Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Technical report
Recycled utility arisings in trench
reinstatement: compaction trial
Quality control during trench reinstatement works is important in ‘getting it right first time’. This research investigated the compaction behaviour of hydraulically bound mixtures (HBMs) manufactured from recycled trench arisings, and evaluated a range of in situ test devices for their suitability for compliance and/or control testing. The study found that HBMs can require less compaction than GSB1/Type1, and that a range of portable in situ test devices are suitable for quality control testing, but a proven compaction method specification is essential.
Project code: MRF106 ISBN: [Add reference]
Research date: August 2008 to March 2009 Date: August 2009
WRAP helps individuals, businesses and
local authorities to reduce waste and
recycle more, making better use of
resources and helping to tackle climate
change.
Document reference: WRAP, 2009, Recycled trench arisings in trench reinstatement: compaction trial (WRAP project
MRF106). Report prepared by J Edwards, P Edwards, L Robinson and A Buttress. Banbury, WRAP
Written by: J Edwards (Scott Wilson Ltd), P Edwards (Lafarge A&C UK), L Robinson (Scott Wilson Ltd) and
A Buttress (Scott Wilson Ltd)
Front cover photography: In situ testing during the compaction trial (courtesy of Scott Wilson Ltd)
WRAP (Waste and Resources Action Programme) and Scott Wilson Ltd believe the content of this report to be correct as at the date of writing. However, factors such as
prices, levels of recycled content and regulatory requirements are subject to change and users of the report should check with their suppliers to confirm the current
situation. In addition, care should be taken in using any of the cost information provided as it is based upon numerous project-specific assumptions (such as scale,
location, tender context, etc.).
The report does not claim to be exhaustive, nor does it claim to cover all relevant products and specifications available on the market. While steps have been taken to
ensure accuracy, WRAP cannot accept responsibility or be held liable to any person for any loss or damage arising out of or in connection with this information being
inaccurate, incomplete or misleading. It is the responsibility of the potential user of a material or product to consult with the supplier or manufacturer and ascertain
whether a particular product will satisfy their specific requirements. The listing or featuring of a particular product or company does not constitute an endorsement by
WRAP and WRAP cannot guarantee the performance of individual products or materials. This material is copyrighted. It may be reproduced free of charge subject to the
material being accurate and not used in a misleading context. The source of the material must be identified and the copyright status acknowledged. This material must
not be used to endorse or used to suggest WRAP’s endorsement of a commercial product or service. For more detail, please refer to WRAP’s Terms & Conditions on its
web site: www.wrap.org.uk
Recycled utility arisings in trench reinstatement: compaction trial 1
Executive summary
Trench arisings have the potential to be recycled into hydraulically bound mixtures (HBMs) for use as backfill and
in the structural layers of pavements. Using recycled trench arisings within HBMs offers the potential to:
� improve the performance of the reinstatement;
� reduce the cost and carbon associated with the works by replacing asphalt materials; and
� increase the recycling of trench arisings, diverting recoverable materials from landfill.
All non flowable materials (including GSB1/Type1, unbound backfill materials, asphalt and HBMs) require
adequate compaction; and acceptable performance within a trench reinstatement is dependent on this. For HBMs
and asphalts, inadequate compaction will affect durability (ability to withstand long term environmental
degradation), while for GSB1/Type1, inadequate compaction is likely to result in settlement.
The aim of this research is to investigate the compaction behaviour of HBMs to assess the suitability of adopting a
method specification for their installation (compaction); and evaluate a range of in situ test devices to assess
their suitability for compliance and/or control testing. Therefore, the following objectives were defined:
� objective 1: to evaluate compaction behaviour of two HBMs with regard to the compaction methodology; and
� objective 2: to evaluate available indirect test methods for compliance and/or control testing of HBM
compaction.
The research was initiated with a desk study that reviewed available compaction methodologies and in situ test
methods to facilitate the development of a test matrix for the compaction trial. Two HBMs were produced from
recycled trench arisings; the ‘HBM Fine’ was produced from screened trench arisings while the ‘HBM Coarse’
comprised screened trench arisings with additional recycled concrete aggregate. A typical primary GSB1/Type1
reinstatement material was used as the control. The influence of workmanship factors such as lift thickness,
compactive effort and the effect of delayed compaction were evaluated to assess the compaction behaviour.
Adequate compaction in accordance with Clause 870 of the manual of contract documents for highways works,
volume 1 - MCHW1 (HA, 2009a) was taken as a suitable target. This is defined as; compaction to an average wet
density of not less than 95% of the average wet density of test specimens compacted to refusal.
A calibrated nuclear density meter (NDM) was used, in direct transmission mode as specified in MCHW1 (HA,
2009a), to determine the in situ density of the materials, hence degree of compaction achieved, for comparative
analysis across the range of compaction methodologies. It also provided a baseline against which to compare the
data from selected in situ test devices. The eight different in situ test devices used in the trial are:
� nuclear density meter (NDM);
� asphalt pavement quality indicator (PQI);
� lightweight deflectometer (LWD);
� German dynamic plate (GDP);
� bearing capacity and deflectometer (BC&D);
� Clegg impact hammer;
� Panda2 variable energy dynamic penetrometer; and
� dynamic cone penetrometer (DCP).
All the test devices selected were portable, rapid to use and could, with varying degrees of ease and speed, be
operated by one or two operatives. However, the parameters measured by the majority of devices did not
correlate with relative density, and as such did not indicate when adequate compaction had been achieved.
As a result of this project, the following conclusions are made concerning the placement and assessment of
adequate compaction of HBMs for trench reinstatement:
� the two HBMs selected for the trial required less compactive effort than the GSB1/Type1 to achieve adequate
compaction, defined as complying with Clause 870 of MCHW1 (HA, 2009a);
� only direct density determination can be used to demonstrate adequate compaction and compliance with a
density specification, however, in situ test devices may be used to monitor consistency as part of a quality
control test regime, this requires development of material specific threshold values (pass/fail); and
� compliance with a suitable compaction method specification for HBMs should be sufficient to achieve
adequate compaction. Development of such a method specification would require a demonstration trial
utilising direct density determination to prove adequate compaction (Appendix A).
Recycled utility arisings in trench reinstatement: compaction trial 2
Contents
1.0 Introduction ................................................................................................................................7
1.1 Background........................................................................................................................ 7
2.0 Compaction and in situ testing equipment ................................................................................9
2.1 Compaction methodologies.................................................................................................. 9
2.1.1 Roller compactors .................................................................................................. 9
2.1.2 Vibrating plate compactors.................................................................................... 10
2.1.3 Vibrotampers ....................................................................................................... 10
2.2 Direct density determination.............................................................................................. 10
2.2.1 Nuclear density meter........................................................................................... 10
2.2.2 Sand replacement or sand cone method................................................................. 10
2.2.3 The water replacement method ............................................................................. 11
2.2.4 The rubber balloon method ................................................................................... 11
2.2.5 The core cutter or drive cylinder method ................................................................ 11
2.3 Indirect determination of compaction ................................................................................. 12
2.3.1 Electrical impedance............................................................................................. 12
2.3.2 Surface stiffness................................................................................................... 12
2.3.3 Resistance to impact ............................................................................................ 13
2.3.4 Resistance to penetration...................................................................................... 14
2.4 Selection of test methodologies for site trial ........................................................................ 15
3.0 Site work ...................................................................................................................................16
3.1 Methodology .................................................................................................................... 16
3.1.1 Materials ............................................................................................................. 16
3.1.2 Test matrix .......................................................................................................... 18
3.1.3 Material reinstatement and test positions ............................................................... 19
3.2 Site trial construction ........................................................................................................ 20
3.3 In situ testing of reinstatement layers ................................................................................ 22
4.0 Compaction behaviour ..............................................................................................................23
4.1 Density results ................................................................................................................. 24
4.2 Discussion........................................................................................................................ 26
4.3 Summary......................................................................................................................... 27
5.0 Indirect test device evaluation.................................................................................................27
5.1 Asphalt pavement quality indicator..................................................................................... 27
5.2 Dynamic plates................................................................................................................. 28
5.3 Clegg impact hammer ....................................................................................................... 30
5.4 Cone penetrometers ......................................................................................................... 31
5.5 Discussion........................................................................................................................ 35
6.0 Post compaction trial testing....................................................................................................35
6.1 In situ performance monitoring.......................................................................................... 35
6.2 Laboratory testing ............................................................................................................ 38
7.0 Conclusions ...............................................................................................................................38
7.1 Compaction behaviour ...................................................................................................... 38
7.2 Indirect test device evaluation ........................................................................................... 38
7.3 Post compaction trial testing.............................................................................................. 39
7.4 Summary......................................................................................................................... 39
8.0 References.................................................................................................................................40
Appendix A: Guidance note for establishing a material specific compaction method specification ..43
Recycled utility arisings in trench reinstatement: compaction trial 3
Figures
Figure 1: Primary GSB1/Type1.................................................................................................................. 17
Figure 2: Envirosand................................................................................................................................ 17
Figure 3: Recycled aggregate ................................................................................................................... 17
Figure 4: ‘As received’ trench arisings........................................................................................................ 18
Figure 5: Processed trench arisings ........................................................................................................... 18
Figure 6: Particle size distribution of the HBM Fine and HBM Coarse............................................................. 18
Figure 7: Schematic trench profile for 150 and 300 mm lifts and the in situ testing equipment ....................... 19
Figure 8: Testing bays for allocated test devices......................................................................................... 20
Figure 9: Schematic plan view of site trial construction ............................................................................... 20
Figure 10: Excavation of the trial trenches................................................................................................. 21
Figure 11: Vibrotamper used in the trial .................................................................................................... 21
Figure 12: Measuring the compacted HBM lift ............................................................................................ 21
Figure 13: Compaction of material (HBM Fine) ........................................................................................... 21
Figure 14: Nuclear Density Meter (NDM) gauge block for calibration and NDM.............................................. 22
Figure 15: Asphalt pavement quality indicator (PQI) ................................................................................... 22
Figure 16: Lightweight deflectometer (LWD).............................................................................................. 22
Figure 17: German dynamic plate (GDP) ................................................................................................... 22
Figure 18: Bearing capacity & deflectometer (BC&D) .................................................................................. 23
Figure 19: Clegg impact hammer .............................................................................................................. 23
Figure 20: Variable energy dynamic penetrometer (Panda2) ....................................................................... 23
Figure 21: Dynamic cone penetrometer (DCP) ........................................................................................... 23
Figure 22: Relative density after 3, 5 and 8 passes on 150 mm lifts ............................................................. 25
Figure 23: Relative density following 3, 5 and 8 passes on 300 mm lifts ....................................................... 26
Figure 24: Lightweight dynamic plate data against relative density .............................................................. 28
Figure 25: German dynamic plate data against relative density.................................................................... 29
Figure 26: Bearing capacity and deflectometer data against relative density ................................................. 29
Figure 27: Mean Clegg impact hammer derived CBR value against relative density........................................ 31
Figure 28: DCP profile on the HBM Coarse reinstated in 300 mm lifts ........................................................... 32
Figure 29: DCP derived CBR value against relative density .......................................................................... 32
Figure 30: Panda2 resistance to penetration against relative density ............................................................ 33
Figure 31: Panda2 data for one 300 mm lift of GSB1/Type1 following 3, 5 and 8 passes................................ 34
Figure 32: Panda2 profile of reinstated trench of GSB1/Type1 compacted in 300 mm lifts.............................. 34
Figure 33: LWD data sets from within 2 hours of reinstatement to 28 days age ............................................ 36
Figure 34: Panda2 data for GSB1/Type1 compacted in 300 mm lifts after reinstatement and 28 days ............. 37
Figure 35: Panda2 data for HBM Fine compacted in 150 mm lifts after reinstatement and 28 days.................. 37
Tables
Table 1: Harmonised British and European standard BS EN 14227 Specifications for HBMs ............................... 8
Table 2: Test matrix for HBMs and GSB1/Type1 control compacted in 150 mm and 300 mm lifts .................... 19
Table 3: NDM determined in situ density after 3, 5 and 8 passes of vibrotamper ........................................... 24
Table 4: Laboratory determined refusal wet densities.................................................................................. 25
Table 5: PQI and NDM density readings undertaken on NDM gauge block..................................................... 28
Table 6: Effect of increasing compactive effort on CBR measured using the Clegg impact hammer.................. 30
Table 7: In situ testing up to 28 days age .................................................................................................. 36
Table 8: 28 day compressive strength data ................................................................................................ 38
Recycled utility arisings in trench reinstatement: compaction trial 4
Glossary
Aggregate
Granular material used in construction. Aggregate may be naturally occurring, manufactured or recycled.
ARM
Alternative reinstatement material as defined in the SROH (HAUC, 2002).
Asphalt pavement quality indicator (PQI)
Portable device designed for the in situ determination of density and water content of bituminous mixtures
(asphalts).
Bearing capacity and deflectometer (BC&D)
Portable dynamic plate used for the determination of surface modulus (stiffness); a variation of LWD.
BS EN
Harmonised British and European Standard. All British and European Standards for HBMs are harmonised;
however, for some materials (notably for concrete) the Standard is not harmonised and does not contain Annex A
which sets out clauses addressing essential requirements and other provisions of EU Directives.
California bearing ratio (CBR)
The relative resistance to penetration of a 50 mm diameter plunger. The force is typically recorded at 2.5 and
5 mm penetration and the result expressed as a percentage by relating it back to a standard material.
Clegg impact hammer
Portable, in situ test device where an accelerometer (hammer) is manually dropped from a set height, recording
the resultant impact value (deceleration of the hammer); results can be converted to CBR values on a material
specific basis.
Dynamic cone penetrometer (DCP)
A portable cone penetrometer used to determine resistance to penetration with depth; the resistance to
penetration can be correlated to CBR on a material specific basis.
German dynamic plate (GDP)
Portable dynamic plate apparatus used to determine surface modulus (stiffness), designed for use on unbound
subbase layers; a variation of LWD.
GSB1 (Granular Subbase 1)
The SROH (HAUC, 2002) equivalent to a Clause 803 Type 1 unbound mixture as defined in MCHW1 (HA, 2009a);
under certain circumstances, GSB1 can be used as a reinstatement material up to and including base layers – see
SROH (HAUC, 2002).
HA
The Highways Agency (and its equivalent for the devolved administrations); responsible for the motorways and
trunk road network.
HAUC (UK)
Highway Authorities and Utilities Committee (UK).
Hydraulically bound mixture
A mixture comprising soil or aggregate with a hydraulic binder.
Hydraulic Binder
Material (or a combination of materials) that react with water to harden.
Layer
Structural horizon within the pavement.
Recycled utility arisings in trench reinstatement: compaction trial 5
Lift
Construction elevation. Thickness specified to achieve required compaction for particular plant and or material
type.
Lightweight deflectometer (LWD)
Portable dynamic plate apparatus used to determine surface modulus (stiffness), and included within Highways
Agency guidance IAN73/06 (HA, 2009c) for the performance assessment of materials.
MCHW
Manual of contract documents for highway works. Volume 1. Specification for highway works; also taken to
include Volume 2 (Notes for guidance on the Specification for highway works).
NRSWA
New roads and street works Act 1991.
Nuclear density meter (NDM)
Portable device for the determination of in situ density and water content which uses a radioactive isotope within
its measurement system.
Pavement
The surface on which vehicles or pedestrians travel, including footways and roads, it provides friction for vehicles
and transfers stresses to the underlying soils.
Recycled aggregate
Aggregate derived from the processing of inorganic material previously used in construction.
Recovered product
Material which is deemed to have been recovered from the reprocessing of waste and is no longer a waste.
Secondary aggregate
By-products of industrial processes not previously used in construction.
Stabilised material for fill (SMF )
A naming convention within the SROH (HAUC, 2002), SMF is an ad-hoc grouping of materials, which includes
processed, improved, modified and hydraulically bound and stabilised materials. SMFs can be derived from any
source materials (including primary materials) and are not necessarily hydraulically bound. SMFs are defined by a
set of compositional and performance requirements rather than by feedstock and production.
Structural material for reinstatement (SMR)
A naming convention within the SROH (HAUC, 2002), SMR is a material group generally falling within the
hydraulically bound mixture family and extending to specific materials such as foamed concrete. They are broadly
defined as materials that include ‘cementitious, chemical, hydraulic binder or are inherently self-cementing’
(HAUC, 2002) and are differentiated from SMF by superior performance.
SROH
Specification for the reinstatement of openings in highways, is the statutory document for street works in
England, under the new roads and street works Act (1991), there is a suite of parallel documents for the
devolved administrations.
Subgrade
The made ground or naturally occurring soil that is found below pavements.
Trench arisings
Material excavated during trench works.
Variable energy dynamic penetrometer (Panda2)
Portable intrusive penetrometer, similar in concept to the DCP, but with a microprocessor attached to provide real
time data and a user defined depth warning system. The resistance to penetration is reported in MPa and can be
correlated to CBR on a material specific basis.
Recycled utility arisings in trench reinstatement: compaction trial 6
Acknowledgements
The work described in this research report was carried out by Scott Wilson Limited and funded by WRAP. The
work reported here was directly supported by:
� Corehard Ltd
� Derbyshire County Council
� Independent Stabilising Company Ltd
� Lafarge A&C UK
� Loughborough University
� Mid Sussex Testing Ltd
� NT Killingley Ltd
� The University of Nottingham
The technical advisory group for this work comprised:
� Mr John Barritt (WRAP)
� Mr Andrew Dawson (The University of Nottingham)
� Dr Joanne Edwards (Scott Wilson Ltd)
� Dr Paul Edwards (Lafarge A&C UK)
� Mr Martyn Jones (Scott Wilson Ltd)
� Mr John Kennedy (JK Pavement Consulting)
The wider project was also supported by a working group comprising local authorities, utility companies, material
producers and testing companies. WRAP are grateful for all this support, and would like to thank the
organisations listed above and those below:
� Balfour Beatty Utility Services
� Clancy Docwra
� Dorset County Council
� First Intervention
� Lancashire County Council
� Morrison Utilities Services
� National Grid Gas
� Norfolk County Council
� Sheffield City Council
� SMR (UK) Ltd
� Staffordshire County Council
� Stent Foundations
� Tarmac
� Three Valleys Water
� Wales & West Utilities
Recycled utility arisings in trench reinstatement: compaction trial 7
1.0 Introduction Water and gas maintenance trench works generate around 4.8 million tonnes of trench arisings across Great
Britain, equivalent to 4.5% of the national construction, demolition and excavation waste (WRAP, 2005a). Trench
arisings have the potential to be recycled into hydraulically bound mixtures (HBMs) for use as backfill and in the
structural layers of the pavement. Using recycled trench arisings within HBMs offers the potential to:
� improve the performance of the reinstatement;
� reduce the cost and carbon emissions associated with the works by reducing asphalt materials used; and
� increase the recycling of trench arisings, diverting recoverable materials from landfill.
All non flowable materials (including GSB1/Type1, unbound backfill materials, asphalt and HBMs) will require
adequate compaction; and acceptable performance within a trench reinstatement is dependent on this. For HBMs
and asphalts, inadequate compaction will affect durability (ability to withstand long term environmental
degradation), while for GSB1/Type1, inadequate compaction is likely to result in settlement.
Specifications for compaction can be either;
� end product, where a quantifiable property (such as relative density) has to be achieved; or
� method, where a procedure (such as plant number of passes and lift thickness) is detailed.
Adequate compaction of GSB1/Type1 is currently achieved through compliance with a method specification and is
not generally tested, whereas HBMs are compacted to achieve a specified threshold relative density as specified
in the MCHW1 (HA, 2009a) and compliance testing is conducted. This is not due to a generic difference in
materials, but rather the fact that HBMs cover a wide range of potential material types and; therefore, behaviour
during compaction. Certain HBMs could require more or less compactive effort than GSB1/Type1, dependent on a
range of variables. During road construction the density of the placed HBM is directly determined in situ, using
either a nuclear density meter or a replacement method, to ensure it complies with the specification. This practice
does not readily transfer to trench reinstatement work, mainly due to the scale of works. In addition, there are
health and safety implications associated with the gamma radiation source in the NDM.
To circumvent this problem a method specification for HBM compaction, as with GSB1/Type1, may be adopted or
an alternative test device to the NDM utilised to determine whether adequate compaction has been achieved. To
follow the former approach requires an understanding of the compaction behaviour of an HBM, while to follow
the latter approach requires an evaluation of currently available in situ test methods and an assessment of their
suitability for compliance testing.
The aim of this research is to investigate the compaction behaviour of HBMs to assess the suitability of adopting a
method specification for their installation (compaction); and evaluate a range of in situ test devices to assess
their suitability for compliance and/or control testing. Therefore, the following objectives were defined:
� objective 1: to evaluate compaction behaviour of two HBMs with regard to the compaction methodology; and
� objective 2: to evaluate available indirect test methods for control testing of HBM compaction.
The previous WRAP project, Trench reinstatements: recycled materials and performance testing - AGG105-005
(WRAP, 2008), evaluated the performance and durability of HBMs composed of recycled trench arisings using test
methods and specifications given in harmonised British and European standards and in UK specifications. Other
work, funded by WRAP and running concurrent with the compaction trial reported herein, has developed a quality
manual for hydraulically bound mixtures (WRAP, 2009a), general guidance on the use of HBMs for the range of
stakeholders within the street works supply chain (WRAP, 2009b) and a template quality management scheme for
the production of a hydraulically bound mixture (WRAP, 2009c).
1.1 Background The recycling and use of trench arisings within HBMs for trench reinstatements is desirable as there is a potential
performance improvement over unbound equivalents, such as the ‘traditional’ GSB1/Type1 (WRAP, 2008), and as
it can benefit the sustainability of the associated street works.
The main specifications for the use of HBMs as trench reinstatement materials in subbase and base layers are:
� specification for the reinstatement of openings for highways (SROH). Second edition (HAUC, 2002);
� manual of contract documents for highways works. Volume 1. Specification for highways works. Series 800
road pavements - unbound, cement and other hydraulically bound mixtures (HA, 2009a);
Recycled utility arisings in trench reinstatement: compaction trial 8
� manual of contract documents for highways works. Volume 2. Notes for guidance. Series NG 800 road
pavements - unbound, cement and other hydraulically bound mixtures. (HA, 2009b);
� harmonised British and European standards (BS ENs) for hydraulically bound mixtures (the BS EN 14227
series – see Table 1), and the relevant aggregate standard - BS EN 13242:2002 (BSI, 2007); and
� design manual for roads and bridges. Volume 7 pavement design and maintenance. Section 1 part 2.
Conservation and the use of secondary and recycled material (HA, 2004).
Table 1: Harmonised British and European standard BS EN 14227 Specifications for HBMs
Part Part Title Report Reference
-1:2004 Cement bound granular mixtures BSI, 2004a
-2:2004 Slag bound mixtures BSI, 2004b
-3:2004 Fly ash bound mixtures BSI, 2004c
-4:2004 Fly ash for hydraulically bound mixtures BSI, 2004d
-5:2004 Hydraulic road binder bound mixtures BSI, 2004e
-10:2006 Soil treated by cement BSI, 2006a
-11:2006 Soil treated by lime BSI, 2006b
-12:2006 Soil treated by slag BSI, 2006c
-13:2006 Soil treated by hydraulic road binder BSI, 2006d
-14:2006 Soil treated by fly ash BSI, 2006e
Hydraulically bound mixtures are classed as alternative reinstatement materials (ARMs) in the SROH (HAUC,
2002). The term ARM does not mean that the materials are necessarily new or untried; for example HBMs
including stabilised soils and foamed concrete have been used for over 40 years in mainstream civil engineering
and construction projects (WRAP, 2008).
Within the SROH (HAUC, 2002), the majority of HBMs are classed as non flowable structural materials for
reinstatements (NFSMR), a subcategory of structural materials for reinstatements (SMR). NFSMR differ from
flowable material in that they require compaction, and generally have lower hydraulic binder contents. HBMs can
also be categorised as stabilised material for fill (SMF). The aggregate feedstock can be similar between NFSMRs
and SMFs; however, the level of processing, quality control and end performance demanded vary. In accordance
with the SROH (HAUC, 2002):
� NFSMRs are required to achieve a 90 day strength of between 2 and 10 MPa, or 4 and 10 MPa dependent on
the class of road and their position within the pavement; but
� SMFs are classed according to the soaked CBR value of specimens tested at 90 day.
This report focuses on NFSMRs, which require higher levels of durability and performance than SMFs,
corresponding to their use as structural layers within the pavement.
Previous work (WRAP, 2008) concentrated on determining the performance and durability of HBMs manufactured
from processed trench arisings, and the specification and compliance of these materials with BS ENs, in a similar
manner to concrete and bituminous bound mixtures (asphalts). In that work, trench arisings were sourced from
recycling centres across the UK. These were incorporated into non-proprietary BS EN HBMs, such as fly ash
bound mixtures (FABMs) and cement bound granular mixtures (CBGMs) and into HBMs using proprietary products
such as Ecofill, Trenchmod and SMR soil stabiliser. The laboratory performance and durability of these HBMs were
assessed in accordance with the requirements of MCHW1 (HA, 2009a) and the SROH (HAUC, 2002). The
performance of these HBMs was compared to GSB1/Type1. Using a conservative estimation of long term layer
stiffness, the HBMs were shown to be suitable as a direct replacement for applications up to and including those
permitted for NFSMR and GSB1/Type1.
The site performance of NFSMRs and unbound subbase mixtures (GSB1/Type1) were also assessed in the
previous project. Lightweight deflectometer (LWD), Clegg impact hammer and dynamic cone penetrometer (DCP)
tests were used immediately after compaction and at several months age. The work indicates that, the higher the
coarse aggregate content (with all other variables being equal), within the overall context of the HBM particle size
distribution (PSD), the more mechanically stable the mixture. In summary, the previous project demonstrated
that appropriately designed HBMs perform better than GSB1/Type1, and that they should be designed for both
short and long term performance.
Recycled utility arisings in trench reinstatement: compaction trial 9
Performance of HBMs relies not only on a suitable mixture design but also on adequate compaction of the
material. Therefore, an assessment of the various methods of determining in situ density (and hence compaction)
of the reinstated material was undertaken within this project.
2.0 Compaction and in situ testing equipment The selection of a compaction methodology for trench reinstatement depends on a number of factors including:
� type and size of the compaction plant that is practical for use in particular reinstatement works;
� lift thickness and number of passes required to achieve compaction of each lift; and
� the tolerances in lift and layer thickness.
In general terms, the heavier the compaction equipment, the more compaction energy is transferred into a
material when it is operating. If insufficient energy is used when compacting the material, then it will not be
compacted to the correct density. For trench reinstatements (and other applications), this could result in:
� high air voids content;
� poor particle interlock;
� settlement issues; and
� insufficient performance and durability, resulting in the need to undertake remedial action.
However, the use of compaction equipment that has excessive compaction energy could result in:
� crushing of aggregate particles in granular soils;
� shearing of fine grained soils;
� less stable backfill;
� risk of damage to buried services; and
� unnecessary exposure of the operative to hand arm vibration.
Inadequate compaction could result in reduced durability or, in the case of unbound mixtures (GSB1/Type1)
settlement. Therefore, it is a key requirement for reinstatement works that a reliable method (including
appropriate plant, trained operators, suitable materials, effective installation procedure and necessary health and
safety considerations) is used to achieve adequate compaction.
In situ testing can be used either to ensure compliance with a specification requirement, as a quality control
check or to determine performance (for example stiffness or resistance to rutting). Compliance testing may
include control test methods but control testing will not necessarily indicate compliance unless the test is included
in the specification. For example CBR tests may be conducted to produce comparative data for control testing (to
check consistency), but will only be utilised as a compliance test if a minimum and/or mean CBR value is given in
the specification.
Given the importance of adequate compaction, the direct measurement of in situ density is highly desirable. The
nuclear density meter (NDM) is the test device specified by the Highways Agency in MCHW1 (HA, 2009a) for the
determination of in situ wet density. However, there are health and safety issues related to exposure to a
radioactive source (see Section 4.3) and many authorities involved in road construction are seeking an
alternative. In addition, the operation of this device is generally impractical for all but the largest of trench
reinstatement projects; and robust and reliable in situ test devices which measure a property that can be used for
quality control monitoring are preferred.
The following Sections review the compaction plant available (Section 4.1) and the tests for directly measuring
density (Section 4.2) or providing some indirect measure for use in quality control (Section 4.3). Section 4.4
explains the selection of compaction methodology and in situ test methods used in the compaction trial.
2.1 Compaction methodologies This Section reviews the available compaction plant for trench reinstatement works; the review was used to
select representative plant for use in the site compaction trial.
2.1.1 Roller compactors Roller compactors, also known as road rollers, include pedestrian and ride on rollers which can be a single or twin
drum set up. The type of drum is selected for the material to be compacted and the range includes smooth-
wheeled and pneumatic-tyre rollers, dead-weight sheep’s foot, tamping and grid rollers. They are generally
Recycled utility arisings in trench reinstatement: compaction trial 10
utilised on large scale works such as road construction and earthworks and unless specifically designed for the
narrow confines of a trench are not suitable for trench reinstatement. Therefore, roller compactors were not
considered further for this study.
2.1.2 Vibrating plate compactors Vibrating plates are low amplitude and high frequency compactors, designed to compact granular soils and
asphalt. They have a flat plate in contact with the soil on which either one or two eccentrically weighted shafts
are mounted. The power unit and control handles for a pedestrian operator are attached to a chassis supported
above the base plate on flexible mountings, usually in the form of springs or rubber cushions. Many machines are
equipped with some form of wheeled undercarriage which can be used to assist in transit between working areas
(Parsons, 1992). The SROH (HAUC, 2002) specifies a weight category over 1800 kg/m2 for a compacted lift
thickness up to 150 mm. However, vibrating plate compactors are not generally used for reinstating the lower
layers of a trench so were not incorporated into the compaction trial.
2.1.3 Vibrotampers The vibrotamper is also known as a power rammer, trench compactor, elephant’s foot or jumping jack. A high
impact force (high amplitude) is delivered from an engine driven reciprocating mechanism, which acts on a spring
system through which vertical oscillations are set up in a base plate (Parsons, 1992). Vibrotampers have masses
between 50 and 150 kg, and usually operate at a frequency of about 10 Hz. They are often used in confined and
small areas due to their portability and manoeuvrability. For situations where fumes from internal combustion
engines may be a hazard, electrically driven vibrotampers are available (Parsons, 1992).
Vibrotampers are commonly used in trench reinstatements. The compaction effort applied to the reinstatement
can be controlled by varying the number of passes with the vibrotamper and by changing the size of the foot; a
smaller foot will result in higher compactive effort compared to a large foot fitted to the same vibrotamper.
Compaction using a vibrotamper was selected for the trial as it is representative of current practice and
technically suitable for compacting the selected materials (Section 3.0).
2.2 Direct density determination Few devices are able to give a direct determination of density. Those tests that are used to determine in situ
density are often labour intensive and time consuming. Hence, widespread adoption of the methods described in
the following sections would not be practical in all but the largest scale utility reinstatement works.
2.2.1 Nuclear density meter The principal method for the direct determination of in situ density within MCHW1 (HA, 2009a) is the use of a
nuclear density meter (NDM), also referred to as a nuclear density gauge. The gamma radiation source may be
hazardous to the health of users, unless proper precautions are taken; therefore, the NDM requires a special
licence, with accompanying legal requirements for its storage and operation. The NDM is operated by technicians
who are specially trained in its use so is not widely adopted for routine assessment of compaction and in situ wet
density within trench reinstatement work. However, this test was used in the compaction trial to provide baseline
data of direct density measurements for comparative analysis.
Ionizing radiation is applied to the material with the amount of radiation detected decreasing in proportion to the
wet density of the material between source and receiver. Measurement is possible in two modes; backscatter and
direct transmission. Backscatter mode can be used to determine density, but is generally used to determine water
content while direct transmission is the favoured method for density. In backscatter mode, the source and
detector are placed on the surface of the material, and the test only penetrates to a depth of 70 to 80 mm (BSI,
1990a), whereas in direct transmission mode the radioactive source is positioned within the soil profile rather
than at the surface. The dry density is calculated from wet density and determined water content.
The device is portable and testing is relatively quick (less than 5 minutes) once the NDM has been calibrated for
the material to be tested. NDM calibration is specified in BS 1924 -2 (BSI, 1990a) and comprises compacting the
test material into a gauge block of known volume to calculate the wet density.
2.2.2 Sand replacement or sand cone method The sand replacement method is designed for testing soils and is specified in BS 1377-9 (BSI, 1990b), it involves
the excavation of a cylindrical hole in the material, which is then filled with dry uniformly graded sand. The
excavated material is carefully collected and weighed, and the water content is determined. The apparatus is
Recycled utility arisings in trench reinstatement: compaction trial 11
calibrated prior to testing, and includes determination of the density of the sand and the amount of sand
contained within the cone.
The test utilises a pouring cylinder incorporating a cone at the base, designed to generate a cone of sand of
known volume above the sand filled void to ensure that the excavated hole is completely filled. The sand
remaining in the pouring cylinder and cone is used to calculate the mass required to fill the hole and determine
the exact volume of the excavated hole. This in combination with the mass of the soil excavated from the hole, is
used to calculate and the wet and dry density of the soil. For accuracy, it is essential that the surface in which the
hole is excavated is flat and that the soil surrounding the excavated hole remains undisturbed.
The size of hole required for the test, and corresponding pouring cylinder, depends on the particle size
distribution of the soil. The British standard test method (BSI, 1990b) specifies a 100 mm diameter hole for fine
and medium grained soils, whereas a 200 mm diameter hole is required for coarse grained soils. This test method
was developed for the assessment of soils and has potential limitations associated with excavation into bound
materials and coarse aggregates.
2.2.3 The water replacement method The water replacement method is another test for soils described in BS 1377-9 (BSI, 1990b). It is intended for
use with coarse and very coarse soils, where other methods for determining the field density (Section 2.2.2)
would be unsuitable because the volume sampled would be too small to be representative. A circular density ring
is placed on the surface of the ground and a hole is excavated inside the ring, the excavated material is weighed
as in the sand replacement method. Flexible plastic sheets are used to line the excavated hole to retain water
which is poured into the hole. The volume of the hole is then determined from the mass of water; the wet density
(and dry density) of the soil can be calculated using the excavated material. This test method is not widely used
as it is relatively complicated and time consuming. Its application to bound materials will have similar limitations
to the sand replacement method.
2.2.4 The rubber balloon method The rubber balloon method was used during early research into the quality of compaction in trenches (Fleming
and Cooper, 1995). It is specified as ASTM D 2167-08 (ASTM, 2008); there is no British standard available for this
method. In principle, this method of determining the density of soil in situ is very similar to the sand replacement
method. A hole is excavated in the compacted material and the removed soil is weighed. Its dry weight is
measured as part of the water content determination. The volume of the excavated hole is determined from the
volume of water required to inflate a rubber balloon to completely fill the hole. The water is contained in a
calibrated vessel, enclosed at its base by the rubber balloon. A means of pressurising the liquid is normally
incorporated in the design of the apparatus. An initial reading is taken by placing the rubber balloon apparatus on
the surface of the soil at the location of the intended test hole, and the final reading is taken after excavation of
the hole. The difference between the two readings gives the volume of the excavated hole; which is then used to
calculate the density of the soil.
The test is not recommended for soils that deform easily (particularly where the apparatus is used under
pressure). It is also noted that the test may not be suitable for soils that contain sharp edges that could puncture
the rubber balloon (ASTM, 2008), which further limits its use.
2.2.5 The core cutter or drive cylinder method The core cutter method outlined in BS 1377-9 (BSI, 1990b) consists of driving a thin walled open ended steel
cylinder (100 mm diameter and 130 mm long) into the soil. The cylinder containing the soil is excavated and
removed from the ground, and the soil protruding from the cylinder ends is trimmed off. The mass of soil
contained in the known volume of the cylinder provides the wet density, and the dry density is calculated
following a water content determination. The method is only applicable to unhardened fine grained materials
(BSI, 1990a), therefore, not suitable for coarse granular material or many HBMs.
Recycled utility arisings in trench reinstatement: compaction trial 12
2.3 Indirect determination of compaction A variety of in situ test devices are available for pavement and subgrade evaluation, however, these devices are
designed to measure specific parameters and do not measure density directly. The following sections describe
methodologies which were assessed for use in the compaction trial. The devices selected for the trial have been
grouped into four main categories based on the parameter measured by each test:
� electrical impedance:
o asphalt pavement quality indicator (PQI);
� surface stiffness (modulus):
o lightweight deflectometer (LWD);
o German dynamic plate (GDP);
o bearing capacity and deflectometer (BC&D);
� resistance to impact:
o Clegg impact hammer;
� resistance to penetration:
o Panda2 variable energy dynamic penetrometer;
o dynamic cone penetrometer (DCP).
This list covers a range of test devices used within UK civil, utility and construction works. Static plate loading
devices, typically used for determination of resistance to monotonic loading, such as a crane outrigger or building
foundation, have not been included since they are not portable and can require significant kentledge.
A number of the test devices selected report values in terms of CBR (California bearing ratio). The CBR was
introduced into the UK following the Second World War (Croney and Croney, 1997) and has been in continued
use since then (although not in California where it has been replaced). The CBR is limited in its application as it is
only suitable for testing soils with a maximum particle size not exceeding 20 mm (BSI, 1990b), due to the size of
the plunger (50 mm diameter). The CBR test is relatively robust and cost effective; however, it is not a
fundamental property of the material and does not replicate the behaviour of materials under pavement type
loadings (Brown, 1997). In addition, its application to HBM is relatively limited (WRAP, 2008) so it was not
included in this study. For the test devices which provide a CBR value, the CBR should only be seen as a test
specific output and any quality control testing using target CBR values would need to be based upon an
established relationship between the test device and material type.
2.3.1 Electrical impedance The asphalt pavement quality indicator (PQI) was designed to give measurements of the density of bituminous
bound mixtures in a non destructive, non nuclear format (Sawchuk, 1998). The density of asphalt pavement is
directly proportional to the measured dielectric constant and composite resistivity of the material. The PQI uses a
low frequency constant current and a toroidal (doughnut shaped) electrical sensing field to measure changes in
electrical impedance of the material matrix. The electronics in the PQI convert the field signals into material
density readings and displays the results. Once calibrated, direct density readings can be consistently obtained for
bituminous bound mixtures. The company behind the PQI have recently developed a soil density gauge (SDG)
based on the same principles as the PQI; although it was not available at the time of this work.
2.3.2 Surface stiffness Lightweight dynamic plate tests are used to determine a surface modulus (stiffness), where the response of a
material under dynamic loading (its transient deflection) and the applied stress approximates to those
experienced in-service. A variety of designs are available such as the lightweight deflectometer (LWD), the
German dynamic plate (GDP) and the bearing capacity and deflectometer (BC&D). The test involves dropping a
weight onto a bearing plate which is placed on to the surface of the material. The area of loading and applied
stress can be readily controlled, and usually a damping mechanism is incorporated to control the loading time.
Lightweight dynamic plates may not be suitable to test thicker and/or stiffer materials as the result is dependent
on deflection during the test. It is important to note that the equipment cannot provide stiffness values for
individual lifts compacted in a trench reinstatement or discrete layers because the stress applied to the surface
travels down through the material. Therefore, the actual stiffness reading is a composite value of the material
down to the bottom of the zone of influence. The zone of influence of these tests is typically considered to be a
bulb of significant stress which is dependent on the size of the bearing plate and applied load.
Recycled utility arisings in trench reinstatement: compaction trial 13
LWD and GDP have been correlated against the larger falling weight deflectometer (FWD), which is used in the
UK to assess the condition of pavements; whereas the BC&D has been recently introduced to the UK so limited
data are available. The BC&D provides a measurement of surface modulus similar to the LWD and GDP.
The LWD testing procedure used in the compaction trial was developed for pavement foundations and is detailed
in the Highways Agency design guidance - IAN73/06 (HA, 2009c). The key requirements are:
� the equipment is capable of delivering a load pulse of peak magnitude in the range 4 to 15 kN, of total
duration 15 to 60 ms, to a rigid circular plate of 300 mm diameter;
� both the applied load and the transient deflection are measured directly on the tested surface. The deflection
measurement transducer must be capable of measuring deflections up to 2000 microns;
� the peak stress applied during each test is within the range 50 to 200 kPa. A peak stress of 100 kPa was
targeted, unless the deflection measurement fell outside the range 100 to 1000 microns, in which case the
applied stress was increased in order to achieve a realistically measurable deflection;
� at each test point, three initial seating drops, to bed the plate into the surface are undertaken. Three further
drops are then carried out. The results (measured load and deflection) from the last set of three drops are
then averaged to give the surface modulus applicable to that test point; and
� the surface modulus is computed at each point tested, using the following formula:
D
RPE
)1(2 2ν−
=
where:
E = surface modulus (MPa)
v = Poisson’s ratio (default = 0.35)
P = contact pressure (kPa)
D = deflection (microns)
R = plate radius (150 mm)
A disadvantage of the lightweight dynamic plate tests, with regard to testing trench reinstatement, is the
standard 300 mm plate diameter (although this can be reduced) due to the influence of lateral support. Testing
of the same material within two different trenches (with varying degrees of lateral support), is likely to give
different results (surface modulus being higher for higher levels of lateral support). However, the influence of
lateral support (and the trench itself) is likely to be more significant when assessing the performance of unbound
materials (GSB1/Type1) than HBMs. The stiffness value derived from the test is a composite stiffness value of the
layers and lateral support within the zone of influence. The zone of influence is (typically taken to be 10% of the
applied stress) and normally reaches a depth of around 1.5 times the plate diameter, so for a 300 mm diameter
bearing plate (standard for the LWD and GDP) it typically reaches to a depth of 450 mm below the surface.
2.3.3 Resistance to impact The Clegg impact hammer was developed in Australia during the 1970s by Dr Baden Clegg for the in situ
evaluation of granular base course (Clegg. 1976). It works by measuring the deceleration of a drop hammer,
registering the deceleration in units of ‘Impact Value’ (IV). It is specified as ASTM D 5874- 02 (ASTM, 2007) and
AS 1289.6.9.1 (AS, 2000). A range of hammer masses is available, namely a standard (4.5 kg), medium (2.25
kg), light (0.5 kg) and heavy (20kg). The drop height for the standard and medium hammers is 450 mm, and 300
mm for the light and heavy hammers. The standard hammer is most commonly used for trench reinstatement, so
was selected for the compaction trial.
The impact value is dependent upon a combination of density, soil type and water content. Snowdon (1992)
reported that the Clegg impact hammer should only be used for the purpose it was devised for, and that any
application of the Clegg to control or monitor dry density and state of compaction for earthworks should not be
considered.
The advantages of this in situ test are practicality, ease of use and reliability. The rapid nature of the test means
that large data sets can be readily gathered, which is ideal for checking material consistency. In addition, the
latest models convert IV to CBR in real time.
Recycled utility arisings in trench reinstatement: compaction trial 14
The main disadvantage of the standard test is the ratio between the contact area and maximum aggregate size.
The standard hammer diameter of 50 mm (which makes contact with the test surface) has the following
limitations:
� testing of materials which contain coarse aggregates (such as GSB1/Type1 and the majority of the recycled
trench arisings) is questionable, high variability might be expected depending on whether the test area is
wholly, partly or not at all over a large aggregate particle;
� the hammer diameter means that the zone of influence (the zone that is significantly stressed under the drop
hammer) is likely to be in the order of 50 to 150 mm; so low density zones at the base of thick lifts may be
missed.
The heavy (20 kg) Clegg hammer has a diameter of 130 mm so can be used to test coarser material and has a
larger zone of significant stress than the other Clegg hammers, however, the standard Clegg hammer is more
commonly used within the UK.
2.3.4 Resistance to penetration Cone penetrometers are used to determine resistance to penetration and give a profile with depth; and as with
any intrusive test, care should be taken when using it to assess material under which buried services are present.
Various sizes of portable dynamic and static field cone penetrometers are available. Static cone devices, such as
the mexicone, are generally designed to assess soft and medium fine grained subgrades, as the cone is pushed
into the soil, and are not considered suitable for assessing reinstatement materials.
The dynamic cone penetrometer (DCP), developed by CNS Farnell in conjunction with TRL (Clark, 2000), is
commonly used for the evaluation of subgrade strength under pavements – see IAN73/06 (HA, 2009c). The DCP
is driven into the ground under the action of an 8 kg steel drop weight falling vertically through 575 mm and
making contact with a steel anvil, attached via steel rods (less than 20 mm diameter) to a 20 mm diameter 60o
steel cone, which is thus driven vertically into the ground. It is suited to stronger and coarser materials than other
penetrometers, such as the Mackintosh probe. The Mackintosh probe is a light dynamic cone penetrometer that
can be operated manually and is generally suited to site investigation in soft deposits. The probe has a 4.5 kg
hammer which falls through a 300 mm drop height. Due to the relatively low energy, the Mackintosh probe is
limited in the depths and materials it can penetrate (Clayton et al, 1995); therefore, it is not considered practical
for testing bound material.
The accumulated blows are recorded against depth (rate of penetration), expressed as mm/blow, which for the
DCP can be converted to a CBR value using the following relationship (HA, 2009c):
Log10(CBR) = 2.48 – 1.057 × Log10(mm/blow)
Subsequent plots of accumulated blows (or CBR) against depth can be used to characterise layer depth and are
indicative of material strength. However, this strength measure will not normally be specified for materials other
than subgrade, since results are highly dependent on particle size and can, without calibration to specific
materials, be misleading (HA, 2009c).
The type of cone penetrometer specified in IAN73/06 (HA, 2009c) is the DCP developed by CNS Farnell and TRL,
if any alternative is to be used, then it should be carefully calibrated against equipment complying with the
specification for the specific types of material encountered.
The variable energy dynamic penetrometer (Panda2), unlike the DCP, has been designed specifically for use in
trench reinstatements, and can be used above utility services as it includes a user defined depth warning system.
The Panda2 was developed in France where it has become widely used and accepted (Langton, 1999), and is
supported by the French standard XP P 94-105 (AFNOR, 2000). A hammer is used to apply the force necessary to
drive the cone into the ground, rather than a fixed weight being dropped a standard height. This variable energy
input is determined using the output from two sensors, which measure the speed of impact and the depth of
cone penetration, recorded by a microprocessor. The results may be correlated to a CBR but this is likely to be a
material specific relationship. The dynamic cone resistance (qd) and its current depth are then calculated and
displayed in real time on the screen of the microprocessor. The dynamic cone resistance is calculated using the
formula given in XP P 94-105 (AFNOR, 2000) and has been modified to the following (Langton, 1999):
Recycled utility arisings in trench reinstatement: compaction trial 15
°
⋅
+
⋅=
90
2
1
1
.21
1
x
M
P
VM
Aqd
where:
x90° = penetration from one drop (90° cone)
A = area of the cone
M = weight of the striking mass
P = weight of the struck mass
V = speed of impact (of the hammer)
The software can also plot material specific reference curves (tolerance lines) which may be applied as pass/fail
criteria when assessing compaction compliance, as they are based on the relationship between the degree of
compaction and the dynamic cone resistance for a given material. The tolerance lines are established in
accordance with XP P 94-105 (AFNOR, 2000) and correspond to the maximum dry density, obtained from the
Proctor test. A choice of materials with predefined tolerance lines is available in the software (Langton, 1999).
2.4 Selection of test methodologies for site trial A number of test methodologies for the determination of in situ density both directly and indirectly have been
reviewed. For the purposes of the trial and the development of appropriate guidance, the selection of test
methodologies was based on the ease of use, cost and availability.
The NDM is the most commonly available method for directly measuring in situ density. It is the direct density
measurement procedure specified for use with HBMs within MCHW1 (HA, 2009a). Although the test is accurate, it
does have the following disadvantages; it:
� requires calibration using material compacted in gauge blocks;
� can only be used by a trained operator; and
� has special health and safety requirements for transport, storage and operation.
For these reasons, it is not a practical test for reinstatement crews to use in the field for standard reinstatement
works and industry are looking for a non nuclear alternative. However, for the detailed assessment of the
behaviour of an HBM at the design stage (for example, evaluating compaction behaviour or developing a
compaction method specification) or as part of a two year approval trial (HAUC, 2002), the use of an NDM for
density measurement may be appropriate. The other in situ direct density measurement methods, such as the
various replacement methods specified in BS 1377-9 (BSI, 1990b), are difficult and time consuming to undertake,
and are of limited use for bound and coarse grained materials. This makes them impractical and is most likely the
principal reason they have not been widely adopted in the UK. For the purposes of the compaction trial, the NDM
was selected for the direct method of determining in situ density.
The following test methodologies were selected for assessment during the compaction trial:
� asphalt pavement quality indicator (PQI);
� lightweight deflectometer (LWD);
� German dynamic plate (GDP);
� bearing capacity and deflectometer (BC&D);
� Clegg impact hammer;
� dynamic cone penetrometer (DCP); and
� variable energy dynamic cone penetrometer (Panda2).
The test methodologies selected for assessment measure a number of different material properties which may or
may not correlate to actual in situ material density. The PQI measures density through the direct correlation with
electrical impedance and can present the results in the same units as the NDM. However, the PQI is principally
used for asphalt testing; its application to HBMs in reinstatements is beyond the purpose that it was designed for.
The appropriate counterpart of the PQI, the soil density gauge (SDG) was not available in the UK during the trial
so was not included.
The LWD, GDP and BC&D are used to measure surface stiffness. Surface stiffness is a fundamental property of
the material’s ability to support the overlying pavement IAN73/06 (HA, 2009c). The LWD and GDP are commonly
Recycled utility arisings in trench reinstatement: compaction trial 16
used in pavement evaluation so were included in the trial. The relatively new BC&D works on a similar principle to
the LWD and GDP so was also incorporated into the trial.
The Clegg impact hammer and Panda2 were incorporated into the trial to assess compaction because they have
been adopted by UK practitioners for the assessment of trench reinstatements. The DCP was included in the trial
as it is a specified test for the evaluation of subgrades (HA, 2009c). Only the DCP and Panda2 have the ability to
give a profile with depth. The remainder of the tests are undertaken on the surface of the installed material, with
the zone of influence dependent on the material tested, the contact diameter and stress level applied during the
test.
3.0 Site work The compaction trial was designed to meet the research project aim and objectives (Section 1.0). The following
Section details the trial methodology, materials, test matrix and construction.
3.1 Methodology The in situ test devices selected for use on the trial comprised the nuclear density meter (NDM) and a range of
portable test devices, with potential for use as compliance and/or control tests during the installation of trench
reinstatement materials (Section 2.4). The testing requirements were rationalised to permit the works to be
undertaken within the duration of the trial.
The same compaction equipment was used for the site trial construction, for consistency (Section 5.3), the
variables included:
� the degree of compaction (controlled by the number of compaction passes);
� reinstatement material;
� lift thickness; and
� time delay between material production and use.
All testing and installation works were undertaken by appropriately trained and accredited staff.
Three reinstatement materials were selected for use in the compaction trial including:
� two generic HBMs produced from recycled trench arisings; and
� GSB1/Type1 control from a primary source, representing a ‘traditional’ reinstatement material.
A pipe bedding material was also selected to produce a consistent and representative starting layer for the
installation of the HBMs and the GSB1/Type1. Further details of the materials are given in Section 3.1.1.
A site was selected to permit experimentation and levels of compaction that would not be allowed in a highway.
The site selected for the trial was located on brownfield land owned by Derbyshire County Council, adjacent to
the offices of the Markham Vale Environment Centre near Chesterfield. The site consisted of made ground to a
depth of approximately two metres, the upper 200 mm of which had been in situ stabilised with cement to serve
as a working platform. The made ground comprised firm cohesive soils, which are representative of the lower
layers (subgrade) of many trench reinstatements. The ground conditions, in terms of confinement during the
material installation, are considered representative of full depth in typical highway or footpath reinstatements.
Details of the site trial construction are given in Section 3.2.
An additional complication with testing HBMs is their strength gain over time, which is related to a number of
variables; therefore, additional testing was undertaken after the trial to monitor potential strength gains. Post
compaction trial testing was undertaken at 7 days using the LWD and GDP; and 28 days using the LWD, GDP,
Clegg impact hammer, Panda2 and DCP. This testing provided additional information on the performance of the
trench reinstatement materials (Section 6.0).
3.1.1 Materials The following three materials were selected for use in the compaction trial:
� HBM Fine;
� HBM Coarse; and
� GSB1/Type1.
Recycled utility arisings in trench reinstatement: compaction trial 17
The HBMs were stored for 24 hours before compaction to investigate the influence of delayed compaction (DC):
� HBM Fine (DC); and
� HBM Coarse (DC).
The two HBMs were chosen as representative particle size distributions of processed trench arisings from a range
of sources across the UK (WRAP, 2008). The HBM Fine grading represents trench arisings where the oversize
material (>63 mm material comprising asphalt, concrete and bricks) has been screened out (removed). This
recycling option has the advantage that the oversize material can be crushed to produce recycled aggregate
feedstock for other products, but the lack of coarse aggregate (dependent on the feedstock) has to be
accommodated within the HBM design, typically by increasing the hydraulic binder additions.
The HBM Coarse grading represents the same trench arisings, but where the oversized material has been
screened and crushed to produce a <40 mm coarse aggregate, which has been blended back into the processed
trench arisings. Other sources of coarse aggregate can also be introduced to enable the production of a more
consistent material.
The previous study indicated that the HBM Coarse grading would be advantageous during installation and in the
short term, when compared to the HBM Fine grading. This is because; in the short term HBM performance is
related to the unbound aggregate mixture performance, which is dependent on the grading and quality of fines
(WRAP, 2008). In short, the higher the coarse aggregate content, within the overall context of the HBM particle
size distribution (PSD), the more mechanically stable the mixture is in the short term (WRAP, 2008).
The GSB1/Type1 aggregate (Figure 1) control material was taken from a local primary source to represent
‘traditional’ reinstatement practice. In addition, Envirosand (Figure 2), produced from recycled glass, was selected
for use as bedding sand in the trial, and recycled aggregate (Figure 3) required to optimise the grading of the
HBM Coarse, were sourced from the Lafarge A&C UK recycling hub located at Chaddesden, Derbyshire. All the
recycled aggregates were produced in accordance with the WRAP Quality Protocol for production of aggregates
from inert waste (WRAP, 2005b). Pulverised-fuel ash (PFA) and lime were selected for the binder additions to
produce the HBMs.
Figure 1: Primary GSB1/Type1
Figure 2: Envirosand
Figure 3: Recycled aggregate
The ‘as received’ trench arisings (Figure 4) were processed (by screening out the oversize materials) at the
Lafarge A&C UK recycling hub in accordance with the WRAP Quality Protocol (WRAP, 2005b) to produce the
feedstock material (Figure 5) for both HBMs. The coarse grading was achieved by combining the processed
trench arisings with additional recycled aggregate (Figure 3) to optimise the grading, while the fine grading
utilised the screened arisings without further modification.
Recycled utility arisings in trench reinstatement: compaction trial 18
Figure 4: ‘As received’ trench arisings
Figure 5: Processed trench arisings
The particle size distribution (PSD) of both HBM Fine and HBM Coarse were determined in accordance with BS EN
933-1 (BSI, 1997) and are shown in Figure 6. The grading curves are shown plotted against the grading
envelope taken from BS EN 14227-3 (BSI, 2004c), for a fly ash bound mixture (FABM 1) produced using 0/31.5
mm aggregate and siliceous fly ash. The HBM Coarse falls within the envelope and is compliant as an FABM 1,
while the HBM Fine falls outside the envelope; however, it is compliant as a soil treated with fly ash (SFA)
covered in BS EN 14227-14 (BSI, 2006e). Therefore, both HBMs are compliant with the appropriate BS EN,
MCHW1 (HA, 2009a) and the SROH (HAUC, 2002).
Figure 6: Particle size distribution of the HBM Fine and HBM Coarse
31.5
20.0
16.0
14.0
10.0
8.0
6.3
4.0
2.0
1.0
0.5
0
0.2
5
0.1
25
0.0
63
0
10
20
30
40
50
60
70
80
90
100
0.0 0.1 1 .0 1 0.0 1 00.0
Sieve size (mm)
Mass %
passing
HBM Fine
HBM Coarse
FABM 1 envelope min
FABM 1 envelope max
3.1.2 Test matrix The test matrix for the trial is shown in Table 2. The typical trench depth was 1000 mm, and the reinstatement
comprised 100 mm of sand to provide a consistent layer on which to place the test materials. The test materials
were placed in either 150 mm or 300 mm lifts, and testing was generally undertaken after 3, 5 and 8 passes with
the vibrotamper. These two construction profiles and the associated testing regimes were used for each of the
material types. Consultation with industry and the desk based review (Section 2.1) indicated that 8 passes would
be a pragmatic upper limit, as this is the method specification for GSB1/Type1 (HAUC, 2002).
Recycled utility arisings in trench reinstatement: compaction trial 19
Table 2: Test matrix for HBMs and GSB1/Type1 control compacted in 150 mm and 300 mm lifts
In Situ Testing
Material
Lift
thickness
(mm)
Lift
Number LWD GDP NDM PQI Clegg Panda2 or
BC&D DCP*
Sand 100 Test No test
1 Test after 3, 5 and 8 passes No test
2 No test
3 Test after 8 passes No test
4 No test
5 Test after 3, 5 and 8 passes No test
HBM Fine
HBM Coarse
GSB1/Type1
HBM Fine (DC)
HBM Coarse (DC)
150
6 Test after 8 passes
1 Test after 3, 5 and 8 passes No test
2 Test after 8 passes No test
HBM Fine
HBM Coarse
GSB1/Type1
HBM Coarse (DC)
300
3 Test after 3, 5 and 8 passes Test after
8 passes
*Note – DCP only undertaken on completed trench due to sensitivity of the test.
The test matrix (Table 2) allowed comparison of compaction behaviour and also the evaluation of the potential
of a range of in situ testing devices to be used for compliance and/or quality control purposes. A calibrated
nuclear density meter (NDM) was used, in direct transmission mode, to determine the in situ density and hence
degree of compaction achieved for the various materials across the range of installations.
Post construction testing using the LWD and GDP was undertaken on the reinstated trenches after 7 and 28 days.
In addition, Clegg, DCP and Panda2 testing was conducted on all the trenches after 28 days (Section 6.0).
3.1.3 Material reinstatement and test positions The trenches were divided into five test sections (Bays) to which an in situ test was allocated. A schematic
representation of the reinstatement profiles and test positions is given in Figure 7. The assigned bay remained
the same throughout the reinstatement for both 150 mm lifts and 300 mm lifts (Figure 8). The test bay
allocation was devised to ensure minimum modification/disruption to the surface prior to in situ testing and
facilitate on site management of the testing operations.
Figure 7: Schematic trench profile for 150 and 300 mm lifts and the in situ testing equipment
Recycled utility arisings in trench reinstatement: compaction trial 20
Figure 8: Testing bays for allocated test devices
3.2 Site trial construction The site trial was conducted over three days (12 to 14 November 2008 inclusive). The HBMs were mixed at the
site by the Independent Stabilising Company Ltd. The layout of the site trial is shown in Figure 9. The weather
varied between sunny and overcast on Day 1, to intermittent showers during Days 2 and 3.
Figure 9: Schematic plan view of site trial construction
HBM Fine
Lift thickness 150 mm
Trench number 1
Construction day 1
HBM Fine
Lift thickness 300 mm
Trench number 2
Construction day 1
GSB1/Type1
Lift thickness 150 mm
Trench number 6
Construction day 2
HBM Coarse
Lift thickness 300 mm
Trench number 4
Construction day 2
HBM Fine - delayed compaction
Lift thickness 150 mm
Trench number 7
Construction day 2
HBM Coarse - delayed compaction
Lift thickness 150 mm
Trench number 8
Construction day 3
HBM Coarse
Lift thickness 150 mmTrench number 5
Construction day 2
GSB1/Type1
Lift thickness 300 mm
Trench number 3
Construction day 1
HBM Coarse - delayed compaction
Lift thickness 300 mm
Trench number 9
Construction day 3
The trenches were excavated using a tracked excavator, supplied by NT Killingley Ltd (Figure 10), to the
nominal dimensions of 1 m deep and 0.8 m wide. The length of each trench was 2 m to allow each test device to
be assigned its own bay (Section 3.1.3) and to have an additional section that allowed safe access/egress to the
trench to conduct in situ testing.
Recycled utility arisings in trench reinstatement: compaction trial 21
Figure 10: Excavation of the trial trenches
Figure 11: Vibrotamper used in the trial
Compaction was achieved using a Wacker BS60-2 vibrotamper, shown in Figure 11. The specifications of the
particular model used in the trial are:
� Dimensions (length x width x height) 675 mm x 345 mm x 965 mm.
� Shoe size (width x length) 280 mm x 330 mm.
� Operating weight 66 kg.
� Compaction depth (depending on soil) 580 mm.
� Impact energy 85 J.
� Force/blow per CIMA - LEMB 13.4 kN.
� Percussion rate up to 700 blows/min.
� Travel speed up to 17.4 m/min.
The water content of the HBMs was assessed during the trial by moisture condition value (MCV) testing. The
excavator bucket was used to place the reinstatement material into the trench, which was then spread evenly by
an operator before an un-compacted measurement of the lift thickness was taken. Thickness was adjusted, prior
to compaction, to achieve a lift of reinstatement material of 150 mm or 300 mm. The reinstatement materials
were compacted using the designated number of passes with the vibrotamper. Testing was conducted with each
apparatus (Section 2.4) in the designated bay before the next allocation of passes was undertaken or the next lift
installed. The width of the vibrotamper foot in relation to the width of the trench meant that one pass comprised
three traverses.
The thickness of the lift was measured after 8 passes of the vibrotamper to check that the target lift thickness
had been achieved (Figure 12). Six measurements were taken along each trench and averaged. Control of the
lift thickness was readily achieved with the selected compaction methodology for the sand, GSB1/Type1, and
HBM Coarse. The HBM Fine proved more problematic and was noted to be more mobile (the material was easily
pushed around during compaction, Figure 13); although producing an acceptable mean layer thickness, there
was variability in the surface level of the individual lifts.
Figure 12: Measuring the compacted HBM lift
Figure 13: Compaction of material (HBM Fine)
Recycled utility arisings in trench reinstatement: compaction trial 22
3.3 In situ testing of reinstatement layers Eight in situ test devices were selected for use in the trial (Section 2.4) these were provided by members of the
project team and operated by trained technicians; the test devices and layout for testing within each trench are
shown in Figure 7 (Section 3.1.3). The NDM and PQI were calibrated for each material type using a gauge block
(Figure 14) with each material compacted to refusal in accordance with BS 1924-2 (BSI, 1990a).
Measurement of in situ density was determined using an NDM (Figure 14) in direct transmission and
backscatter, by trained operatives from Mid Sussex Testing Ltd. The results from the test in direct transmission
provided the baseline measurement of in situ density to which the other methodologies were compared. The
NDM was used on every lift that the indirect methodologies were used. The PQI (Figure 15) was provided by
Lafarge A&C UK and was trialled on Day 1 and part of Day 2. Its use was abandoned on Day 2 following initial
feedback from the interpretation of its data sets. In short, the usefulness of this device for testing asphalt did not
translate to other materials. Lafarge A&C UK also provided and operated an LWD (Figure 16). The surface
stiffness of the reinstated materials was measured using an LWD and a GDP (Figure 17), which were supplied by
Scott Wilson.
Figure 16: Lightweight deflectometer (LWD)
Figure 17: German dynamic plate (GDP)
The bearing capacity & deflectometer (BC&D) was provided by Loughborough University and operated by
research staff who visited on Day 2 (Figure 18). Tests were undertaken on the HBM Coarse compacted in 150
mm and 300 mm lifts and GSB1/Type1 compacted in 150 mm lifts. The Clegg impact hammer (Figure 19) was
used on Days 1 and 2. Measurements were taken at three points within its assigned bay. Unfortunately the
device malfunctioned on Day 3, so it was not possible to complete all the scheduled testing with it. However, it
was repaired and used to conduct the 28 day testing.
Figure 14: NDM gauge block for calibration and NDM
Figure 15: Asphalt pavement quality indicator (PQI)
Recycled utility arisings in trench reinstatement: compaction trial 23
Figure 18: Bearing capacity & deflectometer (BC&D)
Figure 19: Clegg impact hammer
The Panda2 (Figure 20) was provided and used by trained operatives from Corehard Ltd who visited the site on
Day 1; measurements were taken on the HBM Fine reinstated in 150 mm and 300 mm lifts and GSB1/Type1
reinstated in 300 mm lifts. It was also used to test all the trenches at 28 days. Dynamic cone penetrometer
testing (Figure 21) is not designed for use on single lifts; therefore, testing was carried out on the fully reinstated
trenches immediately after construction by trained operatives from Scott Wilson Ltd. Dynamic cone penetrometer
testing was also undertaken 7 and 28 days after compaction.
Figure 20: Variable energy dynamic penetrometer (Panda2)
Figure 21: Dynamic cone penetrometer (DCP)
Post compaction, the materials were left unsurfaced and untrafficked. All the in situ test methodologies proved
sufficiently portable and robust for use by trained operatives on both the HBMs and GSB1/Type1. The only
exception being the PQI, for which its usefulness for testing asphalt did not translate to HBMs and GSB1/Type1.
The site testing data sets are given in Section 4.0 along with laboratory test data sets for the HBMs.
4.0 Compaction behaviour Installation factors such as lift thickness, compactive effort and the effect of delayed compaction were included in
the test matrix (Section 3.1.2) to evaluate the compaction behaviour of HBMs and GSB1/Type1 control. The
nuclear density meter (NDM) was used to determine the in situ density of the HBMs and GSB1/Type1, and the
data related to a laboratory determined refusal density to assess whether adequate compaction was achieved.
The compaction methodologies adopted aimed to investigate the influence of factors such as lift thickness,
compactive effort and the effect of delayed compaction; and give a range of relative densities (compaction), for
subsequent assessment with the indirect test devices (Section 5.0).
Recycled utility arisings in trench reinstatement: compaction trial 24
4.1 Density results The NDM was calibrated for each material using a gauge block sample compacted to refusal in accordance with
BS 1924-2 (BSI, 1990a). Density data were collected in both backscatter and direct transmission modes, the
backscatter results were generally 5 to 10% lower than the values yielded from the direct transmission mode.
Direct transmission is the preferred method due to the deeper zone of influence, as backscatter measurements
only penetrate to a depth of 70 to 80 mm (BSI, 1990a). Therefore, the values in direct transmission mode have
been used as the reference density in the subsequent analysis of the results. A summary of the direct
transmission NDM results is presented in Table 3.
Table 3: NDM determined in situ density after 3, 5 and 8 passes of vibrotamper
In situ wet density (kg/m3)
Number of passes Material Lift thickness
(mm) Lift number
3 5 8
1 2052 2090 2071
3 - - 2121
5 2111 2157 2127 150
6 - - 2156
1 2008 2007 1972
2 - - 2091
HBM Fine
300
3 2021 2052 2080
1 2110 2180 2019
3 - - 2176
5 2173 2217 2236 150
6 - - 2202
1 2042 2036 2050
2 - - 2088
HBM Coarse
300
3 2031 2002 2125
1 2052 2188 2218
3 - - 2111
5 2059 2128 2223 150
6 - - 2141
1 2009 2075 2093
2 - - 2088
GSB1/Type1
300
3 2033 2056 2143
1 1949 1960 1998
3 - - 2124
5 2105 2104 2105
HBM Fine
(DC) 150
6 - - 2150
1 2174 2194 2173
3 - - 2224
5 2147 2195 2222 150
6 - - 2179
1 2034 2034 2089
2 - - 2198
HBM Coarse
(DC)
300
3 2028 2091 2120
Laboratory specimens were manufactured in 150 mm diameter cylinders. The material was compacted to refusal
using a vibrating hammer in accordance with BS EN 13286-51 (BSI, 2004f). The density of cylinders prepared in
this way is termed the refusal density. The refusal densities for the specimens, manufactured from the material
sampled at the time of installation, are given in Table 4. The hydraulic binder used in both HBMs can alter the
reference density of the material over time; therefore, care was taken to ensure that the material sampled for the
laboratory specimens was representative, and the specimens were manufactured in a timely manner.
The ratio between the refusal density and the in situ density, which was determined using the NDM, is termed
the relative density, and is expressed as a percentage:
In situ wet density (kg/m3) / Refusal wet density (kg/m3) x 100 = Relative density (%)
Recycled utility arisings in trench reinstatement: compaction trial 25
Table 4: Laboratory determined refusal wet densities
Material Laboratory wet density (Mg/m3)
HBM Fine 2.22
HBM Coarse 2.24
GSB1/Type1 2.28
HBM Fine (DC) 2.22
HBM Coarse (DC) 2.24
Compaction of HBMs for use in subbase and base layers of a pavement foundation is specified in Clause 870 of
the MCHW1 (HA, 2009a), which states that, the relative density must be ≥95%. In addition, a minimum value is
required for base layers; the result of each single determination of in situ wet density shall be not less than 92%
of the wet density of the HBM at its optimum water content (HA, 2009a).
The requirement for unbound mixtures (GSB1/Type1) is normally to follow a method specification for on site
compaction, which states the maximum lift thickness, compaction plant type and number of passes. As such,
achievement of adequate compaction of GSB1/Type1 is implied through compliance with the method specification
and compliance testing is not normally undertaken. The method specification, which comprised 8 passes over 150
mm lifts using the selected plant, within the SROH (HAUC, 2002) was followed. This means that adequate
compaction for the GSB1/Type1 should be achieved after 8 passes, whilst thicker lifts or a lower number of
passes would not be expected to achieve adequate compaction.
For comparative purposes, and the use of GSB1/Type1 as a control material, the threshold target of ≥95%
relative density (as specified for HBMs) was selected for the GSB1/Type1. GSB1/Type1 which does not meet this
requirement would potentially rut/deform to excess under trafficking, and/or potentially settle over time. The
target density for compaction of the GSB1/Type1 was determined using samples compacted to refusal using a
vibrating hammer, in accordance with BS EN 13286-4 (BSI, 2003a). The in situ density was determined using the
NDM and the refusal density (Table 4) achieved using the vibrating hammer was used as the reference, to
determine the relative density. The relative densities after 3, 5 and 8 passes of the vibrotamper are plotted for
the materials reinstated in 150 mm lifts in Figure 22.
Figure 22: Relative density after 3, 5 and 8 passes on 150 mm lifts
3 passes 5 passes 8 passes40%
45%
50%
55%
60%
65%
70%
75%
80%
85%
90%
95%
100%
Number of passes over 150 mm lifts
Relative Density
HBM Fine
HBM Fine (DC)
HBM Coarse (DC)
HBM Coarse
GSB1/Type 1
Target density
It can be seen from Figure 22 that the GSB1/Type1 material requires all 8 passes to achieve the target density,
while the HBM Coarse has achieved the target density after 3 passes and the HBM Fine after 5 passes. The
standard deviations indicate the threshold target is consistently achieved with the HBM Coarse and HBM Coarse
(DC), but not necessarily with the GSB1/Type1, HBM Fine and HBM Fine (DC). Retaining the material for 24 hours
Recycled utility arisings in trench reinstatement: compaction trial 26
prior to compaction (delayed compaction) did not adversely affect the ease of compaction of either the HBM
Coarse or HBM Fine, as both achieved similar relative densities.
When compacted in 300 mm lifts none of the materials achieved the target relative density after 3 and 5 passes
(Figure 23) and only the HBM Coarse (DC) material achieved the target relative density after 8 passes. The
increased in situ densities measured for the HBM Coarse (DC) may be related to a reduction in plastic character
associated the addition of lime to the feedstock material. However, the standard deviation around the HBM
Coarse (DC) material indicates that this value would not necessarily be consistently achieved. A mean relative
density less than 95% indicates a potential for subsequent settlement or durability issues. Therefore, more
compactive effort, either from heavier compaction plant or a greater number of passes, is required to consistently
achieve an acceptable degree of compaction for all materials when compacted in 300 mm lifts.
Figure 23: Relative density following 3, 5 and 8 passes on 300 mm lifts
8 passes 5 passes3 passes40%
45%
50%
55%
60%
65%
70%
75%
80%
85%
90%
95%
100%
Number of passes over 300 mm lifts
Relative Density
HBM Fine
HBM Coarse
HBM Coarse (DC)
GSB1/Type 1
Target density
The compaction methodology adopted gave a range of relative densities for the different materials, which are
suitable for evaluating the indirect test devices (Section 5.0).
4.2 Discussion The site trial and laboratory testing provided useful information on the compaction behaviour of two slow curing
HBMs with different particle size distributions, in comparison to a GSB1/Type1 unbound subbase aggregate. The
HBMs were installed on the same day of manufacture, and also following a delay between manufacture and
installation (termed delayed compaction) to assess the influence of storing the material for 24 hours.
The materials were compacted using the same plant and operatives, while the number of passes and lift
thickness were varied to assess the ease of compaction of each material. The GSB1/Type1 (150 mm lifts)
required all 8 passes with the vibrotamper, as specified in the SROH (HAUC, 2002), to achieve the target relative
density. The HBM Coarse (fresh and delayed compaction) proved to be the easiest material to compact, as the
target relative density was achieved after only 3 passes with the vibrotamper, when compacted in 150 mm lifts.
The HBM Fine achieved the target relative density after 5 passes with the vibrotamper, when compacted in 150
mm lifts. However, the HBM Fine behaved like a cohesive material, and the mobility of the soil under the
vibrotamper made it difficult to achieve a level surface (Section 3.2).
Surface tolerance is critical to compliant reinstatement works, and poor control of the lower layers of the
pavement will impact on the thickness of asphalt overlay or the surface tolerance of the reinstatement. Therefore,
given the nature of the HBM Fine it would be prudent to lay the material low to ensure the specified minimum
thickness of asphalt overlay, or lay coarser material above to achieve the desired level for asphalt overlay. The
Recycled utility arisings in trench reinstatement: compaction trial 27
HBM Coarse showed comparable surface level control and ability to be immediately overlaid by asphalt to
GSB1/Type1. The higher coarse aggregate content of the HBM Coarse is advantageous, since it allows greater
mechanical interlock of the aggregate particles which is associated with a higher bearing capacity and the ability
to support early trafficking without permanent deformation. HBM Coarse allowed for easier laying and
compaction, as shown by the increase in density measured for a given compactive effort (Section 6.1).
Only the HBM Coarse met the requirement in Clause 870, MCHW1 (HA, 2009a) for achieving ≥95% wet density
under compaction, when compacted in 300 mm lifts. The other materials (GSB1/Type1, HBM Fine, HBM Fine (DC)
and HBM Coarse) compacted in 300 mm lifts did not achieve the ≥95% criteria so their placement did not
achieve compliance for subbase and base applications for road foundations. However, these reinstatements
provided a range of densities against which to assess the in situ test devices (Section 5.0).
The compaction trial demonstrated that adequate compaction of GSB1/Type1 was only achieved following 8
passes of the vibrotamper over 150 mm lifts, which highlights the importance of compliance with the compaction
method specification. Conversely, the specific HBMs included in this study did not require the full compactive
effort to achieve the threshold density. The SROH (HAUC, 2002) recognises that some alternative reinstatement
materials may not require the full compaction specified for GSB1/Type1 and may be damaged if compaction is
continued; and recommends that such materials should be placed and compacted in accordance with the
manufacturer’s recommendations and with due regard to the requirements of Appendix A9. A guidance note for
the establishing a material specific compaction method specification is given in Appendix A.
4.3 Summary In summary, the site trial demonstrated that, for the given ground conditions and compaction plant:
� the two HBMs compacted in 150 mm lifts required less compactive effort than the GSB1/Type1;
� GSB1/Type1 achieved the target relative density after 8 passes when compacted in 150 mm lifts, in
accordance with the method specification for GSB1/Type1 stated in the SROH (HAUC, 2002);
� the HBM Coarse in 150 mm lifts was most easily compacted, achieving the target relative density in the least
number of passes;
� only the HBM Coarse (DC) material achieved the target relative density after 8 passes when compacted in 300
mm lifts, adequate compaction was not achieved for GSB1/Type1, HBM Coarse and HBM Fine material when
compacted in 300 mm lifts; and
� a 24 hour delay in compaction, with appropriate storage and hydraulic binder selection, was found to not
detrimentally affect the HBMs or GSB1/Type1.
5.0 Indirect test device evaluation This Section focuses on the potential to use selected test devices to determine adequate compaction during
installation. The relationship between compactive effort (lift thickness or number of passes) and relative density is
illustrated in Section 4.1. Test devices that produce a CBR value (discussed in Section 2.3) determined from
another measured parameter are not expected to correlate between tests. The in situ test devices selected for
further investigation were used by trained operatives during the site trial. The following Sections detail output
and evaluate suitability of selected test devices for compliance and/or control testing.
5.1 Asphalt pavement quality indicator The asphalt pavement quality indicator (PQI) was developed to measure the wet density (Section 2.3.1), or
compaction level, of an asphalt by the response of the electrical sensing field to changes in electrical impedance
of the material matrix, which in turn is a function of the composite resistivity and dielectric constant of the
material (Sawchuk, 1998). The PQI was assessed by measuring the density of the NDM gauge block samples
(Figure 14) and in situ during Day 1 and part of Day 2 of the site trial. The PQI output is reported as density
(kg/m3) and percentage water content.
The results from the direct comparison of the NDM and PQI on the gauge block samples are presented in Table
5; on the basis of these results the PQI was anticipated to yield 10 to 15% higher density results for the HBMs
and GSB1/Type1 compared to the results generated by the NDM.
Recycled utility arisings in trench reinstatement: compaction trial 28
Table 5: PQI and NDM density readings undertaken on NDM gauge block
Material PQI Mean Density (kg/m3) NDM Mean Density (kg/m3)
HBM Fine 2489 2224
HBM Coarse 2466 2236
GSB/Type1 2702 2280
During the site trial, the operator encountered difficulties in using the PQI on GSB1/Type1 and HBMs due to the
silt/sand sized fraction of these materials affecting the operating mechanism. This difficulty became more severe
during short periods of inclement weather and the test was abandoned; indicating its unsuitability for use with
HBMs and unbound mixtures.
In summary, the PQI is a proven highly advantageous in situ density testing apparatus for asphalt (being quick,
portable and avoiding health and safety considerations of the NDM), but this version was not suitable for testing
materials placed in the lower levels of trench reinstatements.
5.2 Dynamic plates The dynamic plate test methods are detailed in (Section 2.3.2). The objective of this testing was to see if a
correlation between degree of compaction and surface stiffness could be obtained. Therefore, this Section
primarily concentrates on the use of the dynamic plate tests for compliance testing during installation. Post
installation performance testing using the dynamic plate tests is discussed in Section 6.0. The post installation
testing focuses on monitoring stiffness development over time, rather than compliance and/or quality control
during installation. Quality control for HBMs is detailed in the quality manual for hydraulically bound mixtures
(WRAP, 2009a) and the template quality management scheme for their production (WRAP, 2009c).
The lightweight deflectometer (LWD), German dynamic plate (GDP) and the bearing capacity and deflectometer
(BC&D) were used in this trial. All three test devices where found to be portable, robust and relatively easy to
use. The main practical disadvantage being that they require a level test surface which may not always be
available – particularly for coarse granular materials. Lightweight dynamic plate (LWD) test results, for 3, 5 and 8
passes with the vibrotamper are plotted against relative density in Figure 24.
Figure 24: Lightweight dynamic plate data against relative density
1
10
100
1000
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
Surface stiffness (Mpa)
150 mm HBM Fine 150 mm HBM Coarse 150 mm GSB1/Type1 150 mm HBM Coarse (DC) 150 mm HBM Fine (DC)
300 mm HBM Fine 300 mm HBM Coarse 300 mm GSB1/Type1 300 mm HBM Coarse (DC)
Target relative
density (95%)
The relative density is derived from in situ NDM wet density data and laboratory refusal wet density data (Section
4.1). The scatter in data for the HBMs and GSB1/Type1 (Figure 24) does not indicate a correlation between
relative density (incurred from 3, 5 and 8 passes of the vibrotamper or change in lift thickness) and surface
stiffness measured with the LWD. However, the data show a relationship with material type.
Recycled utility arisings in trench reinstatement: compaction trial 29
Figure 25: German dynamic plate data against relative density
1
10
100
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
Surface stiffness (Mpa)
150 mm HBM Fine 150 mm HBM Coarse 150 mm GSB1/Type1 150 mm HBM Coarse (DC) 150 mm HBM Fine (DC)
300 mm HBM Fine 300 mm HBM Coarse 300 mm GSB1/Type1 300 mm HBM Coarse (DC)
Target relative
density (95%)
A correlation between material type and surface stiffness recorded with the GDP is also evident from Figure 25,
although the values recorded are lower than those from the LWD. The GSB1/Type1 has a generally higher
surface stiffness than the HBMs (Figure 25); however, there is no clear distinction between materials compacted
in 150 mm or 300 mm lifts, or relationship with increasing relative density. The generally consistent surface
stiffness values recorded throughout the trial, irrespective of number of passes or lift thickness, indicate that the
GDP, like the LWD, does not readily highlight where different compactive effort has been applied (that results in
varying densities) and/or where material has been placed in excessively thick lifts. The BC&D was used to test the
HBM Coarse compacted in 150 mm and 300 mm lifts and GSB1/Type1 compacted in 150 mm lifts (Figure 26).
Again there is no distinct correlation between the BC&D data and the increase in relative density, associated with
greater compaction.
Figure 26: Bearing capacity and deflectometer data against relative density
1
10
100
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
Surface stiffness (Mpa)
150 mm HBM Coarse 300 mm HBM Coarse 150 mm GSB1/Type1
Target relative
density (95%)
Recycled utility arisings in trench reinstatement: compaction trial 30
The testing procedure, assumptions in the analysis and equipment for each dynamic plate testing device have
several key differences, and as a result, they are not expected to produce identical outputs. For example, the
GDP only measures deflection and assumes a contact stress of 100 kPa, whereas the LWD and BC&D measure
the actual contact stress. Furthermore, the geophone which measures the deflection is fixed to the bearing plate
on the GDP, but is in direct contact with the material surface with the LWD and BC&D; therefore, the GDP would
generally be expected to measure a higher deflection, thus resulting in a lower surface stiffness. The most widely
used dynamic plate device in the UK is the LWD, and testing was undertaken to the protocol given in Highways
Agency guidance IAN73/06 (HA, 2009c). As the BC&D is currently being used in research projects at
Loughborough University, UK-wide experience with equipment is limited.
Although the dynamic plates did not demonstrate a discernable correlation with relative density, they do clearly
differentiate between the immediate potential for the materials to be overlaid with asphalt. The HBM Coarse,
being specifically designed for immediate overlay, gives some indication of the range of acceptable surface
stiffness values which could be adopted. The immediate surface stiffness values, recorded during site trial
construction, do not reflect the long term stiffness values anticipated from a bound material, once setting and
hardening has occurred. Therefore, further dynamic plate testing was undertaken after 7 and 28 days to monitor
the gain in surface stiffness. Material performance and the 7 and 28 day dynamic plate testing data sets are
discussed further in Section 6.0 (Post compaction trial testing).
5.3 Clegg impact hammer The Clegg impact hammer (Section 2.3.3) measures resistance to impact, through deceleration of the drop
hammer (accelerometer) when it makes contact with the test surface. The test yields an impact value which may
be converted to a CBR. The Clegg impact hammer was used on all trenches except 8 and 9 (HBM Coarse Delayed
Compaction) as a mechanical fault prevented measurements being taken. A total of five drops were carried out
for each measurement, and three measurements were taken within the designated bay. The results are
summarised in Table 6 and illustrated in Figure 27.
Table 6: Effect of increasing compactive effort on CBR measured using the Clegg impact hammer
Mean CBR (%)
Number of passes Material Lift thickness
(mm) Lift number
3 5 8
1 9 8 8
3 - - 8
5 7 7 5 150
6 - - 9
1 7 6 5
2 - - 6
HBM Fine
300
3 11 11 8
1 18 17 16
3 - - 12
5 14 15 12 150
6 - - 17
1 18 14 13
2 - - 14
HBM Coarse
300
3 16 18 19
1 37 ABORT 44
3 - - 48
5 26 57 53 150
6 - - 49
1 32 44 48
2 - - 71
GSB1/Type1
300
3 26 ABORT 63
1 7 8 7
3 - - 5
5 7 7 7
HBM Fine
(DC) 150
6 - - 17
Recycled utility arisings in trench reinstatement: compaction trial 31
The Clegg impact values were automatically converted to a CBR value by the hand set. Where the readout
displayed ‘abort’ for all three measurements is also recorded. The readout displays abort when the impact value
for successive hammer drops falls outside the limit of 2.
Figure 27: Mean Clegg impact hammer derived CBR value against relative density
1
10
100
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
Clegg derived CBR (%
)
150 mm HBM Fine 150 mm HBM Coarse 150 mm GSB1/Type1 150 mm HBM Fine (DC)
300 mm HBM Fine 300 mm HBM Coarse 300 mm GSB1/Type1
Target relative
density (95%)
The Clegg impact hammer results illustrated in Figure 27 do not indicate a correlation between CBR and relative
density. The CBR values are generally consistent for each material irrespective of the lift thickness or relative
density achieved. Therefore the Clegg impact hammer does not discriminate where the material has been laid
excessively thick (in this case 300 mm) or where ≥95% relative compaction has not been achieved (Section 4.1).
However, a correlation between material type and Clegg impact hammer derived CBR values is evident from
Figure 27.
5.4 Cone penetrometers Resistance to penetration is measured by a cone penetrometer, which is an intrusive test apparatus, described in
Section 2.3.4. The two selected for inclusion in the trial are the dynamic cone penetrometer (DCP) and Panda2.
The DCP test is not suitable for assessing a thin layer of material or individual lifts from the surface because of
the seating requirement, which means that the readings for the top part of the lift are questionable. Therefore,
this test was conducted on each trench upon completion of the reinstatement and a median DCP derived CBR
value was calculated for each lift for subsequent analysis. An example results sheet for the DCP test is shown in
Figure 28 for the HBM Coarse reinstated in 300 mm lifts.
The advantage of penetrometer devices is that they give a relative profile with depth. However, an increased
resistance to penetration with depth is expected due to additional confinement and potential friction on the rods.
Increased penetration resistance with depth in reinstatements can also be due to cumulative compaction from
compactive effort applied to overlying layers. Therefore, any trend of increasing CBR value with depth (Figure
28) is considered to be an artefact of testing a completed trench.
Recycled utility arisings in trench reinstatement: compaction trial 32
Figure 28: DCP profile on the HBM Coarse reinstated in 300 mm lifts
The median DCP derived CBR values for each lift, calculated from the reinstated trench data, plotted against
relative density, determined from in situ NDM wet density data and laboratory refusal wet density data, (Figure
29) do not indicate a distinct correlation. There is also no distinction between the trenches reinstated in 150 mm
or 300 mm lifts; therefore, the DCP does not differentiate between adequate and inadequate relative density and
compaction.
Figure 29: DCP derived CBR value against relative density
1
10
100
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
DCP derived CBR (%
)
150 mm HBM Fine 150 mm HBM Coarse 150 mm GSB1/Type1 150 mm HBM Coarse (DC) 150 mm HBM Fine (DC)
300 mm HBM Fine 300 mm HBM Coarse 300 mm GSB1/Type1 300 mm HBM Coarse (DC)
Target relative
density (95%)
The Panda2 variable energy dynamic penetrometer has several advantages over the DCP. The ability to apply a
lower force means it can be more sensitive to changes with depth giving a more accurate profile and seating is
not as problematic as with the DCP allowing the test to be carried out on individual lifts (Section 2.3.4).
Recycled utility arisings in trench reinstatement: compaction trial 33
The Panda2 was used to test the HBM Fine compacted in both 150 mm and 300 mm lifts and the GSB1/Type1
compacted in 300 mm lifts. The data was supplied in both resistance to penetration (expressed as a stress in
MPa) and CBR. Expression as CBR assumes a material specific relationship between penetration resistance and
CBR; therefore, for the purposes of the analysis, the resistance to penetration data has been used.
The mean resistance to penetration values for each lift plotted against relative density are given in Figure 30. A
trend of increased resistance with increasing density is evident for the GSB1/Type1, but not for the HBM Fine.
This may be due to the nature of the HBM Fine which lacked coarse aggregate. Therefore, the Panda2 is the only
device tested which correlated with relative density, albeit for only one material. However, the advantage of the
Panda2, like the DCP, is the profile of resistance to penetration with depth, which may be of more use than
taking a mean penetration value.
Figure 30: Panda2 resistance to penetration against relative density
1
10
86% 88% 90% 92% 94% 96% 98% 100%
Relative density
Mean resistance to penetration (MPa)
300 mm HBM Fine 300 mm GSB1/Type1
Target relative
density (95%)
The GSB1/Type1 data were supplied with reference curves for granular subbase, available with the software
(Figure 31); these criteria have been adopted for the GSB1/Type1 in the absence of material specific data sets.
The green line represents the chosen level of compaction for the material (line of reference), and values plotting
to the right of this line are greater than that level of compaction. The red line represents a 2 to 3% margin of
error below which the compaction of the material should not fall (line of failure).
Recycled utility arisings in trench reinstatement: compaction trial 34
Figure 31: Panda2 data for one 300 mm lift of GSB1/Type1 following 3, 5 and 8 passes
0
50
100
150
200
250
300
0.1 1 10 100Penetration resistance (MPa)
Penetration depth (mm)
3 passes
5 passes
8 passes
'Line of reference'
'Line of failure'
The Panda2 resistance data following 3 passes (Figure 31) indicate that adequate compaction has not been
achieved, as the profile plots to the left of the line of failure. This corresponds well with a relative density,
determined from in situ NDM wet density data and laboratory refusal wet density data, of 88%, which falls below
the target mean of ≥95% and the minimum value of 92% (HA, 2009a). The resistance data for 8 passes plot to
the right of the line of reference (Figure 31), which corresponds with a determined relative density of 92%. The
full depth profile of the GSB1/Type1 reinstatement (Figure 32) indicates two zones of inadequate compaction,
which correspond with the base of the first and second 300 mm lifts.
Figure 32: Panda2 profile of reinstated trench of GSB1/Type1 compacted in 300 mm lifts
0
100
200
300
400
500
600
700
800
900
0.1 1 10 100Penetration resistance (MPa)
Penetration depth (mm)
GSB1/Type1
'Line of failure'
'Line of reference'
The tolerance lines are determined to take into account variables associated with the material and depth,
including resistance due to confinement and the minimal contribution from frictional resistance on the shaft.
Recycled utility arisings in trench reinstatement: compaction trial 35
Due to the material specific nature of the tolerance lines calibration with the reference material is important when
applying the pass/fail criteria for compliance and control testing. However, as a minimum the Panda2 testing
allows a relative comparison of penetration through the depth of a reinstatement.
5.5 Discussion In general, the indirect test methods did not demonstrate a correlation with the range of densities recorded
during the study. Therefore, they are not suitable proxies for the NDM for compliance testing of in situ density.
However, the parameters measured demonstrated material specific correlations and the test devices are suitable
for monitoring consistency as part of a quality control test regime, if material specific relationships are
determined.
Unless direct density is determined in situ, compliance with Clause 870 MCHW1 (HA, 2009a), cannot be
demonstrated. Therefore, compliance with a compaction method specification, rather than targeting a threshold
density which may not be readily measured in situ, is a more pragmatic way forward. Material specific compaction
method specifications should be generated for particular HBMs to achieve adequate compaction. Material
suppliers of an HBM would be required to supply such a compaction method specification for their material
(HAUC, 2002). This would have to be based on a demonstration trial utilising a direct measurement of density
such as the NDM, to prove that the threshold density (adequate compaction) is achieved with the specified
compaction methodology. The indirect test devices trialled in this study would not be suitable for developing a
compaction method specification as the data are unlikely to be sufficiently robust. However, such a demonstration
trial would be an ideal opportunity to calibrate the indirect test devices for material specific correlations.
6.0 Post compaction trial testing Post compaction performance testing measured in situ at 7 and 28 days, based upon measuring fundamental
properties that can be used in design, is discussed in Section 6.1 and has been covered by a previous project
(WRAP, 2008). The laboratory data sets were used to classify the HBM with regard to the SROH (HAUC, 2002)
designation (Section 6.2). Performance testing is normally undertaken on materials that have complied with the
specification requirements.
6.1 In situ performance monitoring Performance monitoring of the trench reinstatements was undertaken up to 28 days age. The trenches were left
uncovered, untrafficked and open to the weather. None of the trenches displayed visible settlement or softening
of the surface material over this time period.
The SROH (HAUC, 2002) does not specify any particular tests for the performance monitoring of trench
reinstatements. The Highways Agency guidance IAN73/06 (HA, 2009c) only permits deflectometer tests for the in
situ assessment of surface stiffness (LWD or the trailer mounted falling weight deflectometer) and DCP for the
assessment of subgrade CBR. The other test devices have been included to allow a relative assessment of
material performance over time. The limitations of using CBR to describe a material’s performance are discussed
in Section 2.3.
The development of surface stiffness over time, measured using the LWD, is shown in Figure 33. Although the
GSB1/Type1 yielded higher stiffness values on the day of construction all the HBMs achieved higher stiffness
values than the GSB1/Type1 after 28 days (Figure 33), indicating that HBMs outperform GSB1/Type1 in the long
term. Furthermore, given the nature of the hydraulic binder used to manufacture the HBMs (a combination of PFA
and lime), it would be expected that the HBMs will continue to gain strength and stiffness beyond 28 days.
The HBMs subject to delayed compaction, display an increased stiffness compared to their immediately laid
counterparts (Figure 33), which indicates that the mellowing period prior to compaction was beneficial to their
performance.
Recycled utility arisings in trench reinstatement: compaction trial 36
Figure 33: LWD data sets from within 2 hours of reinstatement to 28 days age
0
50
100
150
200
250
300
350
400
450
500
0 7 14 21 28
Time post compaction (Day)
Surface stiffness (MPa)
HBM Fine
HBM Fine (DC)
HBM Coarse
HBM Coarse (DC)
GSB1/Type1
LWD data from the day after compaction indicate that the surface stiffness of the HBM Fine had increased and
was suitable for overlay on the day after compaction. However, there are no definitive published threshold values
to indicate when an HBM is suitable for overlay for any of the devices tested, and material specific demonstration
trials would be required to establish threshold values.
The in situ test data sets collected over a 28 day period are given in Table 7. The data indicate an improvement
in relative performance (increase in surface stiffness and CBR) during the monitoring period for all the materials.
Table 7: In situ testing up to 28 days age
LWD Stiffness (MPa) GDP Stiffness
(MPa)
DCP
Mean CBR (%)
Clegg
Mean CBR (%)
Number of days post construction Material
Lift
Thickness (mm)
0 1 7 28 0 7 28 0 28 0 28
150 4 21 41 145 22 19 125 17 26 9 34 HBM Fine
300 15 28 139 181 10 30 98 13 24 8 26
150 55 55 125 228 15 60 121 26 50 17 31 HBM Coarse
300 30 25 131 475 14 75 >150 26 34 19 34
150 34 - 118 155 65 51 >150 55 70 49 71 GSB1/Type1
300 58 - 107 105 64 60 >150 24 66 69 98
HBM Fine (DC) 150 13 - 59 291 7 24 106 9 55 17 26
150 30 - 188 301 11 56 >150 17 39 - 39 HBM Coarse (DC)
300 19 - 154 588 14 80 >150 17 32 - 43
Key: - = test not undertaken, Day 0 = testing undertaken with 2 hours of installation
Panda 2 testing was also undertaken on the trenches following 28 days. Figure 34 illustrates the resistance to
penetration with depth, for the GSB1/Type1 compacted in 300 mm lifts, following reinstatement and 28 days. The
similarity between the profiles indicates that there has been no significant improvement in performance following
28 days. Furthermore, the 28 day profile plots behind the pass/fail criteria indicating that adequate compaction
has not been achieved.
Recycled utility arisings in trench reinstatement: compaction trial 37
Figure 34: Panda2 data for GSB1/Type1 compacted in 300 mm lifts after reinstatement and 28 days
0
100
200
300
400
500
600
700
800
900
0.1 1 10 100
Penetration resistance (MPa)Penetration depth (mm)
GSB1/Type1 after reinstatement
GSB1/Type1 after 28 days
'Line of failure'
'Line of reference'
Figure 35 illustrates the resistance to penetration with depth, for the HBM Fine compacted in 150 mm lifts,
following reinstatement and 28 days. The two profiles indicate a general improvement in performance following
28 days.
Figure 35: Panda2 data for HBM Fine compacted in 150 mm lifts after reinstatement and 28 days
0
100
200
300
400
500
600
700
800
900
1 10 100
Penetration resistance (MPa)
Penetration depth (mm)
HBM Fine after reinstatement
HBM Fine after 28 days
For both the HBM Fine and HBM Coarse, stiffness and CBR values (measured by Clegg impact hammer and DCP)
increased with time as the materials cured. While in comparison, the GSB1/Type1 showed marginal increases,
which may be attributed to self cementing properties of the feedstock material.
Recycled utility arisings in trench reinstatement: compaction trial 38
6.2 Laboratory testing Specimens with a slenderness ratio of 1:1 (height: diameter) were manufactured for laboratory testing for
compressive strength (Rc) testing in accordance with BS EN 13286-41 (BSI, 2003b) and tested at 28 days age.
This testing regime is in accordance with MCHW1 (HA, 2009a) and is more onerous (the material has less time to
gain strength) than that given in the SROH (HAUC, 2002). A minimum compressive strength of 2 MPa after 90
days is required for compliance with the structural material for reinstatement (SMR) category detailed in the
SROH (HAUC, 2002). The results of the laboratory testing are presented in Table 8; all four HBMs achieved or
exceeded 2 MPa after 28 days.
Table 8: 28 day compressive strength data
Material Compressive Strength Rc (MPa)*
HBM Fine 3.1
HBM Fine (DC) 2.4
HBM Coarse 2.4
HBM Coarse (DC) 2.0
Note: * testing undertaken at 28 days age
7.0 Conclusions The compaction trial allowed comparison of both compaction behaviour of a selection of HBMs and GSB1/Type1,
and also the potential for a range of in situ testing devices to be used for compliance and/or quality control
purposes. A calibrated nuclear density meter (NDM) was used, in direct transmission mode as specified in
MCHW1 (HA, 2009a), to determine the in situ density of the materials, hence degree of compaction achieved, for
comparative analysis. It also provided a baseline against which to compare the other in situ test devices. All the
in situ test devices selected were portable, rapid to use and could, with varying degrees of ease and speed, be
operated by one or two operatives.
7.1 Compaction behaviour Non flowable materials achieve the required density via a suitable compaction methodology. The compaction trial
reported here supports the requirements specified in the SROH (HAUC, 2002) for reinstatement of GSB1/Type1 in
terms of lift thickness (150 mm) and compactive effort (8No. passes of a vibrotamper). Whereas, the NDM testing
showed that the HBMs compacted in 150 mm lifts required less compactive effort than the GSB1/Type1, to
achieve a relative density greater than 95%. The trial demonstrated that adequate compaction was not achieved
for the majority of materials compacted in 300 mm lifts, and more compactive effort (either heavier plant or a
greater number of passes) was necessary. Delayed compaction of the HBMs was not found to have an adverse
effect on the ease of compaction.
The SROH (HAUC, 2002) recognises that different materials require different compactive effort and requires
manufacturers of alternative reinstatement materials to give details of the compaction methodology for their
product. The achievement of adequate density for all material used within trench reinstatements is critical to
achieving short term performance requirements, such as an ability to support overlying layers and immediate
trafficking, and long term performance, such as avoiding settlement and providing durable reinstatements. The
MCHW1 Clause 870, (HA, 2009a) requirement for HBMs, used in subbase and base layers, to achieve compaction
to ≥95% of the mean wet density of laboratory specimens compacted to refusal, was adopted as the threshold
for the trial.
7.2 Indirect test device evaluation The PQI is only considered suitable for use on asphalt, although the manufacturers are developing a version of
this equipment which may also be suitable for testing GSB1/Type1 and HBMs. The main advantage of this test is
that it is rapid, non intrusive and does not have the health and safety issues associated with the nuclear source
required by the NDM.
All the plate testing devices were quick and easy to use and gave an immediate assessment of surface stiffness,
however, the results did not correlate with relative density so did not indicate if adequate compaction had been
achieved. The various plate testing devices have advantages and disadvantages; however, the LWD is the only
piece of equipment specified for use by the HA (HA, 2009c). The main concern in using these testing devices for
control testing is that the test is influenced by other unrelated factors. These factors include water content, the
size of the zone of influence which may include materials outside the trench, the confinement provided by the
surrounding materials and age of the material. The LWD is suitable for measuring performance, specifically the
surface stiffness and allows the performance of materials to be benchmarked (WRAP, 2008).
Recycled utility arisings in trench reinstatement: compaction trial 39
The Clegg impact hammer is commonly used in trench reinstatement work; however, the data from the trial did
not correlate with relative density, and therefore, the test did not indicate when adequate compaction had been
achieved. The impact value obtained from the device is dependent on the water content, material type and the
size of the zone of influence, which is dependent on the drop weight and contact area. The rapid nature of the
test means large data sets can be readily gathered which is ideal for checking consistency; however, the device
should be calibrated for the material type.
The intrusive nature of the DCP and Panda2 test devices means that the profile of the complete reinstatement
can be evaluated, the Panda2 has the added benefit of being lightweight enough to test a single lift (if required).
During the course of the trial it was found that the Panda2 is more suitable for trench reinstatements than a
standard DCP since it has a depth warning system and is more sensitive. It also has an advantage that the results
are given in real time, and pass/fail criteria can be selected for a range of materials, making it suitable for control
testing in the field. However, material specific calibration of the pass/fail criteria is important if the device is to be
used for compliance testing. The Panda2 did show a correlation between MPa and relative density for the
GSB1/Type1, but not for the HBM Fine. This is thought to be due to the nature of the HBM Fine which lacked
coarse aggregate. It is important with any penetration devices, which report in any values other than resistance
to penetration, that they are adequately calibrated for different materials.
7.3 Post compaction trial testing Performance of the materials was not addressed in detail in this study; however, as part of the assessment of the
materials some performance test data were garnered. For more information on material performance testing refer
to WRAP (2008). The laboratory compressive strength testing demonstrated compliance of both HBMs with the
SROH (HAUC, 2002) requirements for an SMR, achieving or exceeding the minimum strength of 2 MPa. However,
strength testing did not indicate whether the performance of the two HBMs would differ in the short term. The
HBMs used in this trial were slow curing (utilising lime and PFA); therefore, short term performance was
dependent on their unbound behaviour (hence coarse aggregate content was very important), while they gained
strength/stiffness over time. The long term performance of the HBMs would be dependent on their bound
performance. LWD performance testing at 28 days illustrated that the HBM Coarse was significantly stiffer than
the comparable GSB1/Type1 and HBM Fine. The HBMs would be expected to continue gaining strength/stiffness
over time, potentially up to and beyond 365 days age (WRAP, 2008).
7.4 Summary In summary, the compaction trial and in situ testing demonstrate:
� the two HBMs selected for the trial required less compactive effort than the GSB1/Type1 to achieve adequate
compaction, defined as complying with Clause 870 of MCHW1 (HA, 2009a);
� only direct density determination can be used to demonstrate adequate compaction and compliance with a
density specification, however, a range of in situ test devices may be used to monitor consistency as part of a
quality control test regime, this requires development of material specific threshold values;
� compliance with a suitable compaction method specification for HBMs should be sufficient to achieve
adequate compaction. Development of such a method specification would require a trial utilising direct density
determination to demonstrate adequate compaction (Appendix A);
� in the absence of in situ direct density determination compliance with a proven compaction method
specification is essential;
� the importance of developing and controlling a robust compaction methodology – this includes designing any
HBM to be as easily compacted and not overly sensitive to water content;
� in the absence of an alternative method for measuring density, the nuclear density meter should be used to
establish a suitable compaction methodology;
� materials that require less compactive effort will reduce health and safety issues related to hand arm vibration
syndrome;
� any quality control testing using target CBR values would need to be based upon an established relationship
between the test device and material type;
� the available in situ test devices have a range of advantages and limitations, these need to be understood in
the context of their use;
� it is important to design an HBM to achieve both short and long term performance;
Recycled utility arisings in trench reinstatement: compaction trial 40
� HBMs can be designed to permit immediate overlay, this is largely dependent on either using a rapid setting
hydraulic binder and/or designing a mechanically stable mixture (controlling the particle size distribution)
which contains suitable particles to provide aggregate interlock;
� HBMs gain strength and stiffness, and can be designed to achieve a range of performances;
� HBMs can be produced and installed to be compliant with the appropriate BS EN, MCHW1 (HA, 2009a) and
the SROH (HAUC, 2002);
� HBMs can be designed to require a reduced number of compaction passes, when compared to GSB1/Type1 –
however, any such proposal by the material producer must be supported by direct density data sets from a
demonstration trial.
8.0 References
AFNOR, 2000. XP P 94-105 Quality control of compaction. Association Française de Normalization.
AS, 2000. AS 1289.6.9.1: Methods for testing soils for engineering purposes- Method 6.9.1: Soil strength and
consolidation test - Determination of stiffness of soil - Clegg impact value (CIV). Australian Standards.
ASTM, 2007. ASTM D 5874 - Standard test method for determination of the impact value (IV) of a soil. American
society for testing materials, West Conshohocken, USA.
ASTM, 2008. ASTM D2167 – 08. Standard test method for density and unit weight of soil in place by the rubber
balloon method. American society for testing materials, West Conshohocken, USA.
Brown, SF, 1997. Achievements and challenges in asphalt pavement engineering. ISAP - 8th International
conference on asphalt pavements - Seattle, 1997.
BSI, 1990a. BS 1924-2 1990. Stabilized materials for civil engineering purposes. Methods of test for cement-
stabilized and lime-stabilized materials. British standards institution, London.
BSI, 1990b. BS 1377-9:1990. Methods of test for soils for civil engineering purposes. In situ tests. British
standards institution, London.
BSI, 1997. BS EN 933-1:1997. Tests for geometrical properties of aggregates. Determination of particle size
distribution. Sieving method. British standards institution, London.
BSI, 2003a. BS EN 13286-4:2003 - Unbound and hydraulically bound mixtures. Test methods for laboratory
reference density and water content. Vibrating hammer. British standards institution, London.
BSI, 2003b. BS EN 13286-41:2003. Unbound and hydraulically bound mixtures. Test method for determination of
the compressive strength of hydraulically bound mixtures. British standards institution, London.
BSI, 2004a. BS EN 14227-1:2004, Hydraulically bound mixtures. Specifications. Cement bound granular mixtures.
British standards institution, London.
BSI, 2004b. BS EN 14227-2:2004, Hydraulically bound mixtures. Specifications. Slag bound mixtures. British
standards institution, London.
BSI, 2004c. BS EN 14227-3:2004, Hydraulically bound mixtures. Specifications. Fly ash bound mixtures. British
standards institution, London.
BSI, 2004d. BS EN 14227-4:2004, Hydraulically bound mixtures. Specifications. Fly ash for hydraulically bound
mixtures. British standards institution, London.
BSI, 2004e. BS EN 14227-5:2004, Hydraulically bound mixtures. Specifications. Hydraulic road binder bound
mixtures. British standards institution, London.
Recycled utility arisings in trench reinstatement: compaction trial 41
BSI, 2004f. BS EN 13286-51:2004. Unbound and hydraulically bound mixtures. Method for the manufacture of
test specimens of hydraulically bound mixtures using vibrating hammer compaction. British standards institution,
London.
BSI, 2006a. BS EN 14227-10:2006, Hydraulically bound mixtures. Specifications. Soil treated by cement. British
standards institution, London.
BSI, 2006b. BS EN 14227-11:2006, Hydraulically bound mixtures. Specifications. Soil treated by lime. British
standards institution, London.
BSI, 2006c. BS EN 14227-12:2006, Hydraulically bound mixtures. Specifications. Soil treated by slag. British
standards institution, London.
BSI, 2006d. BS EN 14227-13:2006, Hydraulically bound mixtures. Specifications. Soil treated by hydraulic road
binder. British standards institution, London.
BSI, 2006e. BS EN 14227-14:2006, Hydraulically bound mixtures. Specifications. Soil treated by fly ash. British
standards institution, London.
BSI, 2007. BS EN 13242:2002 +A1:2007, Aggregates for unbound and hydraulically bound materials for use in
civil engineering work and road construction. British standards institution, London.
Clark, T, 2000. Site investigation for permanent way maintenance design. Railway engineering 2000.
Clayton, CRI, Matthews, MC, and Simons, NE, 1995, Site investigation, p. 584. Blackwell science, Oxford, 2nd Ed.
Clegg, B, 1976. An impact testing device for in situ base course evaluation. Proceedings, 8th ARRB conference,
Vol 8, Perth, Australia.
Croney, D, and Croney, P, 1998. The design and performance of road pavements. 3rd Ed. New York McGraw-Hill.
Fleming, PR and Cooper, MR, 1995. Investigations into the settlement characteristics of trench reinstatements.
The proceedings of the institution of civil engineers, municipal, 109(2), June 1995, pp 112-119, ISSN 0965 0903.
HA, 2004. Design manual for roads and bridges. Volume 7. Pavement design and maintenance. Section 1
Preamble. Part 2. Conservation and the use of secondary and recycled materials. The stationery office, London.
HA, 2009a. Manual of contract documents for highways works. Volume 1. Specification for highways works.
Series 800. Road pavements - unbound, cement and other hydraulically bound mixtures. The stationery office,
London.
HA, 2009b. Manual of contract documents for highways works. Volume 2. Notes for guidance. Series NG 800.
Road pavements - unbound, cement and other hydraulically bound mixtures. The stationery office, London.
HA, 2009c. Interim advice note IAN73/06 Revision 1 (2009). Design guidance for road pavement foundations.
(Draft HD25). Highways agency. London.
HAUC, 2002. Specification for the reinstatement of openings in highways. The stationary office, London.
Langton, DD, 1999. The Panda lightweight penetrometer for soil investigation and monitoring material
compaction. Ground engineering, Vol 32:99.
Parsons, AW, 1992. Compaction of soils and granular materials: A review of research performed at the Transport
Research Laboratory. The Stationery Office, London.
Sawchuk, P, 1998. Pavement quality indicator field operational testing and product transfer. IDEA project final
report. Contract NCHRP-47. IDEA Program. Transportation Research Board National Research Council.
Snowdon, RA, 1992. Compaction monitoring devices for earthworks. Research Report RR361. Transport Research
Laboratory, Crowthorne.
Recycled utility arisings in trench reinstatement: compaction trial 42
WRAP, 2005a. Identifying opportunities for recycling of excavated spoil from utility works within local authority
areas, and promoting the use of recycled materials through good practice in procurement. Banbury, WRAP.
WRAP, 2005b. The Quality Protocol for the production of aggregates from inert waste. Banbury, WRAP.
WRAP, 2008. Trench reinstatements: recycled materials and performance testing (WRAP project AGG105-005).
Report prepared by P Edwards. Banbury, WRAP.
WRAP, 2009a. Quality Manual for hydraulically bound mixtures: Notes for Guidance (WRAP project MRF106).
Report prepared by J Edwards, J Kennedy and P Edwards. Banbury, WRAP.
WRAP, 2009b. Recycled materials in trench reinstatement: Guidance document (WRAP project MRF106). Report
prepared by J Edwards and P Edwards. Banbury, WRAP.
WRAP, 2009c. Template quality management scheme for the production of a hydraulically bound mixture (WRAP
project MRF106). Report prepared by J Edwards. Banbury, WRAP.
Recycled utility arisings in trench reinstatement: compaction trial 43
Appendix A: Guidance note for
establishing a material specific compaction
method specification
A method specification must detail the precise procedure required to achieve adequate compaction for the stated
material. Development of a compaction method specification will require a demonstration trial to define method
and material variables. The variables to be defined from the demonstration trial include:
� method specification;
o plant – mass and relevant dimensions and other factors influencing performance;
o number of passes; and
o lift thickness.
� material / mixture range;
o grading;
o binder content;
o water content; and
o constituents – quality of fines.
Control of mixture parameters is covered by production control information contained in Annex A of the relevant
BS EN, further information is also given in the quality manual for hydraulically bound mixtures (WRAP 2009a).
Direct determination of density is required to assess the material and establish when adequate compaction has
been achieved.
Representative ground conditions for trench reinstatement are required to accommodate effects of lateral
confinement and underlying material.
The compaction method specification should be supplied with the material, along with workability and production
data.
Material specific threshold values and pass/fail criteria for control testing may be determined at the same time as
the method specification development. Methods suitable for control testing include the Clegg impact hammer,
Panda2 and dynamic plates. NOTE: these are not suitable as a proxy for direct density determination but will
monitor consistency as part of a quality control regime.
www.wrap.org.uk/construction