 2: SANRAL PMS Traffic Data E80s / worst lane /

day Year East West

1996 693.4 665.4

1999 841.6 372.9

2000 862.4

2001 1339.8

2002 1176.4

2003 1105.2 1107.8

2004 1104.4

0

500

1000

1500

2000

2500

1996 1998 2000 2002 2004 2006 2008 2010Year

E80s

per

wor

st la

ne p

er d

ay

EastWestRef [1]Estimated

Factual Observations • After 5 years of service, the roughness does not exceed an IRI of 2.6. • After 5 years of service the average rut is 5 mm and the 95th percentile rut is less than

8.2 mm. • FWD deflections measured in 1999 (2 years of service) show an average peak deflection

of 240 microns and a 95th percentile of 345 microns on the Eastbound lane, and an average peak deflection of 220 microns and a 95th percentile of 340 microns on the Westbound lane.

• FWD deflections measured in 2005 (8 years of service) show an average peak deflection of 292 microns and a 95th percentile of 436 microns on the Eastbound lane, and an average peak deflection of 326 microns and a 95th percentile of 454 microns on the Westbound lane.

Observations Based on Estimates or Interpretations • The 1996 assessment estimated that the 15 and 20 year design traffic would be between

3 and 12 million E80’s or 12 to 50 million E80s depending on the growth rates. Using all the traffic data available, and growth rates of 4 and 8%, the cumulative traffic was estimated as follows:

o 1997 to 2005 (8 years): 2.8 to 4.4 million E80s o 1997 to 2011 (15 years): 5.3 to 8.4 million E80s o 1997 to 2017 (20 years): 8.1 to 15 million E80s

• In July 2006, after 9 years of service, the road was in a good condition.

Key Persons: • Hugh Thompson of WSP • Mike Hughes (previously WSP) • Mias Wiese of SANRAL • Anthony Fourie of WSP (RE)

Reviewed by: • Mike Hughes (previously WSP)

Page 93: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 5 N4 Section 5X, Km 20 to 25

Road Type: ETB Road Code: N4 Section 5X Wonderfontein to Crossroads Constructed: 1996 (reconstruction) Limits Considered: Km 20 to 25, Eastbound Project Description: N4 Section 5X Wonderfontein to Crossroads Available Documentation: 1. Mackintosh, Bergh and Sturgess, “Die Rehabilitasie van Nasionale Roete 4 Seksies 5X

and 6X Tussen Wonderfontein en Cross Roads”, Project Document, February 1997. 2. As-built for reconstruction (from WSP and SANRAL). 3. Deflections, traffic data, visual assessment data from TRAC PMS. 4. SANRAL PMS 5. Discussion with Gawie Jordaan, TRAC. 6. Discussion with Mike Hughes, previously WSP.

Area and Climate The road is situated in flat terrain in Mpumalanga between Wonderfontein and Crossroads. This area has a moderate to wet climate, with an average annual rainfall of approximately 880 mm.

Construction History The date of the construction of the original pavement is not known. However, the road was reconstructed in 1996. The reconstruction or rehabilitation work done under this contract covers N4-5X from Km 9 to 56.6, and also includes a portion of N4-6X. From the project specifications [1], it appears that varying rehabilitation methods were used. However, the work done on approximately Km 12 to 27, as well as the pavement performance of this section, differs from the rest of the contract and thus the contract was split into two sections for the purpose of this analysis. The SANRAL PMS [4] does not show an ETB layer, and does not indicate any differences in construction between Km 12 and 27.5. The design information shown in the as-built documents suggest that the bitumen emulsion treatment was done primarily on the shoulder and climbing lanes. However, the as-built materials sheets suggest the emulsion treated base was applied over most parts of the project length. The depth of treatment varied considerably, with the most intensive treatment being applied to climbing lane areas. From the contract document, it appears that depth and type of treatment would be determined on site, depending on the pavement situation. The as-built records support this, and show a varying depth of treatment, especially for the subbase and selected layer. For the purposes of this analysis, a part of the project with a relatively uniform construction throughout was selected, in order to allow meaningful and consistent observations to be made. The section selected for this analysis ranges from Km 20 to 25 of the eastbound lane. For this section, the base layer was stabilized with cement and emulsion throughout, while the subbase was stabilized with cement and emulsion on all parts except between Km 23 and 24.3. A full depth construction up to subgrade level was done on a climbing lane between Km 22 and 23. Key observations to emerge from a scan of the as-built information for this sub-section of the project are summarized in Table 1.

Page 94: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 5 N4 Section 5X, Km 20 to 25

Table 1: Summary of As-Built Information, Km 20 to 25, Eastbound slow lane [2] Layer Property Specification Observations from As-Built Records

Aggregate 13.2 mm 13.2 mm Quartzite 13.2 mm Single Seal Binder Not Shown B8 (80/100 pen) Natref Bitumen

Thickness 25 mm Generally close to or above 25 mm. Some areas have average thickness closer to 30 mm.

25 mm Wearing Course Binder 5.4% B12 (60/70 pen)

Natref Average binder content seems close to 5.2%, and generally varies between 5.0 and 5.7%. Few tests show binder content below 5%.

Thickness Not shown Thickness shown constantly as 150 mm, with no standard deviation indicated. Suspect this is not a measured value.

Material Type G1 or BT1 Most of this section received a BT1 treatment. PI after treatment is constantly 1, with no standard deviation indicated.

Stabilizers Not Shown Generally 15 l/m3 of 60% anionic emulsion added. (See Note 1) There is no indication of cement being added for this sub-section, but it may be an omission in the records, as most other batches included 1% cement.

Grading / Fineness

Not shown Percentage passing the 0.425 sieve is generally below 28 and on average around 25%, which would conform to part of the grading envelope for G1 to G3 material.

Compaction 88% Compaction is on average above 88 and often above 90%. One test point showed a density of 86.3%.

150 mm G1 or BT1

UCS Not shown None reported Thickness Not Shown Thickness shown constantly as 150 mm, with no

standard deviation indicated. Suspect this is not a measured value

Material Type C3 The subbase consisted of in situ material stabilized with 3% Portland Blast Furnace Cement. Material class is shown as C2 or C3.

Grading / Fineness

Not Shown The percentage passing the 0.425 sieve was 30% or lower, which suggests a relatively coarse material as for G1 to G3. The PI after treatment is shown as 4 or 1.

150 mm C2 or C3

UCS Not shown Reported UCS values range from 3180 to 4800 kPa.

Note 1. Mike Hughes states 24 to 36 l/m3 would have been applied. This is not in agreement with the as-builts.

Based on these observations, it appears that the constructed pavement structure consisted of the following: • 13.2 mm single seal • 25 mm continuously graded asphalt with 5.4% binder; • 150 mm emulsion and cement stabilized crushed stone (roughly 0.4% binder and 1%

cement♣); • 150 mm C3 subbase, consisting of in-situ material stabilized with 3 per cent cement, and • Thickness and class of selected layers and subgrade is not known.

Behaviour and Material Quality Indicators • FWD deflections were measured in 2000, 2002 and 2004, as part of the TRAC pavement

management system data collection process. These deflections are summarized in Figure 1, which plots the range of the measured distribution for each year by showing the 15th percentile, median and 85th percentile values. The deflections generally range from 250

♣ The application rate of the 60% anionic emulsion is indicated as 15 l/m3. The average density of the compacted material is roughly 2500 kg/m3. If the relative density of the emulsion is assumed to be 1.0, this translates to roughly 0.6% emulsion by mass. For a 60% emulsion, this means a net binder content of 0.4%.

Page 95: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 5 N4 Section 5X, Km 20 to 25

micron to 450 micron, and show a stable median value of roughly 330 micron over the three years of measurement.

Figure 1. Maximum Deflections (normalized to 566 kPa) measured on N4-5X, Eastbound, Km 20 to 25.

Performance Indicators • Figure 2 summarizes the available rut depth data extracted from the TRAC PMS and the

SANRAL PMS. The TRAC data, which are measured at 10 m intervals, show an apparent decreasing trend in the observed rut. This discrepancy is most likely due to the precision or accuracy of the network level surveillance equipment. Overall, however, the median and 85th percentile rut depths are close to or below 10 mm for both data collection periods. Investigations conducted with different measuring equipment suggest that the rut depth data collected in 2002 and 2004 may be conservative (i.e. showing higher ruts than are actually present). The SANRAL data are recorded every 100 metres and show an overall increase in the rut depths from 1995 to 2002. The median rut depths for 2002 in the SANRAL database and TRAC database are fairly close, however the range of measurements are much higher for the TRAC data, probably because of the increased measurement frequency.

• A visual assessment, conducted in 2004 as part of the TRAC PMS data collection, showed no significant structural distress (i.e. no crocodile or longitudinal cracks, pumping or rutting). Distress reported was generally confined to the surfacing and consisted of ravelling, aggregate loss and dryness. The visual condition index reported for segments within the section under consideration were always greater than 85 (lowest value of 87). The visual assessment therefore suggests a pavement in the a very good condition.

• Roughness measurements recorded in the SANRAL database are illustrated in Figure 3. The roughness remained consistent from 1999 to 2002, and shows the pavement is in a very good condition.

• In June 2006, 10 years after the rehabilitation, this section of the road was in a good condition [5].

Page 96: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 5 N4 Section 5X, Km 20 to 25

0

1

2

3

4

5

6

7

8

9

10

11

12

1999S 2000S 2001S 2002S 2002T 2004TYear

Rut

dep

th (m

m)

Figure 2. Left Wheel Path Rutting measured on N4-5X, Eastbound, Km 20 to 25 (S

data from SANRAL PMS and T from TRAC PMS)

0.0

1.0

2.0

3.0

4.0

5.0

1999 2000 2001 2002Year

IRI

Figure 3. Left Wheel Path IRI measured on N4-5X, Eastbound, Km 20 to 25

Traffic Loading Indicators The project document for the 1996 reconstruction documents traffic data recorded in 1994. For the eastbound, direction, Km 23, the following traffic counts were stated: 5 Day Average: 3041 vehicles per day 7 Day Average: 2872 vehicles per day The TRAC PMS Traffic data consists of monthly counts in 2003 and some of 2004. The AADT is recorded, and using an E80 per heavy factor of 2, the E80s in the slow lane were calculated. The average counts are shown in Table 2 and all the data are illustrated in Table 2. The SANRAL PMS Traffic data are also shown in Table 2. The 2002 data seems incorrect.

Page 97: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 5 N4 Section 5X, Km 20 to 25

Using both set of traffic data, and growth rates of 7 and 12%, the initial traffic was estimated, as shown. The cumulative traffic to December 2005 was calculated as the area under the curves, and gives 2.4 to 5.3 million E80s in the 10 year service life. Table 2. SANRAL and TRAC PMS Traffic Data

E80s per slow lane per day

TRAC PMS* (Station) Year SANRAL PMS

Wonderfontein Farrefontein

2001 1241.5

2002 1943.2

2003 1502.5 891 1152

2004 1774.2 889 1069

2005 1627

2006 1209

0

500

1000

1500

2000

2500

1997 1998 1999 2001 2002 2003 2005 2006Year

E80s

in s

low

lane

SANRAL PMSTRAC PMS (Farrefontein)EstimatedTRAC PMS (Wonderfontein)

* Average for the year is shown calculated from available data. An 85-15% lane distribution was assumed.

Factual Observations • The pavement has been in service for 9 years and has not had any maintenance or

rehabilitation. • The condition after 9 years in service is good to very good. Rut depths measured with

high speed surveillance equipment are generally below 10 mm, the IRI are less than 2.5 and network level visual assessment showed no serious structural distresses.

Observations Based On Estimates of Interpretations • The estimated cumulative traffic to 2005 (9 years of service life) is 1.4 to 5 million E80s. • After 6 to 8 years in service, the maximum FWD deflection (normalized to 566 kPa, or

40 kN loading) is generally below 450 micron, with a median deflection that remains stable at roughly 330 micron over three years of measurement.

• In June 2006, 10 years after the rehabilitation, this section of the road was in a good condition [5].

Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 98: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 6 N4 Section 5X, Km 27.6 to 29.0

Road Type: ETB Road Code: N4 Section 5X Wonderfontein to Crossroads Constructed: 1996 (reconstruction) Limits Considered: Km 27.6 to 29.0, Eastbound Project Description: N4 Section 5X Wonderfontein to Crossroads Available Documentation: 1. Mackintosh, Bergh and Sturgess, “Die Rehabilitasie van Nasionale Roete 4 Seksies 5X

and 6X Tussen Wonderfontein en Cross Roads”, Project Document, Edenvale, February 1997.

2. As-built for reconstruction (from WSP and SANRAL). 3. Deflections, traffic data, visual assessment data from TRAC PMS. 4. SANRAL PMS. 5. Discussion with Gawie Jordaan, TRAC. 6. Discussion with Mike Hughes, previously WSP. 7. SNA Civil and Development Engineers, “Preliminary Design Report, Section 5a,

Wonderfontein to Belfast”, December 2005.

Area and Climate The road is situated in flat terrain in Mpumalanga between Wonderfontein and Crossroads. This area has a moderate to wet climate, with an average annual rainfall of roughly 880 mm.

Construction History The date of the construction of the original pavement is not known. However, the road was reconstructed in 1996. The reconstruction or rehabilitation work done under this contract covers N4-5X from Km 9 to 56.6, and also includes a portion of N4-6X. From the project specifications [1], it appears that various rehabilitation methods were used. However, the work done on roughly Km 27 to 56, as well as the pavement performance of this section, differs from the rest of the contract and thus the contract was split into two sections for the purpose of this analysis. The SANRAL PMS [4] does not show an ETB layer, and does not indicate any differences in construction between KM 12 and 27.5. From the contract document, it appears that depth and type of treatment would be determined on site, depending on the pavement situation. The as-built records support this, and show a varying depth of treatment, especially for the subbase and selected layer. For the purposes of this analysis, a part of the project with a relatively uniform construction throughout was selected, in order to allow meaningful and consistent observations to be made. The section selected for this analysis is situated on an eastbound climbing lane, between from Km 27.6 and 29.0. Key observations to emerge from a scan of the as-built information for this sub-section of the project are summarized in Table 1. Based on these observations, it appears that the constructed pavement structure consisted of the following: • 13.2 mm single seal • 25 mm continuously graded asphalt with 5.4% binder; • 150 mm emulsion and cement stabilized natural gravel (roughly 0.5% binder and 1 or 2%

cement♣; • 150 mm C3 upper subbase, consisting of coarse imported material stabilized with 1 to 3%

cement; • 150 mm C3 lower subbase, consisting of fine in-situ material stabilized with 3 to 4 per

cent cement. For the last 1 km of the section the lower subbase consisted of an in-situ G5 natural gravel;

• 150 mm Selected layer or prepared subgrade of a G5 or G6 class material, and • Thickness and class of lower subgrade is not known.

♣ The application rate of the emulsion is 20 l/m3. The average density of the compacted material is roughly 2500 kg/m3. If the relative density of the emulsion is assumed to be 1.0, this translates to roughly 0.8% emulsion by mass. For a 60% emulsion, this results in a 0.5% net binder content.

Page 99: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 6 N4 Section 5X, Km 27.6 to 29.0

Table 1: Summary of As-Built Information, Km 27.6 to 30.6, Eastbound Climbing Lane (see Note 1) Layer Property Specification Observations from As-Built Records

Aggregate 13.2 mm 13.2 mm Quartzite 13.2 mm Single Seal Binder Not Shown B8 (80/100 pen) Natref Bitumen

Thickness 25 mm Generally close to or above 25 mm. Some samples show thickness of around 34 mm.

25 mm Wearing Course Binder 5.4% B12 (60/70 pen)

Natref Binder content for samples in this area vary from 5.3 to 5.6 %.

Thickness Not shown Thickness shown constantly as 150 mm, with no standard deviation indicated. Suspect this is not a measured value.

Material Type BT2 The base was constructed as a BT2 material. PI after treatment is constantly 1, with no standard deviation indicated.

Stabilizers Not Shown Generally 20 l/m3 of emulsion (type not shown) added (see Note 2), plus 1 or 2% cement (GPS or PHOS).

Grading / Fineness

Not shown Percentage passing the 0.425 sieve is generally below 28, and on average around 27%, which would conform to part of the grading envelope for G1 to G3 material.

Compaction 88% (probably ARD) Compaction is always above 99% for samples in this section. (probably Mod AASHTO)

150 mm BT2

UCS Not shown UCS ranges from 2433 kPa to 6300 kPa. The one sample with 6300 kPa received 2% cement, while the others only 1%. For 1% cement content samples, the average UCS is around 3000 kPa.

Thickness Not Shown Thickness shown constantly as 150 mm, with no standard deviation indicated. Suspect this is not a measured value

Material Type C3 The subbase consisted of imported material stabilized with 1 to 3% Portland Blast Furnace Cement

Grading / Fineness

Not Shown The percentage passing the 0.425 sieve was around 30%, which suggests a relatively coarse material as for G1 to G3 or G4. The PI after treatment is shown as 3 or 1.

150 mm C3 Subbase

UCS Not shown Reported UCS values range from 2900 to 9900 kPa. Average is around 5300 kPa.

Thickness Not Shown Thickness shown constantly as 150 mm, with no standard deviation indicated. Suspect this is not a measured value

Material Type C3 This layer consisted of the in-situ material stabilized with 3 or 4 per cent cement. The last 1 km consisted of G5 insitu material that was not stabilized.

Grading / Fineness

Not Shown The percentage passing the 0.425 sieve was generally between 40 and 50%, which suggests a relatively fine natural gravel material.

150 mm C3 Subbase

UCS Not shown Average reported UCS for the sections that were stabilized is around 2300 kPa.

Selected General Not Applicable Roughly 150 mm of the in situ material was reworked as a selected layer and prepared subgrade. This material appears to be quite good, with a G5 or G6 classification. A short 300 mm section consisted of imported material treated with cement.

Note 1: This was the original section investigated, however, Km 29 onwards was rebuilt in 1998 when the Belfast interchange was rebuilt. The as-built results are still representative of the shorter section.

Note 2: Mike Hughes states 24 to 36 l/m3 would have been applied. This is not in agreement with the as-builts.

.

Page 100: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 6 N4 Section 5X, Km 27.6 to 29.0

Behaviour and Material Quality Indicators • FWD deflections were measured in 2000, 2002 and 2004, as part of the TRAC pavement

management system data collection process. These deflections are summarized in Figure 1, which plots the range of the measured distribution for each year by showing the 15th percentile, median and 85th percentile values. The deflections generally range from 170 micron to 570 micron, and show a median value between 290 and 380 micron for the three years of measurement. The deflections measured in 2002 are somewhat higher than those measured in 2000 and 2004, and also show a wider range. A possible explanation for this observation is that the 2000 and 2004 data was measured later into the dry winter season. All three years of data show relatively high deflections of around 500 to 700 micron at some points. However, there is no significant rate of change in the observed deflections for this section.

Figure 1: Maximum Deflections (normalized to 566 kPa) measured on N4-5X, Eastbound, Km 27.6 to 29. • Backcalculated stiffnesses done on deflections measured in October 2004 show that the

base layer stiffness is significantly lower on this section than on the remainder of N4-5X. The lower subbase is also lower than the rest of the section, but the difference is not as large as for the base [7].

• DCP measurements done in September 2005 show that of the 8 measurements done (both Eastbound and Westbound, the different directions cannot be distinguished), at 5 locations the penetration rate in the top layer is over 2 mm/blow, and in one case is as high as 4.68 mm/blow. This indicates a weak base layer. [7]

Performance Indicators • Figure 2 summarizes the available rut depth data extracted from the TRAC and SANRAL

PMSs. The TRAC data, which are measured at 10 m intervals, show an apparent decreasing trend in the observed rut. This discrepancy is most likely due to the precision or accuracy of the network level surveillance equipment. TRAC data for both dates show a low rut depth with an 85th percentile value of between 5 and 7 mm. Although both sets of data show a maximum rut of around 20 mm, these ruts occur at isolated positions only, as confirmed by the 95th percentile value for both TRAC dates, between 8 and 12 mm.

Page 101: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 6 N4 Section 5X, Km 27.6 to 29.0

The SANRAL data are recorded every 100 metres and show an overall increase in the rut depths from 1995 to 2002. The median rut depths for 2002 in the SANRAL database and TRAC database are fairly close, however the range of measurements are higher for the TRAC data, probably because of the increased measurement frequency.

• A visual assessment, conducted in 2004 as part of the TRAC PMS data collection, showed that crocodile cracking, longitudinal cracking and pumping was observed on one of the two assessment segments in this area. These distresses are generally of degree 1 or 2, and extent 2. The overall visual condition index for the two assessment segments, however, was 81 and 93, which suggests a pavement in a good to very good condition.

• Roughness measurements recorded in the SANRAL database are illustrated in Figure 3. The roughness remained consistent from 1999 to 2002, and shows the pavement was in a very good condition. In 2004 the roughness increased, indicating the pavement may be deteriorating.

0

1

2

3

4

5

6

7

8

9

10

11

12

1999S 2000S 2001S 2002S 2002T 2004TYear

Rut

dep

th (m

m)

Figure 2: Left Wheel Path Rutting measured on N4-5X, Eastbound, Km 27.6 to 29

(S data from SANRAL PMS and T from TRAC PMS) • This section is in a bad condition (July 2006), with rutting and potholes present. It is not

currently being rebuilt as this part of the road will be widened, and will be rebuilt at that time. [5].

• Between 2004 and 2005, the section deteriorated as a visual assessment in February 2005 showed that the visual condition index is below 60, which indicates a fair or poor condition. The minimum concession contract value is 60. A Reseal Needs Index was also calculated, and falls below 50 at all locations, indicating a high or very high need to reseal. [7]

• From the assessment in reference [7], it appears as if the emulsion treated base layer is the weak layer in the pavement structure.

Page 102: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 6 N4 Section 5X, Km 27.6 to 29.0

0.0

1.0

2.0

3.0

4.0

5.0

1999S 2000S 2001S 2002S 2002T 2004TYear

IRI

Figure 3: Left wheel path roughness measured on N4-5X, Eastbound, Km 27.6 to 29

Traffic Loading Indicators The project document for the 1996 reconstruction documents traffic data recorded in 1994. For the eastbound, direction, Km 23, the following traffic counts were stated: 5 Day Average: 2502 vehicles per day 7 Day Average: 2293 vehicles per day The TRAC PMS Traffic data consists of monthly counts in 2003 and some of 2004. The AADT is recorded, and using an E80 per heavy factor of 2, the E80s in the slow lane were calculated. The average count is shown in Table 2 and all the data are illustrated in Table 2. The SANRAL PMS Traffic data are also shown in Table 2. The 2002 data seems incorrect. Using both set of traffic data, and growth rates of 7 and 12%, the initial traffic was estimated, as shown. The cumulative traffic to December 2005 was calculated as the area under the curves, and gives 1.4 to 5 million E80s in the 9 year service life. Table 2. SANRAL and TRAC PMS Traffic Data

E80s per slow lane per day Year

SANRAL PMS TRAC PMS*

2001 1241.5

2002 1943.2

2003 1502.5 1049

2004 1774.2 1046

0

500

1000

1500

2000

2500

1997 1998 1999 2001 2002 2003 2005Year

E80s

in s

low

lane

SANRAL PMSTRAC PMSEstimated

* Average for the year is shown

Page 103: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 6 of 6 N4 Section 5X, Km 27.6 to 29.0

Factual Observations • The pavement has been in service for 10 years and has not had any maintenance or

rehabilitation; • The condition after 8 years in service (2004) is good to very good. Rut depths are

generally below 10 mm. • Network level visual assessment show degree 2, extent 2 structural distress (crocodile

cracking and pumping). • Roughness measurements after 2 to 5 years of service are less than 2.5, and show the

pavement is in a good condition. • After 2004, the section deteriorated and was in a poor condition in 2006, 10 years of

service, and needs to be resealed.

Observations Based On Estimates or Interpretations • The estimated cumulative traffic to 2005 (9 years of service life) is 1.4 to 5 million E80s. • After between 6 and 8 years in service, the deflections generally range from 170 micron

to 570 micron, and show a stable median value of between 230 and 300 micron over three years of measurement.

Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 104: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 N7 Section 7

Road Type: ETB Road Code: N7 Section 7 Constructed: 1987 (reconstruction) Limits Considered: Km 47 to 47.2 of north and southbound lanes Project Description: N7-7 Garies to Okiep (near Kammieskroon) Available Documentation: 1. Wright, B.G., et al, “Long Term Performance of emulsion treated base pavements”,

PR88/014, RDAC, CSIR, Pretoria, 1991. 2. As-built records contained in the project file for project PR88/014. 3. SANRAL PMS. 4. Steyn, WJvdM, “Summary of field and laboratory test results on selected ETB

pavements”, CR-98/073, CSIR, Pretoria, 1997. 5. Theyse, H.L., “Towards Guidelines for the Structural Design of Pavements with Emulsion-

treated Layers”, CR-97/045, CSIR, Pretoria, 1998. 6. Discussion with Chris Roux, WSP, Maintenance Route Manager. 7. Discussion with Andries Bester, WSP. 8. Review by M Shaheen Nackerdien, ASCH Consulting, formerly Resident Engineer for

Jeffares and Green.

Area and Climate The road is situated near Kammieskroon in the North Western Cape. The area is dry, with an approximate Weinert n-value of 10, and an average annual rainfall of around 160 mm [1].

Construction History This ETB pavement was constructed as part of a rehabilitation of an existing pavement, which was carried out in 1986 and 1987. The available information does not clearly show the original pavement configuration, but suggests that the pavement consisted of a 200 mm thick cement treated base. The rehabilitation consisted of milling the top 150 mm of the existing base including the old Cape Seal which was then mechanically blended with gabbro fines from a nearby quarry. At least 2% anionic emulsion with 1% cement was added with the compaction water [1, 8]. A dilute emulsion was sprayed on top of the ETB to enrich the layer for traffic accommodation [8]. The rehabilitated pavement consisted of the following [1,8]:

• Seal (see next paragraph) • 150 mm emulsion treated base; • 50 mm cement stabilized subbase; • 150 mm natural gravel selected layer or lower subbase, and • 250 mm selected subgrade. Reference [1] states that a Cape seal surfacing was used, possibly also including a bitumen rubber seal. However, Mr Andries Bester indicates that there is no evidence of a Cape Seal on the section. Rather, the surfacing appears to consist of a 19 mm seal and then either a 6.7 mm well graded seal or a grid seal. Further, Mr Shaheen Nackerdien indicated that a 16 mm bitumen rubber seal was applied in 1986 and possibly a 9.5 mm single seal in 1995. The SANRAL PMS does not reflect ETB in the base layer. The LTPP information contained some as-built records for the rehabilitation project [1, 2]. Although the material control sheets of these records did not cover the section under consideration (Km 47 to 47.2), some general observations could be made on the basis of the available as-built records for the emulsion treated base [2]:

• The legend shown in the as-built records for the area under investigation states the treatment consisted of “milled CTB treated with emulsion and sealed with bitumen rubber”. This suggests that a bitumen rubber seal may have been placed over the section under consideration. However, there is no sign of a bitumen rubber seal on the specific section [7].

• The summarized comments shown in the as-built records state the following:

Page 105: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 N7 Section 7

“Generally the milled CTB conformed to the specified grading and the admixing of a soil binder to improve grading was never found necessary”

“The milled material was satisfactorily recompacted after the addition of 1.0% nett bitumen, using 60% stable grade emulsion, to the compaction water. The grading of the recycled base material improved somewhat with respect to the finer fractions after mixing and compaction. The compaction requirement was generally met and recompaction was only occasionally found necessary.

• The source material for the ETB consisted of milled out cemented crushed granite; • The percentage passing the 0.425 mm sieve is generally below 20, and always below 30.

This suggests a fairly coarse material with a low fines fraction. The percentage retained on the 19 mm sieve is generally between 4 and 28 (i.e. between 96 and 72 % passing);

• The field density (sand replacement) is generally between 1900 and 2200 kg/m3. • The optimum moisture content is surprisingly high at around 8%; • For all of the sections treated with emulsion, and for which materials control sheets were

found, the layer thickness specified is 100 mm, and not 150 mm as indicated in the records. However, a control sheet for the section under investigation could not be found, and it is therefore not certain that the layer thickness was 100 mm instead of 150 mm. Shaheen Nackerdien recollected that ETB layer thickness was never less than 150 mm.

A 1997 investigation [4], during which cores were extracted from the base at two locations, inside and outside the wheel path (i.e. to provide four cores in total) showed a residual binder content of 0.5%.

Behaviour and Material Quality Indicators • Deflection measurements (confirmed RSD by Mr Andrew Laatz) taken in 1989 [1], show

that the average deflection for the north and south directions was around 380 and 400 micron, respectively. The 95th percentile deflections for these two directions were both around 520 micron.

• DCP measurements taken around 1989 [1] and the assessment of these data concluded that the ETB could not easily be penetrated and exhibited high shear strength [1]. This documentation [1], also states that at a depth of 100 mm the tests had to be halted, which suggests that the remaining part of the original cement treated subbase was still largely intact and cohesive.

• DCP measurements recorded around 1997 [4] show an average penetration rate in the ETB of 1.1 mm/blow, with a standard deviation of 0.4 mm/blow.

• A 1997 investigation [4] found that the UCS value of two cores extracted from the emulsion treated base was 10.2 and 6.8 MPa.

• The 1997 investigation [4] showed an average in-situ moisture content of 6.8%, with a standard deviation of 0.7%.

• FWD deflections recorded in 1997 [4,5] show that the deflection ranges roughly from 225 to 600 micron, with an average of around 370 micron (three isolated points exhibited deflections of between 600 and 900 micron).

• FWD deflections recorded on the northbound direction showed peak deflections of 300 and 470 microns in 2000 and 354 and 470 microns in 2005 (only 2 points measured in the section under consideration).[3] These deflections are in line with the kilometre surrounding the section. No deflections were recorded on the southbound direction.

Performance Indicators and Treatment History • Rut depth measurements were taken around 1989 [1] and show that the average rut for

the north and southbound directions was roughly 6 and 7 mm, respectively, with a 95th percentile rut of 7 and 9 mm, respectively.

• The average rut depth measured in a 1997 study [4] was 6.8 mm (direction of measurement is not stated), with a standard deviation of 1.8 mm. This corresponds to a 95th percentile rut of around 10 mm.

• Rut depth measurements extracted from the SANRAL PMS in 2005 are shown in Figure 1. Because only 2 measurements are included in the section under consideration, the kilometre surrounding the section is also analysed. The lines in Figure 1 represent

Page 106: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 N7 Section 7

the 5th to 95th percentile of the kilometre section. The rutting is consistently less than 9 mm, and shows a drop off from 2001 to 2002.

• Riding quality measurements taken around 1989 [1] show that the road was in a sound condition.

• Riding quality measurements from the SANRAL PMS for 1999 to 2002 show the left wheelpath IRI is low, 2.1 (1999), 2.2 (2000), 2.2 (2001) and 2.3 (2002). Only two measurements are available in the section under consideration, but analysis of 1 km surrounding the section shows the roughness is consistent over the kilometre.

• Based on a survey carried out in 1989 [1] it was concluded that the road was in a sound condition with respect to cracking, and that a small amount of bleeding of the surface was observed. It was, however, noted that the rutting was approaching a warning level.

• Based on information in the SANRAL PMS, it appears that this road received a 9.5 mm single seal in 1995, when the road had been in service for 8 years [3]. There is however, no evidence of this on the road [7].

• At the end of 2005 a surface enrichment (fog spray) of Cat70 bitumen emulsion was applied at an application rate of 0.35 l/m2 net bitumen. [7].

• The current condition of the section is good. There are some small surface cracks, but no major surface cracking. There is no rutting. The wheel tracks have “fattened up”, but this occurred a while ago and the condition is currently stable. [7] On the 10 km section surrounding the specific section under investigation, there is some severe transverse cracking which has been repaired with geotextile patching.

0

2

4

6

8

10

12

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003Year

Rut

dep

th (m

m)

Section KM 47 to 47.2Section KM 46.5 to 47.5

Figure 1. Left Wheelpath Rut Depths from SANRAL PMS (2005)

Traffic Loading Indicators The information recorded in the 1989 study of this section’s history showed the following [1]:

• In 1980 the number of heavy vehicles on this road was 120 per day, while the light vehicles count was 283 per day. The total number of equivalent vehicle units (e.v.u.’s) was 643.

• In 1984 the number of heavy vehicles on this road was 90 per day, while the light vehicles count was 433 per day. The total number of equivalent vehicle units (e.v.u.’s) was 703.

In the 1989 study, a 50/50 directional split was assumed, with an E80 per heavy factor of 0.6, and a growth in traffic and heavy vehicles of 2%. Using these assumptions, the cumulative traffic between 1987 (the date of construction) and 2007 (a 20 year design period), was 0.68 million E80’s. Traffic data in the SANRAL PMS in 1997 show an estimated number of E80’s per day of 130 (assumed to be for both lanes). Traffic data obtained from the SANRAL PMS in 2005 is shown in Table 1. Using these data, and the 1980 and 1984 data (E80 per heavy assumed same as calculated for 1997 and 2004

Page 107: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 N7 Section 7

SANRAL data), the cumulative traffic from construction to end 2005 (19 years of service) is between 1 and 1.4 million E80s (areas under the lines in figure in Table 1). Table 1. 2005 SANRAL PMS Traffic Data

E80s per slow lane per day Year

Northbound Southbound

1997 201.8 183.4

2000 217.8* 200.2*

2003 208.6* 191.8*

2004 239.9 221.9 0

50

100

150

200

250

300

1980 1985 1990 1995 2000 2005Year

E80s

per

slo

w la

ne p

er d

ay

2005 SANRAL PMS, Northbound2005 SANRAL PMS, Southbound1997 SANRAL PMS, north and south1989 Study, north and south

* E80s calculated using number of heavy vehicles per day and E80 per heavy factor calculated for 1997 and 2004.

Factual Observations • Average rut depth measurements are

– 6 and 7 mm (1989) – 6.8 mm (1997) – 6.8 mm (2000) – 3.5 mm (2001)

• Riding quality measurements taken around 1989 and 1999 to 2002 show that the road was in a sound condition.

• Based on a survey carried out in 1989 [1] it was concluded that the road was in a sound condition with respect to cracking, and that a small amount of bleeding of the surface was observed. Rutting, however, was approaching a warning level.

• The road received a 9.5 mm single seal in 1995 after 8 years of service and a fog spray after 18 years of service.

• DCP measurements recorded around 1997 [4] show an average penetration rate in the ETB of 1.1 mm/blow, with a standard deviation of 0.4 mm/blow.

Observations Based on Estimates or Interpretations • The cumulative traffic on the section until 2005 (18 years of service is) 1 and 1.4 million

E80s. • FWD deflections recorded in 1997 average 370 micron and in 2000 are 300 and 470

microns.

Reviewed by: • M Shaheen Nackerdien

Page 108: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 N12-19 Section 1

Road Type: ETB Road Code: N12-19 near Daveyton, Experimental Section No 1 Constructed: 1974 Limits Considered: Km 27.320 to 27.470, Eastbound Slow Lane (Km 30.073 to 30.226

prior to 1997) Available Documentation: 1. 1990 ATC Paper by M Hughes, H Thompson and J du Plessis (1990). 2. Jones, D., “Collection of ETB Data on the N12-19 for calibration of mechanistic design

models: Phase 1 – Inception Report”, CSIR Report CR-2004/54, Pretoria, 2004. 3. Detail Evaluation Report by MBS for Rehabilitation and Widening, Feb, 1991. 4. Pavement Assessment Report by Ndodana Consulting Engineers, August, 2003. 5. Discussion with and review by Mike Hughes, previously MBS/WSP. 6. Visual assessment from Ndodana Consulting Engineers, approximately late 2006. 7. Discussion with Mr Henk Diedericks, Ndodana Consulting Engineers.

Area and Climate The road is situated near Daveyton in the Gauteng province. The road is located in an area that TRH4 (1996) classifies as moderate, with a Weinert-n value of approximately 2.4 and an annual rainfall of between 724 to 750 mm [3].

Construction History The section under consideration forms part of a long term pavement performance (LTPP) monitoring experiment. The section was originally constructed between 1968 and 1969 as a cement treated crushed stone base pavement. The section failed prematurely and was rehabilitated in 1974. The rehabilitation consisted of breaking up of the cemented base using a 15 tonne grid roller and a D8 bulldozer [2]. Bitumen emulsion was then added to the base layer which was then re-compacted and resurfaced. Records indicate that the emulsion content was between 0.7 and 1.2%, with a value of 1.0% net bitumen being most often mentioned in the available literature [2, 3]. The pavement structure after the rehabilitation in 1974 consisted of the following [1, 2]:

• 13.2 mm single seal; • 50 mm continuously graded asphalt; • Approximately 100 mm emulsion treated base (1% net bitumen); • Roughly 100 mm crushed stone subbase; • Roughly 75 to 100 mm cement stabilized gravel subbase, and • 300 to 350 mm selected subgrade. Table 1 summarizes the material indicators that were retrieved from the available literature [1]. It should be noted that the section under consideration accommodated bi-directional traffic between 1969 (when the road was opened to traffic) and February 1976, when the westbound carriageway was constructed. After February 1976, the section under consideration only accommodated traffic on the eastbound carriageway of the N12 Section 19. Centreline survey tests reported in the 1991 report by MBS [3] showed that the subgrade (700 to 1100 mm depth) had soaked laboratory determined CBR values of 36% (average of 6 tests along the centreline) and 33% (average of 7 tests along the median). Corresponding CBR values obtained from DCP tests were 36% and 23%, respectively.

Page 109: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 N12-19 Section 1

Table 1: Layer Quality and Thickness Indicators [1] Recorded At Time Of Layer Parameter

Original Construction (1969)

Monitoring (1989)

Thickness (mm) Not yet constructed 120 % Passing 0.075 4 PI NP

ETB

CBR (see note1)

Not yet constructed

DCP Refusal Thickness (mm) Not reported 120 % Passing 0.075 5 12 PI 3 NP

Crushed Stone Subbase

CBR 98 135 Thickness (mm) Not reported 80 % Passing 0.075 11 15 PI 5 SP

Stabilized Gravel Subbase

CBR (see note1) 93 85 Thickness (mm) Not reported 140 % Passing 0.075 18 26 PI 9 6

Upper Selected Subgrade

CBR (see note1) 58 67 Thickness (mm) Not reported 280 % Passing 0.075 32 31 PI 15 12

Lower Selected Subgrade

CBR (see note1) 76 34 Note 1: The CBR recorded in 1969 is the in situ determined CBR (method not known). The CBR for

1989 is a weighted average of the soaked laboratory CBR at 95% Mod AASHTO and the CBR determined using the DCP (using a weighting ratio of 0.7 and 0.3, respectively);

Behaviour and Material Quality Indicators • The average Benkelman Beam deflection measured in 1973, the year before the ETB

rehabilitation, was 330 micron [1]; • Deflectograph deflections recorded in September, 1988 showed an average deflection of

190 micron [1]; • FWD deflections measured in March, 1990 showed an average deflection of 216 micron

[1]. • A DCP test conducted during road monitoring in 1989 showed that the ETB layer could

not be penetrated [1]; • FWD deflections were measured in 2000 at 100 m spacings. Three deflections fell within

or at the edges of the section under consideration. Maximum deflections for these deflection bowls were 229, 229 and 273 micron.

• The SANRAL PMS has FWD deflections measured in 2001 every 100 metres. The maximum deflections measured in the section were 255 and 257 microns.

Performance Indicators and Treatment History • Available records suggest that apart from a fog spray applied in 1990/1991, no other

maintenance or rehabilitation actions were applied between 1974 and 2000 [2]. However, the SANRAL PMS states that a single seal was placed in 1986, and another single seal and 30 mm asphalt was placed in 1990. Mike Hughes is of the opinion that the seal was placed in 1984 and not 1986, and was on Km 17 – 23, i.e., not on the section under consideration. The 1990 seal was placed in 1989/1990 and also did not include the section under consideration [5]. The SANRAL PMS data are therefore considered incorrect, and are not used.

• A 1990 report of the condition of the section [1] mentions no cracking, and indicates that the pavement was sound with slight undulations from the underlying clay.

Page 110: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 N12-19 Section 1

• Since 2000, patching was performed on the section. A survey conducted in 2004 [2] showed a patch approximately 12 m long over the slow lane. Crack sealing was also conducted on the section.

• Base patching was initiated on the road in December 2003 [2]. • The recorded rut depth in 1988 (i.e. after 14 years in service) was roughly 3 mm. It is not

known which statistic this value represents, or how large the sample was [1]. • The SANRAL PMS shows that the roughness on this section remains constant from 1999

to 2002, with an IRI less than 3.5. These measurements are reported for 100 m sections, so only 2 readings are recorded on this section.

• The rutting on the section, as recorded in the SANRAL PMS from 1995 to 2002, shows no indication of the road deteriorating. The highest rut depth, 7 mm, measured was in 1995. In 2002 the rut depth averaged 5.9 mm.

• The visual condition in 1995, 1996 and 1997 is fairly good, with none of the structural parameters, such as cracking, exceeding degree 3, and extent 2.

• The visual condition in 2004 is reasonably good. There is degree 1, extent 3 rutting over the whole section, degree 2/3 and extent 1/3 cracking over most of the section and degree 1, extent 2 or 3 bleeding over the most of the section. [6]

• The condition of the pavement has deteriorated rapidly from the last visual assessment in 2004 to the current condition in 2006. [7]

Traffic Loading Indicators A rehabilitation assessment performed in 1990 [3] showed that the road has four distinct traffic patterns. The pattern of interest to the segment under consideration lies between Km 23.6 and 30.6. Axle weight analyses conducted as part of the assessment in 1990 measured 1714 E80’s in the Eastbound slow lane. For the analysis of traffic loading, the 1990 assessment adopted the following assumptions:

• Percentage Trucks: 5.9% • Percentage Buses and Light Delivery: 2.3% • Percentage Light Vehicles: 91.8% Using the above assumptions, together with the counted vehicles per day of 28 400 to 33 300, the 1990 assessment concluded that the 20 year design traffic was between 21 and 30 million E80s. Another assessment reported in 1991 [1] provided the following indicators of traffic loading:

• The total cumulative traffic carried on the eastbound slow lane traffic from the time of opening (1976) to December 1989 was estimated at 5 million E80s.

• The number of daily E80s on the eastbound slow lane was estimated at 1200 to 1350 per day (reported in 1991, assumed measured in December 1989);

• A growth rate of 5% was used as an appropriate illustrative value in this assessment. An assessment performed in 2003 [4] provided the following traffic loading indicators (measured in 2002):

• The average daily truck traffic in both directions at Km 26.8 was 3 271 (both directions, 4 lanes). The ADT was 37 539 (both directions) which provide a heavy vehicle volume of 8.7%.

• The average daily truck traffic in both directions at Km 28.0 was 2 301 (both directions, 4 lanes). The ADT was 17 638 (both directions) which provide a heavy vehicle volume of 13.0%.

The SANRAL traffic data yield the following results: • Traffic data are available from 1996 to 2004, as shown in Table 1. In the figure, the

historical values mentioned above are plotted for 1989 and 1990. Using growth rates of 2 and 7%, the initial traffic in 1974 was back-calculated. The cumulative E80s from 1974 to 2004 (30 years) were calculated as the areas under the dashed and solid lines in the figure, respectively. These give a range of 16.4 to 26.8 million E80s.

Page 111: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 N12-19 Section 1

Table 1. SANRAL Traffic Data Year E80s per slow lane per day 1996 2549.1 1997 3130.7 1998 650.6 1999 2312.0 2000 2413.6 2001 2432.3 2002 2575.3 2003 2550.8 2004 3192.2

0

500

1000

1500

2000

2500

3000

3500

1974 1979 1984 1989 1994 1999 2004Year

E80s

per

wor

st la

ne p

er d

ay

SANRAL PMSHistorical [1]Estimated

Factual Observations • The only maintenance to occur between 1974 and 2000 was a fog seal in 1991. The road

was patched and cracks sealed after 2000. • Rut depths in 2002 were less than 7 mm (28 years of service). • Riding quality up until 2002 did not exceed an IRI of 3.5 (28 years of service). • The visual condition in 2004 (30 years of service_ showed the road is in a reasonable

condition.

Observations Based on Estimates or Interpretations • The different traffic estimates range from 21 and 30 million E80s for a 20 year service, to

5 million E80s in 15 years, and 16.4 to 26.8 million E80s for 30 years. • Maximum deflections recorded on the sections in 1990, 2000, 2001 are less than 300

microns. • The road has apparently deteriorated rapidly from 2004 to 2006 (30 to 32 years of

service). Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 112: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 N12-19 Section 2

Road Type: ETB Road Code: N12-19 near Daveyton, Experimental Section No 2 Constructed: 1974 Limits Considered: Km 27.170 to 27.320, Eastbound Slow Lane (Km 29.920 to 30.073

prior to 1997) Available Documentation: 1. 1990 ATC Paper by M Hughes, H Thompson and J du Plessis (1990). 2. Jones, D., “Collection of ETB Data on the N12-19 for calibration of mechanistic design

models: Phase 1 – Inception Report”, CSIR Report CR-2004/54, Pretoria, 2004. 3. Detail Evaluation Report by MBS for Rehabilitation and Widening, Feb, 1991 4. Pavement Assessment Report by Ndodana Consulting Engineers, August, 2003. 5. Discussion with and review by Mike Hughes, previously MBS/WSP. 6. Visual assessment from Ndodana Consulting Engineers, approximately late 2006. 7. Discussion with Mr Henk Diedericks, Ndodana Consulting Engineers.

Area and Climate The road is situated in near Daveyton in the Gauteng province. The road is located in an area that TRH4 (1996) classifies as moderate, with a Weinert-n value of approximately 2.4 and an annual rainfall of between 724 to 750 mm [3].

Construction History The section under consideration forms part of a long term pavement performance (LTPP) monitoring experiment. The section was originally constructed between 1968 and 1969 as a cement treated crushed stone base pavement. The section failed prematurely and was rehabilitated in 1974. The rehabilitation consisted of breaking up of the cemented base using a 15 tonne grid roller and a D8 bulldozer [2]. Bitumen emulsion was then added to the base layer which was then re-compacted and resurfaced. Records indicate that the emulsion content was between 0.7 and 1.2%, with a value of 1.0% net bitumen being most often mentioned in the available literature [2, 3]. The pavement structure after the rehabilitation in 1974 consisted of the following [1, 2]:

• 13.2 mm bitumen rubber single seal (the use of bitumen rubber is disputed [5], as rubber was not yet on the market);

• 50 mm continuously graded asphalt; • Approximately 135 mm emulsion treated base (1% net bitumen); • Roughly 110 to 150 mm cement stabilized gravel subbase; • 300 to 330 mm selected subgrade; Table 1 summarizes the material indicators that were retrieved from the available literature [1]. It should be noted that the section under consideration accommodated bi-directional traffic between 1969 (when the road was opened to traffic) and February 1976, when the westbound carriageway was constructed. After February 1976, the section under consideration only accommodated traffic on the eastbound carriageway of the N12 Section 19. Centreline survey tests reported in the 1991 report by MBS [3] showed that the subgrade (700 to 1100 mm depth) had soaked laboratory determined CBR values of 36% (average of 6 tests along the centreline) and 33% (average of 7 tests along the median). Corresponding CBR values obtained from DCP tests were 36% and 23%, respectively.

Page 113: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 N12-19 Section 2

Table 1: Layer Quality and Thickness Indicators[1] Recorded At Time Of Layer Parameter

Original Construction (1969)

Monitoring (1989)

Thickness (mm) Not yet constructed 150 % Passing 0.075 Not reported PI Not reported

ETB

CBR (see note1)

Not yet constructed

Not reported Thickness (mm) Not reported 110 % Passing 0.075 10 26 PI NP 5

Stabilized Gravel Subbase

CBR (see note1) 94 86 Thickness (mm) Not reported 220 % Passing 0.075 15 25 PI 10 11

Upper Selected Subgrade

CBR (see note1) 85 60 Thickness (mm) Not reported 100 % Passing 0.075 15 50 PI 10 12

Upper Selected Subgrade

CBR (see note1) 85 108 Note 1: The CBR recorded in 1969 is the in situ determined CBR (method not known). The CBR for

1989 is a weighted average of the soaked laboratory CBR at 95% Mod AASHTO and the CBR determined using the DCP (using a weighting ratio of 0.7 and 0.3, respectively);

Behaviour and Material Quality Indicators • The average Benkelman Beam deflection measured in 1973, the year before the ETB

rehabilitation, was 175 micron [1]; • Deflectograph deflections recorded in September, 1988 showed an average deflection of

190 micron [1]; • FWD deflections measured in March, 1990 showed an average deflection of 149 micron

[1]. • FWD deflections were measured in 2000 at 100 m spacings. Three deflections fell within

or at the edges of the section under consideration. Maximum deflections for these deflection bowls were 110, 179 and 229 micron.

• The SANRAL PMS has deflections measured in 2001 every 100 metres. The maximum deflections measured in the section were 109, 104 and 232 microns. The deflection on the second half of the section is higher than on the first, indicating a marginally weaker pavement structure.

Performance Indicators • Available records suggest that apart from a fog spray applied in 1990/1991, no other

maintenance or rehabilitation actions were applied between 1974 and 2000 [2]. However, the SANRAL PMS states that a single seal was placed in 1986, and another single seal and 30 mm asphalt was placed in 1990. Mike Hughes is of the opinion that the seal was placed in 1984 and not 1986, and was on Km 17 – 23, i.e., not on the section under consideration. The seal in 1990 was placed in 1989/1990 and also did not include the section under consideration [5]. The SANRAL PMS data are considered incorrect, and are therefore not used.

• A 1990 report of the condition of the section [1] mention no cracking, and indicate that the pavement was sound with slight undulations from the underlying clay.

• Since 2000, patching was performed on the road but during a survey conducted in 2004 [2], no patches were noted within the section under consideration. Crack sealing was, however, conducted on the section.

• Base patching was initiated on the road in December 2003 [2]. • The recorded rut depth in 1988 (i.e. after 14 years in service) was roughly 2 mm [1]. It is

not known which statistic this value represents, or how large the sample was.

Page 114: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 N12-19 Section 2

• The SANRAL PMS shows that the roughness on this section remains constant from 1999 to 2002, with an IRI less than 3. These measurements are reported for 100 m sections, so only 2 readings are recorded on this section. The second half of the section has lower roughness than the first.

• The rutting on the section, as recorded in the SANRAL PMS from 1995 to 2002, shows no indication of the road deteriorating. The highest rut depth, 8 mm, measured was in 1995. In 2002 the rut depth averaged 4.7 mm.

• The visual condition in 1995, 1996 and 1997 is fair, with none of the structural parameters, such as cracking, exceeding degree 3, extent 2.

• The visual condition in 2004 is reasonably good. There is degree 1, extent 3 rutting over the whole section, degree 2/3 and extent 1-3 cracking over most of the section, degree 1, extent 2 or 3 bleeding over the most of the section and some degree 1 or 3, extent 1 patching. [6]

• The condition of the pavement has deteriorated rapidly from the last visual assessment in 2004 to the current condition in 2006. [7]

Traffic Loading Indicators A rehabilitation assessment performed in 1990 [3] showed that the road has four distinct traffic patterns. The pattern of interest to the segment under consideration lies between Km 23.6 and 30.6. Axle weight analyses conducted as part of the assessment in 1990 measured 1714 E80’s in the Eastbound slow lane. For the analysis of traffic loading, the 1990 assessment adopted the following assumptions:

• Percentage Trucks: 5.9% • Percentage Buses and Light Delivery: 2.3% • Percentage Light Vehicles: 91.8% Using the above assumptions, together with the counted vehicles per day of 28 400 to 33 300, the 1990 assessment concluded that the 20 year design traffic was between 21 and 30 million E80s. Another assessment reported in 1991 [1] provided the following indicators of traffic loading:

• The total cumulative traffic carried on the eastbound slow lane traffic from the time of opening (1976) to December 1989 was estimated at 5 million E80s.

• The number of daily E80s on the eastbound slow lane was estimated at 1200 to 1350 per day (reported in 1991);

• A growth rate of 5% was used as an appropriate illustrative value in this assessment. An assessment performed in 2003 [4] provided the following traffic loading indicators (measured in 2002):

• The average daily truck traffic in both directions at Km 26.8 was 3 271 (both directions, 4 lanes). The ADT was 37 539 (both directions) which provide a heavy vehicle volume of 8.7%.

• The average daily truck traffic in both directions at Km 28.0 was 2 301 (both directions, 4 lanes). The ADT was 17 638 (both directions) which provide a heavy vehicle volume of 13.0%.

The SANRAL traffic data yield the following results: • Traffic data are available from 1996 to 2004, as shown in Table 1. In the figure, the

historical values mentioned above are plotted for 1989 and 1990. Using growth rates of 2 and 7%, the initial traffic in 1974 was back-calculated. The cumulative E80s from 1974 to 2004 (30 years) were calculated as the areas under the dashed and solid lines in the figure, respectively. These give a range of 16.4 to 26.8 million E80s.

Page 115: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 N12-19 Section 2

Table 1. SANRAL Traffic Data Year E80s per slow lane per day 1996 2549.1 1997 3130.7 1998 650.6 1999 2312.0 2000 2413.6 2001 2432.3 2002 2575.3 2003 2550.8 2004 3192.2

0

500

1000

1500

2000

2500

3000

3500

1974 1979 1984 1989 1994 1999 2004Year

E80s

per

wor

st la

ne p

er d

ay

SANRAL PMSHistorical [1]Estimated

Factual Observations • The only maintenance to occur between 1974 and 2000 was a fog seal in 1991. The road

was patched and cracks sealed after 2000. • Rut depths in 2002 were less than 8 mm (28 years of service). • Riding quality up until 2002 did not exceed an IRI of 3 (28 years of service). • The visual condition in 2004 (30 years of service) showed the road is in a reasonable

condition.

Observations Based on Estimates or Interpretations • The different traffic estimates range from 21 and 30 million E80s for a 20 year service, to

5 million E80s in 15 years, and 16.4 to 26.8 million E80s for 30 years. • Maximum deflections recorded on the sections in 1990, 2000, 2001 are less than 250

microns. • The road has deteriorated rapidly from 2004 to 2006, 30 to 32 years of service. Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 116: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 P23/1

Road Type: ETB Road Code: P23/1 Kroonstad to Steynsrus Constructed: 1992 (rehabilitation of base) Limits Considered: Km 3.0 to 4.0 Project Description: P23/1 Kroonstad to Steynsrus, Km 3.0 to 4.0 Available Documentation: 1. As-built records of 1992 construction (studied in the Freetrans plan room, some copies

are available). 2. Investigation report (1989), prior to rehabilitation, by MBS (the report is contained and

was studied in the Freetrans plan room. A full reference is not available). 3. Freetrans PMS traffic information (graphs obtained in electronic format). 4. Discussions with Deirdre Elsmere, Freetrans. 5. Discussion with and review by Mike Hughes, previously MBS/WSP.

Area and Climate The road is situated between Kroonstad and Steynsrus in the Free State Province. Rainfall records for this road could not be obtained, but according to TRH4, the road is situated in a moderate area with a Weinert n-value between 2 and 5.

Construction History This road consisted of a two lane single carriageway originally constructed in 1963 using local dolerites [2]. The investigation report noted that for Km 1.24 to 20.5 the construction utilized good quality dolerite, crushed and used as is. For the original construction, the seal varied from a light sand seal to “stone-bitumen” seals. An investigation on the condition of the original pavement (conducted in 1989 [2]), reported that the pavement showed major failures between Km 2.34 and 4.00. The seal was reported as being dry and losing aggregate. In the area under consideration, “general failure” was indicated, with high rutting of between 10 and 20 mm. Isolated areas showed rutting in excess of 20 mm. There is some indication that the road was flooded from time to time in the area under consideration (assumed from the comment “blocked road floods” in the investigation report [2]). The quality of the materials in the pavement, as assessed at the time of the rehabilitation investigation in 1989 (i.e. before the ETB rehabilitation) was as follows [2]: Base: The base consisted of natural crushed and decomposed dolerite. The soaked

CBR values varied from 50 to greater than 100 per cent. The CBR values derived from DCP testing were consistently above 80 per cent. The material was generally classified as a G5. There was no correlation between the condition of the pavement and the quality of the base layer materials. Using materials sampled from two trial pits in the area of interest, it was shown that the percentage passing the 0.425 mm sieve varied from 15 to 18, while the percentage passing the 0.075 mm sieve was around 9 per cent. The PI was indicated as SP or NP.

Subbase: The subbase consisted of a decomposed dolerite, well graded with a PI of 6 to 10. The CBR values from the DCP were consistently above 45 per cent, and the material was generally classified as a G6. Using materials sampled from two trial pits in the area of interest, it was shown that the percentage passing the 0.425 mm sieve varied from 23 to 29, while the percentages passing the 0.075 mm sieve were 12 and 17. The PI values were 8 and 9 for the two trial pits.

Subgrade: The upper subgrade consisted of sand and ferricrete and sand (trial pit 1, depth 270 to 420 mm) and sand and decomposed dolerite (trial pit 2, depth 330 to 480 mm). The lower subgrade consisted of sandstone and loam (trial pit 1, depth 420 to 640 mm) and decomposed dolerite (trial pit 2, depth 480 to 700 mm). The subgrade was generally classified as a G8 material. The grading modulus for the upper subgrade varied from 0.69 to 1.68 and for the lower subgrade from 0.75 to

Page 117: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 P23/1

2.27. This latter value was for trial pit 2 which consisted of decomposed dolerite. The CBR values from the DCP were consistently above 25 per cent.

For the base layer material, laboratory testing showed that the addition of 2 to 3 per cent lime yielded UCS values of 1400 to 1800 kPa at 100 per cent Mod AASHTO density. The addition of 1 per cent cement and 0.7 per cent bitumen yielded UCS values of 2250 to 3000 kPa at 100 per cent Mod AASHTO density. The pavement was rehabilitated in 1992, using a construction which reworked the base with the addition of 2 per cent lime and 20 l/m3 of 60 per cent anionic bitumen emulsion (roughly 0.5 per cent binder by mass♣). A 30 mm thick asphalt surfacing and a 13.2 mm single seal were then added over the reworked base. Key material indicators extracted from the as-built data for the section under consideration are summarized in Table 1. Table 1: Summary of As-Built Information [1] Layer Property Specification Observations from As-Built Records

Thickness Not Known Varied from 18 to 22 mm (3 cores) % Binder 5.5 % Varied from 4.9 to 5.8%, with average of 5.25% % Binder 3 to 6 Varied from 3.8 to 5.2%, with average of 4.5%. Stability 8 to 16 kN Varied from 11.1 to 12.7

Fine Continuously Graded Asphalt, 60/70 pen binder

Compaction > 93 per cent Varied from 92.9 to 95.1, average 93.7 per cent.

Material Brown Weathered Dolerite

Stabilized with 2 per cent lime and 20 l/m3 of 60% anionic emulsion.

Thickness 150 mm Layer thickness is constantly indicated as 150 mm with no variation (suspect not measured)

Grading Not Shown The percentage passing the 0,425 mm sieve varied between 12 and 18%, and on average around 13. The percentage passing the 0,075 mm sieve varies from 6 to 10%, and on average around 7. This indicates a relatively coarse material meeting the specification for G1 to G3 material.

PI <= 6 Indicated as SP or NP in all cases. It is not certain whether this is before or after stabilization (suspected after stabilization)

Compaction >= 95% Generally achieved or exceeded. Isolated values of 94% exist, but most are above 96%, and many are above 100%.

150 mm In situ treated crushed stone base

UCS 1500 kPa Varied between 1450 and 3500. Average for 21 samples is approximately 2270 kPa.

Behaviour and Material Quality Indicators No post-construction data could be found for this pavement.

Performance Indicators The only performance indicators available for this section are an informal visual assessment done in July 2006 [4]. This assessment indicated that in some parts of the section there is some bleeding, but it is not critical. In the Steynsrus direction, there have been some recent shoulder repairs up to Km 3.7. After Km 3.7 the shoulders are not in a good condition and edge breaks are present. There is some rutting on the section, of about 10 and 20 mm. In the Kroonstad direction, there is also some bleeding, but it is again not critical. At approximately Km 3.5 there is some crocodile cracking in the outer wheelpath, and some associated pumping. The rutting in this direction is between 10 and 20 mm, however increases to 20 to 30 mm near some patching. There is also shoving of the patches. The condition of this section is similar to that of the road on either side of the section. From Km 4 a fog spray was applied recently, but was not applied on the section under consideration. ♣ Assuming untreated material density of 2500 kg/m3 and emulsion relative density of 1.0 kg/l, the percentage binder added is (0.6 * 20 * 1.0)/2500 = 0.48 per cent binder.

Page 118: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 P23/1

The general condition is fair to poor, however the seal is still in a good condition. The road will need to be repaired, and the likely treatment is filling of the ruts and placement of a seal. [4]

Traffic Loading Indicators The 1989 study [2] of this section’s history contained the following traffic indicators [2]: Count Date: 6/1989 7/1989 ADT 1764 1862 Percentage Heavy Vehicles 14.6% 9.5%

In the investigation report, the traffic analysis assumed approximately 185 E80s per lane per day, with zero and 6 per cent growth rates. Using these values, the 20 year design traffic was classified as the (then) E2 class, with expected cumulative traffic ranging from 0.8 to 3 million E80s per lane. For a 6 per cent growth rate, the 20 year design traffic was approximately 2.5 million E80s. Additional traffic data obtained from the Freetrans Traffic Management System show the following: • Only two counts are available for this route. One count was taken in 1996 at Km 44.7

and another in 2004 at Km 41.35 [3]. • A strip map of heavy vehicle distributions over the route suggest that the heavy vehicle

volumes do not vary significantly over the route length. If possible, this should be confirmed, since most counts do not have a count date displayed on the strip map;

• The 1996 count at Km 44.75 show 419 vehicles in the most heavily loaded lane, with 15.4 per cent heavy vehicles;

• The 2004 count at Km 41.35 show 839 vehicles in the most heavily loaded lane, with 12.5 per cent heavy vehicles;

The available traffic data is summarized in Figure 1, which shows the available measured data at different periods, together with a lower and higher best estimate of the traffic increase over time assuming different E80 per heavy factors and different growth rates. It will be noted from the figure that the traffic measured in 1989 does not agree with the 1996 and 2004 estimates, but only for the higher E80 per heavy factor. It is likely that the E80 per heavy factor in 1989 was significantly lower than for 1996 to 2005. If the 1989 traffic is plotted only for a E80 per heavy factor of 1.2, then there is a rough agreement between the various traffic counts. Using the lower and higher best estimates, the cumulative traffic on this section between 1992 and 2005 is estimated to be roughly between 0.5 and 1.3 million E80s. Bearing in mind that the road has been in service for 13 years, this estimate correlates roughly with the 1989 estimate of WSP [2], which estimated between 0.8 and 3 million E80s over a 20 year design period.

Page 119: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 P23/1

0

50

100

150

200

250

300

350

400

450

500

1988 1990 1992 1994 1996 1998 2000 2002 2004 2006Year of Measurement

E80'

s pe

r da

yE80/Heavy = 1.2E80/Heavy = 2.5

E80/Heavy = 3.5Upper EstimateLower Estimate

Figure 1: Traffic Data Summary with Estimated Growth Scenarios

Factual Observations • The rehabilitated road consisted of a 20 mm thick asphalt surfacing over a 150 mm

emulsion treated gravel and crushed stone combination (generally G3 to G4 quality) treated with 1 per cent cement and roughly 1 per cent bituminous binder (the latter applied in the form of a 60 per cent emulsion). The subbase was left untreated and generally consists of G6 quality weathered dolerite.

• The pavement has been in service for 13 years and has not received any treatment after the rehabilitation in 1992.

• In July 2006, the general condition of the road is fair to poor. [4]

Observations Based on Estimates or Interpretations • Using the lower and higher best estimates, the cumulative traffic on this section

between 1992 and 2005 (13 years of service) is estimated to be roughly between 0.5 and 1.3 million E80s.

Reviewed by: • Mike Hughes (previously WSP) • Hugh Thompson (WSP)

Page 120: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 Road 1386, Moloto Rd, Stabilised subbase

Road Type: ETB Road Code: Road 1386 Constructed: 2001 (rehabilitation) Limits Considered: Km 14.7 to 17.9 and Km 19.7 to 20.6 Project Description: Road 1386 Roodeplaat Dam towards Moloto (stabilised subbase) Available Documentation: 1. MBS Consulting Engineers, “Rehabilitation and Improvement of Road 1386 from Km 11.0

to Km 33.7 (Roodeplaat Dam towards Moloto)”, Design Report, August 1999. 2. MBS Consulting Engineers, “The Rehabilitation and Improvement of Road 1386 from Km

11.0 to Km 33.7 (Roodeplaat Dam towards Moloto) and Road P207/1 from Km 23.98 to Km 28.62”, As-Built Data, 2001.

3. Gauteng Department of Public Transport, Roads and Works (GDPTRW) PMS.

Area and Climate The road is situated north-east of Pretoria and runs towards Moloto. The mean annual rainfall is 750 mm. Weinert’s N-value is approximately 2.4, which indicates a moderate area. [1]

Construction History • The original pavement structure, as built in 1975 was as follows: [1, 3]

o 110 – 190 mm C3 – C4 base o 80 – 180 mm C3 – C4 upper subbase o 70 – 250 mm G5 – G8 lower subbase o 140 – 370 mm G5 – G10 selected layer

• The following treatments were applied to the original pavement structure [1, 3]: o 13.2 mm single seal (1978/1980) o 9.5 mm single seal (1981) o 9.5 mm single seal (1989)

• From test pits opened on the section prior to the rehabilitation, the pavement structure shown in Table 1 was present:

Table 1: Original Pavement Structure and Material Classification

Layer Average thickness (mm)

Min – max thickness range (mm)

Classification

Seal 20 Base 164 110 – 190 G5 – G7 Upper subbase 157 80 – 200 G6 – G7 Lower subbase 206 120 – 410 G5 – G8 Selected layer 303 150 – 580 G5 – G10 Subgrade 185 170 – 200 G6 – G8

• The pavement was rehabilitated in 2001. The existing base layer was reworked and

treated with cement for the subbase, and new material was imported and treated with emulsion and cement for the base layer. The resulting pavement structure is as follows:

o 13.2 mm seal with polymer modified binder o 25 mm continuously graded asphalt o 150 mm G5 base with 2% cement and 1% bitumen o Existing base with 3% cement

Page 121: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 Road 1386, Moloto Rd, Stabilised subbase

Table 2: Summary of Information from As-builts [2] Layer Property Specification Measured

Description 13.2 mm Quartzite chips precoated with 12l/m3 of petrosote 80/100 pen SBR 3 bitumen

Aggregate 0.011 m3/m2 Average = 0.011 m3/m2

Seal

1st spray 13.1 l/m2 Average = 1.303 l/m2 Material description

AC Medium with 60/70 pen binder. Meta quartzite from Alpha Ferro Crusher. Drum mixed.

Thickness 25 mm Average = 26 mm, min = 25 mm P 19.0 P 4.75 P 0.425 P 0.075

Average 0 68 22 7 Grading

Standard deviations range from 0 to 3% Binder content 5% Average = 4.9%, std dev. ranges from 0.1 to 0.2 Reference density

Average = 2.321, std dev. ranges from 0.01 to 0.02

Compaction Average = 100.8, std dev. ranges from 0.4 to 1.4 Marshall stability

Average = 9.2, std dev. ranges from 9 to 9.6

Marshall flow Average = 2.5, std dev. ranges from 0.1 to 0.4 ITS (kN) Average = 1059, std dev. ranges from 980 to 1092

Asphalt

Tack 0.55 l/m2 (suspect not measured value) Material Imported material Binder content 1% (suspect not measured value) Cement content

2% (suspect not measured value)

Thickness 150 mm (suspect not measured value) P19 P4.75 P0.425 P0.075 Average 90 73 33 12

Grading

Std dev ranges from 1.7 to 11.9%. PI (before) Slightly plastic or non-plastic Reference density

Average = 2038 kg/m3, 10th percentile = 2010 kg/m3

OMC Average = 8.66%, max = 8.2% UCS @ 97% of Mod. Average = 1688 kPa, 10th percentile = 1492 kPa ITS @ 97% of Mod. Average = 434 kPa, 10th percentile = 363 kPa

Base

Compaction 98% of Mod. Average = 98.5%, 10th percentile = 98.1% Description In situ (original base) Thickness 150 mm (suspect not measured value) Cement content

3% Ranges from 1 to 3%. 70% of measurements = 2% cement. P19 P4.75 P0.425 P0.075 Average 76 53 29 13

Grading

Std dev ranges from 0 to 9.7%. PI (before) Slightly plastic Reference density

Average = 2026 kg/m3, 10th percentile = 1970 kg/m3

OMC Average = 9.8%, max = 10.4% UCS @ 97% of Mod. Average = 2644 kPa, 10th percentile = 2189 kPa

Subbase

Compaction 96% of Mod. Average = 97.3%, 10th percentile = 96.9%

Behaviour and Material Quality Indicators • There are no behaviour or material quality data in the Gautrans PMS. [3]

Performance Indicators and Treatment History • Before the rehabilitation, in 1997, the following visual observations were made [1]:

o Dry bitumen o Crocodile cracking and pumping o Major base failures o Patch/repair work o Road edge breaking o Isolated wheel rutting

Page 122: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 Road 1386, Moloto Rd, Stabilised subbase

• Before the rehabilitation, the VCI decreased from the fair to poor to very poor zones between 1985 and 1997. [1]

• No treatments have been applied to the road since the 2001 rehabilitation. [3] • The visual condition index (VCI) is shown in Figure 1 for two segments of the road which

incorporate the specific sections under investigation in this summary. On the first section, which incorporates the short Km 14.7 to 14.9 section, the VCI is steadily decreasing to unacceptable levels. On the second section, which incorporates Km 19.7 to 20.6, the VCI is at more acceptable levels, although it is also decreasing and by 2006 is changing from good to fair. [3]

0

10

20

30

40

50

60

70

80

90

100

2002 2003 2004 2005 2006Year

Visu

al C

ondi

tion

Inde

x

Km 8.72 to 15.20Km 18.96 to 21.13

Figure 1. Visual Condition Index • Riding quality measured at 100 meter intervals in 2004 gave an average IRI of 3.1 with a

90th percentile of 6, indicates fair to poor performance. [3] • Rutting measured at 100 meter intervals in 2004 gave an average rut depth of 4.6 mm

with a 90th percentile of 6.4 mm, which indicates good performance. [3]

Traffic Loading Indicators • Traffic data obtained from the GDPTRW PMS is shown in Figure 2. Using these data, the

traffic carried from the 2001 rehab to the end of 2006 (estimated from the areas under the dotted lines in the figure) is 1.87 to 4.84 million E80s. [3]

0

500

1000

1500

2000

2500

3000

3500

1982 1984 1987 1990 1992 1995 1998 2001 2003 2006Date

E80'

s pe

r slo

w la

ne

E80s/heavy = 2E80s/heavy = 3.5Rehabilitation date

Figure 1. E80s per day per slow lane [3]

Page 123: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 Road 1386, Moloto Rd, Stabilised subbase

Factual Observations • Since the 2001 rehabilitation, no treatments have been applied to the road in the 5 years

of service. • Since the rehabilitation, the visual condition index has been decreasing and is reaching

warning levels. • Roughness measured in 2004 (3 years of service) indicates that the road in a fair to poor

condition. • Rutting measured in 2004 (3 years of service) indicates that the road is in a good

condition.

Observations Based on Estimates or Interpretations • Based on the factual observations, the road is approaching a warning condition. • The road has carried an estimated 1.87 to 4.84 million E80s in each direction after 5

years of service. • For the purposes of development and validation of the structural design method, the

pavement structure is assumed to be: o 13.2 mm seal with polymer modified binder o 25 mm continuously graded asphalt o 150 mm G5 base treated with 2% cement and 1% bitumen o 150 mm G5-G7 subbase treated with 2% cement o 350 mm G5-G8 selected layer o G5-G10 subgrade

Reviewed by: No review obtained.

Page 124: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 Road 1386, Moloto Rd, Unstabilised subbase

Road Type: ETB Road Code: Road 1386 Constructed: 2001 (rehabilitation) Limits Considered: Km 17.9 to 18.8 and Km 19.4 to 19.7 Project Description: Road 1386 Roodeplaat Dam towards Moloto (unstabilised

subbase) Available Documentation: 1. MBS Consulting Engineers, “Rehabilitation and Improvement of Road 1386 from Km 11.0

to Km 33.7 (Roodeplaat Dam towards Moloto)”, Design Report, August 1999. 2. MBS Consulting Engineers, “The Rehabilitation and Improvement of Road 1386 from Km

11.0 to Km 33.7 (Roodeplaat Dam towards Moloto) and Road P207/1 from Km 23.98 to Km 28.62”, As-Built Data, 2001.

3. Gauteng Department of Public Transport, Roads and Works (GDPTRW) PMS.

Area and Climate The road is situated north-east of Pretoria and runs towards Moloto. The mean annual rainfall is 750 mm. Weinert’s N-value is approximately 2.4, which indicates a moderate area. [1]

Construction History • The original pavement structure, as built in 1975 was as follows: [1, 3]

o 110 – 190 mm C3 – C4 base o 80 – 180 mm C3 – C4 upper subbase o 70 – 250 mm G5 – G8 lower subbase o 140 – 370 mm G5 – G10 selected layer

• The following treatments were applied to the original pavement structure [1, 3]: o 13.2 mm single seal (1978/1980) o 9.5 mm single seal (1981) o 9.5 mm single seal (1989)

• From test pits opened on the section prior to the rehabilitation the pavement structure shown in Table 1 was present:

Table 1: Original Pavement Structure and Material Classification

Layer Average thickness (mm)

Min – max thickness range (mm)

Classification

Seal 20 Base 157 150 – 160 G5 – G6 Upper subbase 123 100 – 150 G6 – G7 Lower subbase 167 150 – 200 G5 Subgrade 397 270 – 570 G5 – G6

• The pavement was rehabilitated in 2001 by deep in situ recycling of the base layer and

treating it with emulsion and cement. The rehabilitated pavement structure is as follows: o 13.2 mm seal with polymer modified binder o 25 mm continuously graded asphalt o Existing G5-G6 base treated with 2% cement and 1% bitumen

Page 125: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 Road 1386, Moloto Rd, Unstabilised subbase

Table 2: Summary of Information from As-builts [2] Layer Property Specification Measured

Description 13.2 mm Quartzite chips precoated with 12l/m3 of petrosote 80/100 pen SBR 3 bitumen

Aggregate 0.011 m3/m2 Average = 0.012 m3/m2

Seal

1st spray 1.31 l/m2 Average = 1.304 l/m2 Material description

AC Medium with 60/70 pen binder. Meta quartzite from Alpha Ferro Crusher. Drum mixed.

Thickness 25 mm Average = 28.8 mm, min = 24 mm P 19.0 P 4.75 P 0.425 P 0.075

Average 0 65 22 7 Grading

Standard deviations range from 0 to 6% Binder content 5% Average = 5.1%, std dev. ranges from 0.1 to 0.4 Reference density

Average = 2.322, std dev. ranges from 0.01 to 0.04

Compaction Average = 100.9, std dev. ranges from 0.5 to 2.7 Marshall stability

Average = 8.6, std dev. ranges from 7.3 to 11.2

Marshall flow Average = 2.5, std dev. ranges from 0.1 to 0.3 ITS (kN) Average = 1116 std dev. ranges from 877 to 1378

Asphalt

Tack 0.55 l/m2 (suspect not measured value) Material In situ material Binder content 1% (suspect not measured value) Cement content

2% (suspect not measured value)

Thickness 150 mm (suspect not measured value) P19 P4.75 P0.425 P0.075 Average 91 59 29 13

Grading

Std dev ranges from 1.4 to 9.7%. PI (before) Slightly plastic or non-plastic Reference density

Average = 2006 kg/m3, 10th percentile = 1988 kg/m3

OMC Average = 9.7%, max = 10.2% UCS 97% of Mod. Average = 2720 kPa, 10th percentile = 2514 kPa ITS 97% of Mod. Average = 255 kPa, 10th percentile = 133 kPa

Base

Compaction 98% of Mod Average = 99%, 10th percentile = 98.8%

Behaviour and Material Quality Indicators • There are no behaviour or material quality data in the GDPTRW PMS. [3]

Performance Indicators and Treatment History • Before the rehabilitation, in 1997 the following visual observations were made [1]:

o Dry bitumen o Crocodile cracking and pumping o Major base failures o Patch/repair work o Road edge breaking o Isolated wheel rutting

• Before the rehabilitation, the VCI decreased from the fair to poor to very poor zones between 1985 and 1997. [1]

• After the rehabilitation, the VCI is steadily decreasing but remains at acceptable levels. The VCI is shown in Figure 1 for three segments of the road which incorporate the specific section under investigation in this summary. [3]

Page 126: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 Road 1386, Moloto Rd, Unstabilised subbase

0

10

20

30

40

50

60

70

80

90

100

2002 2003 2004 2005 2006Year

Visu

al C

ondi

tion

Inde

x

Figure 1. Visual Condition Index • Riding quality measured at 100 meter intervals in 2004 gave an average IRI of 2.4 with a

90th percentile of 3.8, which indicates fair performance. [3] • Rutting measured at 100 meter intervals in 2004 gave an average rut depth of

6.4 mm with a 90th percentile of 8.8 mm, which indicates good performance. [3]

Traffic Loading Indicators • Traffic data obtained from the GDPTRW PMS is shown in Figure 2. Using these data, the

traffic carried from the 2001 rehabilitation to the end of 2006 is estimated (from the areas under the dotted lines in the figure) as 1.87 to 4.84 million E80s. [3]

0

500

1000

1500

2000

2500

3000

3500

1982 1984 1987 1990 1992 1995 1998 2001 2003 2006Date

E80'

s pe

r slo

w la

ne

E80s/heavy = 2E80s/heavy = 3.5Rehabilitation date

Figure 1. E80s per day per slow lane [3]

Factual Observations • Since the 2001 rehabilitation, no treatments have been applied to the road in the 5 years

of service. • Since the rehabilitation, the visual condition index has been decreasing, however is still

within reasonable limits. • Rutting and roughness measured in 2004 indicates the road was in a fair to good

condition after 3 years of service.

Page 127: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 Road 1386, Moloto Rd, Unstabilised subbase

Observations Based on Estimates or Interpretations • Based on the factual observations, the road is in an acceptable condition. • The road has carried an estimated 1.87 to 4.84 million E80s in each direction after 5

years of service. • For the purposes of development and validation of the structural design method, the

pavement structure is assumed to be: o 13.2 mm seal with polymer modified binder o 25 mm continuously graded asphalt o 150 mm G5-G6 base treated with 2% cement and 1% bitumen o 300 mm G5-G7 subbase o G5-G6 subgrade

Reviewed by: No review obtained.

Page 128: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 6 MR504

Road Type: Foamed Bitumen Road Code: MR504 near Shongweni Constructed: 1995 Limits Considered: Km 3.0 to 3.7, both lanes Available Documentation: 1. Collings, Lindsay and Shunmugan, “LTPP Exercise on a Foamed Bitumen Treated Base

– Evaluation of Almost 10 Years of Heavy Trafficking on MR 504 in Kwazulu-Natal ”, 8th CAPSA, 2004.

2. Theyse, H.L., “Preliminary Assessment of the Structural Properties of Pavements with Base Layers Treated with Foamed Bitumen”, CR-97/087, CSIR, Pretoria, 1998.

3. Discussions with Mike Proudfoot, KZN DoT. 4. Review by Dave Collings, Loudon International. 5. Informal visual assessment by Fritz Jooste, MAS. 6. Review by Pat Dorkin, Kwazulu-Natal Department of Transport.

Note: The data reported for this project are extensive, and potentially confusing within the context of the present study. To clarify and simplify the information reported here, the summary will focus primarily on the uphill lane for the project (also reported as the right hand side lane). This lane carries significantly higher traffic than the downhill lane, and is more significant for this study.

Area and Climate The road is situated near Key Ridge in Kwazulu Natal, approximately half-way between Durban and Pietermaritzburg. In TRH4 (1996), the area is classified as wet, with a Weinert-n value of less than 2.

Construction History [1] This road was originally constructed as a gravel road to serve the Shongweni local communities near Key Ridge. A 700 m section of this road is situated on a steep gradient (greater than 10%), and showed a constant need for maintenance in the rainy season. This 700 m section of road was upgraded in mid 1995 using different variations of a foamed bitumen base. The different construction types used are summarized in Table 1. It should be noted from Table 1 that sections of the road received no surfacing, while other sections were treated with a fog spray or a slurry. The treatment of the three sections summarized in Table 1 can be summarized as follows [1]: • Section A: Source material was reclaimed asphalt (RAP) which was blended with 20

per cent (by volume) crusher dust and then treated with 1.7% (by mass) foamed bitumen.

• Sections B and C: Source material was weathered granite from borrow pits treated with 3.5% foamed bitumen (by mass) and 1% hydrated lime (by mass).

Based on informal records reported in [1], the slurry sections of the road started to strip around 1998, and a nominal 30 mm thick asphalt was placed over sections B and C (as defined in Table 1), while Section A received a 19 mm single seal. There are no records of work being done on the section since 1998.

Page 129: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 6 MR504

Table 1: Subsection Configuration on MR504 [1] Parameter Section A1 Section A2 Section B1 Section B2 Section C Location Km 3.0 to 3.1 Km 3.1 to 3.2 Km 3.2 to 3.4 Km 3.4 to 3.6 Km 3.6 to 3.7 Average Gradient 12% 9.6% 11.5% 14.4% 14.2% Surfacing None None Slurry/fog spray

and slurry1 Fog Spray/slurry1

Slurry

Thickness (as designed) 175 175 175 175 150 Thickness (from Test Pits2) 160 150 120 130 180 Source Material RAP RAP Weath. Granite Weath. Granite Weath. Granite Source Material Dry Stability (kN) 23.7 23.7 12.53 12.5 12.5

Foamed Bitumen Base

Compaction Level Normal High Medium Medium High Thickness 150 150 150 150 150 Subbase

Source G6 Weathered Granite (imported from borrow pit)

Reworked In-situ Material/G6 Weathered Granite4

Subgrade Reworked In-situ Material (nominal 150 mm thickness) In-situ Material 1. According to Dave Collings, Section B1 had a slurry, Section B2 a fog spray and Section C a slurry [4]. According to Pat Dorkin, Section B1 had a fog

spray and slurry, Section B2 a slurry and Section C a slurry [6]. 2. Based on thicknesses obtained from one test pit per section, measured in 2003 3. The RAP and weathered granite materials were stockpiled without cover for roughly 3 months prior to construction. In this time the dry stability of the

weathered granite material reduced from approximately 13.7 kN (after 2 weeks) to 12.5 kN (after 2 months). The dry stability of the reclaimed asphalt pavement (RAP) material showed no reduction over a similar period. Laboratory mix design tests for the weathered granite and RAP reported dry stabilities of 23 kN and 28 kN respectively. [4]

4. According to Pat Dorkin, the subbase for all the sections was the G6 weathered granite. Additional Notes: • The average field density of the foamed bitumen material was close to or above 100% of Mod. AASHTO density for all sections, except for the upper

100 m portion of Section B1, which showed a density of 95.6% Mod AASHTO. • Sections A1 and A2 received a 19 mm single seal in 1998. • Sections B1, B2 and C received a nominal 30 mm asphalt surfacing in 1998. • Traffic on the left-hand side lane travels downhill and is southbound. Traffic loading is significantly lower on the downhill side compared to the uphill side. • Traffic on the right-hand side lane travels uphill and is northbound. Traffic loading is significantly higher on the uphill side compared to the downhill side. • Photographs of the road show an unpaved shoulder, and Dave Collings states there is no “formal” shoulder [4].

Page 130: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 6 MR504

Behaviour and Material Quality Indicators [1] A significant amount of rutting and deflection data was collected on the section since the foamed bitumen construction. For clarity, only the data collected on the uphill (right-hand side) lane will be summarized here, as this lane carries by far the heaviest traffic of the two lanes. Key aspects of data recorded in 1996 and 2004 are summarized in Table 2.

Table 2: Summary of Benkelman Beam Deflections 1996 Deflection, in Micron 2004 Deflection, in Micron Section Average 90th Percentile Average 90th Percentile

A (Km 3.0 to 3.2) 394 680 330 380 B (Km 3.2 to 3.6) 387 510 659 782 C (Km 3.6 to 3.7) 317 370 744 876

The material used for the foamed bitumen base was stockpiled for roughly three months prior to construction, and in this period significant decreases in strength and stiffness indicators were observed. Some of the final indicators, as measured prior to construction (i.e. after stockpiling for some months) are summarized in Table 3.

Table 3: Summary of Strength and Stiffness Parameters (after stockpiling) [1] Parameter Section A (RAP Source

Material) Sections B and C

(Weathered Granite Source Material)

Dry Stability (kN) 23.7 12.5 Soaked Stability (kN) 19.1 7.3 Retained Stability (%) 80.6 58.4 Resilient Modulus (MPa) 1453 No Test Dynamic Creep (MPa) 14.5 83.3

Observations following from Tables 2 and 3 and from other reported data [1] are:

• The Benkelman beam deflections do not show a consistent trend over the three sections. In the case of Section A, a decrease in deflections was noted between 1996 and 2004. However, in the case of Sections B and C, a significant increase in deflection was observed over this period.

• The intended compaction effort and resulting densities achieved from construction are: o Section A was compacted using a high and normal compaction effort, and a 100.9%

density was reported for this section. o Section B1 (first part) was compacted using a high compaction effort, and a 95.6%

density was reported for this section. o Section B1 (second part) was compacted using a normal compaction effort, and a

99.3% density was reported for this section. o Section B1 (third part) was compacted using a high compaction effort, and a 102.6%

density was reported for this section. o Section C was compacted using a high compaction effort, and a 101.4% density

was reported for this section. • Material samples taken from test pits opened in 2004 suggested that the subbase was of

G7 quality, with PI values ranging from 9 to 11 (average around 11), and CBR at 95% compaction of 23 and 32 per cent.

• Material samples taken from test pits opened in 2004 suggested that the subgrade was of G7 quality, with PI values ranging from 7 to 15 (average around 11), and CBR at 95% compaction of 20 and 26 per cent.

UCS, Resilient Modulus and ITS data from cores sampled in 2004 are summarized in Figures 1 and 2.

Page 131: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 6 MR504

0

1000

2000

3000

4000

5000

6000

7000

8000

UCS ResMod UCS ResMod UCS ResMod

UC

S (k

Pa) o

r Res

ilien

t Mod

ulus

(MPa

)10th PercentileAverage90th Percentile

Note: UCS is in kPa, Resilient Modulus in MPa

Section A Section B Section C

Figure 1: UCS and Resilient Modules Values from Cores Sampled in 2004

0

200

400

600

800

1000

1200

ITS ITS ITS

ITS

(kPa

)

10th Percentile

Average

90th Percentile

Section A Section B Section C

Figure 2: ITS Values from Cores Sampled in 2004

Performance Indicators [1] Rut depth measurements were recorded in 1997 and in 2004. However, the rut depths measured in 1997 were apparently only measured in the downhill lane, which is not the focus of this summary. Rut depths measured in 2004 showed the following [1]:

• For Section A, the rut depths ranged from 1 to 6 mm, with an average of approximately 3 mm.

• For Section B, the rut depths ranged from 0 to 11 mm, with an average of approximately 5 mm. Note that the asphalt overlay placed in 1998/1999 is likely to have filled the ruts present at the time.

• For Section C, the rut depths ranged from 0 to 2 mm, with an average of approximately 1 mm. Note that the asphalt overlay placed in 1998/1999 is likely to have filled the ruts present at the time.

Page 132: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 6 MR504

A visual assessment performed in 2004, roughly nine years after construction, showed that overall the road was in a reasonable condition. However, isolated failures were observed. In the uphill lane, two failures were noted on Sections B1 and B2 and consisted of deformation in the wheel path with severe shoving extending over areas of 10 and 20 m2 [1]. In Section C a crack was noted, running at a 45 degree angle towards the outer edge. It appears that this crack is not associated with fatigue. Dave Collings is of the opinion that these are subgrade failures due to inadequate drainage [4].

A 30 mm thick asphalt overlay was placed in 1988 or 1989 over the sections which had originally received a slurry surfacing or Fog Spray (Sections B and C). Apparently the slurry surfacing had started to strip and had exposed the base to abrasion and potholing. The section which had been left un-surfaced (i.e. Section A, in which RAP material was used in the Foamed Bitumen base) received a 19 mm single seal, also in 1988 or 1989. In July 2006, the general condition of the road is acceptable. However, there are two major edge failures at Km 3.2 and Km 3.3 that extend approximately 1 meter into the road and several (approximately 3 to 4) areas of structural failure about 20 metres in length. [3,5] The failure is so severe that the area cannot be patched and will need to be repaired relatively soon. These failures correspond to those mentioned earlier. The failures are still likely to be related to drainage inadequacies and poor maintenance. [4] There is minimal cracking and rutting on the section, but the road is in general need of rehabilitation. [5]

Traffic Loading Indicators [1] The traffic analysis presented in the 2004 paper by Collings et al [1] relied on the production figures of heavy industries in the area. The paper concluded that the cumulative E80’s that used the uphill lane over a nine year period was “almost 1 million”. After taking overloading estimates into account, the final estimated traffic accommodated by the uphill lane was 2 million E80’s over a nine year period since the foamed treated sections were opened to traffic [1]. The paper also quotes traffic counts taken in 2002 by the Kwazulu Natal DOT which states the AADT as 1234, with 33 per cent heavy vehicles [1]. It is noted that this estimate was more than double that measured at the end of 1995 (traffic data not shown), which suggests a significant growth rate. Data obtained from Kwazulu Natal DOT showed that in 2004 the AADT was 892, with 25% heavy vehicles in one direction [3]. Figure 3 shows the upper and lower estimates for the traffic loading on the section between 1996 and 2005. The points plotted in this figure represent the 2002 traffic count data (assuming a 50:50 directional split) and the 2004 traffic count data for different E80 per heavy factors. The lines represent lower and upper estimates of traffic growth. The cumulative traffic estimates for the upper and lower lines are roughly between 0.7 and 1.6 million E80s for the period 1996 to 2005. This figure agrees roughly with the analysis stated in the paper by Collings et al. [1]. Dave Collings is of the opinion that if a WIM station was on the road the actual traffic would be much higher than that shown as he convinced gross overloading occurs on the road. [4]

Page 133: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 6 of 6 MR504

0

100

200

300

400

500

600

700

800

900

1000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Year of Measurement

E80s

Per

Day

2002 Data, E80/Heavy = 2.02002 Data, E80/Heavy = 2.52002 Data, E80/Heavy = 3.02004 Data, E80/Heavy = 1.52004 Data, E80/Heavy = 2.52004 Data, E80/Heavy = 3Low er EstimateHigher Estimate

Figure 3: Summary of Traffic Growth Estimates

Factual Observations • A 30 mm thick asphalt was placed on Sections B and C and a 19 mm seal was placed on

Section A in 1998 (3 years of service). No rehabilitation has been performed since. • Rut depths averaged approximately 3 mm (Section A), 5 mm (Section B) and 1 mm

(Section C) in 2004, 9 years of service.

Observations Based on Estimates or Interpretations • Average Benkleman beam deflections on Section A decrease from 394 to 300 microns

from 1996 to 2004. On Section B and C the deflections increase between 1996 and 2004 from 387 to 659 microns and 317 to 744 microns, respectively.

• Visual assessment in 2004 (9 years of service) show the majority of the road was in a good condition with some isolated failures. By 2006 the road is in need of rehabilitation.

• The cumulative traffic estimates are 0.7 to 1.6 million E80s from 1996 to 2005 (calculated in this summary) and 2 million E80s (calculated by Collings, et al [1]).

Reviewed by: • Dave Collings, Loudon International. • Pat Dorkin, Kwazulu-Natal Department of Transport

Page 134: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 N11 Section 8

Road Type: Foamed bitumen Road Code: N11 Section 8 Constructed: 2003/2004 (rehabilitation) Limits Considered: Km 44.4 to 45.9 of northbound lane Project Description: N11-8 Ermelo to Hendrina Available Documentation: 1. Arcus Gibb and Simunye Consulting and Implementation Partnership (SCIP), “Reseal of

the N11-08 from Ermelo to Hendrina, Project N011-08-2002/1”, Preliminary Assessment Report, August 2002.

2. Arcus Gibb and Simunye Consulting and Implementation Partnership (SCIP), “Special Maintenance Hendrina Ermelo N11/08”, As-built Materials Test Results, Contract Number N011-08-2002/1, January 2003 to April 2004.

3. SANRAL PMS. 4. CTO Traffic Data. 5. Discussion with, and review by, Ben Botes, SCIP. 6. Mikros Traffic Data.

Area and Climate The road is situated between Hendrina and Ermelo in Mpumulanga, and was previously numbered R65. The average rainfall for the Hendrina region is 713 mm per year (1961 to 1990) with the maximum rainfall experienced in December. Frost occurs in winter and snow is experienced once in 5 years. According to TRH4, area is in the moderate region with a Weinert N-value of approximately 2 to 5. The area is, however, close to the wet region boundary, in which the N-value is 2.

Construction History • The original pavement was constructed in 1969 and consisted of the following structure:

[2, 3] o 150 C3 base o 150 C4 subbase o 150 G8 selected layer

• There is no information on the subgrade in the references [1,2,3]. However, from observations made during the rehabilitation, Ben Botes would classify the subgrade as a deeply weathered ferricrete with sandstone with a G8 classification. The moisture content was also above optimum. [5]

• The original pavement was subject to a 13.2 mm S1 seal in 1979, a S9 seal in 1987 and a 13.2 mm S1 seal in 1992. [2, 3]

• The road was rehabilitated in 2003 to 2004 using deep in situ recycling with the addition of foamed bitumen and cement. The pavement structure is as follows [2]:

o 20 mm AO (UTFC) o 180 mm BT2 (1.5% foamed bitumen and 1% cement)

• On this section, a portion of the pavement was rehabilitated with a G2 overlay. This overlay did not meet the construction density specifications, and was rebuilt by recycling with foam bitumen and emulsion. This section under consideration in this summary does not include any part of the section where the G2 overlay was rebuilt.

Page 135: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 N11 Section 8

Table 1: Summary of As-Built Information for ~Km 44.4 to 45.9 and trial pits [1, 2] Layer Property Specification Observations from As-Builts

Material Ultra thin friction course with crushed dolerite. Thickness 20 mm 20 mm, not measured.

Sieve size 19.0 4.75 0.60 0.075 Average 100 39 14 5

Grading

Std dev 0 7.8 2.1 0.6 Binder content 5.0% Average = 5.2%, Std dev = 0.3% Ref density MTRD Average = 2640 kg/m3, Std dev = 3.0 kg/m3 Voids Average = 7.1%, Std dev = 0.8% Marshall stability

Average = 10.2 kN, Std dev = 1.5 kN

Marshall flow Average = 2.6 mm, Std dev = 0.4 mm

Asphalt surfacing

Tack application

Average = 150 l/m2

Material Ferricrete in dark brown/brown silty sand Thickness Average = 110 mm Grading Average percentage passing 2.0 mm = 53%

Average percentage passing 0.075 mm = 26% Grading modulus = 1.83

PI Non-plastic or PI = 3 CBR @ 98% Mod Average = 81% Stabilisation No indication of any residual stabilisation.

Base, prior to rehabilitation1

TRH14 class Estimated G5/G6 based on available data (not shown in as-builts or assessment report)

Material Ferricrete and old seals Binder 1.5% foamed bitumen (suspect not measured value) Cement 1% Cem II 32.5 MPa (suspect not measured value) Thickness 180 mm 180 mm (suspect not measured value)2

Sieve size (mm) 19 4.75 0.425 0.075 % passing 95 59 32 10

Grading

The standard deviations given range from 1.5 to 7.7. UCS @ 100% Mod Average = 1709 kPa, std dev 363 to 444 kPa. ITS (dry @ 100% Mod Average = 404 kPa, std dev 27 to 49 kPa. PI Non-plastic Reference density

Average = 2078 kg/m3

Reference OMC

Average = 9.2%

Base, after rehabilitation

Compaction 100 % Average = 101.2%, Std dev ranges 0.8 to 2. Material Ferricrete and sandstone in dark brown/brown silty

sand Thickness Average = 95 mm Grading Average percentage passing 2.0 mm = 49%

Average percentage passing 0.075 mm = 25% Grading modulus = 1.90

PI 3 CBR @ 93% Mod 88% (one sample only)

Subbase, prior to rehabilitation1

TRH 14 class Estimated G5-G6 based on available data (not shown in as-builts or assessment report)

Material Ferricrete and sandstone in light yellow brown silty sand

Thickness Average = 200 mm Grading Average percentage passing 2.0 mm = 70%

Average percentage passing 0.075 mm = 36% Grading modulus = 1.37

PI Average = 3 CBR 71% (one sample only)

Selected layer prior to rehabilitation1

TRH 14 class Estimated G6-G8 based on available data (not shown in as-builts or assessment report)

Notes 1. From trial pits at Km 40.4 RHS and Km 46.2 LHS 2. It is difficult to measure the thickness of a recycled layer. The recycling machine is

however set to the thickness required, and is fairly accurate in achieving the desired thickness.

Page 136: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 N11 Section 8

Behaviour and Material Quality Indicators • Before the rehabilitation, DCP penetration rates for the base ranged between 1.09 and

6.42 mm/blow (2.6 mm average) and for the subbase ranged between 1.21 and 8.49 mm/blow (2.6 mm average). The DCP results indicate that the pavement layers were in an overstressed condition generally with the base in a failed condition. [1]

• Before the rehabilitation, in 2001 the average FWD deflection on the section was 864 microns with a 90th percentile of 988 microns. [3]

• After the rehabilitation, in 2005 the average FWD deflection was 528 microns with a 90th percentile of 775 microns. These deflections are high for a new pavement. [3]

Performance Indicators and Treatment History • Before the rehabilitation, the visual condition of the section showed the riding quality and

smoothness of texture in a warning state, and cracking, rutting, surface disintegration, deformation and drainage in a sound condition. [1]

• After the rehabilitation there are no roughness, rutting or visual condition data in the SANRAL PMS. [3]

• An informal visual survey by Ben Botes in late 2006 indicated that the pavement is still in a good condition, although in 1 or 2 places rutting is beginning. It is not clear whether this rutting is on the specific section under investigation. [5]

Traffic Loading Indicators • Traffic data from the CTO traffic counts between 2001 and 2005 are shown in Table 2

with the Mikros recorded data for 2001 to 2006. [4,6] The areas under the lines in the shaded area were calculated to roughly estimate the range of cumulative E80s carried since the rehabilitation. Based on those calculations, the estimated traffic between 2004 and 2007 is 0.62 to 1.13 million E80s. [4]

• In the Assessment Report, the estimated traffic for two design lives was calculated using growth rates of 2, 4 and 6% and yielded the following results [1]:

o 7 years: 1 to 1.25 million E80s o 15 years: 2.3 to 3.5 million E80s

• Ben Botes noted that since the rehabilitation the traffic growth has greatly exceeded expectations, which is confirmed by the Mikros traffic counts. [5]

Table 2. Traffic data, Average E80s per Day (Northbound)

Year CTO Data1

Mikros Data1,2

2001 282 203

2002 427 309

2003 320 260

2004 402 326

2005 832 781

2005 801

1. Station 892, N11-8 Km 38.8 2. Average shown for months

recorded

0

300

600

900

1200

1500

2000 2001 2002 2003 2004 2005 2006 2007Year

E80s

in N

orth

boun

d la

ne p

er d

ay

CTO DataMikros Data

Factual Observations • In 2005 (after 2 years of service) the average FWD deflection was 528 microns with a 90th

percentile of 775 microns. These deflections are high for a new pavement.

Page 137: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 N11 Section 8

Observations Based on Estimates or Interpretations • The road has carried an estimated 0.62 to 1.13 million E80s in the northern direction after

3 years of service. • In general, after 3 years of service the section is still in a relatively good condition visually,

although the deflections are relatively high. • For the purpose of calibration and validation of the materials and structural design

method, the pavement structure is as follows: o 20 mm AO (UTFC) o 180 mm foamed bitumen (previously C4, equivalent G5/G6, 1.5% foamed

bitumen and 1% cement) o 250 mm G6 to G8 subbase (previously selected layer and small part of

subbase) o G8 subgrade

Reviewed by: • Ben Botes, SKIP

Page 138: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 3 P24/1

Road Type: Foamed Bitumen Road Code: P 24/1 near Vereeniging Constructed: 1999 Limits Considered: Km 65.9 to 66.6, Westbound Slow Lane Available Documentation: 1. Keeve Steyn, “Reseal and Light Rehabilitation of Road P24-1 from km 65.50 to km

94.94”, Design Report, Braamfontein, August 1997. 2. Keeve Steyn, “The Rehabilitation of Roads P24/1 and P88/1 between Road P73/1 (R553)

and Km 6,20, District Vereeniging”, Project Document, Sunninghill, December 1998. 3. As-built records containing materials control data and mix design information. 4. Gautrans PMS.

Area and Climate The road is situated near Vereeniging in the Gauteng province. The road is located in an area that TRH4 (1996) classifies as moderate, with a Weinert-n value of approximately 2.4 and an annual rainfall of 750 mm [1].

Construction History According to Gautrans records, this road was constructed in 1964 as a natural gravel base over a natural gravel subbase. The road was rehabilitated in 1999, and part of this rehabilitation consisted of a foamed bitumen recycling operation. The investigation that was coupled with the 1999 rehabilitation [1] showed that – in the area under consideration – the road structure differed from that shown by the Gautrans records, and consisted of the following:

• 110 mm crusher run base; • 250 mm cemented natural gravel subbase (in two layers); • 330 mm selected layer (in two layers) consisting of clayey sand, and • Silty clay subgrade. The road was showing various levels of distress by 1997. In the area under consideration, the main forms of distress were extensive cracking and pumping. Rut depths varied from 4 to 11 mm, with an average of around 8 mm. The rehabilitation of this road consisted of various treatment types, which varied from a surface treatment to base reconstruction using foamed bitumen and emulsion treated crushed stone. Many of these treatment areas are short (less than 50 m in length), but the area selected for this assessment is a part of the westbound slow lane, located approximately between Km 66.0 and 66.7. This area was identified from the as-built records [3], and show that a foamed bitumen base treatment was placed in the westbound slow lane between Km 65.908 and Km 66.660 (752 m length) on 9 April 1999. Bitumen rubber asphalt was placed over this section on 27 May, 1999. For this segment, the following material indicators for the foamed bitumen base were retrieved from the as-built records [3]:

• Average layer thickness was 150 mm (no standard deviation is provided); • 2% O.P.C and 1.5% net 80/100 bitumen were used in the construction. • The source material consisted of the in-situ base layer, which should be crushed stone

according to [1]. • The percentages passing the 19, 4.75, 0.425 and 0.075 mm sieves are 98, 61, 10 and

4.8%, respectively. The material is thus relatively coarse with little fines. • The PI after treatment is noted as non plastic. • The reference density is indicated as 2096 kg/m3; • The reference optimum moisture content is indicated as 7.9%, and • The compaction is indicated as 104.8% with a standard deviation of 2.284%. The materials design sheets in the as-built records contain additional information on the source material (i.e. the existing crushed stone base) which shows the grading generally conforms to that of a G1/G2 material, with a Grading Modulus of around 2.65. The CBR at 100% compaction varied from 124 to 60% for the untreated base material.

Page 139: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 3 P24/1

The materials control sheets for the bitumen rubber asphalt show an average thickness of 32 mm with a standard deviation of 5.9 mm. The pavement after rehabilitation can thus be assumed to be as follows:

• 30 mm bitumen rubber asphalt surfacing; • 150 mm foamed bitumen crushed stone base, using 2% OPC and 1.5% net bitumen; • 210 mm cemented natural gravel subbase; • 330 mm selected layer (in two layers) consisting of clayey sand, and • Silty clay subgrade. No rehabilitation has been performed on this road since the major rehabilitation in 1999.

Behaviour and Material Quality Indicators • FWD deflections measured before the foamed bitumen rehabilitation varied from

approximately 140 to 500 micron, with an average of around 290 micron in the area under consideration. [1]

• A single DCP test conducted in the area under consideration before the foamed bitumen rehabilitation showed a penetration rate of around 2 mm/blow in the crushed stone base. The cemented subbase could not be penetrated and the selected layers had a penetration rate of around 4.5 mm/blow. The penetration rate in the subgrade was around 14 mm/blow [1].

Performance Indicators • Riding quality measurements from the Gautrans PMS [4] are given in Table 1. The road

is in a good condition based on these measurements. There is no apparent reason for the decrease from 1999 to 2005.

Table 1. Riding quality (HRI) [4]

1989* 1999 2005 95th percentile 2.1 3.6 4.6 5th percentile 1.5 2.1 1.0 Average 1.8 2.7 1.9

* before rehabilitation

• The visual condition index (VCI) is recorded in the Gautrans database for Km 65.87 to 69.94, which is a longer section than the section under consideration in this summary. The VCI is above 75 for every year from 2000 to 2006, indicating the pavement is in a good condition.

Traffic Loading Indicators • Based on two 24 hour classified counts of traffic, conducted in 1994, the AADT was

estimated as 4655 and 4736 with 14.7 and 23.8% heavy vehicles reported for the two counting stations, respectively. The estimated number of heavy vehicles thus varied from 696 to 1110 [1].

• Based on the above-noted traffic count data, the designers estimated that the number of heavy vehicles in 1997 were 1000 (assumed to be for all lanes, both directions) [1]. If a 50/50 directional split is assumed with a lane distribution factor of 0.95, then the estimated number of daily heavy vehicles over the area under consideration was 475 in 1997.

• Gautrans traffic data were collected from 1998 to 2005, and are shown in Table 1. The E80s per day in the slow lane were calculated using the percentage of heavies in the ADT, and multiplying by 0.5 (reduce to one direction), 0.95 (lane distribution factor) and E80 per heavy factors of 2 and 3.5. The resulting E80s in the slow lane per day are shown in the figure in Table 2. The cumulative E80s from 1999 to end 2005 are estimated as the area under the lines in the figure, and give a range of 2 to 4.5 million E80s.

Page 140: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 3 P24/1

Table 2: Gautrans PMS Traffic Data Date AADT %Heavy

1999/09/13 5618 17 2000/09/11 8901 10 2000/11/01 5512 18 2001/11/05 4802 15 2002/11/11 5822 19 2002/11/11 5711 19 2004/06/09 5380 20 2004/06/09 5398 20 2005/04/08 4626 26 2005/04/08 4410 16

0

500

1000

1500

2000

2500

3000

1999 2000 2001 2003 2004 2005Date

E80

per s

low

lane

E80s/heavy = 2E80s/heavy = 3.5

Factual Observations • No rehabilitation has been performed on this road since the major rehabilitation in 1999, 7

years of service. • Riding quality (HRI) measurements in 1989, 1999, and 2005 are less than 2.7.

Observations Based on Estimates or Interpretations • The average FWD deflection prior to rehabilitation was 290 microns. • Based on riding quality, the road is in a good condition after 6 years of service • Based on visual assessments, the road is in a good condition after 7 years of service. • Cumulative traffic estimated from 1999 to 2005 (6 years of service) is 2 to 4.5 million

E80s. Reviewed by: • Andre Ungerer, Goba • Pablo Balmaceda, formerly Goba

Page 141: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 P243/1

Road Type: Foamed Bitumen Road Code: P243/1 Constructed: 2000 (rehabilitation) Limits Considered: Km 14.4 to 21.140 of East- and Westbound Lanes Project Description: P243/1 near Vereeniging Available Documentation: 1. Biyana, Van Tonder and Associates, “Evaluation Report on Roads D1884 and P243/1 in

the Vereeniging/Benono Districts”, April 1998. 2. Biyana, Van Tonder and Associates, “The Rehabilitation of a Section of Road P243/1 in

the Vereeniging District, Materials Control Data Sheets”, January 2000 to October 2000. 3. Gauteng Department of Public Transport, Roads and Works PMS.

Area and Climate The road is situated near Vereeniging and the Vaal Dam in Gauteng. The specific road starts at road D83 and ends at the Gauteng and Mpumulanga border.[1] According to TRH4, the area is in a moderate climatic area with a Weinert N-value of 2 to 5.

Construction History • The road is a two lane surfaced rural road with gravel shoulders [1]. • The original pavement was constructed in 1988 and consisted of: [1]

o Stone seal o 150 mm C4 stabilised gravel base o 150 mm C4 stabilised gravel subbase o 150 mm G6 natural gravel upper selected layer o 150 mm G9 natural material lower selected layer

• No maintenance occurred on the road between 1988 and the rehabilitation in 2000. • As part of the evaluation assessment, 2 test pits were opened on the specific section.

Material results from these test pits are given in Table 1. • The pavement was rehabilitated from January 2000 to October 2000. The existing

surfacing, base and some of the subbase were deep in situ recycled and treated with foamed bitumen to create a new base layer. Relevant information from the as-built records is shown in Table 2. The rehabilitated pavement structure is as follows:

o 13.2 mm single seal o 20 mm ACM o 250 mm foamed bitumen treated base layer

• Based on the test pits and as-built information, the pavement structure is assumed to be: o 13.2 mm single seal o 20 mm ACM o 250 mm foamed bitumen treated base layer (2% cement, 1.8% foamed

bitumen, G5/G6 previously cemented parent material) o ~150 mm subbase (G8) o Subgrade (G8-G10)

Page 142: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 P243/1

Table 1: Summary of test pit information prior to rehabilitation [1] Layer Property Test pit 4 (Km 16.2) Test pit 5 (Km 19.6) Asphalt 13.2 and 6.7 mm double seal

Material description

Loose to medium dense sandy gravel, weakly cemented dolerite gravel. Slightly moist, dark brown.

Thickness 255 mm 165 mm Grading GM = 2.28 GM 2.26 PI 8 SP MC/OMC 8.2%/10.1% 9.0%/11.5% CBR @ 98% Mod

82 87

Base

TRH14 class G6 G5 Material Medium dense sandy gravel,

weakly cemented ferricrete and quartz. Slightly moist, pale red.

Thickness 215 mm Grading GM = 1.37 PI 8 MC/OMC 21.7%/15.5% CBR @ 93% Mod

63

Subbase

TRH 14 class

Included in base layer.

G6 Material description

Firm sandy clay with gravel, chert. Slightly moist, dusky red.

Firm sandy clay, ferricrete and quartz. Slightly moist, pale red.

Thickness 165 mm 90 mm Grading GM = 1.18 GM = 0.84 PI 14 18 MC/OMC 16.0%/14.0% 23.1%/15.7% CBR @ 93% Mod

37 30

Selected subgrade

TRH14 class G8 G8 Material description

Medium dense clay with gravel, andesite. Slightly moist, pale red

Clay with gravel, ferricrete and quartz. Slightly moist, pale red.

Thickness 150+ mm 100+ mm Grading GM = 1.46 GM = 1.1 PI 15 19

Subgrade

MC 19.5% 17.4%

Page 143: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 P243/1

Table 2: Summary of test pit information prior to rehabilitation [1] Layer Property Specification Value

1st spray 2.1 to 2.2 l/m2

Bitumen rubber TSG 80 Average spray rate= 2.16 l/m2, 10th percentile=2.09

Aggregate 0.011 m3/m2

13.2 mm Quartzite Average spread rate = 0.01 m3/m2, min=0.011

Seal

Final spray 30% fogspray 0.65l/m2

SS60 Average spray rate = 0.66 l/m2, 10th percentile=0.64

Description ACM thin asphalt layer Thickness 20 mm Average = 26.6, all averages in spec.

Sieve 13.2 4.75 06 0.075 Ave 100 72 24 7 Min 100 63 22 6 Max 100 72 24 7

Grading

Standard deviation in segments is 0 to 4.5%. Binder content 5.3% Average = 5.3%, 10th percentile = 5.2% Compaction 92% of Rice Average = 92.5%, 10th percentile = 91.2% Marshall stab. 8-12 Average = 12.9%, 70% out of spec Marshall flow 2-4 Average = 3.6, 6% out of spec ITS Min 800 Average = 1482 kPa, all in spec

Asphalt

Voids Average = 3.6%, 10th to 90th percentile=2.9 to 4.9% Material description

Weathered dolerite and dec ferricrete

Thickness 250 mm 250 mm (suspect not measured value) Binder Foamed bitumen, 1.8% 80/100 pen bitumen Cement 2% Cem 1 42.5 MPa

Sieve (mm) 19 4.75 0.425 0.075 Average 93.9 67.2 13.8 5.7 Minimum 88 60 9 3 Maximum 100 76 31 13

Grading

Standard deviation in segments is 0 to 18%. PI Average = 5.1, 90th percentile = 7. Ref OMC Average = 12, 90th percentile = 12.6. UCS @ 100% Average = 2549, 10th percentile = 2118.

Base

Compaction 100% mod AASHTO

Average = 100.8, 10th percentile = 99.32. 23% out of specification.

Behaviour and Material Quality Indicators • Before the rehabilitation, FWD deflections measured in April 1998 were very variable with

an average peak deflection of 733 microns and a 90th percentile of 1218 microns. [1] • Before the rehabilitation, the DCP DSN800 was 69 and 114 at the two test pit locations.

[1] • No FWD, DCP or other behaviour indicators have been measured subsequent to the

2000 rehabilitation.

Performance Indicators and Treatment History • Before the rehabilitation, in 1997, the visual condition of the road was rated as fair to

poor. [1] • Before the rehabilitation, in 1998, the visual condition of the road showed fairly extensive

crocodile and longitudinal cracking. Pumping of the cracks was evident along the section. Failure was observed at various parts of the section. [1]

• After the rehabilitation, the VCI decreased steadily from approximately 93 in 2001 to 72 in 2006. [3]

• After the rehabilitation, rutting was measured in 2005. The average rut depth is 4.7 mm with a 10th and 90th percentile of 3.7 mm and 5.8 mm, respectively. [3]

• After the rehabilitation, riding quality has only been measured in 2005. The average IRI in the left wheel path is 1.3 with a 10th and 90th percentile of 1.0 and 2.0, respectively. The riding quality of the road is therefore good. [3]

• No treatments have been applied to the road since the 2000 rehabilitation. [3]

Page 144: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 P243/1

Traffic Loading Indicators • The evaluation report indicated that the total expected E80s in a 20 year design life is

410 000 to 670 000. [1] • Traffic data contained in the GDPTRW PMS is shown in Figure 1 [3]. The traffic was

converted to E80s using E80/heavy factors of 2 and 3.5. Using the shaded areas the cumulative traffic carried from the rehabilitation to April 2007 ranges from 170 000 to 525 000 E80s.

0

20

40

60

80

100

120

140

1980

/01/

01

1982

/09/

27

1985

/06/

23

1988

/03/

19

1990

/12/

14

1993

/09/

09

1996

/06/

05

1999

/03/

02

2001

/11/

26

2004

/08/

22

2007

/05/

19

E80s

(one

dire

ctio

n)

E80/heavy = 2E80/heavy = 3.5

Figure 1. E80s per day per direction [3]

Factual Observations • Since the 2000 rehabilitation, no treatments have been applied to the road in the 6 years

of service. • Rutting measured in 2005 (5 years of service) averaged 4.7 mm with a 90th percentile of

5.8 mm, indicating a good condition. [3] • Riding quality (IRI) measured in 2005 (5 years of service) averaged 1.3 with a 90th

percentile of 2.0, indicating a good condition. [3] • The VCI after 6 years of service is 72 and has been steadily decreasing since the

rehabilitation.

Observations Based on Estimates or Interpretations • Based on the factual observations, the road is still in a reasonable condition. The road

has carried an estimated 170 000 to 525 000 E80s in each direction after 6 years of service.

Reviewed by: Review not received in time for inclusion in report.

Page 145: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 4 R22-4 (MR439)

Road Type: Foamed Bitumen Road Code: R22-4 (MR439) Constructed: 2002/2003 (rehabilitation) Limits Considered: Km 0 to 31.2 (Km 85.04 to 116.27 on MR439) Project Description: R22-4 (MR439) Mseleni to Phelandaba Available Documentation: 1. Millar, D. and B. Nothard, “The Blending of Marginal Materials using Foamed Bitumen”,

8th CAPSA, 2004. 2. Siyenza Engineers, “The Construction of MR 439 (R22-4) from Mseleni to Phelandaba”,

As-built Materials Test Data, Contract No. NRA S 5803001/1, January 2002 to July 2003. 3. SANRAL PMS. 4. Discussion with, and review by, Barry Nothard, Siyenza Engineers. 5. AA Loudon and Partners and SNA Civil and Development Engineers, “The Construction

of MR439 From Mseleni (km 82.70) to Phelandaba (km 116.27) in Northern Kwazulu Natal”, Project Document, Materials Investigation and Utilisation, Volume 6, July 2001.

Area and Climate The road is situated between Mseleni and south of Phelandaba in Kwazulu-Natal. The road was named MR439, but has subsequently been renamed R22 Section 4. The average rainfall for the closest weather station is 1000 mm per year (1952 to 1997). According to TRH4, area is in the wet region with a Weinert N-value of less than 2.

Construction History • The original road was a gravel track. To construct the pavement, the existing materials

on the track, which contained some calcrete, were ripped up and stockpiled on the side of the road. The dune sands underlying the calcrete were left in place. Additional dune sands were excavated from borrow pits adjacent to the side of the road and placed on the road and compacted to a formation level. Up to 1.5 meters of material was imported. Selected dune sand was then brought in to form the selected layers. The stockpiled calcrete and a blend of other materials were then brought in to form a working platform for construction traffic. These materials were then recycled in situ to a depth of 250 mm with the addition of foamed bitumen and lime to form the treated base layer. The base was trafficked to allow the bitumen fines to flush up and seal the surface which allowed the slurry to be placed without a prime. [4]

• The final pavement structure was as follows: [2, 4] o 6 mm course slurry seal o 250 mm foamed bitumen treated base (G6 and G7 parent material with 3.5 or

4% foamed bitumen and 2% lime). o G7-G9 sand subgrade (not a typical G7 material)

• The final slurry thickness ranged from between 2 mm and 10 mm because the COLTO base tolerance is 20 mm, which made construction of the thin slurry difficult.

• Soon after the slurry seal was applied it showed distress (cracking and stripping) in some areas, specifically where heavy timber trucks turned, in areas where the base was not trafficked before the application of the slurry seal, i.e., at bus bays, and in areas where the slurry was only 2 mm thick, which occurred along some longitudinal joints. [4] It was suggested that an inverted seal or thin asphalt may have been a more appropriate surfacing.[1] A prime was not applied before the slurry as described above, which in hindsight may have provided an improved bond for the slurry seal [4].

• The blends of materials used in the base layer varied across the section. Table 1 contains a summary of the various blends used in the sections. All the blends essentially produced a final material with similar enough behaviour so that the materials can be classified in the same class. Therefore, in this summary the 5 sections shown in Table 1 are combined into one section and the material properties for all 5 sections are averaged in Table 2.

Page 146: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 4 R22-4 (MR439)

Table 1. Materials blends used to construct base layer [2] Kilometres TRH14

class Material Blend Foam

(%) Lime (%)

R22-4: 0 to 8.94 (MR 439: 85.04 to 93.98)

G7 40% existing calcrete wearing course 20% existing silty sand 40% red silty sand ex Tshongwe borrow pit

4 2

R22-4: 8.94 to 10.12 (MR 439: 83.98 to 95.16)

G7 40% existing calcrete wearing course 30% sandstone 30% sand ex Kwa Joba borrow pit

4 2

R22-4: 10.12 to 20.4 (MR 439: 95.16 to 105.800)

G7 50% existing calcrete wearing course 50% existing sand

4 2

R22-4: 20.4 to 23.4 (MR 439: 105.800 to

108.440)

G7 60% existing calcrete wearing course 40% crusher dust ex Nyawo Stone

4 2

R22-4: 23.4 to 31.20 (MR 439: 108.44 to 116.27)

G7 60% existing calcrete wearing course 40% crusher dust ex Nyawo Stone

3.5 2

Behaviour and Material Quality Indicators • The only pavement behaviour data in the SANRAL PMS are FWD deflections measured

in 2005. The average peak deflection was 501 microns, with a 90th percentile of 723 microns. [3]

Performance Indicators and Treatment History • In 2007 the condition of the base layer was good. In approximately 4% of the section the

slurry seal was wearing thin and pealing. This may be because the slurry was not primed, which in hindsight is recommended. [4]

• In 2007 a 13 mm and 6.7 m double seal was applied to the whole section to protect the slurry seal. [4]

• There are no performance data in the SANRAL PMS for the section under discussion.

Page 147: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 4 R22-4 (MR439)

Table 2: Summary of As-Built Information for R22-4 Km 0 to 31.2 (MR439: Km 85.04 to 116.27) [2]

Layer Property Specification Observations from As-Builts Slurry seal No as-built information given.

Binder 80/100 pen bitumen

Average penetration = 87, all values in spec.

Lime 2% Carlime Thickness 250 mm Average = 251 mm, Std dev ranges from 3 - 12 mm

Sieve size (mm) 19 4.75 0.425 0.075 Average1 97 89 72 7 Minimum 87 62 28 4 Maximum 100 100 90 13

Grading

Std dev ranges from ITS (dry) @ 100% Mod 2 Average = 297 kPa, Std dev ranges from 8-187 kPa ITS (soaked) @ 100% Mod 2 Average = 187 kPa, Std dev ranges from 8-172 kPa PI Non-plastic or slightly plastic (before and after

recycling) Reference density

Average = 1859 kg/m3

Reference OMC

Average = 9%

Base, after rehabilitation

Compaction 98% of Mod AASHTO

Average = 100% kPa, Std dev ranges from 1-9%

Material Sand Light brown silty sand Thickness 300 mm Average = 302 mm, Std dev ranges from 2.2-9.6mm

Sieve size (mm)

19 4.75 0.425 0.075

Average1 100 100 90 4 Minimum 93 92 81 2 Maximum 100 100 99 7

Grading

Std dev ranges from 0 to 8%. PI <12 Non-plastic CBR Average = 20%, Std dev ranges from 3-13.8% Reference density

Average = 1712 kg/m3, 10th percentile = 1676 kg/m3

Reference OMC

Average = 12%, Std dev ranges from 9 to 16%

Subgrade

Compaction 100% Mod. AASHTO

Average = 103%, Std dev ranges from 0.4 to 7%

Notes 1. The grading of the 5th section is slightly more well graded than the other four sections. 2. ITS tests done using 100mm Marshall briquettes

Traffic Loading Indicators • There are no CTO traffic counting stations on or near the section. • Traffic data from a count in 1993 are included in the project document [5], and are shown

in Table 2. The E80s per day are calculated using E80/heavy factors of 2 and 3.5. Table 2. Traffic data

E80/day Count Position ADT (veh/day)

No of Heavies E80/heavy = 2 E80/heavy = 3.5

Lower Mkuze – Mbaswana 381 76 152 266 Mbaswana – Mseleni 431 100 200 350 Mseleni – Phelandaba 450 188 376 658 • Estimates of the traffic made in October 2005 were 600 vehicles per day with 22%

heavies. For E80/heavy factors of 2 and 3.5 this gives 264 and 462 E80/day, respectively. [4]

• Using these traffic estimates, a 50/50 directional split and growth rates of 6 and 10%, the range of traffic carried from 2002/2003 to 2007 is 0.3 to 2.2 million E80s. It is recommended that if further monitoring of this section is done, a traffic count is included to obtain more accurate traffic estimates.

Page 148: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 4 R22-4 (MR439)

Factual Observations • In 2005 (after 3 years of service) the average FWD deflection was 501 microns with a 90th

percentile of 723 microns. These deflections are high for a relatively new pavement. The deflection measurement may have been influenced by the slurry thickness.

Observations Based on Estimates or Interpretations • In 2007, the road is in a good condition as a seal was applied. • The road has carried an estimated 0.3 to 2.2 million E80s in the northern direction after 4

years of service. • For the purposes of development and validation of the structural design method, the

assumed pavement structure is as follows: o 6 mm course slurry seal o 250 mm foamed bitumen treated base (G6 and G7 parent material with 3.5/4%

foamed bitumen and 2% lime). o G7-G9 sand subgrade (not a typical G7 material)

Reviewed by: • Barry Notnard, Siyenza Engineers

Page 149: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 3 Same-Himo

Road Type: Foamed Bitumen Road Code: Same-Himo Road Constructed: 1992 (reconstruction) Limits Considered: Entire Project Project Description: Same-Himo Pilot Project, Tanzania Available Documentation: 1. Overby, Johansen and Mataka, “Bitumen Foaming: An Innovative Technique used on a

Large Scale for Pavement Rehabilition in Africa. Case Study: Same-Himo Monitored Pilot Project”, 8th CAPSA, 2004.

2. Communication with Mr Charles Overby. 3. Overby, C., “Detailed investigation of a bitumen stabilised pavement”, I 006 (Institutional

Cooperation between Ministry of Works, Tanzania and the Norwegian Public Roads Administration, Norway), November 1996.

Area and Climate [1] The road is situated in Tanzania, in an area that receives an average annual rainfall of 850 mm. The area can therefore be classified as wet.

Construction History [1] This road was originally constructed in 1968 to 1969, and consisted of a cement stabilized base course and a double seal. Reconstruction was performed in 1992, and followed two approaches. These were:

A. Milling of the existing surface and cement stabilized base in one operation. The material was then stabilized with 4.4% bitumen using the foamed bitumen process. The thickness of the foam stabilized base was 175 mm.

B. Construction of a premixed base consisting of natural gravel premixed with foamed bitumen in a cold process at the borrow pit. This material was then transported to the road and placed with pavers to a thickness of 150 mm.

The design requirements for the foamed bitumen base were as follows:

• Marshall stability greater than 6 kN; • Marshall flow between 2 and 4 mm; • Minimum E-modulus of 1600 MPa; • Bitumen content of 4.4% (this translated to 80 to 100 kg/m3 of bitumen in the compacted

mix). Evaluation of test data showed that the voids in the compacted base varied considerably and were generally around 10%. The subbase is approximately 100 mm thick and consists of scarified or imported materials, with a quality roughly corresponding to a G6 class. Based on more than 100 samples, the properties of these materials were determined as:

• Average CBR of 40% (4 days soaked at 95% mod AASHTO); • Average field density of around 96.7% Mod AASHTO, and • Optimum moisture content roughly between 5 and 13 %. Subgrade investigations were performed before and after construction, and indicated that the subgrade had the following properties:

• CBR greater than 10% at 93% Mod AASHTO; • CBR swell maximum of 1.5%; • Compaction of 93% Mod AASHTO. A fairly detailed monitoring of the performance of this road was performed since construction, and the results of this programme are summarized in [1]. The observations were taken on

Page 150: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 3 Same-Himo

three 100 m long sections, each of which represented a different construction method. These methods or sections are defined as follows: • Section 1: Base consisting of natural gravel premixed with foamed bitumen. • Section 2: Premixed foamed bitumen base consisting of natural gravel base material with

imported material. The old cement treated based was milled and left in place to act as the subbase. [2]

• Section 3: In-situ milled material, mixed with foamed bitumen. The sections were surfaced with a double chip seal (19 mm and 12 mm) with an 80/100 penetration grade bitumen. Key observations from the monitoring programme of these three sections are noted below.

Behaviour and Material Quality Indicators [1] • Benkelman Beam deflection measurements were recorded at 100 m intervals over a

section which included Section 1. The measurements used the rebound method at a load of 8100 kg and a tyre pressure of 590 kPa. These deflections vary considerably depending on the lane and wheel-path. Overall, the deflections vary between 250 micron and roughly 1000 micron.

• Core samples were extracted at different times after construction, and were split and tested using Indirect Tensile Tests at a one hour preconditioning temperature of 29 degrees Celcius. The indirect tensile test stiffnesses show a fairly consistent trend for all three test sections. From graphs included in [1], it appears that the stiffness of the foamed bitumen material increased slightly over the first three years in service. Thereafter, the stiffness gradually decreased. Overall, and with the exception of one data point, the stiffnesses for all sections and all years remain in the range of 3000 to 5000 MPa, which suggests an intact, cohesive material.

Performance Indicators [1] • Rut depth measurements were taken at various stages over a 11 year service period. The

90th percentile rut is generally below 9 mm. Section 2, as defined above, showed a 90th percentile rut of 12 mm. In all instances, most of the rut initiating soon after construction. The rutting that developed over the bulk of the service period appears to be less than 6 mm in all instances.

• Visual assessments show no pothole or significant crack development. An exception was noted in an area that is periodically flooded, and which showed some crack development. Overall, the pavement condition at the end of an 11 year service period is described as excellent.

• The pavement has received no resealing in roughly an 11 year service period. On some sections, crack sealing has been performed.

• In 2006 and 2007, the road is in a fair condition with some hairline cracking. [2]

Traffic Loading Indicators [1] • The design traffic estimate is for 5 million E80’s in a 15 year design period. [1,3] • The AADT in 1991 (time of completion of construction) is 580 with 43% heavy vehicles

and in 2000 is 870 with 36% heavy vehicles. These values are for both directions. [3] • The number of E80s accumulated by 2000 is 2.1 million in the northbound lane and 1.4

million in the southbound lane. [3] • The number of loads that had used to road by the end of the monitoring period (roughly

2003), appear to be close to 4 million E80s. It is noted that the pavement accommodated a “considerable amount of severely overloaded axles in the early part of the design period”. [1]

Page 151: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 3 Same-Himo

Factual Observations • The pavement received some crack sealing, but no seals before 2003. • The number of E80s accumulated by 2000 (8 years of service) is 2.1 million in the

northbound lane and 1.4 million in the southbound lane. [3] • The E80s accumulated by the end of 2003 (11 years of service), the end of the monitoring

period, appear to be close to 4 million E80s. • The 90th percentile rut is generally below 9 mm. Section 2 showed a 90th percentile rut of

12 mm. In all instances, most of the rut initiating soon after construction. The rutting that developed over the bulk of the service period appears to be less than 6 mm in all instances.

• Visual assessments show no pothole or significant crack development, apart from some cracking in a section which is periodically flooded. Overall, the pavement condition at the end of an 11 year service period is described as excellent.

• In 2006, the road is still in a good condition. [2]

Observations Based on Estimates or Interpretations • Benkelman Beam deflections vary between 250 micron and roughly 1000 micron. • ITT tests showed that the stiffness of the foamed bitumen material increased slightly over

the first three years in service. Thereafter, the stiffness gradually decreased. Overall, and with the exception of one data point, the stiffnesses for all sections and all years remain in the range of 3000 to 5000 MPa, which suggests a intact, cohesive material.

Reviewed by: Review not received in time for inclusion in report.

Page 152: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 1 of 6 TR16-3 (R27)

Road Type: Foamed bitumen Road Code: TR 16 Section 3 (R27) Constructed: 2003/4 Limits: East- and Westbound Km 7.4 to Km 14.4 and Km 25.0 to Km 30.0 Project Description: Trunk Road 16 Section 12 (R27) between Nieuwoudtville and

Calvinia Available Documentation: 1. SNA Civil and Development Engineers (Pty) Ltd and Letaba Consulting Engineers CC,

“The Rehabilitation of Trunk Road 16 Section 3 Between Nieuwoudtville and Calvinia”, Detail Assessment Report, Project No: NC 068, November 2002.

2. As-Built Materials Data, The Rehabilitation of 21.3 km of Trunk Road 16 Section 3 between Nieuwoudtville and Calvinia”, Contract No. PWH 11/03.

3. SANRAL PMS. 4. Discussion with, and review by, Johan Opperman, SNA. 5. Discussion with Kobus van der Walt, SANRAL. 6. Discussion with Adrian Skea, UWP, Route Manager.

Area and Climate The road, which is part of Route R27, is situated in the Northern Cape between Nieuwoudtville and Calvinia in the Namakwa District Municipality. The average annual rainfall from 1961 to 1990 is 228 mm.[1] In TRH4 the area is classified as a dry region with a Weinert N value exceeding 5.

Construction History The original pavement was constructed by the then CPA between 1958 and 1964. The pavement structure was as follows: [1] • 19 mm seal with slurry (Cape Seal) • 100 mm G3 base (graded crushed stone with fines added) • 100 mm G5 subbase (natural gravel, min CBR 45@ 95%, PI 10 max) • 150 mm G5 upper selected subgrade (natural gravel, min CBR 45@ 95%, PI 10 max) • 250 mm G9 lower selected subgrade (gravel-soil, min CBR 5 @ 90%) The following seals were applied to the original pavement [1]: • Km 0 – 9: 7 mm seal in 1978, 13 mm seal in 1988 • Km 9 – 11: 13 mm seal in 1978, 13 mm seal in 1988 • Km 11 – 13: 7 mm seal in 1979, 13 mm seal in 1988 • Km 13 – 14: 13 mm seal in 1979, 13 mm seal in 1988 • Km 15 – 28: 13 mm seal in 1979, 7 mm seal in 1988 • Km 28 – 47: 13 mm seal in 1980, 7 mm seal in 1990 • Km 21 – 28: fogspray, date not given Although the 2003/4 construction involved Km -0.06 to 16.1 and Km 25 to 30, there is some evidence of reconstruction in 1997 to 1998 from Km 0 to 7.15 and Km 14.7 to 23.5. This entailed reconstruction of the upper 150 mm of the pavement layers and treating with 2% lime and sealing with a 13.2 mm/grit seal [1]. These sections are omitted from this summary due to the uncertainty of exactly what the reconstruction involved. Prior to the rehabilitation, as part of the detailed assessment, 3 trial pits were opened on the sections under discussion. The pavement structure indicated by the trial pits is shown below. Material data from the trial pits are summarized in Table 1 in the shaded rows. • G5 base • G5 subbase • G5/G6/G8 subgrade

Page 153: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 2 of 6 TR16-3 (R27)

Table 1: Summary of material properties from detailed assessment [1] and as-builts [2] Layer Property Observations from As-Builts

Description 13.2 mm and 6.7 mm seal with SBR modified binder Binder Penetration: Average = 92, Range 87 to 97 (whole section)

Softening point: Average=52%, Range 50 to 61 (whole section) Prime Spray rate average = 0.75 l/m2, spec = 0.75 l/m2, 27% of measurements less than spec.4 1st spray Spray rate average = 0.86 l/m2, spec = 0.85 l/m2, 24% of measurements less than spec. 4 Aggregate 1st layer (13.2 mm stone) Spread rate average = 100.7 m2/m3, spec = 105 m2/m3, 64% of measurements less than spec. 4 2nd spray Spray rate average = 0.76 l/m2, spec = 0.75 l/m2, 27% of measurements less than spec. 4 Aggregate 2nd layer (6.7 mm stone) Spread rate average = 201.6 m2/m3, spec = 200 m2/m3, 42% of measurements less than spec. 4

Seal

Final spray Spray rate average = 0.81 l/m2, spec = 0.80 l/m2, all measurements in spec. 4 Description Crushed dolerite/tillite/shale and weathered dolerite in silty sand Grading Grading close to G1-G3 in almost all cases, and meets G4 in most cases.

Grading modulus: 2.02 – 2.38. Thickness Base: 75 – 95 mm

Subbase: 160 – 230 mm Plasticity Index Ranges from 4 - 7 Class A-1-a, A-1-b, A-2-4

Base and subbase: G5

Base and subbase1 (before rehab)

CBR Not measured on specific sections under discussion. Ranges from 64 - 92 @ 98% density on adjacent sections. Thickness Average reported as 200 mm, no standard deviation given, therefore assumed not measured value.5 Binder Content: 2.5% (assumed not measured value)6

Penetration: Spec 80 – 100 pen, Average = 90, 6% out of spec (whole section) Softening point: Average2=47.2%, Range 46 to 50.4 (whole section)

Cement 1.5% Cem 1 42.5 MPa7 Material class Majority is A-2-6

P 19.0 P 4.75 P 0.425 P 0.075 Average 86 60 24 8 Range 63-97 27-78 12-44 3-15

Grading

Standard deviations are high. GM spec = 1.5

UCS (at 97% compaction) Average = 2078 kPa, 10th - 90th percentile = 1462 to 2599 kPa (using modified mould) ITS (at 97% compaction) Average = 227 kPa, 10th - 90th percentile = 150 to 314 kPa PI (After treatment) Non-plastic Reference density Average = 2162 kg/m3, Range 2083 to 2220 kg/m3 Reference OMC Average = 6.6%, Range 4.2 to 9.0%

Foamed bitumen base (after rehab)

Compaction Average = 98.5%, Range 96.3 to 100.8% (spec is 98%, 28% out of spec)4

Page 154: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 3 of 6 TR16-3 (R27)

Layer Property Observations from As-Builts Description Natural gravel Thickness Average reported as 200 mm, no standard deviation given, therefore assumed not measured value.5,8

P 19.0 P 4.75 P 0.425 P 0.075 Average 90 73 28 10

Grading

Range 82-97 55-97 20-36 7-14 CBR (@95%) Average = 64%, 10th - 90th percentile = 43 to 83% PI (After treatment) Average = 3.2, Range SP to 6 Reference density Average = 2092 kg/m3, Range 1124 to 2270 kg/m3 Reference OMC Average = 7.0%, Range 5.2 to 8%

Subbase (after rehab)

Compaction Average = 97 %, Range 96 to 109 % (spec is 95%, all in spec4)

Description Weathered dolerite/mudstone and silty sand/sand Thickness 2 layers, total thickness = 220 to 300 mm Grading Grading modulus = 1.65 to 2.28 Plasticity Index Ranges from non-plastic to 8

Subgrade (before rehab)

Class A-1-a, A-1-b, A-2-4, A-2-6 G5

Note: 1. Of the three trial pits, in two the base and subbase were combined into 1 layer, hence the results are reported together. 2. Averages given are determined from averages given in as-builts for short sections. 3. Of the three trial pits, in two the subgrade was reported in two layers. The results are reported as one layer herein. 4. Allowable tolerances were not considered when determining the percentage of measurements not meeting the specification. 5. Thickness of layer monitored during recycling process by means of dips. [4] 6. Binder content monitored by means of dips from bulk tanker supplies. [4] 7. Monitored in terms of number of bags applied. [4] 8. Johan Opperman states layer thickness was 175 mm, which disagrees with as-builts. [4]

Page 155: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 4 of 6 TR16-3 (R27)

The road was rehabilitated from October 2003 to August 2004 and involved reconstruction in half-widths by recycling the existing surfacing, base layer and the underlying existing subbase with the addition of foamed bitumen and cement to create a new base layer 200 mm thick. Traffic was accommodated in half-widths on either the existing surface or on the completed half-width construction. The original surfaced lane widths of approximately 3.2 m were widened to 3.7 m, consisting of a 3.4 m lane and a 0.3 m surfaced shoulder. Gravel shoulders of 0.95 m width were reinstated adjacent to the surfaced lanes. The crossfall of the road was corrected to 2% or superelevation. [4] In the widened section, a 175 mm subbase was reconstructed prior to construction of the overlying base. A G10 fill and 150 mm roadbed prep were constructed below the subbase. The surfaced area of the road was a 13.2 mm and 6.7 mm modified binder double seal. A summary of the relevant material data from the 2003/4 as-built records is included in Table 1 in the unshaded rows. The pavement structure is as follows [2, 4]: • 13.2 mm and 6.7 mm modified binder double seal • 200 mm foamed bitumen treated natural gravel (1.5% cement, 2.5% binder) • 175 mm natural gravel subbase • Natural gravel subgrade. Problems with Construction According to Johan Opperman, the riding quality of the reconstructed sections became a problem. This is attributed largely to the half-width construction, areas where the base appeared coarser than desirable and continual regrading during construction. The half-width construction resulted in differences in levels, especially in areas of super-elevation and mainly on the centreline where the two half-widths of construction abutted. Level differences were also attributed to the difficulty of cutting levels on the second half-width of construction to the same levels of the first half-width, within the narrow width of overlap between the two half-widths determined by the accommodation of traffic. As opposed to an asphalt, the relatively thin double seal also did not facilitate the accommodation of the level differences within the bituminous surfacing. The attempted remedial measure was to mill out the affected areas and to apply a slurry which was thereafter resealed. In the areas that exhibited a coarse surface texture, it is suspected that the in situ material was coarse and due to the lack of lateral mixing by the recycling machines these areas remained coarse. In these areas the seal exhibited a tendency to flush. The attempted remedial measure for localised depressions and level differences on the centreline was the application of an emulsion and 7.6 mm aggregate. This measure was successful initially in certain areas, but later, in other areas it is likely that the emulsion was over applied, resulting in bleeding, which was carried forward by traffic. Another factor, which in restropect, may have contributed to the poor riding quality was continual regrading, over portions of the vertical alignment, which was undertaken in an effort to both limit the importing of material and to adjust levels on the second half-width of construction to that of the initial half-width. Although this factor is suggested, regrades were subsequently tested against permissible tolerances in terms of levels between portions of regrade and were generally found to be acceptable. The specified seal is also considered inadequate for the specific application and it is proposed that the seal should have consisted of a larger aggregate size or that an asphalt should have been applied. Financial constraints, however, determined the use of the thinner seal.

Page 156: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 5 of 6 TR16-3 (R27)

Behaviour and Material Quality Indicators • Before the rehabilitation, FWD measurements obtained in 2001 as part of the detailed

assessment indicated that road was generally in a warning condition, with the section between Km 0 and 11.0 in a severe condition. [1].

• Before the rehabilitation, DCP measurements on the specific section in 2001 gave a DSN800 of 114 to 261, indicating a warning or sound condition [1].

• After the rehabilitation, FWD measurements taken in 2005 gave an average peak deflection of 529 microns with a 95th percentile of 1 016 microns [3]. These deflections are high for a road that has been in service for 1 to 2 years.

Performance Indicators and Treatment History • Before the rehabilitation, the riding quality was measured in 2000 (PMS) and ranged from

2.2 mm/km to 2.7mm/km over the entire section investigated (appears Km 0 – 65) [1]. • Before the rehabilitation, the riding quality was measured as part of the detailed

assessment in 2001. Approximately 95 per cent of the road (appears Km 0 – 65) had an IRI in the sound condition.

• Before the rehabilitation, rutting was measured in 2000. Between Km 2 and 25, the rut depths were less than 5 mm. Between Km 25 and 30, the rut depth was 18 mm and rated as being close to a severe condition [1].

• Before the rehabilitation, rutting was measured as part of the detailed assessment in 2001. On the whole section (assumed Km 0 – 65), on the eastbound lane, 57% was in the sound condition, 34% a warning condition and 9.2% in a severe condition. In the westbound lane, 70% was in the sound condition, 25% a warning condition and 5% in a severe condition. The westbound lane was therefore in an acceptable condition.

• Before the rehabilitation, the Visual Condition was obtained from 2000 PMS data. The Visual Condition Index (VCI) ranged from 25 to 47 for the whole section (assumed Km 0 – 65), which indicates a very poor to poor condition. The road was assessed as very poor between Km 10 to Km 15 and Km 25 to Km 30 [1].

• Before the rehabilitation, a visual assessment was performed in 2001. The condition was rated as good from Km 0 to 7.0, Km 19 to 24; fair from Km 7.0 to 10.0, Km 12.5 to 15.0, Km 16.0 to 19.0 and Km 28.5 to 59.5; and poor from Km 10.0 to 12.5, Km 15.0 to 16.0, and Km 24.0 to 28.5. The main structural distresses observed were rutting, crocodile cracking and pumping, but did not manifest in the form of extensive potholing or severe rutting.

• After the rehabilitation, no riding quality or rutting data are recorded in the SANRAL PMS. • In 2006, the road is not in a good condition with cracking and bleeding at levels which

should not be experienced on a 2 year old road. [5] • According to the route manager, the specific sections are in an acceptable condition, but

not as good as should be for a 2 year old road and the condition is deteriorating rapidly. The first section (Km 7.4 to 14.4) is in a poorer condition than Km 25.0 – 30.0. In the first section there is some degree 3 bleeding, and some shoving in the wheelpaths and along the centreline. On Km 25.0 – 30.0, there is also bleeding and showing, approximately degree 2 and 3 in the inner wheelpath. Isolated surface failures are beginning. On both sections, the riding quality is degree 1 or 2, and is therefore not in a warning condition, however the riding quality is inadequate for a 2 year old road.[6]

Traffic Loading Indicators • There are at present no SANRAL (CTO) counting stations on or near this section and

therefore no traffic data can be extracted from the SANRAL PMS. • The only available traffic data are contained in the 2002 Assessment report [1], and the

relevant details are given in Tables 1 and 2. • Although the traffic levels are low on this section, heavy trucks carrying grapes utilize the

road for a period in the heat of summer. [6] • Using the limited traffic data available, the following assumptions were used to determine

the traffic. These are the best assumptions available. o E80s per heavy: 1 to 3.5 E80s o Growth rate: 1 to 4%

Page 157: Summary of LTTP Emulsion and Foamed Bitumen Treated ... · cover both emulsion and foamed bitumen treated materials. A project was thus initiated to develop such a guideline document

Page 6 of 6 TR16-3 (R27)

o Directional split: 50%: 50% • The design traffic based on the above assumptions was determined to be 0.3 to 1.0 million

E80s for a 15 year design period. • Based on the available data in Table 2, the traffic carried from 2004 to 2006 ranges from

0.02 to 0.18 million E80s and based on Table 3, the cumulative traffic carried is 0.06 to 0.61 million E80s. A reasonable range then for the estimated traffic carried is 0.06 to 0.61 million E80s.

Table 2: Traffic from Department PMS for 2000 [1]

PMS Km references

Intersection references

From To

AADT range: vpd in both directions

% Heavies

Nieuwoudtville/Loeriesfontein – Matjiesfontein 5 25 488 – 504 15.2

Matjiesfontein - Clanwilliam 25 35 307 – 388 13.2 Clanwilliam – Kieskei (Calvinia) 35 65 571 – 585 8.9

Table 3: Traffic from Department PMS for 1996 to 1998 [1] Traffic Count (both

directions) Traffic Count Section Date of

Count Total

Vehicles Heavy

Vehicles

% Heavy Vehicles

Calvinia to DR2279 Keiskei 14/11/96 588 112 19.0 DR2279 Keiskei to MR316 Ceres 14/11/96 469 112 19.7 MR316 Ceres to MR542 Clanwilliam 30/07/98 585 52 8.9 MR542 Clanwilliam to MR280 Matjiesfontein 07/02/94 307 46 15.0

MR280 Matjiesfontein to DR2280 Rondekop 09/03/96 493 76 15.4

DR2280 Rondekop to MR572 Nieuwoudtville 09/03/96 521 79 15.2

Observations Based On Estimates or Interpretations • The riding quality after construction was poor, although no quantitative measurements are

available. • FWD measurements after 1 year of service showed an average peak deflection of

530 microns, with a 95th percentile of 1015 microns, which is high for such a young pavement.

• The performance of the road is adequate at present, however the condition is deteriorating rapidly. As such, the pavement can be said to be in a warning state.

• The estimated traffic carried is as follows: 2004 to 2006 (2 years of service): 0.06 to 0.61 million E80s.

• For the purposed of calibration and validation of the structural design method, the pavement structure is assumed to be as follows [2, 4]:

o 13.2 mm and 6.7 mm modified binder double seal o 200 mm foamed bitumen treated natural gravel (1.5% cement, 2.5% binder) o 175 mm natural gravel subbase (G5) o Natural gravel subgrade (G5-G8)

Reviewed by: • Johan Opperman, SNA


Recommended