Steel Design Guide Series Industrial Buildings - صفحه اصلی · PDF fileSteel Design Guide Series Industrial Buildings Roofs to Column Anchorage Roofs to Column Anchorage James

Steel Design Guide Series

Industrial BuildingsRoofs to Column Anchorage

Steel Design Guide Series

Industrial BuildingsRoofs to Column Anchorage

Roofs to Column AnchorageJames M. FisherComputerized Structural Design, Inc.Milwaukee, WI

A M E R I C A N I N S T I T U T E O F S T E E L C O N S T R U C T I O N

© 2003 by American Institute of Steel Construction, Inc. All rights reserved.This publication or any part thereof must not be reproduced in any form without permission of the publisher.

Administrator

Rectangle

Copyright 1993

by

American Institute of Steel Construction, Inc.

All rights reserved. This book or any part thereofmust not be reproduced in any form without the

written permission of the publisher.

The information presented in this publication has been prepared in accordance with rec-ognized engineering principles and is for general information only. While it is believedto be accurate, this information should not be used or relied upon for any specific appli-cation without competent professional examination and verification of its accuracy,suitablility, and applicability by a licensed professional engineer, designer, or architect.The publication of the material contained herein is not intended as a representationor warranty on the part of the American Institute of Steel Construction or of any otherperson named herein, that this information is suitable for any general or particular useor of freedom from infringement of any patent or patents. Anyone making use of thisinformation assumes all liability arising from such use.

Caution must be exercised when relying upon other specifications and codes developedby other bodies and incorporated by reference herein since such material may be mod-ified or amended from time to time subsequent to the printing of this edition. TheInstitute bears no responsibility for such material other than to refer to it and incorporateit by reference at the time of the initial publication of this edition.

Printed in the United States of America

Second Printing: October 2003


Administrator

Rectangle

TABLE OF CONTENTS

Part 1

INDUSTRIAL BUILDINGS -GENERAL

1. INTRODUCTION TO PART 1 . . . . . 1

2. LOADING CONDITIONS ANDLOADING COMBINATIONS . . . . . . . . 1

3. OWNER ESTABLISHED CRITERIA 23.1 Slab-on-Grade Design . . . . . . . . . . . . . . . . 23.2 Jib Cranes . . . . . . . . . . . . . . . . . . . . . . . . . . 33.3 Interior Vehicular Traffic . . . . . . . . . . . . . . 33.4 Future Expansion . . . . . . . . . . . . . . . . . . . . 33.5 Dust Control/Ease of Maintenance . . . . . . . 3

4. ROOF SYSTEMS . . . . . . . . . . . . . . . 34.1 Steel Deck for Built-up or Membrane

Roofs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44.2 Metal Roofs . . . . . . . . . . . . . . . . . . . . . . . . . 44.3 Insulation and Roofing . . . . . . . . . . . . . . . . 64.4 Expansion Joints . . . . . . . . . . . . . . . . . . . . . 64.5 Roof Pitch, Drainage and Ponding . . . . . . . 84.6 Joists and P u r l i n s . . . . . . . . . . . . . . . . . . . . . 9

5. ROOF T R U S S E S . . . . . . . . . . . . . . . . 95.1 General Design and Economic

Considerations . . . . . . . . . . . . . . . . . . . . . . . 95.2 Connection Considerations . . . . . . . . . . . . 115.3 Truss Bracing . . . . . . . . . . . . . . . . . . . . . . 115.4 Erection Bracing . . . . . . . . . . . . . . . . . . . . 145.5 Other Considerations . . . . . . . . . . . . . . . . . 15

6. WALL SYSTEMS . . . . . . . . . . . . . . 156.1 Field Assembled Panels . . . . . . . . . . . . . . 166.2 Factory Assembled Panels . . . . . . . . . . . . 166.3 Precast Wall Panels . . . . . . . . . . . . . . . . . . 176.4 Masonry Walls . . . . . . . . . . . . . . . . . . . . . 176.5 Girts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186.6 Wind Columns . . . . . . . . . . . . . . . . . . . . . 19

7. FRAMING SCHEMES . . . . . . . . . . 197.1 Braced Frames vs. Rigid Frames . . . . . . . 207.2 Tube Columns vs. W Sections . . . . . . . . . 207.3 Economic Considerations . . . . . . . . . . . . . 20

8. BRACING SYSTEMS . . . . . . . . . . . 218.1 Rigid Frame Systems . . . . . . . . . . . . . . . . 218.2 Roof Bracing Systems . . . . . . . . . . . . . . . . 22

8.3 Temporary Bracing . . . . . . . . . . . . . . . . . . 24

9. COLUMN ANCHORAGE . . . . . . . . 259.1 Resisting Tension Forces with Anchor

Bolts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259.2 Resisting Shear Forces using Shear Friction

Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299.3 Resisting Shear Forces through Bearing .. 339.4 Column Anchorage Examples . . . . . . . . . . 34

10. SERVICEABILITY CRITERIA . . . . 4010.1 Serviceability Criteria for Roof Design . . . 4010.2 Metal Wall Panels . . . . . . . . . . . . . . . . . . . 4110.3 Precast Wall Panels . . . . . . . . . . . . . . . . . . 4110.4 Masonry Walls . . . . . . . . . . . . . . . . . . . . . 41

Part 2

INDUSTRIAL BUILDINGS -WITH CRANES

11. INTRODUCTION TO PART 2 . . . . . 4211.1 AISE Building Classifications . . . . . . . . . 4211.2 CMAA Crane Classifications . . . . . . . . . . 42

12. FATIGUE . . . . . . . . . . . . . . . . . . . . 43

13. CRANE INDUCED LOADS ANDLOAD COMBINATIONS . . . . . . . . . . . 44

13.1 Vertical Impact . . . . . . . . . . . . . . . . . . . . . 4413.2 Side T h r u s t . . . . . . . . . . . . . . . . . . . . . . . . . 4413.3 Longitudinal or Tractive Force . . . . . . . . . 4513.4 Crane Stop Forces . . . . . . . . . . . . . . . . . . . 4513.5 Eccentricities . . . . . . . . . . . . . . . . . . . . . . . 4513.6 Seismic Loads . . . . . . . . . . . . . . . . . . . . . . 4613.7 Load Combinations . . . . . . . . . . . . . . . . . . 46

14. ROOF SYSTEMS . . . . . . . . . . . . . . . 47

15. WALL SYSTEMS . . . . . . . . . . . . . . 47

16. FRAMING SYSTEMS . . . . . . . . . . . 47

17. BRACING SYSTEMS . . . . . . . . . . . 4817.1 Roof Bracing . . . . . . . . . . . . . . . . . . . . . . . 4817.2 Wall Bracing . . . . . . . . . . . . . . . . . . . . . . . 49

18. CRANE RUNWAY DESIGN . . . . . . 5118.1 Crane Runway Beam Design Procedure . . 5118.2 Plate Girders . . . . . . . . . . . . . . . . . . . . . . . 5618.3 Simple Span vs. Continuous Runways . . . 60


Administrator

Rectangle

18.4 Channel Caps . . . . . . . . . . . . . . . . . . . . . . 6118.5 Runway Bracing Concepts . . . . . . . . . . . . 6118.6 Crane Stops . . . . . . . . . . . . . . . . . . . . . . . . 6218.7 Crane Rail Attachments . . . . . . . . . . . . . . 62

18.7.1 Hook Bolts . . . . . . . . . . . . . . . . . . 6218.7.2 Rail Clips . . . . . . . . . . . . . . . . . . . 6318.7.3 Rail Clamps . . . . . . . . . . . . . . . . . 6318.7.4 Patented Rail Clips . . . . . . . . . . . . 6318.7.5 Design of Rail Attachments . . . . . 63

18.8 Crane Rails and Crane Rail Joints . . . . . . . 64

19. CRANE RUNWAY FABRICATION& ERECTION TOLERANCES . . . . . . . 64

20. COLUMN DESIGN . . . . . . . . . . . . . 6720.1 Base Fixity and Load Sharing . . . . . . . . . . 6720.2 Preliminary Design M e t h o d s . . . . . . . . . . . 70

20.2.1 Obtaining Trial Moments of Inertiafor Stepped Columns: . . . . . . . . . 71

20.2.2 Obtaining Trial Moments of Inertia

for Double Columns: . . . . . . . . . . 7220.3 Final Design Procedures . . . . . . . . . . . . . . 7220.4 Economic Considerations . . . . . . . . . . . . . 78

21. OUTSIDE CRANES . . . . . . . . . . . . . 79

22. UNDERHUNG CRANE SYSTEMS . 79

23. MAINTENANCE AND REPAIR . . . 80

24. SUMMARY AND DESIGNPROCEDURES . . . . . . . . . . . . . . . . . . . 80

Acknowledgments . . . . . . . . . . . . . . . 81

REFERENCES . . . . . . . . . . . . . . . . . 82

APPENDIX A . . . . . . . . . . . . . . . . . . 84

APPENDIX B . . . . . . . . . . . . . . . . . . 94

Note: This Design Guide is generally based on AISC ASD provisions. Any LRFD provisions havebeen specifically referenced as LRFD.


Administrator

Rectangle

PREFACE

This booklet was prepared under the direction of theCommittee on Research of the American Institute ofSteel Construction, Inc. as part of a series of publica-tions on special topics related to fabricated structuralsteel. Its purpose is to serve as a supplemental referenceto the AISC Manual of Steel Construction to assistpracticing engineers engaged in building design.

The design guidelines suggested by the author that areoutside the scope of the AISC Specifications or Codedo not represent an official position of the Institute andare not intended to exclude other design methods andprocedures. It is recognized that the design of struc-tures is within the scope of expertise of a competent li-censed structural engineer, architect or other licensedprofessional for the application of principles to a par-ticular structure.

The sponsorship of this publication by the AmericanIron and Steel Institute is gratefully acknowledged.

The information presented in this publication has been prepared in accordance with recognizedengineering principles and is for general information only. While it is believed to be accurate, thisinformation should not be used or relied upon for any specific application without competent professionalexamination and verification of its accuracy, suitability, and applicability by a licensed professionalengineer, designer or architect. The publication of the material contained herein is not intended as arepresentation or warranty on the part of the American Institute of Steel Construction, Inc. or theAmerican Iron and Steel Institute, or of any other person named herein, that this information is suitablefor any general or particular use or of freedom infringement of any patent or patents. Anyone making useof this information assumes all liability arising from such use.


Administrator

Rectangle

Administrator

Rectangle

1. INTRODUCTION

Part 1INDUSTRIAL BUILDINGS -GENERAL

Although the basic structural and architectural com-ponents of industrial buildings are relatively simple,combining all of the elements into a functional economi-cal building can be a complex task. General guidelinesand criteria to accomplish this task can be stated. Thepurpose of this guide is to provide the industrial buildingdesigner with guidelines and design criteria for the de-sign of buildings without cranes, or for buildings withlight-to-medium duty cycle cranes. Part 1 deals withgeneral topics on industrial buildings. Part 2 deals withstructures containing cranes.

Most industrial buildings primarily serve as an en-closure for production and/or storage. The design of in-dustrial buildings may seem logically the province of thestructural engineer. It is essential to realize however,that most industrial buildings involve much more thanstructural design. The designer may assume an expandedrole and may be responsible for site planning, establish-ing grades, handling surface drainage, parking, on-sitetraffic, building aesthetics, and, perhaps, landscaping.Access to rail and the establishment of proper floor ele-vations (depending on whether direct fork truck entry torail cars is required) are important considerations.Proper clearances to sidings and special attention tocurved siding and truck grade limitations are also essen-tial.

2. LOADING CONDITIONS ANDLOADING COMBINATIONS

Loading conditions and load combinations for in-dustrial buildings without cranes are well established bybuilding codes.

Loading conditions are categorized as follows:

1. Dead load: This load represents the weight ofthe structure and its components, and is usuallyexpressed in pounds per square foot. In an in-dustrial building, the building use and industrialprocess usually involve permanent equipmentwhich is supported by the structure. This equip-ment can sometimes be represented by a uni-form load (known as a collateral load), but thepoints of attachment are usually subjected toconcentrated loads which often require a sepa-rate analysis to account for the localized effects.

2. Live load: This load represents the force im-posed on the structure by the occupancy and useof the building. Building codes give minimumdesign live loads in pounds per square foot,which vary with the classification of occupancyand use. While live loads are expressed as uni-form, as a practical matter any occupancy load-ing is inevitably nonuniform. The degree ofnonuniformity which is acceptable is a matterof engineering judgment. Some building codesdeal with nonuniformity of loading by specify-ing concentrated loads in addition to uniformloading for some occupancies. In an industrialbuilding, often the use of the building may re-quire a live load in excess of the code statedminimum. Often this value is specified by theowner or calculated by the engineer. Also, theloading may be in the form of significant con-centrated loads as in the case of storage racks ormachinery.

3. Snow loads: Most codes differentiate betweenroof live and snow loads. Snow loads are afunction of local climate, roof slope, roof type,terrain, building internal temperature, andbuilding geometry. These factors may betreated differently by various codes.

4. Rain loads: These loads are now recognized asa separate loading condition. In the past, rainwas accounted for in live load. However, somecodes have a more refined standard. Rain load-ing can be a function of storm intensity, roofslope, and roof drainage. There is also the po-tential for rain on snow in certain regions.

5. Wind loads: These are well codified, and are afunction of local climate conditions, buildingheight, building geometry and exposure as de-termined by the surrounding environment andterrain. Typically based on a 50 year recurrenceinterval - maximum 5 second gust. Buildingcodes account for increases in local pressure atedges and corners, and often have stricter stan-dards for individual components than for thegross building. Wind can apply both inwardand outward forces to various surfaces on thebuilding exterior and can be affected by size ofwall openings. Where wind forces produceoverturning or net upward forces, there must bean adequate counterbalancing structural deadweight or the structure must be anchored to anadequate foundation.

- 1 -


Administrator

Rectangle

6. Earthquake loads: Seismic loads are estab-lished by building codes and are based on:

a. the degree of seismic risk,

b. the degree of potential damage,

c. the possibility of total collapse, and

d. the feasibility of meeting a given level ofprotection.

Earthquake loads in building codes are usuallyequivalent static loads. Seismic loads are gen-erally a function of:

a. the geographical and geological location ofthe building,

b. the use of the building,

c. the nature of the building structural system,

d. the dynamic properties of the building,

e. the dynamic properties of the site, and

f. the weight of the building and the distribu-tion of the weight.

Load combinations are formed by adding the effectsof loads from each of the load sources cited above. Codesor industry standards often give specific load combina-tions which must be satisfied. It is not always necessaryto consider all loads at full intensity. Also, certain loadsare not required to be combined at all. For example, windneed not be combined with seismic. In some cases only aportion of a load must be combined with other loads.When a combination does not include loads at full inten-sity it represents a judgment as to the probability of si-multaneous occurrence with regard to time and intensity.

3. OWNER ESTABLISHED CRITERIA

Every industrial building is unique. Each is plannedand constructed to requirements relating to building us-age, the process involved, specific owner requirementsand preferences, site constraints, cost, and building regu-lations. The process of design must balance all of thesefactors. The owner must play an active role in passing onto the designer all requirements specific to the buildingsuch as:

1. area, bay size, plan layout, aisle location, futureexpansion provisions,

2. clear heights,

3. relations between functional areas, productionflow, acoustical considerations,

4. exterior appearance,

5. materials and finishes, etc.,

6. machinery, equipment and storage method, and

7. loads.

There are instances where loads in excess of codeminimums are required. Such cases call for owner in-volvement. The establishment of loading conditions pro-vides for a structure of adequate strength. A related set ofcriteria are needed to establish the serviceability behav-ior of the structure. Serviceability design considers suchtopics as deflection, drift, vibration and the relation of theprimary and secondary structural systems and elementsto the performance of nonstructural components such asroofing, cladding, equipment, etc. Serviceability issuesare not strength issues but maintenance and human re-sponse considerations. Serviceability criteria are dis-cussed in detail in Serviceability Design Considerationsfor Low-Rise Buildings which is part of the AISC SteelDesign Guide Series(17) Criteria taken from the DesignGuide are presented in this text as appropriate.

As can be seen from this discussion, the design of anindustrial building requires active owner involvement.This is also illustrated by the following topics, of slab-on-grade design, jib cranes, interior vehicular traffic,and future expansion.

3.1 Slab-on-Grade Design

One important aspect to be determined is the spe-cific loading to which the floor slab will be subjected.Forklift trucks, rack storage systems, or wood dunnagesupporting heavy manufactured items cause concen-trated loads in industrial structures. The important pointhere is that these loadings are nonuniform. The slab-on-grade is thus often designed as a plate on an elastic foun-dation subject to concentrated loads.

It is common for owners to specify that slabs-on-grade be designed for a specific uniform loading (e.g.,500 psf). If a slab-on-grade is subjected to a uniformload, it will develop no bending moments. Minimumthickness and no reinforcement would be required. Thefrequency with which the authors have encountered therequirement of design for a uniform load and the generallack of appreciation of the inadequacy of such a criteriaby many owners and plant engineers has prompted the in-clusion of this topic in this guide. Real loads are not uni-form, and an analysis using an assumed nonuniform loador the specific concentrated loading for the slab is re-quired. An excellent reference for the design of slabs-on-grade is Designing Floor Slabs on Grade by Ringoand Anderson(33).

-2-


Administrator

Rectangle

3.2 Jib Cranes

Another loading condition that should be consideredis the installation of jib cranes. Often the owner has plansto install such cranes at some future date. But since theyare a purchased item - often installed by plant engineer-ing personnel or the crane manufacturer - the owner mayinadvertently neglect them during the design phase.

Jib cranes which are simply added to a structure cancreate a myriad of problems, including column distortionand misalignment, column bending failures, crane run-way and crane rail misalignment, and excessive columnbase shear. It is essential to know the location and size ofjib cranes in advance, so that columns can be properly de-signed and proper bracing can be installed if needed.Columns supporting jib cranes should be designed tolimit the deflection at the end of the jib boom to boomlength divided by 225.

3.3 Interior Vehicular Traffic

The designer must establish the exact usage to whichthe structure will be subjected. Interior vehicular trafficis a major source of problems in structures. Forklifttrucks can accidentally buckle the flanges of a column,shear off anchor bolts in column bases, and damagewalls.

Proper consideration and handling of the forklifttruck problem may include some or all of the following:

1. Use of masonry or concrete exterior walls inlieu of metal panels. (Often the lowest sectionof walls is masonry or concrete, and metal pan-els are used above.)

2. Installation of fender posts (bullards) for col-umns and walls may be required where speedand size of fork trucks are such that a column orload bearing wall could be severely damaged orcollapsed upon impact.

3. Use of metal guard rails or steel plate adjacentto wall elements may be in order.

4. Curbs.

Lines defining traffic lanes painted on factory floorshave never been successful in preventing structural dam-age from interior vehicular operations. The only realisticapproach for solving this problem is to anticipate poten-tial impact and damage and to install barriers and/or ma-terials that can withstand such abuse.

3.4 Future Expansion

Except where no additional land is available, everyindustrial structure is a candidate for future expansion.

Lack of planning for such expansion can result in consid-erable expense.

When consideration is given to future expansion,there are a number of practical considerations that re-quire evaluation.

1. The directions of principal and secondary fram-ing members require study. In some cases itmay prove economical to have a principal frameline along a building edge where expansion isanticipated and to design edge beams, columnsand foundations for the future loads. If thestructure is large and any future expansionwould require creation of an expansion joint at ajuncture of existing and future construction, itmay be prudent to have that edge of the buildingconsist of nonload-bearing elements. Obvi-ously, foundation design must also include pro-vision for expansion.

2. Roof Drainage: An addition which is con-structed with low points at the junction of theroofs can present serious problems in terms ofwater, ice and snow piling effects.

3. Lateral stability to resist wind and seismic load-ings is often provided by X-bracing in walls orby shear walls. Future expansion may requireremoval of such bracing. Formal notificationshould be made to the owner of the critical na-ture of wall bracing and its location to preventaccidental removal. In this context, bracing caninterfere with many plant production activitiesand the importance of such bracing cannot beoveremphasized to the owner and plant engi-neering personnel. Obviously, the location ofbracing to provide the capability for future ex-pansion without its removal should be the goalof the designer.

3.5 Dust Control/Ease of Maintenance

In certain buildings (e.g. food processing plants)dust control is essential. Ideally there should be no hori-zontal surfaces on which dust can accumulate. Tube sec-tions as purlins reduce the number of horizontal surfacesas compared to C's, Z's, or joists. If horizontal surfacescan be tolerated in conjunction with a regular cleaningprogram, C's or Z's may be preferable to joists. Thesame thinking should be applied to the selection of mainframing members (i.e. tubes or box sections may be pref-erable to beams or trusses).

4. ROOF SYSTEMS

The roof system is often the most expensive part ofan industrial building (even though walls are more costly

- 3 -


Administrator

Rectangle

per square foot). Designing for a 20 psf mechanical sur-charge load when only 10 psf is required adds cost over alarge area.

Often the premise which guides the design is that theowner will always be hanging new piping or installingadditional equipment, and a prudent designer will allowfor this in the system. If this practice is followed, theowner should be consulted, and the decision to provideexcess capacity should be that of the owner. The designlive loads and collateral (equipment) loads should benoted on the structural plans.

4.1 Steel Deck for Built-up or Membrane Roofs

Decks are commonly 1-1/2 in. deep, but deeperunits are also available. The Steel Deck Institute(9) hasidentified three standard profiles for 1-1/2 in. steel deck,(narrow rib, intermediate rib and wide rib) and has pub-lished load tables for each profile for thicknesses varyingfrom .0299 to .0478 inches. These three profiles, (shownin Table 4.1) NR, IR, and WR, correspond to the manu-facturers' designations A, F and B respectively. TheSteel Deck Institute identifies the standard profile for 3in. deck as 3DR. A comparison of weights for each pro-file in various gages shows that strength to weight ratio ismost favorable for wide rib and least favorable for nar-row rib deck. In general, the deck selection which resultsin the least weight per square foot may be the most eco-nomical. However consideration must also be given tothe flute width because the insulation must span theflutes. In the northern areas of the US, high roof loadsand thick insulation generally make the wide rib (B) pro-file predominant. In the South, low roof loads and thin-ner insulation make the intermediate profile common.Where very thin insulation is used narrow rib deck maybe required, although this is not a common profile. Ingeneral the lightest weight deck consistent with insula-tion thickness and span should be used.

In addition to the load, span, and thickness relationsestablished by the load tables, there are other considera-tions in the selection of a profile and gage for a given loadand span. First, the Steel Deck Institute limits deflectiondue to a 200 1b concentrated load at midspan to span di-vided by 240. Secondly, the Steel Deck Institute has pub-lished a table of maximum recommended spans for con-struction and maintenance loads (Table 4.1), and, finallyFactory Mutual lists maximum spans for various profilesand gages in its Approval Guide(3) (Table 4.2 ).

Factory Mutual in its Loss Prevention Guide(LPG)l-28 Insulated Steel Deck(23) provides a standardfor attachment of insulation to steel deck. LPG 1-29Loose Laid Ballasted Roof Coverings(23) gives a standardfor the required weight and distribution of ballast forroofs that are not adhered.

LPG 1-28 requires a side lap fastener between sup-ports. This fastener prevents adjacent panels from de-flecting differentially when a load exists at the edge ofone panel but not on the edge of the adjacent panel. Fac-tory Mutual permits an over span from its published ta-bles of 6 inches (previously an overspan of 10 percenthad been allowed) when "necessary to accommodate col-umn spacing in some bays of the building. It should notbe considered an original design parameter." The SteelDeck Institute recommends that the side laps in cantile-vers be fastened at twelve inches on center.

Steel decks can be attached to supports by welds orfasteners, which can be power or pneumatically installedor self-drilling, self-tapping. The Steel Deck Institute inits Specifications and Commentary for Steel RoofDeck(38) requires a maximum attachment spacing of 18in. along supports. Factory Mutual requires the use of 12in. spacing as a maximum; this is more common. The at-tachment of roof deck must be sufficient to provide brac-ing to the structural roof members, to anchor the roof toprevent uplift, and, in many cases, to serve as a dia-phragm to carry lateral loads to the bracing. While thestandard attachment spacing may be acceptable in manycases decks designed as diaphragms may require addi-tional connections. Diaphragm capacities can be deter-mined through use of Ref. 11.

Manufacturers of metal deck are constantly re-searching ways to improve section properties with maxi-mum economy. Considerable differences in cost mayexist between prices from two suppliers of "identical"deck shapes; therefore the designer is urged to researchthe cost of the deck system carefully. A few cents persquare foot savings on a large roof area can mean a sig-nificant savings to the owner.

Several manufacturers can provide steel roof deckand wall panels with special acoustical surface treat-ments for specific building use. Properties of such prod-ucts can be obtained from the manufacturers. Specialtreatment for acoustical reasons must be specified by theowner.

4.2 Metal Roofs

Standing seam roof systems were first introduced inthe late 1960's, and today many manufacturers producestanding seam panels. A difference between the standingseam roof and lap seam roof (through fastener roof) is inthe manner in which two panels are joined to each other.The seam between two panels is made in the field with atool that makes a cold formed weather-tight joint. (Note:some panels can be seamed without special tools.) Thejoint is made at the top of the panel. The standing seamroof is also unique in the manner in which it is attached tothe purlins. The attachment is made with a clip concealedinside the seam. This clip secures the panel to the purlin

-4-


Administrator

Rectangle

NOTE: SEE SDI LOAD TABLES FOR ACTUAL DECK CAPACITIES.

Table 4.1 Steel Deck Institute Recommended Spans (38)

Table 4.2 Factory Mutual Data (3)

and may allow the panel to move when experiencingthermal expansion or contraction.

A continuous single skin membrane results after theseam is made since through-the-roof fasteners havebeen eliminated. The elevated seam and single skinmember provides a watertight system. The ability of theroof to experience unrestrained thermal movementeliminates damage to insulation and structure (caused bytemperature effects which built-up and through fastenedroofs commonly experience). Thermal spacer blocks areoften placed between the panels and purlins in order toinsure a consistent thermal barrier. Due to the superiorityof the standing seam roof, most manufacturers are will-ing to offer considerably longer guarantees than those of-fered on lap seam roofs.

Because of the ability of standing seam roofs tomove on sliding clips, they possess only minimal dia-phragm strength and stiffness. The designer should as-

-5-

Types 1.5A, 1.5F, 1.5B and 1.5BI Deck. Nominal1-1/2 in. (38 mm) depth. No stiffening grooves.

Type 1.5ANarrow Rib

Type 1.5FIntermediate Rib

Type 1.5B, BlWide Rib

22g.

4'10"(1.5m)

4'11"(1.5m)

6'0"(1.8m)

20g.

5'3"(1.6m)

5'5"(1.7m)

6'6"(2.0m)

18g.

6'0"(1.9m)

6'3"(2.0m)

7'5"(2.3m)

Recommended Maximum Spans for Construction andMaintenance Loads Standard 1-1/2 Inch and 3 Inch Roof Deck

NarrowRib Deck(Old Type A)

IntermediateRib Deck(Old Type F)

Wide Rib(Old Type B)

Deep RibDeck

Type

NR22NR22

NR20NR20

NR18NR18

IR22IR22

IR20IR20

IR18IR18

WR22WR22

WR20WR20

WR18WR18

3DR223DR22

3DR203DR20

3DR183DR18

SpanCondition

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

12 or more

SpanFt.-In.

3'-10"4'-9"

4'-10"5'-11"

5'-11"6'-11"

4'-6"5'-6"

5'-3"6'-3"

6'-2"7'-4"

5'-6"6'-6"

6'-3"7'-5"

7'-6"8'-10"

11'-0"13'-0"

12'-6"14'-8"

15'-0"17'-8"

MaximumRecommendedSpans Roof Deck

Cantilever

1'-0"

1'-2"

1'-7"

1'-2"

1'-5"

1'-10"

1'-11"

2'-4"

2'-10"

3'-6"

4'-0"

4'-10"


Administrator

Rectangle

sume that the standing seam roof has no diaphragm capa-bility, and in the case of steel joists specify that sufficientbridging be provided to laterally brace the joists underdesign loads.

4.3 Insulation and Roofing

Due to concern about energy, the use of additionaland/or improved roof insulation has become common.Coordination with the mechanical requirements of thebuilding are necessary. Generally the use of additionalinsulation is warranted, but there are at least two practicalproblems that occur as a result. Less heat loss through theroof results in greater snow and ice build-up and largersnow loads. As a consequence of the same effect, theroofing is subjected to colder temperatures and, for somesystems (built-up roofs), thermal movement, which mayresult in cracking of the roofing membrane.

4.4 Expansion Joints

Although industrial buildings are often constructedof flexible materials, roof and structural expansion jointsare required when horizontal dimensions are large. It isnot possible to state exact requirements relative to dis-tances between expansion joints because of the manyvariables involved, such as ambient temperature duringconstruction and the expected temperature range duringthe life of the buildings. An excellent reference on thetopic of thermal expansion in buildings and location ofexpansion joints is the Federal Construction Council'sTechnical Report No. 65, Expansion Joints in Build-ings.(14)

The report presents the figure shown herein as Fig-ure 4.4.1 as a guide for spacing structural expansionjoints in beam and column frame buildings based on de-sign temperature change. The report includes data fornumerous cities. The report gives modifying factorswhich are applied to the allowable building length as ap-propriate.

The report indicates that the curve is directly appli-cable to buildings of beam-and-column construction,hinged at the base, and with heated interiors. When otherconditions prevail, the following rules are applicable:

1. If the building will be heated only and will havehinged-column bases, use the allowable lengthas specified.

2. If the building will be air conditioned as well asheated, increase the allowable length 15 percent(if the environmental control system will runcontinuously).

3. If the building will be unheated, decrease the al-lowable length 33 percent.

Fig. 4.4.1 Expansion Joint Spacing Graph

[Taken from F.C.C. Tech. Report No. 65,Expansion Joints in Buildings]

4. If the building will have fixed column bases, de-crease the allowable length 15 percent.

5. If the building will have substantially greaterstiffness against lateral displacement in one di-rection decrease the allowable length 25 per-cent.

When more than one of these design conditions pre-vail in a building, the percentile factor to be appliedshould be the algebraic sum of the adjustment factors ofall the various applicable conditions.

Regarding the type of structural expansion joint,most engineers agree that the best method is to use a lineof double columns to provide a complete separation atthe joints. When joints other than the double column typesuch as Fig. 4.4.2 are employed, low friction sliding ele-ments are generally used. Slip connections may inducesome level of inherent restraint to movement due to bind-ing or debris build-up.

Very often buildings may be required to have firewalls in specific locations. Fire walls may be required toextend above the roof or it may be allowed to terminate atthe underside of the roof. Such fire walls become loca-tions for expansion joints. In such cases the detailing ofjoints can be difficult.

Figures 4.4.2 through 4.4.5 depict typical details topermit limited expansion. Additional details are given inarchitectural texts.

-6-


Administrator

Rectangle

Fig. 4.4.3 Joist Expansion Joint

Fig. 4.4.5 Truss Expansion Joint

Expansion joints in the structure should always becarried through the roofing. Additionally, depending onmembrane type, other joints called area dividers are nec-essary in the roof membrane. These joints are membranerelief joints only and do not penetrate the roof deck. Areadivider joints are generally placed at intervals of150-250 feet for adhered membranes, at somewhatgreater intervals for ballasted membranes, and 100-200feet in the case of steel roofs. Spacing of joints should be

- 7 -

Fig. 4.4.2 Beam Expansion Joint

Fig. 4.4.4 Joist Expansion Joint


Administrator

Rectangle

verified with manufacturer's requirements. The range ofmovement between joints is limited by the flexibility andmovement potential of the anchorage scheme and, in thecase of standing seam roofs, the clip design. Manufactur-ers' recommendations should be consulted and followed.Area dividers can also be used to divide complex roofsinto simple squares and rectangles.

4.5 Roof Pitch, Drainage and Ponding

Prior to determining a framing scheme and the direc-tion of primary and secondary framing members, it is im-portant to decide how roof drainage is to be accom-plished. If the structure is heated, interior roof drainsmay be justified. For unheated spaces exterior drains andgutters may provide the solution.

For some building sites it may not be necessary tohave gutters and downspouts, to control stormwater buttheir use is generally recommended or required by theowner. Significant operational and hazardous problemscan occur where water is discharged at the eaves or scup-pers in cold climates, causing icing of ground surfacesand hanging of ice from the roof edge. This is a specialproblem at overhead door locations and may occur withor without gutters. Protection from falling ice must beprovided at all building service entries.

Performance of roofs with positive drainage is gen-erally good. Problems (e.g., ponding, roofing deteriora-tion, leaking) which may result from poor drainage leadto the recommendation that a roof slope of at least 1/4 in.per ft. be provided for all building roof systems. Pond-ing, which is often not understood or is overlooked, is aphenomenon that may lead to severe distress or partial orgeneral collapse.

Ponding as it applies to roof design has two mean-ings. To the roofing industry, ponding describes the con-dition in which water accumulated in low spots has notdissipated within 24 hours of the last rain storm. Pondingof this nature is addressed in roof design by positive roofdrainage and control of the deflections of roof framingmembers. Ponding an issue in structural engineering is aload/deflection situation in which there is incrementalaccumulation of rain water in the deflecting structure.The purpose of a ponding check is to insure that an equi-librium is reached between the incremental loading andthe incremental deflection. This convergence must occurat a level of stress that is within the allowable value.

The AISC Specifications for both LRFD(22) andASD(42) give procedures for addressing the problem ofponding where roof slopes and drains may be inadequate.The direct method is expressed in Eq. K2-1 and K2-2 ofthe Specifications. These relations control the stiffnessof the framing members (primary and secondary) anddeck. This method, however, can produce unnecessarily

conservative results. A more exact method is provided inAppendix K of the LRFD Specification and in Chap. K inthe Commentary in the ASD Specification.

The key to the use of the allowable stress method isthe calculation of stress in the framing members due toloads present at the initiation of ponding. The differencebetween 0.8 F

y and the initial stress is used to establish

the required stiffness of the roof framing members. Theinitial stress ("at the initiation of ponding") is determinedfrom the loads present at that time. These should includeall or most of the dead load and may include some portionof snow/rain/live load. Technical Digest No. 3 publishedby the Steel Deck Institute(41) gives some guidance as tothe amount of snow load which could be used in pondingcalculations.

The amount of accumulated water used is also sub-ject to judgment. The AISC Ponding criteria only appliesto roofs which lack "... sufficient slope towards parts offree drainage or adequate individual drains to prevent theaccumulation of rain water...". However the possibilityof plugged drains means that the load at the initiation ofponding could include the depth of impounded water atthe level of overflow into adjacent bays, or the elevationof overflow drains or, over the lip of roof edges orthrough scuppers. It is clear from reading the AISCSpecification and Commentary that it is not necessary toinclude the weight of water which would accumulate af-ter the "initiation of ponding". Where snow load is usedby the code, the designer may add 5 psf to the roof load toaccount for the effect of rain on snow. Also, considera-tion must be given to areas of drifted snow.

It is clear that judgment must be used in the determi-nation of loading "at the initiation of ponding". It isequally clear that one hundred percent of the roof designload would rarely be appropriate for the loading "at theinitiation of ponding".

A continuously framed or cantilever system may bemore critical than a simple span system. With continu-ous framing, rotations at points of support,due to non-uniformly distributed roof loads, will initiate upwardand downward deflections in alternate spans. The waterin the uplifted bays drains into the adjacent downwarddeflected bays, compounding the effect and causing thedownward deflected bays to approach the deflectedshape of simple spans. For these systems one approach toponding analysis could be based on simple beam stiff-ness, although a more refined analysis could be used.

The designer should also consult with the plumbingdesigner to establish whether or not a controlled flow(water retention) drain scheme is being used. Such an ap-proach allows the selection of smaller pipes because thewater is impounded on the roof and slowly drained away.This intentional impoundment does not meet the AISC

-8-


Administrator

Rectangle

criterion of "... drains to prevent the accumulation ofrainwater..." and requires a ponding analysis.

Besides rainwater accumulation, the designershould give consideration to excessive build-up of mate-rial on roof surfaces (fly ash, and other air borne mate-rial) from industrial operations. Enclosed valleys, paral-lel high and low aisle roofs and normal wind flows cancause unexpected build-ups and possibly roof overload.

4.6 Joists and Purlins

A decision must be made whether to span the longdirection of bays with the main beams trusses or joistgirders which support short span joists or purlins, or tospan the short direction of bays with main framing mem-bers which support longer span joists or purlins. Experi-ence in this regard is that spanning the shorter bay dimen-sion with primary members will provide the most eco-nomical system. However, this decision may not bebased solely on economics but rather on such factors asease of erection, future expansion, direction of craneruns, location of overhead doors, etc.

On the use of steel joists or purlins, experience againshows that each case must be studied. Standard steel joistspecifications (See Ref. 40) are based upon distributedloads only. Modifications for concentrated loads shouldbe done in accordance with the SJI Code Of StandardPractice. Significant concentrated loads should be sup-ported by hot rolled framing members. However in theabsence of large concentrated loads, joist framing cangenerally be more economical than hot rolled framing.

Cold-formed C and Z purlin shapes provide anotheralternative to rolled W sections. The provisions con-tained in the American Iron and Steel Institute's (AISI)Specification for the Design of Cold-Formed SteelStructural Members(43) or the AISI Load and ResistanceFactor Design Specification for Cold-Formed SteelStructural Members.(21) should be used for the design ofcold formed purlins. Additional economy can beachieved with C and Z sections because they can be de-signed and constructed as continuous members. How-ever, progressive failure should be considered if there is apossibility for a loss in continuity after installation.

Other aspects of the use of C and Z sections include:

1. Z sections ship economically due to the fact thatthey can be "nested".

2. Z sections can be loaded through the shear cen-ter, C sections cannot.

3. On roofs with appropriate slope a Z section willhave one principal axis vertical, while a C sec-tion provides this condition only for flat roofs.

4. Many erectors indicate that lap bolted connec-tions for C or Z sections (bolts) are more expen-sive than the simple welded down connectionsfor joist ends.

5. At approximately a 30 ft. span length C and Zsections may cost about the same as a joist forthe same load per foot. For shorter spans C andZ sections may normally be less expensive thanjoists.

5. ROOF TRUSSES

Primary roof framing for conventionally designedindustrial buildings generally consists of wide flangebeams, steel joist girders, or fabricated trusses. For rela-tively short spans 30 to 40 feet steel beams provide aneconomical solution, particularly if a multitude of hang-ing loads are present. For spans greater than 40 but lessthan 80 feet steel joist girders are often used to supportroof loads. Fabricated steel roof trusses are often usedfor spans greater than 80 feet. In recent years little hasbeen written about the design of steel roof trusses. Mosttextbooks addressing the design of trusses were writtenwhen riveted connections were used. Today weldedtrusses and field bolted trusses are used exclusively. Pre-sented in the following paragraphs are concepts and prin-ciples that apply to the design of roof trusses.

5.1 General Design and Economic Considerations

No absolute statements can be made about whattruss configuration will provide the most economical so-lution for a particular situation; however, the followingstatements can be made regarding truss design:

1. Span-to-depth ratios of 15 to 20 generallyprove to be economical; however, shippingdepth limitations should be considered so thatshop fabrication can be maximized. The maxi-mum depth for shipping is conservativelyabout 14 feet. Greater depths will require theweb members to be field bolted which will in-crease erection costs.

2. The length between splice points is also limitedby shipping lengths. The maximum shippablelength varies according to the destination of thetrusses; but lengths of 80 feet are generallyshippable and 100 feet is often possible. Sincemaximum available mill length is approxi-mately 70 feet, the distance between splicepoints is normally set at 70 feet. Greater dis-tances between splice points will generally re-quire truss chords to be shop spliced.

3. In general, the rule "deeper is cheaper" is true;however, the costs of additional lateral bracingfor more flexible truss chords must be carefully

-9-


Administrator

Rectangle

examined relative to the cost of larger chordswhich may require less lateral bracing. The lat-eral bracing requirements for the top and bot-tom chords should be considered interactivelywhile selecting chord sizes and types. Particu-lar attention should be paid to loads which pro-duce compression in the bottom chord. In thiscondition additional chord bracing will mostlikely be necessary.

4. If possible, select truss depths so that tees can beused for the chords rather than wide flangeshapes. Tees can eliminate (or reduce) the needfor gusset plates.

5. Higher strength steels (Fy = 50 ksi or above)usually provides the use of more efficient trussmembers.

6. Illustrated in Figs. 5.1.1 and 5.1.2 are web ar-rangements which generally provide economi-cal web systems.

7. Utilize only a few web angle sizes, and makeuse of efficient long leg angles for greater resis-tance to buckling. Differences in angle sizesshould be recognizable. For instance avoid us-ing an angle 4x3x1/4 and an angle 4x3x5/16 inthe same truss.

Fig. 5.1.2 Economical Truss Web Arrangement

of continuity are being considered in the designprocess, e.g. effective length determination.The designer must be consistent. That is, if thejoints are considered as pins for the determina-tion of forces, then they should also be consid-ered as pins in the design process. The assump-tion of rigid joints in some cases may provideunconservative estimates on the deflection ofthe truss.

12. Repetition is beneficial and economical. Use asfew different truss depths as possible. It ischeaper to vary the chord size as compared tothe truss depth.

13. Wide flange chords with gussets may be neces-sary when significant bending moments exist inthe chords (i.e.: subsystems not supported atwebs or large distances between webs).

14. AISC Engineering for Steel Construction(13) canprovide some additional guidance on truss de-sign and detailing.

15. Design and detailing of long span joists andjoist girders shall be in accordance with SJIspecifications.(40)

-10-

8. Tube, wide flange or pipe sections may prove tobe more effective web members at some web lo-cations especially where subsystems are to besupported by web members.

9. Designs using the AISC LRFD Specification(22)

will often lead to truss savings when heavy longspan trusses are required. This is due to thehigher DL to LL ratios for these trusses.

10. The weight of gusset plates, shim plates andbolts can be significant in large trusses. Thisweight must be considered in the design since itoften approaches 10 to 15 percent of the trussweight.

11. If trusses are analyzed using frame analysiscomputer programs and rigid joints are as-sumed, secondary bending moments will showup in the analysis. The reader is referred to Ref-erence 27 wherein it is suggested that so long asthese secondary stresses do not exceed 4,000psi they may be neglected. Secondary stressesshould not be neglected if the beneficial effects

Fig. 5.1.1 Economical Truss Web Arrangement


Administrator

Rectangle

5.2 Connection Considerations

1. As mentioned above, tee chords are generallyeconomical since they can eliminate gussetplates. The designer should examine the con-nection requirements to determine if the teestem is in fact long enough to eliminate gussetrequirements. The use of a deeper tee stem isgenerally more economical than adding numer-ous gusset plates even if this means an additionin overall weight.

2. Block shear requirements and the effective areain compression should be carefully checked intee stems and gussets (AISC Appendix B).Shear rupture of chord members at panel pointsshould also be investigated since this can oftencontrol wide flange chords.

3. Intermediate connectors (stitch fasteners or fill-ers) may be required for double web members.Examples of intermediate connector evaluationcan be found in the AISC Manual under Col-umn Design; Double Angles.

4. If wide flange chords are used with wide flangeweb members it is generally more economicalto orient the chords with their webs horizontal.Gusset plates for the web members can then beeither bolted or welded to the chord flanges. Toeliminate the cost of fabricating large shim orfiller plates for the diagonals, the use of compa-rable depth wide flange diagonals should beconsidered.

5. When trusses require field bolted joints the useof slip-critical bolts in conjunction with over-size holes will allow for erection alignment.Also if standard holes are used with slip-criticalbolts and field "fit-up" problems occur, holescan be reamed without significantly reducingthe allowable bolt shears.

6. For the end connection of trusses, top chord seattype connections should also be considered.Seat connections allow more flexibility in cor-recting column truss alignment during erection.Seats also provide for efficient erection and aremore stable during erection than "bottom bear-ing" trusses. When seats are used, a simple bot-tom chord connection is recommended to pre-vent the truss from rolling during erection.

7. For symmetrical trusses use a center splice tosimplify fabrication even though forces may belarger than for an offset splice.

8. End plates can provide efficient compressionsplices.

9. It is often less expensive to locate the workpoint of the end diagonal at the face of the sup-porting member rather than designing the con-nection for the eccentricity between the columnCenterline and the face of the column.

5.3 Truss Bracing

Stability bracing is required at discrete locationswhere the designer assumes braced points or wherebraced points are required in the design of the membersin the truss. These locations are generally at panel pointsof the trusses and at the ends of the web members. Tofunction properly the braces must have sufficientstrength and stiffness. Using standard bracing theory,the brace stiffness required (Factor of Safety = 2.0) isequal to 4P/L, where P equals the force to be braced and Lequals the unbraced length of the column. The requiredbrace force equals .004P. As a general rule the stiffnessrequirement will control the design of the bracing unlessthe bracing stiffness is derived from axial stresses only.Braces which displace due to axial loads only are verystiff, and thus the strength requirement will control. Itshould be noted that the AISE Technical Report No. 13requires a .025P force requirement for bracing. More re-fined bracing equations are contained in Reference 20.Requirements for truss bottom chord bracing are dis-cussed in Reference 15. These requirements do not nec-essarily apply to long span joists or joist girders.

Designers are often concerned about the number of"out-of-straight" trusses that should be considered for agiven bracing situation. No definitive rules exist; how-ever, the Australian Code indicates that no more thanseven out of straight members need to be considered. Forcolumns, the International Standards Code of steel struc-ture design (ISO) has postulated the equation:

where is the assumed brace force for n columns, F1

equals .02N where N is the axial force in the column.This equation thus suggests that trussesshould be considered in the bracing design. Thus, if tentrusses were to be braced, bracing forces could be basedon five trusses. Common practice is to provide horizon-tal bracing every five to six bays to transfer bracingforces to the main force resisting system. In this case thebrace forces should be calculated based on the number oftrusses between horizontal bracing.

A convenient approach to the stability bracing oftruss compression chords is discussed in Reference 28.The solution presented is based upon the brace stiffnessrequirements controlled by an X-braced system. The pa-per indicates that as long as the horizontal X-bracing sys-tem is comprised of axially loaded members arranged asshown in Fig. 5.3.1, the bracing can be designed for 0.6percent of the truss chord axial load. Since two truss

-11-


Administrator

Rectangle

length of the truss; however the brace force requirementsdo accumulate based on the number of trusses consideredbraced by the bracing system.

In addition to stability bracing, top and bottom chordbracing may also be required to transfer wind or seismiclateral loads to the main lateral stability system. Theforce requirements for the lateral loads must be added tothe stability force requirements. Lateral load bracing isplaced in either the plane of the top chord or the plane ofthe bottom chord, but generally not in both planes. Sta-bility requirements for the unbraced plane can be trans-ferred to the laterally braced plane by using vertical swaybraces.

EXAMPLE 5.3.1: Roof Truss Stability Bracing

For the truss system shown in Fig. 5.3.2 determine thebrace forces in the horizontal bracing system. Use theprocedure discussed in Reference 28.

Fig. 5.3.2 Horizontal Bracing System

-12-

chord sections are being braced at each bracing strut lo-cation the strut connections to the trusses must be de-signed for 1.2 percent of the average chord axial load forthe two adjacent chords. In the reference it is pointed outthat the bracing forces do not accumulate along the

Fig. 5.3.1 Horizontal X-Bracing Arrangement


Administrator

Rectangle

Solution:Since the diagonal bracing layout as shown in Fig.

5.3.2 forms an angle of 45 degrees with the trusses, the

solution used in Reference 28 is suitable. The bracingforce thus equals .6% of the chord axial load. Memberforces are summarized below.

-13-

DESIGN FORCES(KIPS)

HORIZONTAL TRUSS WEB MEMBER FORCES

MEMBER PANEL SHEAR FORCE = (1.414) (PANEL SHEAR)

C1-D2D1-C2

C2-D3D2-C3

C3-D4D3-C4

.006(6x600) = 21.6

.006(6x800) = 28.8

.006(6x1000) = 36

30.5

40.7

50.9

HORIZONTAL TRUSS CHORD FORCES

MEMBER MEMBERFORCE

C1-C2D1-D2

C2-C3D2-D3

C3-C4D3-D4

21.6

21.6 + 28.8= 50.4

50.4 + 36= 86.4

STRUT FORCES

MEMBER FORCE = (1.2%) (AVE. CHORD FORCE)

A4-B4, E4-F4

B4-C4, D4-E4

C4-D4

A3-B3, E3-F3

B3-C3, D3-E3

C3-D3

A2-B2, E2-F2B2-C2, D2-E2C2-D2

A1-B1, E1-F1B1-C1, D1-E1C1-D1

12.0

24.0

36.0

10.8

21.6

32.4

8.416.825.2

3.67.2

10.8

Note: Forces not shown are symmetrical


Administrator

Rectangle

Administrator

Rectangle

Administrator

Rectangle

5.4 Erection Bracing

The engineer of record is not responsible for the de-sign of erection bracing unless specific contract arrange-ments incorporate this responsibility into the work. Eventhough the designer of trusses is not responsible for theerection bracing, the designer should consider sequenceand bracing requirements in the design of large trusses inorder to provide the most cost effective system. Largetrusses require significant erection bracing not only to re-sist wind and construction loads but also to provide sta-bility until all of the gravity load bracing is installed. Sig-nificant cost savings can be achieved if the required erec-tion bracing is incorporated into the permanent bracingsystem.

Erection is generally accomplished by first connect-ing two trusses together with strut braces and any addi-tional erection braces to form a stable box system. Addi-tional trusses are held in place by the crane or cranes untilthey can be "tied off with strut braces to the alreadyerected stable system. Providing the necessary compo-nents to facilitate this type of erection sequence is essen-tial for a cost effective project.

Additional considerations are as follows:

1. Columns are usually erected first with the lat-eral bracing system (see Fig. 5.4.1). If top chordseats are used, the trusses can be quickly posi-tioned on top of the columns, braced to one an-other.

Bottom chord bearing trusses require that addi-tional stability bracing be installed at ends oftrusses while the cranes hold the trusses inplace. This can slow down the erection se-quence. These concepts are shown in Fig.5.4.1.

Fig. 5.4.1 Wall Bracing Erection Sequence

2. Since many industrial buildings require clearspans, systems are often designed as rigidframes. By designing rigid frames, erection isfacilitated, in that, the side wall columns are sta-bilized in the plane of the trusses once thetrusses are adequately anchored to the columns.This scheme may require larger columns than abraced frame system; however, economy cangenerally be recovered due to a savings in brac-ing and erection time.

3. Wide flange beams, tubes or pipe sectionsshould be used to laterally brace large trusses atkey locations during erection because of greaterstiffness. Steel joists can be used; however, twonotes of caution are advised:a. Erection bracing strut forces must be pro-

vided to the joist manufacturer; and it mustbe made clear whether joist bridging androof deck will not be in place when theerection forces are present. Large angle topchords in joists may be required to controlthe joist Slenderness ratio so that it does notbuckle while serving as the erection strut,

b. Joists are often not fabricated to exactlengths and long slotted holes are generallyprovided in joist seats. Slotted holes forbolted bracing members should be avoidedbecause of possible slippage. Special coor-dination with the joist manufacturer is re-quired to eliminate the slots and to providea suitable joist for bracing. In addition thejoists must be at the job site when the erec-tor wishes to erect the trusses.

4. Wind forces on the trusses during erection canbe considerable. The AISC Code of StandardPractice states that "temporary supports will se-cure the steel framing, or any partly assembledsteel framing, against loads comparable in in-tensity to those for which the structure was de-signed, resulting from wind, seismic forces anderection operations..." The projected area of allof the truss, and other roof framing memberscan be significant and in some cases the windforces on the unsided structure are actuallylarger than those after the structure is enclosed.

5. A sway frame is normally required in order toplumb the trusses during erection. These swayframes should normally occur every fourth orfifth bay. An elevation view of such a truss isshown in Fig. 5.4.2. These frames can be incor-porated into the bottom chord bracing system.Sway frames are also often used to transferforces from one chord level to another as dis-cussed earlier. In these cases the sway framesmust not only be designed for stability forces,but also the required load transfer forces.

-14-


Administrator

Rectangle

Fig. 5.4.2 Sway Frame

5.5 Other Considerations

1. Camber: Large clear span trusses are generallycambered to accommodate dead load deflec-tions. This is accomplished by the fabricator,by either adjusting the length of the web mem-bers in the truss and keeping the top chord seg-ments straight or by curving the top chord. Teescan generally be easily curved during assemblywhereas wide flange sections may require cam-bering prior to assembly. If significant topchord pitch is provided and if the bottom chordis pitched, camber may not be required. The en-gineer of record is responsible for providing thefabricator with the anticipated dead load deflec-tion and special cambering requirements.

The designer must carefully consider the trussdeflection and camber adjacent to walls, orother portions of the structure where stiffnesschanges cause variations in deflections. This isparticularly true at building end walls, wheredifferential deflections may damage continuouspurlins or connections.

2. Connection details which can accommodatetemperature changes are generally necessary.Long span trusses which are fabricated at onetemperature and erected at a significantly dif-ferent temperature can grow or shrink signifi-cantly.

3. Roof deck diaphragm strength and stiffness iscommonly used for strength and stability brac-ing for joists. The diaphragm capabilities mustbe carefully evaluated if it is to be used for brac-ing of large clear span trusses.

6. WALL SYSTEMS

The wall system can be chosen for a variety of rea-sons and the cost of the wall can vary by as much as a fac-tor of three. Wall systems include:

1. field assembled metal panels,

2. factory assembled metal panels,

3. precast concrete panels,

4. masonry walls (part or full height).

A particular wall system may be selected over others forone or more specific reasons including:

1. cost,

2. appearance,

3. ease of erection,

4. speed of erection,

5. insulating properties,

6. fire considerations,

7. acoustical considerations,

8. ease of maintenance/cleaning,

9. ease of future expansion,

10. durability of finish,

11. maintenance considerations.

Some of these factors will be discussed in the fol-lowing sections on specific systems. Other factors arenot discussed and require evaluation on a case by case ba-sis.

-15-


Administrator

Rectangle

6.1 Field Assembled Panels

Field assembled panels consist of an outer skin ele-ment, insulation, and in some cases an inner liner panel.The panels vary in material thickness and are normallygalvanized, galvanized prime painted suitable for fieldpainting, or prefinished galvanized. Corrugated alumi-num liners are also used. When aluminum materials areused their compatibility with steel supports should beverified with the manufacturer since aluminum maycause corrosion of steel. When an inner liner is used,some form of hat section interior subgirts are generallyprovided for stiffness. The insulation is typicallyfiberglass or a foam. If the inner liner sheet is used as thevapor barrier all joints and edges should be sealed.

Specific advantages of field assembled wall panelsinclude:

1. rapid erection of panels,

2. good cost competition, with a large number ofmanufacturers and contractors being capable oferecting panels,

3. quick and easy panel replacement in the eventof panel damage,

4. openings for doors and windows that can be cre-ated quickly and easily,

5. panels that are lightweight, so that heavy equip-ment is not required for erection. Also largefoundations and heavy spandrels are not re-quired,

6. acoustic surface treatment that can be addedeasily to interior panel wall at reasonable cost.

A disadvantage of field assembled panels in high hu-midity environments can be the formation of frost or con-densation on the inner liner when insulation is placedonly between the subgirt lines. The metal to metal con-tact (outside sheet-subgirt-inside sheet) should be bro-ken to reduce thermal bridging. A detail which has beenused successfully is shown in Fig. 6.1.1. Another optionmay be to provide rigid insulation between the girt andliner on one side. In any event, the wall should be evalu-ated for thermal transmittance in accordance withASHRAE90.1.(12)

6.2 Factory Assembled Panels

Factory assembled panels generally consist of inte-rior liner panels, exterior metal panels and insulation.Panels providing various insulating values are availablefrom several manufacturers. These systems are gener-ally proprietary and must be designed according tomanufacturer's recommendations.

Fig. 6.1.1 Wall Thermal Break Detail

The particular advantages of these factory assem-bled panels are:

1. Panels are lightweight and require no heavycranes for erection, no large foundations orheavy spandrels.

2. Panels can have a hard surface interior liner.

3. Panel side lap fasteners are normally concealedproducing a "clean" appearance.

4. Documented panel performance characteristicsdetermined by test or experience may be avail-able from manufacturers.

Disadvantages of factory assembled panels include:

1. Once a choice of panel has been made, futureexpansions may effectively require use of thesame panel to match color and profile, thuscompetition is essentially eliminated.

2. Erection procedures usually require starting inone corner of a structure and proceeding to thenext corner. Due to the interlocking nature ofthe panels it may be difficult to add openings inthe wall.

3. Close attention to coordination of details andtolerances with collateral materials is required.

4. Thermal changes in panel shape may be moreapparent.

-16-


Administrator

Rectangle

6.3 Precast Wall Panels

Precast wall panels for industrial buildings couldutilize one or more of a variety of panel types including:

1. hollow core slabs,

2. double-T sections,

3. site cast tilt-up panels,

4. factory cast panels.

Panels can be either load bearing or nonload bearingand can be obtained in a wide variety of finishes, texturesand colors. Also panels may be of sandwich constructionand contain rigid insulation between two layers of con-crete. Such insulated panels can be composite or non-composite. Composite panels normally have a positiveconcrete connection between inner and outer concretelayers. These panels are structurally stiff and are goodfrom an erection point of view but the "positive" connec-tion between inner and outer layers may lead to exteriorsurface cracking when the panels are subjected to a tem-perature differential. The direct connection can also pro-vide a path for thermal bridging which may be a problemin high humidity situations.

True sandwich panels connect inner and outer con-crete layers with flexible metal ties. Insulation is ex-posed at all panel edges. These panels are more difficultto handle and erect, but normally perform well.

Precast panels have advantages for use in industrialbuildings:

1. A hard surface is provided inside and out.

2. These panels produce an architecturally "clean"appearance.

3. Panels have inherent fire resistance characteris-tics.

4. Intermediate girts are usually not required.

5. Use of load bearing panels can eliminate exte-rior framing and reduce cost.

6. They provide an excellent sound barrier.

Disadvantages of precast wall panel systems in-clude:

1. Matching colors of panels in future expansionmay be difficult.

2. Composite sandwich panels have "cold spots"with potential condensation problems at paneledges.

3. Adding wall openings can be difficult.

4. Panels have poor sound absorption characteris-tics.

5. Foundations and grade beams may be heavierthan for other panel systems.

6. Heavier eave struts are required for steel framestructures than for other systems.

7. Heavy cranes are required for panel erection.

8. If panels are used as load bearing elements, ex-pansion in the future could present problems.

9. Close attention to tolerances and details to coor-dinate divergent trades are required.

10. Added dead weight of walls can affect seismicdesign.

6.4 Masonry Walls

Use of masonry walls in industrial buildings is com-mon. Walls can be load bearing or non-load bearing.

Some advantages of the use of masonry constructionare:

1. A hard surface is provided inside and out.

2. Masonry walls have inherent fire resistancecharacteristics.

3. Intermediate girts are usually not required.

4. Use of load bearing walls can eliminate exteriorframing and reduce cost.

5. Masonry walls can serve as shear walls to bracecolumns and resist lateral loads.

6. Walls produce a flat finish, resulting in an easeof both maintenance and dust control consid-erations.

Disadvantages of masonry include:

1. Masonry has comparatively low material bend-ing resistance. Walls are normally adequate toresist normal wind loads, but interior impactloads can cause damage.

2. Foundations may be heavier than for metal wallpanel construction.

3. Special consideration is required in the use ofmasonry ties, depending on whether the ma-sonry is erected before or after the steel frame.

4. Buildings in seismic regions may require spe-cial reinforcing and added dead weight may in-crease seismic forces.

-17-


Administrator

Rectangle

6.5 Girts

Typical girts for industrial buildings are hot rolledchannel sections or cold-formed light gage C or Z sec-tions. In recent years, cold-formed sections have gainedpopularity because of their low cost. As mentioned ear-lier, cold-formed Z sections can be easily lapped toachieve continuity resulting in further weight savingsand reduced deflections, Z sections also ship economi-cally. Additional advantages of cold-formed sectionscompared with rolled girt shapes are:

1. Metal wall panels can be attached to cold-formed girts quickly and inexpensively usingself-drilling fasteners.

2. The use of sag rods is often not required.

Hot-rolled girts are often used when:

1. Corrosive environments dictate the use ofthicker sections.

2. Common cold-formed sections do not havesufficient strength for a given span or load con-dition.

3. The girts will receive substantial abuse fromoperations.

4. Designers are unfamiliar with the availabilityand properties of cold-formed sections.

Both hot-rolled and cold-formed girts subjected topressure loads are normally considered laterally bracedby the wall sheathing. Negative moment regions in con-tinuous girt systems are typically considered laterallybraced at inflection points and at girt to column connec-tions. Continuous systems have been analyzed by as-suming:

1. a single prismatic section throughout, or

2. a double moment of inertia condition within thelapped section of the cold-formed girt.

Research indicates that an analytical model assum-ing a single prismatic section is closer to experimentallydetermined behavior.(34)

For the design of hot-rolled channels, the beneficiallateral support provided by cladding attached to the ten-sion flange has generally been ignored. The use of sagrods is generally required to maintain horizontal align-ment of hot-rolled sections. The sag rods are often as-sumed to provide lateral restraint against buckling forsuction loads. Lateral stability based on this assumptionis checked using Chap. F of the AISC Specification.

The typical design procedure for hot-rolled girts isas follows:

1. Select the girt size based on pressure loads, as-suming full flange lateral support.

2. Check the selected girt for sag rod requirementsbased on deflections and bending stresses aboutthe weak axis of the girt.

3. Check the girt for suction loads using Chap. F ofthe AISC Specification.

4. If the girt is inadequate, increase its size or addsag rods.

5. Check the girt for serviceability requirements.

Cold-formed girts should be designed in accordancewith the provisions of the American Iron and Steel Insti-tute (AISI) Specification for the Design of Cold-FormedSteel Structural Members,(43) or the AISI Load and Resis-tance Factor Design Specification for Cold-FormedSteel Structural Members.(21) Many manufacturers ofcold-formed girts have provided this design and offerload span tables to aid design.

Section C3.1.2 "Lateral Buckling Strength" of theAISI Specification provides a means for determiningcold-formed girt strength when the compression flangeof the girt is attached to sheeting (fully braced) or whendiscrete point braces (sag rods) are used. For lapped sys-tems, the sum of the moment capacities of the two lappedgirts is normally assumed to resist the negative momentover the support. For full continuity to exist, a lap lengthon each side of the column support should be equal to atleast 1.5 times the girt depth.(34) Additional provisionsare given in Section C3 for strength considerations rela-tive to shear, web crippling, and combined bending andshear.

Section C3.1.3 "Beams with One Flange Attached toDeck or Sheathing" provides a simple procedure to de-sign cold-formed girts subjected to suction loading. Thebasic equation for the determination of the girt strengthis:

where

R = 0.40 for simple span C sections.

= 0.50 for simple span Z sections.

= 0.60 for continuous span C sections.

= 0.70 for continuous span Z sections.

Elastic section modulus of the effectivesection calculated with the extremecompression or tension fiber at

Design yield stress as determined in SectionA5.2.1 of the Specification.

The procedure does not apply to a continuous girtsystem between inflection points adjacent to a support.

-18-


Administrator

Rectangle

Other restrictions relative to girt geometry, wall panels,fastening systems between wall panels and girts, etc. arediscussed in the AISI Specifications.

6.6 Wind Columns

When bay spacings exceed 30 feet additional inter-mediate columns may be required to provide for eco-nomical girt design. Two considerations that should beemphasized are:

1. Sufficient bracing of the wind columns to ac-commodate wind suction loads is needed. Thisis normally accomplished by bracing the inte-rior flanges of the columns with angles whichconnect to the girts.

2. Proper attention should be paid to the top con-nections of the columns. For intermediatesidewall columns, secondary roof framingmembers must be provided to transfer the windreaction at the top of the column into the roofbracing system. Do not rely on "trickle theory"(i.e. "a force will find a way to trickle out of thestructure."). A positive and calculable systemis necessary to provide a traceable load path (i.e.Fig. 6.6.1). Bridging systems or bottom chordextension on joists can be used to dissipate theseforces, but the stresses in the system must bechecked. If the wind columns have not been de-signed for axial load, a slip connection would benecessary at the top of the column.

Small wind reactions can be transferred from thewind columns into the roof diaphragm system as shownin Fig. 6.6.2.

Allowable values for attaching metal deck to struc-tural members can be obtained from screw manufactur-ers. Allowable stresses in welds to metal deck can be de-termined from the American Welding Society Standard,Specification for Welding Sheet Steel in Structures.AWS D1.3,(44) or from the AISI Specifications.(21)(43) Inaddition to determining the fastener requirements totransfer the concentrated load into the diaphragm, the de-signer must also check the roof diaphragm for its strengthand stiffness. This can be accomplished by using Ref. 11.

7. FRAMING SCHEMES

The selection of "the best" framing scheme for an in-dustrial building without cranes is dependent on numer-ous considerations, and often depends on the owners re-quirements. It may not be possible to give a list of rulesby which the best such scheme can be assured. If "best"means low initial cost, then the owner may face maj or ex-penses in the future for operational expenses or problems

with expansion. Extra dollars invested at the outset re-duce potential future costs.

The economics of use of long span vs. short spanjoists and purlins has been mentioned previously in thisguide. This section expands on the selection of the mainframing system. No attempt has been made to evaluatefoundation costs. In general, if a deep foundation system(e.g., piles or drilled piers) is required, longer bay spac-ings are normally more economical.

The consideration of bay sizes must include not onlyroof and frame factors but also the wall system. The costof large girts and thick wall panels may cancel the sav-ings anticipated if the roof system alone is considered.

-19-

Fig. 6.6.1 Wind Column Reaction Load Transfer

Fig. 6.6.2 Wind Column Reaction Load Transfer


Administrator

Rectangle

AISC offers two manuals that may aid in the designof efficient framing details: Engineering for Steel Con-struction(13) and Detailing for Steel Construction.(10)

7.1 Braced Frames vs. Rigid Frames

The design of rigid frames is explained in numeroustextbooks and professional journals and will not be cov-ered here; however, a few concepts will be presentedconcerning the selection of a braced versus a rigid framestructural system. There are several situations for whicha rigid frame system is likely to be superior.

1. Braced frames may require bracing in bothwalls and roof. Bracing frequently interfereswith plant operations and future expansion. Ifeither consideration is important, a rigid framestructure may be the answer.

2. The bracing of a roof system can be accom-plished through X-bracing or a roof diaphragm.In either case the roof becomes a large horizon-tal beam spanning between the walls or bracingwhich must transmit the lateral loads to thefoundations. For large span to width ratios(greater than 3:1) the bracing requirements be-come excessive. A building with dimensions of100 feet by 300 feet with potential future expan-sion in the long direction may best be suited forrigid frames to minimize or eliminate bracingwhich would interfere with future changes.

Use of a metal building system requires a strong in-teraction between the designer and the metal buildingmanufacturer, because of much of the detailing processconcerning the design is provided by the manufacturerand the options open to the buyer may reflect the limits ofthe manufacturer's standard product line and details.

Experience has shown that there are occasions whenbraced frame construction may prove to be more eco-nomical than either standard metal building systems orspecial rigid frame construction when certain sacrificeson flexibility are accepted.

7.2 Tube Columns vs. W Shapes

The design of columns in industrial buildings in-cludes considerations which do not apply to other typesof structures. Interior columns can normally be bracedonly at the top and bottom, thus square tube columns aredesirable due to their equal stiffness about both principalaxes. Difficult connections with tube members can beeliminated in single-story frames by placing the beamsover the tops of the tubes. Thus a simple to fabricate capplate detail with bearing stiffeners on the girder web maybe designed. Other advantages of tube columns includethe fact that they require less paint than equivalent Wshapes, and they are pleasing aesthetically.

W shapes may be more economical than tubes forexterior columns for the following reasons:

1. The wall system (girts) may be used to brace theweak axis of the column. It should be noted thata stiffener or brace may be required for the col-umn if the inside column flange is in compres-sion and the girt connection is assumed to pro-vide a braced point in design.

2. Bending moments due to wind loads predomi-nate about one axis.

3. It is easier to frame girt connections to a Wshape than to a tube section. Because tubeshave no flanges, extra clip angles are required toconnect girts.

7.3 Economic Considerations

As previously mentioned, bay sizes and columnspacing are often dictated by the function of the building.Economics, however, should also be considered. In gen-eral, as bay sizes increase, the weight of the horizontalframing increases. This will mean additional cost unlessoffset by savings in foundations or erection. Studieshave indicated that square or slightly rectangular baysusually result in more economical structures.

The development of the most economical framingscheme for a roof system is a worthy goal. In order toevaluate various framing schemes, a prototype generalmerchandise structure was analyzed using various spansand component structural elements. The structure was a240'-0" x 240'-0" building with a 25'-0" eave height.The total roof load was 48 psf, and beams with Fy = 50 ksiwere used. Plastic analysis and design was used. Col-umns were tubes with a yield strength of 46 ksi.

Variables in the analysis were:

1. Joist spans: 25, 30, 40, 50 and 60 feet.

2. Girder spans, W sections: 25, 30, 40, 48and 60 feet.

Cost data were determined from several fabricators.The data did not include sales tax or shipping costs. Thestudy yielded several interesting conclusions for engi-neers involved in industrial building design.

An examination of the tabular data shows that themost economical framing scheme was the one withbeams spanning 30 feet and joists spanning 40 feet.

Another factor that may be important is that for thelarger bays (greater than 30 ft) normal girt constructionbecomes less efficient using C or Z sections without in-termediate "wind columns" being added. For the 240' x240' building being considered wind columns could add$0.10 per square foot, of roof, to the cost. Interestingly, ifthe building were 120' x 120', the addition of intermedi-

-20-


Administrator

Rectangle

Administrator

Rectangle

Administrator

Rectangle

JOISTDepth(in.)

16

18

24

30

32

RELATIVE COSTS*

MAIN FRAMING MEMBERS - W SECTIONS

DATASpan

(ft.)

25

30

40

50

60

25

1.10

1.12

1.16

1.22

1.33

Span (ft.)

30

1.10

1.07

1.05

1.18

1.30

40

1.25

1.20

1.15

1.20

1.30

48

1.31

1.28

1.28

1.30

1.33

60

1.53

1.50

1.47

1.54

1.60

* Cost included fabrication and erection of primary and secondary framing (no deck). A total gravity load of48 psf was used in all designs. Uplift and lateral bracing requirements were not included.

ate wind columns would add $0.20 per square foot be-cause the smaller building has twice the perimeter to arearatio as the larger structure.

Additional economic and design considerations areas follows:

1. When steel joists are used in the roof framing itis generally more economical to span the joistsin the long direction of the bay.

2. K series joists are more economical than LHjoists; thus an attempt should be made to limitspans to those suitable for K joists.

3. For 30 ft to 40 ft bays, efficient framing mayconsist of continuous or double cantileveredgirders supported by columns in one directionand joists spanning the other direction.

4. If the girders are continuous, plastic design isoften used. Connection costs for continuousmembers may be higher than for cantilever de-sign; however, a plastically designed continu-ous system will have superior behavior whensubjected to pattern load cases. All flat roofsystems must be checked to prevent pondingproblems. See Section 4.5.

5. Simple-span rolled beams are often substitutedfor continuous or double-cantilevered girderswhere spans are short. The simple span beamsoften have adequate moment capacity. Theconnections are simple, and the savings fromeasier erection of such systems may overcomethe cost of any additional weight.

6. For large bay dimensions in both directions, apopular system consists of cold-formed or hot-rolled steel purlins or joists spanning 20 ft. to 30ft. to secondary trusses spanning to the primary

trusses. This framing system is particularlyuseful when heavily loaded monorails must behung from the structure. The secondary trussesin conjunction with the main trusses provide ex-cellent support for the monorails.

7. Consideration must be given to future expan-sion and/or modification, where columns areeither moved or eliminated. Such changes cangenerally be accomplished with greater easewhere simple span conditions exist.

8. BRACING SYSTEMS

8.1 Rigid Frame Systems

There are many considerations involved in provid-ing lateral stability to industrial structures. If a rigidframe is used, lateral stability parallel to the frame is pro-vided by the frame. However, for loads perpendicular tothe main frames and for wall bearing and "post andbeam" construction, lateral bracing is not inherent andmust be provided. It is important to re-emphasize thatfuture expansion may dictate the use of a rigid frame or aflexible (movable) bracing scheme.

Since industrial structures are normally light andgenerally low in profile, wind and seismic forces may berelatively low. Rigid frame action can be easily andsafely achieved by providing a properly designed mem-ber at a column line. If joists are used as a part of the rigidframe the designer is cautioned on the following points:

1. The design loads (wind, seismic, and continu-ity) must be given on the structural plans so thatthe proper design can be provided by the joistmanufacturer. The procedure must be usedwith conscious engineering judgment and fullrecognition that standard steel joists are de-signed as simple span members subject to dis-

-21-


Administrator

Rectangle

tributed loads. (See Standard Specifications forStandard Steel Joists and Long Span Joists.)(40)

Bottom chords are normally sized for tensiononly. The simple attachment of the bottomchord to a column to provide lateral stabilitywill cause gravity load end moments which can-not be ignored. The designer should not try toselect member sizes for these bottom chordssince each manufacturer's design is unique andproprietary.

2. It is necessary for the designer to provide a welldesigned connection to both the top and bottomchords to develop the induced moments withoutcausing excessive secondary bending momentsin the joist chords.

3. The system must have adequate stiffness to pre-vent drift related problems such as crackedwalls and partitions, broken glass, leaking wallsand roofs, and malfunctioning or inoperableoverhead doors.

8.2 Roof Bracing Systems

Roof Diaphragms

The most economical roof bracing system isachieved by use of a steel deck diaphragm. The deck isprovided as the roofing element and the effective dia-phragm is obtained at little additional cost (for extra deckconnections). A roof diaphragm used in conjunctionwith wall X-bracing or a wall diaphragm system is prob-ably the most economical bracing system that can beachieved. Diaphragms are most efficient in relativelysquare buildings; however, an aspect ratio up to three canbe accommodated.

Cold-formed steel diaphragm is analogous to theweb of a plate girder. That is, its main function is to resistshear forces. The perimeter members of the diaphragmserve as the "flanges".

The design procedure is quite simple. The basic pa-rameters that control the strength and stiffness of the dia-phragm are:

1. profile shape,

2. deck material thickness,

3. span length,

4. the type and spacing of the fastening of the deckto the structural members,

5. the type and spacing of the side lap connectors.

The profile, thickness, and span of the deck are typi-cally based on gravity load requirements. The type offastening (i.e., welding, screws, and power driven pins)

is often based on the designers or contractors preference.Thus the main design variable is the spacing of the fas-teners. The designer calculates the maximum shear perfoot of diaphragm and then selects the fastener spacingfrom the load tables. Load tables are most often based onthe requirements set forth in References (11) and (37).

Deflections are calculated and compared with serv-iceability requirements.

The calculation of flexural deformations is handledin a conventional manner. Shear deformations can be ob-tained mathematically, using shear deflection equations,if the shear modulus of the formed deck material makingup the diaphragm is known. Deflections can also be ob-tained using empirical equations such as those found inReferences (11) and (37). In addition, most metal deckmanufacturers publish tables in which strength and stiff-ness (or flexibility) information is presented. In order toillustrate the diaphragm design procedure a design exam-ple is presented below. The calculations presented arebased on Reference (11).

EXAMPLE 8.2.1: Diaphragm Design

Design the roof diaphragm for the structure shown in Fig.8.2.1. The eave wind loads are shown in the figure.

Fig. 8.2.1. Example

Note that the length to width ratio of the diaphragm doesnot exceed 3, which is the generally accepted maximumfor diaphragms.

Assume that a (0.0358 inch thick) intermediate rib deckspanning 5'-6" is used to support the gravity loads. Steeljoists span in the north-south direction. Use welds toconnect the deck to the structural members and #10screws for the sidelaps.

Solution:

1. Calculate the maximum diaphragm shear.

2. Obtain the shear capacity of the deck from the SDIDiaphragm Design Manual, Second Edition.

For a 20 gage (.0358" thickness) deck, spanning 5'-6" theallowable shear is:

-22-


Administrator

Rectangle

(a) 240 plf with a 36/4 weld pattern and one sidelap screw.

(b) 285 plf with a 36/4 weld pattern and two sidelap screws.

(c) 300 plf with a 36/5 weld pattern and one sidelap screw.

Use patterns (a) and (b) as shown in Fig. 8.2.1.1:

Fig. 8.2.1.1

3. Compute the lateral deflection of the diaphragm.

For simplicity assume one sidelap screw for the en-tire roof.

The deflection equations are:

(a) For bending:

(b) For shear:

where w = the eave force (kips/ft)L = the diaphragm length (ft)D = the diaphragm depth (ft)

From the SDI tables:

The moment of inertia, I, can be based on an assumedarea of the perimeter member. Assuming the edge mem-ber has an area of 3.0 in.2, the moment of inertia equals:

The bending deflection equals:

The shear deflection equals:

The total deflection equals:

To transfer the shear forces into the east and west walls ofthe structure the deck can be welded directly to the pe-rimeter beams. The deck must be connected to the pe-rimeter beams with the same number of fasteners as re-quired in the field of the diaphragm. Thus, 5/8 in. dia. arcspot welds 9 inches on center should be specified at theeast and west walls.

The reader is cautioned regarding connecting steel deckto the end walls of buildings. If the deck is to be con-nected to a shear wall and a joist is placed next to the wall,allowance must be made for the camber in the edge joistin order to connect the deck to the wall system. If properdetails are not provided, diaphragm connection may notbe possible, and field adjustments may be required.Where the edge joist is eliminated near the endwall, thedeck can often be pushed down flat on an endwall sup-port. If the joist has significant camber, it may be neces-sary to provide simple span pieces of deck between thewall and the first joist. A heavier deck thickness may berequired due to the loss in continuity. The edge should becovered with a sheet metal cap to protect the roofing ma-terials. This can present an additional problem since thesharp edge of the deck will stick up and possibly damagethe roofing.

Along the north and south walls, a diaphragm chord canbe provided by attaching an angle to the top of the joistsas shown in Fig. 8.2.2. The angle also stiffens the deckedge and prevents tearing of roofing materials along theedge where no parapet is provided under foot traffic. Insome designs an edge angle may also be required for theside lap connections for wind forces in the east-west di-rection. Also, shear connectors may be required to trans-fer these forces into the perimeter beam. Shown in Fig.8.2.3 is a typical shear collector.

-23-


Administrator

Rectangle

Fig. 8.2.2 Eave Angle

Fig. 8.2.3 Shear Collector

Roof X-Bracing

An alternative to the roof diaphragm is to use X--bracing to develop a horizontal truss system. As with themetal deck diaphragm, as the length to width ratio of thebuilding becomes larger than 3 to 1 the diagonal forces inthe truss members may require consideration of an alter-nate bracing method.

An especially effective way to develop an X-bracedroof is to utilize flat bar stock resting on the roof joists.The use of 1/4 in. bar stock does not usually interfere withdeck placement and facilitates erection.

8.3 Temporary Bracing

Proper temporary bracing is essential for the timelyand safe erection and support of the structural frameworkuntil the permanent bracing system is in place. The needfor temporary bracing is recognized by the AISC Specifi-

cation(22,42) (Section M4.2) and by the Code of StandardPractice(6) (Section 7.9).

The Code of Standard Practice places the responsi-bility for temporary bracing solely with the erector. Thisis appropriate since temporary bracing is an essentialpart of the work of erecting the steel framework.

While the requirements of the Code of StandardPractice are appropriate to establishing the responsibilityfor erection bracing, two major issues have the potentialto be overlooked in the process.

First, the broad standard means that it is difficult tojudge the adequacy of temporary bracing in a particularsituation, nor is there a codified standard upon which tojudge whether or not a minimum level of conformity ismet. Secondly, it is not emphasized in the broad standardthat the process of erection can induce forces and stressesinto components and systems which are not part of thestructural steel framework. Unless otherwise specifiedin the contract documents, it is generally the practice ofarchitects and engineers to design the elements and sys-tems in a building for the forces acting upon the com-pleted structure only.

The lack of clear standards makes it difficult for any-one in the design/construction process to evaluate theperformance of the erector relative to bracing without be-coming involved in the process itself, which is inconsis-tent with maintaining the bracing as the sole responsibil-ity of the erector. The lack of emphasis on the need tocheck the effect of erection forces on other elements al-lows erection problems to be interpreted as being causedby other reasons. This is most obvious in the erection ofsteel columns. In order to begin and pursue the erectionof a steel framework it is necessary to erect columns first.This means that at one time or another each building col-umn is set in place without framing attached to it in twoperpendicular directions. Without such framing the col-umns must cantilever for a time from the supporting foot-ing or pier unless they are braced by adequate guys or de-signed as a rigid frame in both directions. The forces in-duced by the cantilevered column on the pier or footingmay not have been considered by the building designerunless this had been specifically requested. It is incum-bent upon the steel erector to make a determination of theadequacy of the foundation to support cantilevered col-umns during erection.

Trial calculations suggest that very large forces canbe induced into anchor bolts, piers and footings by rela-tively small forces acting at or near the tops of columns.Also wind forces can easily be significant, as can be seenin the following example. Fig. 8.3.1 shows a section ofunbraced frame consisting of three columns and twobeams. The beams are taken as pin ended. Wind forcesare acting perpendicular to the frame line.

-24-


Administrator

Rectangle

Fig. 8.3.1 Erection Bracing Example

Using a shape factor of 2.0 for a 40 mph wind di-rected at the webs of the W12 columns, a base moment ofapproximately 18,000 foot-pounds occurs. If a 5 inch by5 inch placement pattern were used with four anchorbolts and an ungrouted base plate, a tension force of ap-proximately 21.6 kips would be applied to the two anchorbolts. The allowable force for a 3/4 in A36 anchor bolt is8.4 kips. Even if the bolts were fully developed (and the33% increase for wind was utilized) in the concrete, theywould be severely overstressed and would likely fail.Four 1-1/8 inch anchor bolts would be required to resistthe wind force. Of course not only the size of the anchorbolt is affected, but the design of the base plate and its at-tachment to the column, the spacing of the anchor boltsand the design of the pier and footing must also bechecked.

Guying can also induce forces into the structure inthe form of base shears and uplift forces. These forcesmay not have been provided for in the sizing of the af-fected members. This must also be checked by the erec-tor. The placement of material such as decking on the in-complete structure can induce unanticipated loadings.This loading must also be considered explicitly. Theforegoing points to the need for a well planned bracingscheme prepared and executed by the erector.

Erection bracing involves other issues as well. First,the Code of Standard Practice distinguishes betweenself-supporting and non-self-supporting structures.The distinction is drawn because the timing of the re-moval of bracing is affected. In a self-supporting struc-tural steel frame, lateral stability is achieved in the designand detailing of the framework itself. Thus the bracingcan be removed when the erector's work is complete. Anon-self-supporting steel framework may rely on ele-ments other than the structural steel to provide lateral sta-bility. Such frames should be identified in the contractdocuments along with the necessary elements to be in-stalled to provide the stability. The coordination of theinstallation of such elements is a matter which must beaddressed between the erector and the owner's agents.

Temporary support beyond the requirements dis-cussed above would be the responsibility of the owneraccording to the Code of Standard Practice.

The timing of column base grouting affects the per-formance of column bases during erection. Although theCode of Standard Practice assigns the responsibility forgrouting to the owner, the erector should be involved inthe coordination of this work.

All of the foregoing points to the need for care, atten-tion and thoroughness on the part of the erector.

9. COLUMN ANCHORAGE

Building columns must be anchored to the founda-tion system to transfer tension forces, shear forces, andoverturning moments. This discussion will be limited tothe design of column anchorages for shear and tensionforces. The principles discussed here can be applied tothe design of anchorages for overturning moments.

Tension forces are typically transferred to the foun-dation system with anchor bolts. Shear forces are trans-ferred to the foundation system through friction, shearfriction, or bearing. Friction should not be considered ifseismic conditions exist. Design for these various an-chorage methods is addressed in the following text.

Improper design, detailing and installation of an-chor bolts have caused numerous structural problems inindustrial buildings. These problems include:

1. inadequate sizing of the anchor bolts,

2. inadequate development of the anchor bolts fortension,

3. inadequate design or detailing of the foundationfor forces from the anchor bolts,

4. inadequate base plate thickness,

5. inadequate design and/or detailing of the anchorbolt - base plate interface,

6. misalignment or misplacement of the anchorbolts during installation, and

7. fatigue.

The following discussion presents methods of de-signing and detailing column bases.

9.1 Resisting Tension Forces with Anchor Bolts

The design of anchor bolts for tension consists offour steps:

1. Determine the maximum net uplift for the col-umn.

2. Select the anchor bolt material and number andsize of anchor bolts to accommodate this uplift.

3. Determine the appropriate base plate size,thickness and welding to transfer the uplift

-25-


Administrator

Rectangle

forces from the column to the anchor bolts. Re-fer to AISC Design Guide 1.(8)

4. Determine the method for developing the an-chor bolt in the concrete (i.e. transferring thetension force from the anchor bolt to the con-crete foundation).

Step 1 - The maximum net uplift for the column is ob-tained from the structural analysis of the building for theprescribed building loads. The use of light metal roofs onindustrial buildings is very popular. As a result of this,the uplift due to wind often exceeds the dead load; thusthe supporting columns are subjected to net uplift forces.In addition, columns in rigid bents or braced bays may besubjected to net uplift forces due to overturning.

Step 2 - Anchor bolts are often specified to be ASTMA307 material. However, these bolts are frequently notavailable (or readily available) for the anchor bolt lengthsrequired. In these instances a threaded rod with a hook ornut is usually used in lieu of an A307 bolt. In anticipationof this, the ASTM A307 specification states that "non-headed anchor bolts, either straight or bent, to be used forstructural anchorage purposes, shall conform to the re-quirements of ASTM Specification A36". In otherwords, if anchor bolts are designated to be A307 materialbut are not available for the lengths required, then non-headed anchor bolts made from threaded rods of A36 ma-terial with hooks or nuts are to be used. The ASTM A36specification states that the thread dimensions and nuts tobe used with these threaded rods are to conform to theASTM A307 specification. Consequently, since the ulti-mate tensile strengths of A36 material and A307 bolts areapproximately equal, this substitution results in anchorbolts of approximately equal strength.

Anchor bolts of higher tensile strength than A307bolts or A36 threaded rods may be used if desired. TableJ3.2 in the AISC Specification may be consulted to ob-tain the allowable stress for the higher strength bolts orthreaded rods.

The number of anchor bolts required is a function ofthe maximum net uplift on the column and the allowabletensile load per bolt for the anchor bolt material chosen.Prying forces in anchor bolts are typically neglected.This is usually justified when the base plate thickness iscalculated assuming cantilever bending about the weband/or flange of the column section (as described in Step3 below). However, calculations have shown that pryingforces may not be negligible when the bolts are posi-tioned outside the column profile and the bolt forces arelarge. A conservative estimate for these prying forcescan be obtained using a method similar to that describedfor hanger connections in the AISC Manual of Steel Con-struction.

Another consideration in selection and sizing of an-chor bolts is fatigue. For most building applications,where uplift loads are generated from wind and seismicforces, fatigue can be neglected because the maximumdesign wind and seismic loads occur infrequently. How-ever, for anchor bolts used to anchor machinery or equip-ment where the full design loads may occur more often,fatigue should be considered. In addition, in buildingswhere crane load cycles are significant, fatigue shouldalso be considered. AISE Technical Report No. 13 forthe design of steel mill buildings recommends that 50percent of the maximum crane lateral loads or side thrustbe used for fatigue considerations.

In the past, attempts have been made to pretensionor preload anchor bolts to prevent fluctuation of the ten-sile stress in anchor bolts and, therefore, eliminate fa-tigue concerns. This is not recommended since creep inthe supporting concrete foundation can eventually lead torelaxation of the pretensioning. Table 9.1.1 shows rec-ommended allowable fatigue stresses for non-preten-sioned steel bolts. These values are based on S-N datafor a variety of different types of bolts. (These data wereobtained from correspondence with Professor W. H.Munse of the University of Illinois and are based on re-sults from a number of test studies.) By examining thesevalues, it can be ascertained that, for the AISE loadingcondition, fatigue will not govern when A36 or A307 an-chor bolts are used. However, fatigue can govern the de-sign of higher strength anchor bolts for this load case.

Number ofLoading Cycles a

20,000 to 100,000

100,000 to 500,000

500,000 to 2,000,000

Over 2,000,000

Allowable TensileStress (psi)

40,000

25,000

15,000

10,000a — These categories correspond to the loadingconditions indicated in Appendix K of the AISCSpecification.

Table 9.1.1 Allowable Bolt Fatigue Stress

Step 3 - Base plate thickness may be governed by bend-ing associated with compressive loads or tensile loads.For compressive loads, the design procedure illustratedin the "Column Base Plates" section of Part 3 of the AISCManual of Steel Construction may be followed. How-ever, for lightly loaded base plates where the dimensions"m" and "n" (as defined in this procedure) are small, thin-ner base plate thickness can be obtained using yield linetheory.

-26-


Administrator

Rectangle

For tensile loads, a simple approach is to assumethe anchor bolt loads generate bending moments in thebase plate consistent with cantilever action about the webor flanges of the column section (one-way bending). Amore refined analysis for bolts positioned inside the col-umn flanges would consider bending about both the weband the column flanges (two-way bending). For thetwo-way bending approach, the derived bending mo-ments should be consistent with compatibility require-ments for deformations in the base plate. In either case,the effective bending width for the base plate can be con-servatively approximated using a 45 degree distributionfrom the centerline of the anchor bolt to the face of thecolumn flange or web. Calculations for required baseplate thickness for uplift (tensile) loads are illustrated inExamples 9.4.1 and 9.4.2.

Step 4 - The development of anchor bolts in tension isusually accomplished by using a hook, nut, or steel plateat the embedded end of the anchor bolt. Although chemi-cal bond developed between the anchor bolt and the sur-rounding concrete may also aid in developing tension inthe anchor bolt, it is typically neglected because anchorbolts are often oiled during manufacturing. Appendix Bof the ACI Code Requirements for Nuclear Safety Re-lated Concrete Structures (ACI 349-85)(7) recommendsthat the design for the development of the bolt in concreteaccommodate the full tensile strength of the anchor bolt

to insure a ductile failure for the anchor. Theauthors suggest that, for non nuclear structures, the de-sign for the development should accommodate 1.25times the yield strength of the anchor boltFor an A307 or A36 anchor bolt, this is equivalent to ap-proximately 3/4 of the full or ultimate tensile strength ofthe bolt. This is consistent with the provisions for devel-opment length, splices and mechanical connectionslisted in Chapter 12 of the ACI Building Code Require-ments for Reinforced Concrete Structures (ACI318-89)(4). The designer may use his/her judgment in ap-plying this criterion when analysis shows that tension inthe anchor bolts is either nonexistent or minimal.

Hooked anchor bolts usually fail by straighteningand pulling out of the concrete. This failure is precipi-tated by a localized bearing failure in the concrete abovethe hook. Calculation of the development load providedby a hook is illustrated in Example 9.4.1. As indicated inthis example, a hook is generally not capable of develop-ing the recommended tensile capacity mentioned in theprevious paragraph Therefore, hooksshould only be used when tension in the anchor bolt iseither nonexistent or minimal.

Tests have shown that a heavy bolt head, or a heavyhex nut on a threaded rod, will develop the full tensile ca-pacity of even high strength anchor bolts when properlyembedded and confined in concrete. Therefore, the de-sign for development for headed anchor bolts (typically

threaded rods with heavy hex nuts) is a matter of deter-mining the required embedment depths, edge distancesand/or steel reinforcement to prevent failure in the con-crete prior to the development of the recommended ten-sile capacity for the bolt.

As presented in Appendix B of ACI 349-85, failureoccurs in the concrete when tensile stresses along the sur-face of a stress cone surrounding the anchor bolt exceedthe tensile strength of the concrete. The extent of thisstress cone is a function of the embedment depth, thethickness of the concrete, the spacing between adjacentanchors and the location of adjacent free edges in the con-crete. The shapes of these stress cones for a variety ofsituations are illustrated in Figures 9.1.1, 9.1.2 and 9.1.3.The tensile strength of the concrete (in ultimate strengthterms) is represented as a uniform tensile stress of

over the surface area of these stress cones. By ex-amining the geometry, it is evident that the ultimate pull-out strength of this cone is equal to times the pro-jected area of the cone at the surface of the concrete(excluding the area of the anchor head). Expressions for

and the ultimate pullout strength are includedin Figures 9.1.1, 9.1.2 and 9.1.3.

The use of plate washers or bearing plates in lieu ofor in conjunction with a heavy hex bolt head or nut canincrease the surface area of the stress cone and the pulloutstrength of the concrete for a given embedment depth.However, this may not be the case where bolts are rela-tively close to the edges of the concrete structure and theextent of the stress cones surrounding the bolts is limitedby these edges. In this case, the use of larger washers orbearing plates at the embedded end of the anchor boltsmay actually reduce the pullout strength of the concrete(in effect, creating a weakened plane in the concrete).This is often the case with anchor bolts embedded in con-crete piers. Appendix B of ACI 349-85 lists dimensionalcriteria that are to be maintained for washers or bearingplates used as anchor heads. These criteria were devel-oped to be consistent with the dimensional characteris-tics of a standard heavy hex bolt head or nut.

The previously described stress cone checks relyupon the strength of plain concrete for developing the an-chor bolts and would typically apply when columns aresupported directly on spread footings, concrete mats orpile caps. However, in some instances the projected areaof the stress cones or overlapping stress cones is ex-tremely limited due to edge constraints. Consequentlythe tensile strength of the anchor bolts cannot be fully de-veloped with plain concrete. This is often the case withconcrete piers. In these instances, steel reinforcement inthe concrete is used to develop the anchor bolts. This re-inforcement often doubles as the reinforcement requiredto accommodate the tension and/or bending forces in thepier. The reinforcement must be sized and developed forthe recommended tensile capacity of the anchor bolts

-27-


Administrator

Rectangle

* If anchor is locatedin concrete undertension (due to axialload or bending) andtensile stressin concretethen for stress = 0.65

Figure 9.1.1 Stress Cone Development forSingle Anchor Bolts in Concrete

on both sides of the potential failure planedescribed in Figure 9.1.4. The anchor bolt embedmentlengths are determined from the required developmentlengths for this reinforcing steel. These embedmentlengths can be reduced by using a larger number ofsmaller diameter reinforcing bars to develop the anchorbolts. Also, hooks or bends can be added to this rein-forcement to minimize development lengths.

Appendix B of ACI 349-85 also lists criteria forminimum concrete side cover on anchor bolts to pre-vent "failure due to lateral bursting forces at the anchorhead". These lateral bursting forces are associated withtension in the anchor bolts. The failure plane or surfacein this case is assumed to be cone shaped and radiatingfrom the anchor head to the adjacent free edge or side ofthe concrete structure. This is illustrated in Figure 9.1.5.As with the pullout stress cones, overlapping of the stresscones associated with these lateral bursting forces shouldbe considered. The ultimate tensile strength of the con-crete for resisting these bursting forces is equal to

over the projected area of these cones at the freeedge of the concrete.

1. This area is an approximation for the areaenclosed by the four overlapping shearcones. The error associated with thisapproximation is typically small.

2. If anchor is located in concrete undertension (due to axial load or bending) andtensile stress is , then for stresscone strength calculation = 0.65

Figure 9.1.2 Stress Cone Limitations Due toOverlapping Stress Cones and Overall Thick-ness of the Concrete

-28-

Stress Reduction Areafor Limited Depthxy (See sections A–A and B–B)Effective Stress Area

Total Area of Anchor Heads

(For Bolt Group)0.85 (See footnote 2)

Rev.3/1/03


d d(a+b)+ab

Administrator

Rectangle

Administrator

Rectangle

Fig. 9.1.3 Limitations on Tensile Stress Conefor an Anchor Bolt Located Near an Edge

Fig. 9.1.4 The Use of Steel Reinforcement forDeveloping Anchor Bolts

As a design aid, Table 9.1.2 lists embedment lengthand edge distance requirements for single

headed anchor bolts without overlapping stress conesi.e., spacing of anchor bolts is greater than 2 xand respectively. In addition, the curves in Fig-ure 9.1.6 show adjustment factors to be applied to these

Fig.9.1.5 Design Check for Lateral BurstingForces for Anchor Bolts in Tension Locatednear an Edge

values of and for two bolts with overlapping stresscones. The curves in Figure 9.1.7 show adjustment fac-tors to be applied to for four bolts (arranged in asquare pattern) with overlapping stress cones. For thistype of bolt arrangement, adjustment factors for arestill obtained from Figure 9.1.6. These figures and tabledo not account for reductions due to lack of thickness inthe concrete or edge distance.

Anchor bolt design must be coordinated with thedesign of the superstructure and the foundations. Calcu-lations for anchor bolt development are illustrated in Ex-ample 9.4.1.

9.2 Resisting Shear Forces using Shear Friction The-ory

Appendix B of ACI 349-85 describes a "shear-friction" mechanism for transferring shear from anchorbolts into the concrete. The commentary to ACI 349-85suggests that this mechanism is developed as follows:

1. Shear is initially transferred through the anchorbolts to the grout or concrete by bearing near thesurface of the concrete.

2. This bearing results in the formation of a con-crete wedge in front of the anchor bolt approxi-mately one-quarter of the bolt diameter in depth(See Fig. 9.2.1). The formation of this wedgeoccurs at loads below the shear capacity of theanchor bolts.

-29-


Administrator

Rectangle

Single A307 & A36 Headed Anchor Bolts or Bolts with Embedded Nuts

BoltDiameter

(in.)

1/2

5/83/4

7/811-1/41-1/21-3/422-1/42-1/22-3/43

TensileStressArea

(in.2)

0.1420.2260.3340.4620.6060.9691.411.902.503.254.004.935.97

HeavyHexWidthAcrossFlat (F)(in.)

0.8751.06251.251.43751.6252.002.3752.753.1253.503.8754.254.625

EffectiveDiameter(C)(=1.05F)

(in.)

0.921.121.311.511.712.102.492.893.283.684.074.464.86

(in.)

2.93.64.65.36.07.79.3

10.712.314.315.617.419.1

(in.)

1.72.12.53.03.44.35.26.06.97.98.89.7

10.7

(in.)

4.45.66.88.09.2

11.614.016.218.621.223.526.128.8

(in.)

3.94.96.07.18.1

10.212.314.316.418.720.823.025.4

(in.)

3.54.45.36.37.29.1

10.912.714.516.618.420.422.5

* Based on value = 0.85Note: values listed are based on concrete with = 3000 psi

Table 9.1.2 Anchor Bolt Properties and Requirements

Fig. 9.1.6 Adjustment Factors for Two Anchor Bolts with Overlapping Stress Cones

3. The shear force on this wedge causes the wedgeto translate. Lateral translation of the wedge isaccompanied by upward movement causing thewedge to push on the bottom of the base plate.The anchor bolts prevent the base plate frommoving upward. Therefore, this push results in

tensile forces in the anchor bolts and, in effect, aclamping force between the concrete surfaceand the bottom of the base plate. Shear transferat this point is derived from friction between thebase plate and the concrete. The relationshipfor this resistance is described as follows:

-30-


Administrator

Rectangle

Fig. 9.1.7 Adjustment Factors for Four Anchor Bolts with Overlapping Stress Cones and Square Spacing Pattern

Fig. 9.2.1 Concrete Wedge Formed byShear Force in Anchor Bolt

where = shear force transferred by shearfriction.

N = normal or clamping forceµ = coefficient of friction between the

concrete and base plate

Appendix B of ACI 349-85 lists the following co-efficients of friction for use with shear friction theory:

a. 0.9 for concrete or grout against as-rolledsteel with the contact plane a full platethickness below the concrete surface (i.e.,the base plate set into the grout or con-crete).

b. 0.7 for concrete or grout against as-rolledsteel with the contact plane coincidentalwith the concrete surface.

c. 0.55 for grouted conditions with the con-tact plane between the grout and as-rolledsteel exterior to the concrete surface (thenormal condition).

The validity of this mechanism has been questionedby some designers. However, according to an article bythe principal authors of Appendix B of ACI-349-85(5),"these design limits have been established using bothanalytical and test methods" and have been subject to"rigorous evaluation" by ACI Committee 349, the ACITechnical Activities Committee (TAC), and the generalmembership of the American Concrete Institute.

Using this concept, the design of anchor bolts forshear forces becomes a design for tension with the tensileforce in the anchor bolt evaluated as:

where = tensile force per bolt due to shear friction.

µ = appropriate coefficient of friction.

-31-


Administrator

Rectangle

n = number of bolts.

The location and placement of anchor bolts in con-crete is not extremely accurate. In anticipation of this,holes for anchor bolts in column bases are typically over-sized. The larger the anchor bolt, the larger the oversizemust be. The AISC Manuals list recommended oversizedimensions for these holes in the "Suggested Details"portion of the "Connections" section. In recognition ofthe oversized holes, it is very possible that all of the an-chor bolts are not in bearing with the base plate and,therefore, do not participate fully in transferring shearforces. For a typical four bolt anchor bolt arrangement, adesigner may wish to assume that only two of the boltsprovide the shear friction resistance. If special platewashers with standard sized holes are added beneath thenuts and welded appropriately to the base plate, all of theanchor bolts could be assumed to provide the shear fric-tion resistance. Special attention should be given to an-chor bolts with thick base plates and oversized holes and/or thick unconfined grout beds. These conditions cancause large eccentricities between point of loading andthe concrete support, which require bolts to resist bend-ing forces in addition to shear.

All of the criteria previously described for the de-sign of anchor bolts in tension apply to the design for an-chor bolts subjected to tension as a result of shear fric-tion. In situations where shear and tension are to be si-multaneously transferred at the base of a column, the de-sign tension for the anchor bolt is equal to the sum of thetensile force from uplift (T) and the tensile force due toshear friction

For cases where shear and compression exist simul-taneously at the base of a column, the compressive forcemay account for a portion or all of the shear transferthrough friction. In this case, the coefficients of friction(µ) referenced earlier may be used with the compressiveforce to determine the magnitude of this frictional resis-tance. The shear force in excess of this frictional resis-tance would be transferred through the anchor bolts usingshear friction as previously described. For seismic de-signs it is recommended that any compression forcebenefit be neglected.

Appendix B of ACI 349-85 contains a criteria forminimum edge distance for bolts when shear fric-tion is used. These criteria are included to prevent local-ized failure in the concrete adjacent to the bolt. Accord-ing to the commentary to ACI 349-85, "when the bolt isnear an edge, the total shearing force must be developedby tensile stress on a potential failure plane radiating at45 degrees toward the free edge from the anchor steel atthe surface of the concrete". This plane is described by a45 degree half-cone as illustrated in Figure 9.2.2.

Fig, 9.2.2 Edge Distance Criteria forShear Forces Resisted by Shear Friction

The ultimate tensile strength of the concrete for resistingthis shear force is equal to on the projected areaof this half-cone at the free edge. Consideration shouldbe given to overlapping cones for bolts spaced relativelyclose together. If this edge distance cannot be met, rein-forcing similar to that shown in Figure 9.2.3 may beadded to prevent this failure. As shown in this figure, thereinforcement is developed on the wedge side of the po-tential failure plane by attachment to an angle bearing onthe surface of the concrete. In many cases it may not bepractical to use a normal steel tie or stirrup to reinforcethe concrete for this failure mode because the tie or stir-rup cannot be adequately developed on the wedge side ofthe potential failure plane. ACI 349-85 states that even ifthis reinforcement is added (as shown in Figure 9.2.3), aminimum edge distance of must be maintained.

Table 9.1.2 lists values of for a variety of A36 orA307 bolts and the three different values of µ. These val-ues apply to single bolts without overlapping stress cones(i.e. spacing of bolts is greater than 2 x The adjust-ment factors in Figure 9.1.6 can be used with these valuesof for two bolts with overlapping stress cones.

ACI 349-85 states that adequate edge distance orreinforcement should be provided to develop the ultimateshear friction capacity of the anchor boltThis requirement is to insure ductile failure in the case of

-32-


Administrator

Rectangle

Fig. 9.2.3 Concrete Reinforcement Due to Lackof Edge Distance for Shear

severe shear overload. As discussed previously a moreappropriate criterion for non nuclear structures may be todesign edge distance or reinforcement to develop a shearfriction capacity equal to This is con-sistent with the previously recommended criteria for de-velopment of anchor bolts in tension. The previouslymentioned design aids (Table 9.1.2 and Figure 9.1.6)were based on this criterion.

Example 9.4.2 illustrates the design for anchorbolts subjected to combined shear and tension.

9.3 Resisting Shear Forces through Bearing

Shear forces can be transferred in bearing by the useof shear lugs or by embedding the column in the founda-tion. These methods are illustrated in Figure 9.3.1. TheAISC Steel Design Guide Series, Column Base Plates(8)

discusses design for both of these methods. Additionalcomments are provided below:

1. For shear lugs or column embedments bearingin the direction of a free edge of the concrete,Appendix B of ACI 349-85 states that in addi-tion to considering bearing failure in the con-crete, "the concrete design shear strength for thelug shall be determined based on a uniform ten-sile stress of acting on an effective stressarea defined by projecting a 45 degree planefrom the bearing edge of the shear lug to the freesurface". The bearing area of the shear lug (orcolumn embedment) is to be excluded from theprojected area. This criterion may control orlimit the shear capacity of the shear lug or col-umn embedment details in concrete piers.

Shear Lug Detail

Column Embedment Detail

Fig. 9.3.1 Transfer of Base Shears Through Bearing

2. Consideration should be given to bending in thebase plate resulting from forces in the shear lug.As a rule of thumb, the authors generally re-quire the base plate to be of equal or greaterthickness than the shear lug.

3. Consideration should be given to bending in thecolumn resulting from forces in the shear lug.This can be of special concern when the baseshears (most likely due to bracing forces) arelarge and bending from the force on the shearlug is about the weak axis of the column.

4. Multiple shear lugs may be used to resist largeshear forces. Appendix B of ACI 349-85 pro-vides criteria for the design and spacing of mul-tiple shear lugs.

A typical design for a shear lug is illustrated in Example9.4.3.

A brief discussion on the use of hairpins or tie rodsis included to complete the discussion on anchorage de-sign. Hairpins are typically used to incorporate the fric-

-33 -


Administrator

Rectangle

tion between the floor slab and the subgrade in resistingthe column base shear when individual footings are notcapable of resisting horizontal forces. The column baseshears are transferred from the anchor bolts to the hairpinthrough bearing. Problems have occurred when the hair-pin bars were placed too low on the anchor bolts (asshown in Figure 9.3.2), thus generating bending in theanchor bolts when the shear friction capacity of the an-chor bolt detail is exceeded. This problem can beavoided as shown in Figure 9.3.3 or by providing shearlugs. Since hairpins rely upon the frictional restraint pro-vided by the floor slab, special consideration should begiven to the location and type of control and constructionjoints used in the floor slab to assure no interruption inload transfer, yet still allowing the slab to move.

Fig. 9.3.2 Improper Location of Hairpin Bars

Fig. 9.3.3 Alternate Hairpin Detail

Tie rods are typically used to counteract large shearforces associated with gravity loads on rigid frame struc-tures. When using tie rods with large clear span rigidframes, consideration should be given to elongation ofthe tie rods and to the impact of these elongations on theframe analysis and design. Again special attention to thedimension between the base plate and the tie rods is re-quired. In addition significant amounts of sagging orbowing should be removed before tie rods are encased orcovered, since the tie rods will tend to straighten whentensioned.9.4 Column Anchorage Examples (Pinned Base)

EXAMPLE 9.4.1: Column Anchorage For TensileLoads

Design a base plate and anchorage for a W 10x45 columnsubjected to a net uplift as a result of the following load-ings:

Fig. 9.4.1 Example

Procedure:

1. Determine the maximum net uplift on the column.

2. Select the type and number of anchor bolts.

3. Determine the appropriate base plate thickness andwelding to transfer the uplift forces from the columnto the anchor bolts.

4. Determine the method for developing the anchorbolts in the concrete.

Solution:

1. Maximum net uplift = 60-22=38 kips.

2. Use four anchor bolts (helps to provide stability dur-ing erection).

Using Table 1-A in the "Connections" Section ofthe ASD Manual of Steel Construction, select a 3/4inch diameter A307 bolt.

Determine the allowable force for the wind case.

Note: Bolts are positioned inside the column pro-file and bolt forces are not extremely large; there-fore, prying forces are negligible.

-34-


Administrator

Rectangle

3. The bolts are positioned inside the column profilewith a 4 in. square pattern. To simplify the analysis,conservatively assume the tensile loads in the an-chor bolts generate one-way bending in the baseplate about the web of the column. This assumptionis illustrated by the bolt load distributions shown inFig. 9.4.2.

in the base plate equals the bolt force times thelever arm to the column web face.

The effective width of base plate for resisting atthe face of web =

Assuming a 45 degree distribution for the bolt loads,

Use a 1 in. thick plate = 36 ksi).

For welding of the column to the base plate:

Maximum weld load =

Minimum weld for a 1 inch thick base plate = 5/16 in.(Table J2.4 of ASD Specification).

Allowable weld load per inch for a 5/16" fillet weldwith E70 electrode:

2.6 < 6.19 5/16" Fillet weld on each side of the col-umn web is o.k.

4. As noted earlier, this column is anchored in the mid-dle of a large spread footing. Therefore, there are noedge constraints on the concrete tensile cones andthere is no concern regarding edge distance to pre-vent lateral bursting.

To insure a ductile failure in the case of overload, de-sign the development for the anchor bolts for

For 3/4" diameter A307 bolts, this isequal to 1.25(0.334)(36) = 15.0 kips/bolt.

Try using a 3 inch hook on the embedded end of theanchor bolt to develop the bolt.

Assuming uniform bearing on the hook,

hook bearing c a p a c i t y ( 1 6 )

where =0.70= concrete compressive strengthd= hook diameter= hook length

Hook bearing capacity

Thus a 3 inch hook is not capable of developing therequired tensile force in the bolt.

Therefore, use a heavy hex nut to develop the anchorbolt.

According to Table 9.1.2, the required embedmentdepth for a single 3/4" diameter A307 bolt = 4.6inches. Since the bolt spacing is less thanthe stress cones for the bolts overlap. Using Figure9.1.7,

Use an embedment depth =

This is the depth of concrete embedment to the top ofthe heavy hex nut.

Fig. 9.4.3 Embedment Depth

This solution assumes that the footing has sufficientthickness to allow for the full development of thestress cones (See Figure 9.1.2).

For this to be valid, the footing thickness required+ Bolt Spacing/2

-35-

Fig. 9.4.2 Bolt Load Distribution


Administrator

Rectangle

EXAMPLE 9.4.2: Column Anchorage for CombinedTension and Shear Loads (Pinned Base)

Design a base plate and anchorage for the W 10x45 col-umn examined in Example 9.4.1 but with an additionalbase shear of 23 kips due to wind. Assume a 2 inch thickgrout bed is used beneath the base plate. For this exam-ple, the column is assumed to be supported on a 20 inchsquare pier.

Fig. 9.4.4 Example

Procedure:

1. Determine the maximum net tension in the anchorbolts (due to net tension and shear loads at the baseof the column).

2. Select the type and number of anchor bolts.

3. Determine the appropriate base plate thickness andwelding to transfer the uplift and shear forces fromthe column to the anchor bolts.

4. Determine the method for developing the anchorbolts in the concrete.

Solution:

1. As determined in Example 9.4.1, the net uplift on thecolumn = 38 kips. The tension in the bolts due toshear loads (using the shear friction concept) is de-fined as

Use four bolts with plate washers welded to the col-umn base plate to insure that all four bolts participatein transferring the shear load (see discussion in Sec-tion 9.2).

Since a grout bed is used beneath the column,µ = 0.55.

Therefore, = 10.5 kips/bolt

2. As noted in Step 1, a total of (4) anchor bolts are to beused. Using Table 1-A in the "Connections" Sec-tion of the AISC Manual of Steel Construction, se-lect 1" diameter, A307 bolts.

For a wind load case,= 20.9 kips/bolt

3. Position the bolts within the profile of the columnwith a 5 in. square pattern. Again, conservatively as-sume the tensile loads in the anchor bolts generateone-way bending in the base plate about the web ofthe column.

The effective width of the base plate at the face of theweb =

Assuming a 45 degree distribution for the bolt loads,

Use a 1-1/2 in. thick plate = 36 ksi).

For welding of the column to the base plate,

Maximum weld load

Minimum weld for a 1-1/2 inch thick base plate= 5/16 in.

Allowable weld load for a 5/16" fillet weld withE70 electrode=0.3125(0.707)(21)(4/3)=6.19 kips/in.

4.49 < 6.19 5/16" fillet weld is o.k.

4. To insure a ductile failure in the case of overload, de-sign the development for the anchor bolts for1.25 x

-36-


Administrator

Rectangle

For 1" diameter A307 bolts, this is equal to:

As noted earlier, the column is supported on a 20inch square pier. The maximum tensile load that canbe developed by an unreinforced concrete pier isequal to:

times the cross sectional area of pier.

The required ultimate tensile strength to insureproper development for the (4) 1" diameter bolts isequal to 4(27.3) = 109.2 kips.

Therefore, reinforcing steel must be used in the con-crete to develop the anchor bolts. For sizing this re-inforcement, T/bolt may be taken as (ratherthan This is appropriate because thereinforcing steel is ductile in nature and because thecalculated development lengths (per ACI 318) forthe reinforcing steel will include the 1.25 factor.

Using grade 60 reinforcing bars,

Use one #6 reinforcing bar (A=0.44 in.2) with eachbolt.

According to Section 12.2 of ACI 318, the basic de-velopment length for a #6 bar (grade 60) in 3000 psiconcrete = 24.6 inches.

Since the area of steel provided (0.44 in.2) exceedsthe area of steel required (0.40 in.2), this develop-ment length may be reduced by a factor equal to(0.40/0.44) = 0.91.

Therefore, the required development length for the#6 bars is calculated as:

An argument can be made that the more conserva-tive lap splice criteria contained in Section 12.15 ofACI 318 should be applied. The authors feel thatthis is not necessary. The rationale for this decision

is that the Commentary to ACI 318 states that themore conservative criteria applied to lap splices is to"encourage the location of splices away from re-gions of high tensile stress". It is therefore evidentthat the more conservative criteria are not associatedwith reductions in development capabilities within alapped splice.

The required embedment depth for the anchorbolts is determined as shown in Fig. 9.4.5.

Fig. 9.4.5 Embedment Depth

It is suggested that, when reinforcing steel is addedto develop anchor bolts, this reinforcement shouldbe enclosed by tie reinforcement near the top of thereinforcing bars. This reinforcement is added to pre-vent splitting failures in the concrete prior to yield-ing the reinforcing steel. The following equationprovides the suggested minimum total area of tie re-inforcement to be used.

where T = sum of yield strengths for the anchorbolts.

s = spacing from anchor bolt toreinforcement,

h = height of ties above top of the anchorhead.

= yield strength of the tiereinforcement.

For the example problem,

T = 4(0.606)(36) = 87.3 kips

s = 4"

h = 26" (assume tie located 2-1/2" from top ofpier)

-37-


Administrator

Rectangle

Fig. 9.4.6 Tie Reinforcement

Use a #3 bar.

Total tie area provided = 2(0.11) = 0.22 in.2

If the required is large, two ties may be used nearthe top of the reinforcing bars and "h" taken as theaverage height of these ties above the top of the an-chor heads.

Also, edge distance requirements must be checked.Using Table 9.1.2 and Fig. 9.1.6, the required edgedistances are as shown below:

= 3.4"(1.26) =4.3"

= 7.2"(1.26) = 9.1"

For a 20 inch square pier and a 5 inch square anchorbolt pattern, edge distance provided= (20"-5")/2=7.5"

Since exceeds 7.5", reinforcement similar to thatshown in Figure 9.2.3 should be considered. Alter-natively, the pier size could be increased to providethe required edge distance.

EXAMPLE 9.4.3: Design for Shear Lugs (PinnedBase)

Design a shear lug detail for the W 10x45 column consid-ered in Example 9.4.2. See Fig. 9.4.7.

For this detail, the anchor bolts are designed to transferthe net uplift from the column to the pier and the shear lugis designed to transfer the 23 kip shear load to the pier.The design for the anchor bolts is similar to Example9.4.2 except that tension in the bolts due to shear frictionis not included. Therefore, calculations for the anchor

Fig. 9.4.7 Example

bolts are not included in this example. As shown, the an-chor bolts are positioned outside the column flanges toprevent interference with the lug detail.

Procedure:

1. Determine the required embedment for the lug intothe concrete pier.

2. Determine the appropriate thickness for the lug.

3. Size the welds between the lug and the base plate.

Solution:

1. Two criteria are used to determine the appropriateembedment for the lug. These criteria are the bear-ing strength of the concrete and the shear strength ofthe concrete in front of the lug. As discussed in Sec-tion 9.3, the shear strength of the concrete in front ofthe lug is evaluated (in ultimate strength terms) as auniform tensile stress of acting on an effec-tive stress area defined by projecting a 45 degreeplane from the bearing edge of the shear lug to thefree surface (the face of the pier). The bearing areaof the lug is to be excluded from the projected area.Since this criterion is expressed in ultimate strengthterms, the bearing strength of the concrete is alsoevaluated with an ultimate strength approach. Ac-cording to ACI 318-89, the ultimate bearingstrength of the concrete in contact with the lug isevaluated as:

where = bearing capacity of the concrete incontact with the lug.

= 0.70= embedded area of the shear lug (this

does not include the portion of the lugin contact with the grout above thepier).

-38-


Administrator

Rectangle

As discussed in the AISC Design Guide 1(8), Col-umn Base Plates, it is not recommended that thisstrength be increased based on confinement from thesurrounding concrete, grout and base plate.

The factored shear load = 1.3x23 = 29.9 kips. (1.3factor is used since this is a wind load case.)

Equating this to the bearing capacity of the concrete,the following relationship is obtained,

0.70(0.85)(3000) = 29900

= 16.75 in.2

Assuming the base plate and shear lug width to be 9in., the required embedded depth (d) of the lug (inthe concrete) is calculated as:

d= 16.75/9 =1.86 in. Say 2".

See Figure 9.4.8.

Fig. 9.4.8 Shear Lug Depth

Using this embedment, the shear strength of the con-crete in front of the lug is checked. The projectedarea of the failure plane at the face of the pier isshown in Fig. 9.4.9.

Assuming the lug is positioned in the middle of thepier and the lug is 1 inch thick,

a = 5.5" due to constraints of pier width

b = 2"+9.5"=11.5"

The projected area of this plane excluding thearea of the lug, is then calculated as:

= (5.5+9+5.5)(11.5)-2(9) = 212 in.2

Using this area, the shear capacity of the concrete infront of the lug is calculated as:

= 39,480 lbs. > 29.9 kips. o.k.

Fig. 9.4.9 Lug Failure Plane

It is concluded that, for a 9 inch wide lug, an embed-ment depth (d) of 2 inches is adequate.

2. Using working loads and a cantilever model for thelug,

= V(G+d/2)

= 23(2+2/2) = 69 kip-in.

Note: G = thickness of grout bed.

Use a 1-1/4" thick lug = 36 ksi)

(As discussed in Section 9.3, a minimum thicknessof 1-1/4" would also be recommended for sufficientstiffness in the base plate.)

3. Most steel fabricators would rather use heavy filletwelds than penetration welds to attach the lug to thebase plate. The forces on the welds are as shown inFig. 9.4.10.

-39-


Administrator

Rectangle

Fig. 9.4.10

Assuming 5/16" fillet welds,

The resultant weld load is calculated as:

For a 5/16" fillet weld using E70 electrode, the al-lowable load is calculated as:

5/16" fillet welds are o.k.

10. SERVICEABILITY CRITERIA

The design of the lateral load envelope (i.e., the roofbracing and wall support system) must provide for thecode imposed loads, which establish the requiredstrength of the structure. A second category of criteriaestablishes the serviceability limits of the design. Theselimits are rarely codified and are often selectively appliedproject by project based on the experience of the partiesinvolved.

In AISC Design Guide No. 3(17) several criteria aregiven for the control of building drift and wall deflection.These criteria, when used, should be presented to thebuilding owner as they help establish the quality of thecompleted building.

To be useful, a serviceability criterion must set forththree items: a) loading, b) performance limit, and c) ananalysis approach. Concerning lateral forces, the loadingrecommended by Design Guide No. 3 is the pressure dueto wind speeds associated with a ten year recurrence in-terval. These pressures are approximately 75% of thepressures for strength design criteria, based on a fiftyyear return period. The establishment of deflection lim-its is explained below, with criteria given for each of thewall types previously presented. The authors recom-mend that frame drift be calculated using the bare steelframe only. Likewise the calculations for deflection of

girts would be made using the bare steel section. Thecontribution of non structural components acting com-positely with the structure to limit deflection is often dif-ficult to quantify. Thus the direct approach (neglectingnon structural contribution) is recommended and theloads and limits are calibrated to this analysis approach.The deflection limits for the various roof and wall sys-tems are as follows.

10.1 Serviceability Criteria for Roof Design

In addition to meeting strength criteria in the designof the roof structure, it is also necessary to provide for theproper performance of elements and systems attached tothe roof, such as roofing, ceilings, hanging equipment,etc. This requires the control of deflections in the roofstructure. Various criteria have been published by vari-ous organizations. These limits are

1. American Institute of Steel Construction:(42)

a. Depth of fully stressed roof purlins shouldnot be less than times the span.

2. Steel Deck Institute:(38)

a. Maximum deflection of deck due to uni-formly distributed live load: span over 240.

b. Maximum deflection of deck due to a 200lb concentrated load at midspan on a onefoot section of deck: span over 240.

3. Steel Joist Institute:(40)

a. Maximum deflection of joists supportingplaster ceiling due to design live load: spanover 360.

b. Maximum deflection of joists supportingceilings other than plaster ceilings due todesign live load: span over 240.

4. National Roofing Contractors Association(NRCA)(29):

a. Maximum deck deflection due to full uni-form load: span over 240.

b. Maximum deck deflection due to 300 lbload at midspan: span over 240.

c. Maximum roof structure deflection due tototal load: span over 240.

5. Factory Mutual:(3)

a. Maximum deck deflection due to a 300 lbconcentrated load at midspan: span over200.

AISC Design Guide No. 3 also presents deflectionlimits for purlins supporting structural steel roofs (boththrough fastener types and standing seam types). First, a

-40-


Administrator

Rectangle

Administrator

Rectangle

limiting deflection of span over 150 for snow loading isrecommended. Secondly, attention is drawn to condi-tions where a flexible purlin parallels nonyielding con-struction such as at the building eave. In this case deflec-tion should be controlled to maintain positive roof drain-age. The appropriate design load is suggested as deadload plus 50 percent of snow load or dead load plus 5 psflive load to check for positive drainage under load.

Mechanical equipment, hanging conveyors, andother roof supported equipment has been found to per-form adequately on roofs designed with deflection limitsin the range of span over 150 to span over 240 but this cri-teria should be verified with the equipment manufacturerand building owner. Consideration should also be givento differential deflections and localized loading condi-tions.

10.2 Metal Wall Panels

Relative to serviceability metal wall panels have twodesirable attributes: 1) Their corrugated profiles makethem fairly limber for out of plane distortions and 2) theirmaterial and fastening scheme are ductile (i.e., distor-tions and possible yielding do not produce fractures).Also, the material for edge and corner flashing and trimgenerally allows moment and distortion without failure.Because of this the deflection limits associated withmetal panel buildings are relatively generous. They are:

1. Frame deflection (drift) perpendicular to thewall surface of frame: eave height divided by 60to 100.

2. The deflection of girts and wind columnsshould be limited to span over 120, unless walldetails and wall supported equipment requirestricter limits.

10.3 Precast Wall Panels

Non load bearing precast wall panels frequentlyspan from grade to eave as simple span members. There-fore drift does not change the statics of the panel. Thelimitation on drift in the building frame is established tocontrol the amount of movement in the joint at the base ofthe panel as the frame drifts. This limit has been pro-posed to be eave height over 100. A special case existswhen precast panels are set atop the perimeter founda-tions to eliminate a grade wall. The foundation anchor-age, the embedment of the panel in the soil and the poten-tial of the floor slab to act as a fulcrum mean that theframe deflections must be analyzed for compatibilitywith the panel design. It is possible to tune frame driftwith panel stresses but this requires interaction betweenframe designer and panel designer. Usually the design ofthe frame precedes that of the panel. In this case theframe behavior and panel design criteria should be care-fully specified in the construction documents.

10.4 Masonry Walls

Masonry walls may be hollow, grouted, solid, orgrouted and reinforced. Masonry itself is a brittle, non-ductile material. Masonry with steel reinforcement hasductile behavior overall but will show evidence of crack-ing when subjected to loads which stress the masonry intension. When masonry is attached to a supporting steelframework, deflection of the supports may inducestresses in the masonry. It is rarely feasible to providesufficient steel (stiffness) to keep the masonry stressesbelow cracking levels. Thus flexural tension cracking inthe masonry is likely and when properly detailed is not aconsidered a detriment. The correct strategy is to imposereasonable limits on the support movements and detailthe masonry to minimize the impact of cracking.

Masonry should be provided with vertical controljoints at the building columns and wind columns. Thisprevents flexural stresses on the exterior force of the wallat these locations from inward wind. Because the top ofthe wall is generally free to rotate, no special provisionsare required there. Most difficult to address is the base ofthe wall joint. To carry the weight of the wall the basejoint must be solid, not caulked. Likewise the mortar inthe joints make the base of the wall a fixed condition untilthe wall cracks.

Frame drift recommendations are set to limit the sizeof the inevitable crack at the base of the wall. Becausereinforced walls can spread the horizontal base cracksover several joints, separate criteria are given for them. Ifproper base joints are provided, reinforced walls can beconsidered as having the behavior of precast walls; i.e.,simple span elements with pinned bases. In that case thelimit for precast wall panels would be applicable. Wherewainscot walls are used, consideration must be given tothe joint between metal wall panel and masonry wain-scot. The relative movements of the two systems in re-sponse to wind must be controlled to maintain the integ-rity of the joint between the two materials.

The recommended limits for the deflection of ele-ments supporting masonry are:

1. Frame deflection (drift) perpendicular to an un-reinforced wall should allow no more than a1/16—inch crack to open in one joint at the baseof the wall. The drift allowed by this criterioncan be conservatively calculated by relating thewall thickness to the eave height and taking thecrack width at the wall face as 1/16-inch andzero at the opposite face.

2. Frame deflection (drift) perpendicular to a rein-forced wall is recommended to be eave heightover 100.

3. The deflection of wind columns and girtsshould be limited to span over 240 but notgreater than 1.5 inches.

-41-


Administrator

Rectangle

Part 2

INDUSTRIAL BUILDINGS -WITH CRANES

11. INTRODUCTION TO PART 2

This section of the guide deals with crane buildings, andwill include coverage of those aspects of industrial buildingspeculiar to the existence of overhead and underhung cranes.In that context, the major differences between crane buildingsand other industrial buildings is the frequency of loadingcaused by the cranes. Thus, crane buildings should be classi-fied for design purposes according to the frequency of load-ing.

Crane building classifications have been established inthe AISE Technical Report No. 13(20)as classes A, B, C and D.Classifications for cranes have been established by the CraneManufacturers Association of America (CMAA).(39) Thesedesignations should not be confused with the building desig-nations as can be gathered from the following descriptions ofthe two classifications.

11.1 AISE Building Classifications

Class A - are those buildings in which members may ex-perience either 500,000 to 2,000,000 repetitions (LoadingCondition 3) or over 2,000,000 repetitions (Loading Condi-tion 4) in the estimated life span of the building of approxi-mately 50 years. Loading condition refers to the fatigue crite-ria given in Appendix K of the AISC Specifications, LRFD(22)

and ASD(42). The owner must analyze the service and deter-mine which load condition may apply. It is recommended thatthe following building types be considered as Class A:

Batch annealing buildingsScrap yardsBillet yardsSkull breakersContinuous casting buildingsSlab yardsFoundriesSoaking pit buildingsMixer buildingSteelmaking buildingsMold conditioning buildingsStripper buildingsScarfing yardsOther buildings as based on predictedoperational requirements

Class B - shall be those buildings in which members mayexperience a repetition from 100,000 to 500,000 cycles of aspecific loading, or 5 to 25 repetitions of such load per day fora life of approximately 50 years (Loading Condition 2).

Class C - shall be those buildings in which members mayexperience a repetition of from 20,000 to 100,000 cycles of aspecific loading during the expected life of a structure, or 1 to5 repetitions of such load per day for a life of approximately50 years (Loading Condition 1).

Class D - shall be those buildings in which no memberwill experience more than 20,000 repetitions of a specificloading during the expected life of a structure.

11.2 CMAA Crane Classifications

The following classifications are taken directly fromCMAA.

"70-2 CRANE CLASSIFICATIONS

2.1 Service classes have been established so thisspecification will enable the purchaser to specifythe most economical crane for the installation.Specific requirements are shown for these compo-nents where design is influenced by classifica-tions. All classes of cranes are affected by the op-erating conditions so for the purpose of these defi-nitions it is assumed that the crane will be operat-ing in normal ambient temperatures (0 to 100°F)and normal atmospheric conditions (free fromdust, moisture and corrosive fumes).

2.2 CLASS A (STANDBY OR INFREQUENTSERVICE)

This service class covers cranes which may beused in installations such as powerhouses, publicutilities, turbine rooms, motor rooms and trans-former stations where precise handling of equip-ment at slow speeds with long, idle period betweenlifts are required. Capacity loads may be handledfor initial installation of equipment and for infre-quent maintenance.

2.3 CLASS B (LIGHT SERVICE)This service covers cranes which may be used in re-pair shops, light assembly operations, service build-ings, light warehousing, etc., where service require-ments are light and the speed is slow. Loads mayvary from no load to occasional full rated loads withtwo to five lifts per hour, averaging 10 feet per lift.

2.4 CLASS C (MODERATE SERVICE)This service covers cranes which may be used in ma-chine shops or papermill machine rooms, etc., whereservice requirements are moderate. In this type of

42


Administrator

Rectangle

service the crane will handle loads which average 50percent of the rated capacity with 5 to 10 lifts perhour, averaging 15 feet, not over 50 percent of thelift at rated capacity.

2.5 CLASS D (HEAVY SERVICE)This service covers cranes which may be used inheavy machine shops, foundries, fabricating plants,steel warehouses, container yards, lumber mills,etc., and standard duty bucket and magnet opera-tions where heavy duty production is required. Inthis type of service, loads approaching 50 percent ofthe rated capacity will be handled constantly duringthe working period. High speeds are desirable forthis type of service with 10 to 20 lifts per hour aver-aging 15 feet, not over 65 percent of the lifts at ratedcapacity.

2.6 CLASS E (SEVERE SERVICE)This type of service requires a crane capable of han-dling loads approaching a rated capacity throughoutits life. Applications may include magnet/bucketcombination cranes for scrap yards, cement mills,lumber mills, fertilizer plants, container handling,etc., with twenty or more lifts per hour at or near therated capacity.

2.7 CLASS F (CONTINUOUS SEVERE SERVICE)This type of service requires a crane capable of han-dling loads approaching rated capacity continuouslyunder severe service conditions throughout its life.Applications may include custom designed spe-cialty cranes essential to performing the criticalwork tasks affecting the total production facility.These cranes must provide the highest reliabilitywith special attention to ease of maintenance fea-tures."

The class of crane, the type of crane, and loadings all af-fect the design. The fatigue associated with crane class is es-pecially critical for the design of crane runways and connec-tions of crane runway beams to columns. Classes E and F pro-duce particularly severe fatigue conditions. The determina-tion of stress levels and load conditions is discussed in moredetail in the next section.

The CMAA crane classifications do not relate directly tothe AISC loading conditions for fatigue. Based on the aver-age number of lifts for each CMAA crane classification, thecrane classes corresponding to the AISC loading conditionsare shown in Table 11.2.1.

The MB MA Low Rise Building Systems Manual(24) alsorelates CMAA crane classifications to the AISC loading con-ditions. The MBMA relationship shown in Table 11.2.2, is

CMAA CraneClassification

A, BC, DEF

AISC LoadingCondition

1234

Table 11.2.1 Crane Loading Conditions

based on equations which further define loading conditionsby the total weight and lifted load of the crane.

ServiceClass

B

C

D

AISC Loading Condition

R .5

-

1

2

R>.5

1

2

3

Table 11.2.2 MBMA Crane Service Classes

where

R = TW/(TW + RC) for underhung monorailcranes.

R = TW/(TW + 2RC) for bridge cranes.TW = The entire crane weight.RC = The rated capacity of the crane.

The MBMA procedure is recommended for classes B, C,and D since the weight of the cranes can have a very detrimen-tal effect on the fatigue of a runway system.

12. FATIGUE

Crane buildings are often loaded to full design loads, andin many cases the design load will occur thousands of times.Thus design stresses must be selected with regard to fatiguelimits.

An eventual failure due to repeated loading and unload-ing (even if the yield point of the material is never exceeded)is known as fatigue. Fatigue may be observed even if all con-ditions are near ideal; i.e., the material exhibits excellentnotch toughness, no stress concentrations from holes ornotches, uniaxial stress condition, ductile microstructure, noresidual strss, etc. However, the existence of multistress con-ditions, conditions affecting ductility, residual stresses, etc.all reduce the fatigue life of a structure. By knowing the maxi-

43


Administrator

Rectangle

mum number of cycles to which a structure is subjected, alongwith the stress ratio, the fatigue strength can be determined.

Contained in Appendix K of the AISC Specifications areguidelines for the design of structures subjected to a cyclicstress variation (fatigue). In Appendix K, allowable stressranges are given for various geometrical conditions depend-ing on the loading condition (expected number of stress cy-cles).

Fatigue failures have been observed in many crane build-ings. Failures generally occur in crane runway beams and canbe associated with an attachment to a beam, such as a stiff-ener, or they may be due to improper detailing of connectionsbetween runway beams or at a runway beam and column.Recommended design procedures and details illustrated inthe section on plate girders are to help prevent such failures.

13. CRANE INDUCED LOADS AND LOADCOMBINATIONS

It is recommended that the designer shows, on the draw-ings, the crane wheel loads, wheel spacing, bumper forces,and the design criteria used to design the structure.

Although loading conditions for gravity, wind, and seis-mic loads are well defined among building codes and stan-dards, crane loading conditions generally are not.

As mentioned previously, crane fatigue loadings are pri-marily a function of the class of service, which in turn is basedprimarily on the number of cycles of a specific loading case.This classification should be based on the estimated life span,rate of loading, and the number of load repetitions, the ownershould specify or approve the classification for all portions ofa building. A maximum life span of 50 years is generally ac-cepted.

The provisions of the AISC and AISE on crane runwayloads are summarized in the following discussion. As an al-ternate the MBMA Low Rise Building Systems Manual(24)

provides a comprehensive discussion on crane loads.

13.1 Vertical Impact

AISC and AISE

The allowances for vertical impact are specified as fol-lows: For traveling cab operated cranes not less than 25% ofmaximum crane wheel loads. The AISC Specifications fur-ther indicate that for pendant operated traveling crane supportgirders and their connections this load may be reduced to 10%of maximum crane wheel loads. The AISE Technical ReportNo. 13 document requires the use of 20% of the maximum

crane wheel loads for motor room maintenance cranes, withadditional requirements for other cranes.

The AISE Report requires impact to be considered incrane columns when one crane is the governing criterion. TheAISC specification does not require this. In all cases, impactloading should be considered in the design of column brack-ets.

13.2 Side Thrust

Horizontal forces exist in crane loadings due to a numberof factors including:

1. Runway misalignment

2. Crane skew

3. Trolley acceleration

4. Trolley braking

5. Crane steering

AISC

The total lateral force on crane runways shall be not lessthan 20% of the sum of the weights of the lifted load and cranetrolley. The force shall be applied to the top of the rail andnormal to the rail direction and shall be distributed with dueregard to the lateral stiffness of the structure supporting thecrane rails.

AISE

The AISE document requires that "The recommended to-tal side thrust shall be distributed with due regard for the lat-eral stiffness of the structures supporting the rails and shall bethe greatest of:

(1) That specified in Table 1 [Shown here as Table13.2.1].

(2) 20% of the combined weight of the lifted load, trol-ley and other lifting devices (i.e. spreader beam,hook, block, rotating mechanism etc.) For stackercranes this factor shall be 40% of the combinedweight of the lifted load, trolley, rigid arm and mate-rial handling device.

(3) 10% of the combined total weight of the lifted loadand the crane weight. For stacker cranes this factorshall be 15% of the combined total weight of thelifted load and the crane weight."

AISE requires that radio-operated cranes be consideredas cab-operated cranes with regard to side thrusts.

44


Administrator

Rectangle

13.3 Longitudinal or Tractive Force

AISC

The longitudinal force, unless specified otherwise, shallbe taken as not less than 10% of the maximum wheel loads ofthe crane applied at the top of the rail.

AISE

The tractive force shall be 20% of the maximum load ondriving wheels.

Table 13.3.1 is provided to illustrate the variation be-tween the AISC Specification and AISE Technical ReportNo. 13 for a particular crane size.

Crane Type

Mill craneLadle cranesClamshell bucket andmagnet cranes(including slab andbillet yard cranes)Soaking pit cranesStripping cranesMotor roommaintenance cranes,etc.Stacker cranes (cab-operated)

Total side thrustpercent of lifted load

4040

100

10010030

200

ingot and mold

Table 13.2.1 AISE Crane Side Thrusts

SIDE THRUST COMPARISON FOR VARIOUS SPECIFICATIONS

(100T MILL CRANE) (TROLLEY WT=60000#, ENTIRE WT=157200#)

SPECIFICATION

AISC

AISE

EQUATION

(Trolley Wt + Lifted Load)

(1) 0.10 (Trolley Wt + Lifted Load)

(2) 0.05 (Entire Crane Wt + Lifted Load)

(3) 0.20 (Lifted Load)

FORCE TO ONE SIDE

26.00 kips

26.00 kips

17.86 kips

40.00 kips

Table 13.3.1 AISC/AISE Side Thrust Comparison

13.4 Crane Stop Forces

The magnitude of the bumper force is dependent on theenergy absorbing device used in the crane bumper. The de-vice may be linear such as a coil spring or nonlinear such ashydraulic bumpers. See Section 18.6 for additional informa-tion on the design of the runway stop.

The crane stop, crane bracing, and all members and theirconnections that transfer the bumper force to the ground,should be designed for the bumper force. It is recommendedthat the designer indicate on the structural drawings the mag-nitude of the bumper force assumed in the design. The bump-

er force is generally specified by the owner or crane supplier.If no information can be provided at the time of design Sec-tion 6.6 of the MBMA Manual(24) can provide some guidance.

13.5 Eccentricities

The bending of the column due to eccentricity of thecrane girder on the column seat must be investigated. Thecritical bending for this case may occur when the crane is notcentered over the column but located just to one side as illus-trated in Figure 13.5.1. Additional consideration for other ec-centricities is discussed in Sections 17.2 and 18.2.

45


Administrator

Rectangle

Fig. 13.5.1 Possible Critical Crane Location

13.6 Seismic Loads

Although cranes do not induce seismic loads to a struc-ture, the crane weight should be considered in seismic loaddetermination. The designer should carefully evaluate thenumber of cranes to be considered in the seismic force deter-mination and their location within the building.

13.7 Load Combinations

In addition to local building codes, the owner may re-quire conformance with AISC or AISE rules. However, in theabsence of such rules, the designer should consider the usageof the structure in determining the criteria for the design.Building codes generally may not contain information on howto combine the various crane loads; i.e. which crane loads, andhow many cranes should be considered loaded at one time, butgenerally they do address how crane loadings should be com-bined with wind, snow, live, seismic, and other loads.

For one crane, each span must be designed for the mostsevere loading with the crane in the worst position for eachelement that is affected. As mentioned, when more than onecrane is involved in making a lift, most codes are silent on adefined procedure. Engineering judgment on the specific ap-plication must be used.

AISE Technical Report No. 13 includes the followingprovisions for the design of members subject to multiplecrane lifts. These provisions are to be used in the design of thesupporting elements.

The design of members (and/or frames), connection ma-terial and fasteners shall be based on whichever one of thethree cases listed hereinafter may govern. Moments andshears for each type loading shall be listed separately (i.e.,dead load, live load, crane eccentricity, crane thrust, wind,etc.). The permissible stress range under repeated loads shallbe based on fatigue considerations with the estimated number

of load repetitions in accordance with the Building Classifica-tion. The owner shall designate an increase in the estimatednumber of load repetitions for any portion of the buildingstructure for which the projected work load or possiblechange in building usage warrants.

Case 1. This case applies to load combinations for mem-bers designed for repeated loads. The stress range shall bebased on one crane (in one aisle only - where aisle representsthe zone of travel of a crane parallel to its runway beams) in-cluding full vertical impact, eccentric effects and 50% of theside thrust. The number of load repetitions used as a basis forthe design shall be 500,000 to 2,000,000 (AISC Loading Con-dition 3) or over 2,000,000 (AISC Loading Condition 4), asdetermined by the owner, for Class A construction. Class Band Class C constructions shall be designed for 100,000 to500,000 (AISC Load Condition 2) and 20,000 to 100,000(AISC Loading Condition 1) respectively. This case does notapply to Class D buildings. Permissible stress range shall bein accordance with the AISC recommendations (AISC Ap-pendix K).

Case 2. All dead and live loads, including roof live loads,plus maximum side thrust of one crane or more than one craneif specific conditions warrant, longitudinal traction from onecrane, plus all eccentric effects and one of the following verti-cal crane loadings:

1. Vertical load from one crane including full impact.

2. Vertical load induced by as many cranes as may bepositioned to affect the member under considera-tion, not including impact.

Full allowable stresses may be used with no reduction forfatigue. This case applies to all classes of building construc-tion.

Case 3. All dead and live loads including impact fromone crane plus one of the following:

1. Full wind with no side thrust but with one crane po-sitioned for maximum vertical load effects.

2. Fifty percent of full wind load with maximum sidethrust and vertical load effects from one crane.

3. Full wind with no live load or crane load.

4. Bumper impact at end of runway from one crane.

5. Seismic effects resulting from dead loads of allcranes parked in each aisle positioned for maximumseismic effects.

For Case 3 permissible stresses may be increased 33-1/3percent. This case applies to all classes of building construc-tion.

46


Administrator

Rectangle

Because the standard AISE building classifications werebased upon the most frequently encountered situations, theyshould be used with engineering judgment. The engineer, inconsultation with the owner, should establish the specific cri-teria. For example, other load combinations that have beenused by engineers include:

1. A maximum of two cranes coupled together withmaximum wheel loads, 50% of the specified sidethrust from each crane, and 90% of the specifiedtraction. No vertical impact.

2. One crane in the aisle and one in an adjacent aislewith maximum wheel loads, specified vertical im-pact, and with 50% combined specified side thrustand specified traction from each crane.

3. A maximum of two cranes in one aisle and one ortwo cranes in an adjacent aisle with maximumwheel loads, and 50% of the specified side thrust ofthe cranes in the aisle producing the maximum sidethrust, with no side thrust from cranes in the adjacentaisle. No vertical impact or traction.

Additional information relative to loading combinationsare contained in the MBMA Low Rise Building SystemsManual. The crane combinations contained in the Manualagree very closely with the AISE combinations.

14. ROOF SYSTEMS

The inclusion of cranes in an industrial building will gen-erally not affect the basic roof covering system. Crane build-ings will "move" and any aspect in the roof system that mightbe affected by such a movement must be carefully evaluated.This generally means close examination of details (e.g. flash-ings, joints, etc.).

A significant difference in the roof support system de-sign for crane buildings as opposed to industrial buildingswithout cranes is that (except for very light duty-lightweightcranes) the use of diaphragm roof bracing is not recom-mended. Whereas wind loads apply rather uniformly distrib-uted forces to the diaphragm, cranes forces are localized andcause concentrated repetitive forces to be transferred from theframe to the diaphragm. These concentrated loads combinedwith the cyclical nature of the crane loadings (fatigue) make itinadvisable to rely upon diaphragm bracing.

15. WALL SYSTEMS

The special consideration which must be given to wallsystems of crane buildings relates to movement and vibration.Columns are commonly tied to the wall system - to provide

bracing to the column or to have the column support the wall.(The latter is applicable only to lightly loaded columns.) Formasonry and concrete wall systems it is essential that properdetailing be used to tie the column to the wall. Figure 15.1 il-lustrates a detail which has worked well for masonry walls.

Fig. 15.1 Masonry Wall Anchorage

The bent anchor rod has flexibility to permit movement per-pendicular to the wall but are "stiff parallel to the wall, ena-bling the wall to brace the column about its weak axis. Theuse of the wall as a lateral bracing system for columns shouldbe avoided if future expansion is anticipated.

If a rigid connection is made between column and walland crane movements and vibrations are not accounted for,wall distress is inevitable.

16. FRAMING SYSTEMS

The same general comments given previously for indus-trial buildings without cranes apply to crane buildings as well.However, the most economical framing schemes are nor-mally dictated by the crane. Optimum bays are usuallysmaller for crane buildings than buildings without cranes andusually fall into the 25 to 30 foot range. This bay size permitsthe use of rolled shapes as crane runways for lower load cranesizes. 50 to 60 foot main bays, with wind columns, are gener-ally more economical when deep foundations and heavycranes are specified.

The design of framing in crane buildings must includecertain serviceability considerations which are used to controlrelative and absolute lateral movements of the runways bycontrolling the frame and bracing stiffness. The source pro-ducing lateral movement is either an external lateral load(wind or earthquake) or the lateral load induced by the opera-tion of the crane. The criteria are different for pendant oper-ated versus cab operated cranes since the operator rides withthe crane in the latter case. In crane bays with gabled roofs,vertical roof load can cause spreading of the eaves and thusspreading of the crane runways. Conversely, eccentricallybracketed runways on building columns can result in inwardtilting of the columns due to the crane loading. This would

47


Administrator

Rectangle

cause an inward movement of the runways toward each other.Lastly, the crane tractive force can cause longitudinal move-ment of the runway either by torsion in the supporting col-umns where brackets are used or flexing of the frame if rigidframe bents are used for the runway columns. Longitudinalrunway movement is rarely a problem where braced framesare used.

Recommended serviceability limits for frames support-ing cranes:

1. Pendant operated cranes: Frame drift to be less thanrunway height over 100, based on 10 year winds orthe crane lateral loads on the bare frame. While thislimit has produced satisfactory behavior, the rangeof movements should be presented to the buildingowner for review because they may be perceived astoo large in the completed building.

2. Cab operated cranes: Frame drift to be less than run-way height over 240 and less than 2 inches, based on10 year wind or the crane lateral loads on the bareframe.

3. All top running cranes: Relative inward movementof runways toward each other to be less than a 1/2inch shortening of the runway to runway dimension.This displacement would be due to crane verticalstatic load.

4. All top running cranes: Relative outward movementof runways away from each other to be not more thanan increase of 1 inch in the dimension between cranerunways. The loading inducing this displacementwould vary depending on the building location. Inareas of roof snow load less than 13 psf, no snowload need be taken for this serviceability check. Inareas of roof snow load between 13 psf and 31 psf,fifty percent of the roof snow load should be used.Lastly, in areas of where the snow loads exceed 31psf, seventy-five percent of the roof snow loadshould be used.

(The discussion of serviceability limits is also presentedin more detail in AISC Steel Design Guide No. 3.)(17)

17. BRACING SYSTEMS

17.1 Roof Bracing

Roof bracing is very important in the design of cranebuildings. The roof bracing allows the lateral crane forces tobe shared by adjacent bents. This sharing of lateral load re-duces the column moments in the loaded bents. This is truefor all framing schemes (i.e. rigid frames of shapes, plates,

trusses, or braced frames). It should be noted, however, thatin the case of rigid frame structures the moments in the framecannot be reduced to less than the wind induced moments.

Figures 17.1.1, 17.1.2 and 17.1.3 graphically illustratethe concept of using roof bracing to induce sharing of lateralcrane loads in the columns. For wind loading all frames andcolumns are displaced uniformly as shown in Fig. 17.1.1. Fora crane building without roof bracing the lateral crane loadsare transmitted to one frame line (Fig. 17.1.2) causing signifi-cant differential displacement between frames. The additionof roof bracing will create load sharing. Columns adjacent tothe loaded frame will share in the load thus reducing differen-tial and overall displacement. (Fig. 17.1.3).

Fig. 17.1.1 Uniform Displacement Due to Wind

Fig. 17.1.2 Displacement of Unbraced FramesDue to Crane Lateral Load

48


Administrator

Rectangle

Fig. 17.1.3 Displacement of Braced FramesDue to Crane Lateral Load

Angles or tees will normally provide the required stiff-ness for this system.

Additional information on load sharing is contained inSection 20.1.

17.2 Wall Bracing

It is important to trace the longitudinal crane forcesthrough the structure in order to insure proper wall and cranebracing (wall bracing for wind and crane bracing may in factbe the same braces).

For lightly loaded cranes, wind bracing in the plane of thewall may be adequate for resisting longitudinal crane forces.(See Fig. 17.2.1) While for very large longitudinal forces, thebracing will most likely be required to be located in the planeof the crane rails. (See Fig. 17.2.2.)

For the bracing arrangement shown in Fig. 17.2.1, thecrane longitudinal force line is eccentric to the plane of the X-bracing. The crane column may tend to twist if the horizontaltruss is not provided. Such twisting will induce additionalstresses in the column. The designer should calculate thestresses due to the effects of the twisting and add thesestresses to the column axial and flexural stresses. A torsionalanalysis can be made to determine the stresses caused bytwist, or as a conservative approximation the stresses can bedetermined by assuming that the twist is resolved into a forcecouple in the column flanges as shown in Fig. 17.2.3. Thebending stresses in the flanges can be calculated from theflange forces. In order to transfer the twist, Pe, into the twoflanges, stiffeners may be required at the location of the forceP.

Fig. 17.2.1 Wall Bracing for Cranes

Fig. 17.2.2 Vertical Bracing for Heavy Cranes

49


Administrator

Rectangle

Fig. 17.2.3 Eccentric Column Forces

The following criteria will normally define the bracingrequirements for longitudinal crane force transfer:

1. For small longitudinal loads (up to 4 kips) use ofwind bracing is generally efficient, where columnsare designed for the induced eccentric load.

2. For medium longitudinal loads (4 kips - 8 kips) ahorizontal truss is usually required to transfer theforce to the plane of X-bracing.

3. For large longitudinal loads (more than 8 kips) brac-ing in the plane of the longitudinal force is generallythe most effective method of bracing. Separate windX-bracing on braced frames may be required due toeccentricities.

Normally the X-bracing schemes resisting these hori-zontal crane forces are best provided by angles or tees ratherthan rods. In cases where aisles must remain open, portal typebracing may be required in lieu of designing the column forweak axis bending. (See Fig. 17.2.4.)

It should be noted that portal bracing will necessitate aspecial design for the horizontal (girder) member, and that thediagonals will take a large percentage of the vertical craneforces. This system should only be used for lightly loaded,low fatigue situations. The system shown in Fig. 17.2.5 couldbe used as an alternate to 17.2.4. Shown in Fig. 17.2.6 are pos-sible details for bracing connections at the crane columns.Additional details on connections and bracing can be found inAISC Engineering for Steel Construction(13).

Fig. 17.2.5 Modified Portal Crane Runway Bracing

Fig. 17.2.6 Column Brace Details for Fig. 17.2.2

50

Fig. 17.2.4 Portal Crane Runway Bracing


Administrator

Rectangle

18. CRANE RUNWAY DESIGN

Strength considerations for crane girder design are pri-marily controlled by fatigue for Class D, E and F cranes.Wheel loads, their spacing, and girder span are required forthe design of crane girders. The expense of crane girder con-struction normally increases when built-up shapes are re-quired. Fatigue restrictions are more severe for built-upshapes (i.e. for built-up sections the difference between arolled shape vs. a built-up member using continuous filletwelds means a shift from AISC category A to B). This meansthat for a frequently cycled built-up girder (Loading Condi-tion 4) the allowable stress range is reduced from 24 ksi to 16ksi.

The following summary of crane girder selection criteriamay prove helpful.

1. Light cranes and short spans - use a wide flangebeam.

2. Medium cranes and moderate spans use a wideflange beam, and reinforce the top flange with achannel.

3. Heavy cranes and longer spans - use a plate girder,with a horizontal truss or solid plate at the top flange.

4. Limit deflections under crane loads as follows:Vertical Deflections of the Crane Beam due to

wheel loads (no impact):L/600 Light and Medium Cranes (CMAAClasses A, B, C and D.)L/1000 Mill Cranes (CMAA Classes E and F.)Lateral Deflection of the Crane Beam due to

crane lateral loads:L/400 All Cranes.

18.1 Crane Runway Beam Design Procedure

As previously explained, crane runway beams are sub-jected to both vertical and horizontal forces from the sup-ported crane system. Consequently, crane runway beamsmust be designed for combined bending about both the X andY axis.

Salmon(35) and Gaylord(19) point out that the equationspresented in the AISC Specifications for lateral-torsionalbuckling strength are based upon the load being applied at theelevation of the neutral axis of the beam. If the load is appliedabove the neutral axis (for instance, at the top flange of thebeam as is the case with crane runway beams), lateral tor-sional buckling resistance is reduced.. In addition, the lateralloads from the crane system are applied at the top flange level,generating a twisting moment on the beam. When vertical

and lateral loads are applied simultaneously, these two effectsare cumulative. To compensate for this, it is common practiceto assume the lateral loads due to the twisting moment are re-sisted by only the top flange. With this assumption, Salmonand Gaylord both suggest that the lateral stability of a beam ofthis type subject to biaxial bending is otherwise typically notaffected by the weak axis bending moment Conse-quently the appropriate allowable bending stress for com-bined bending is based on a yield criterion and is equal to

for the unbraced section.

Another criterion related to crane runway beam design isreferred to in the AISC Specifications as "sidesway webbuckling" (Section K1.5). This criterion is included to pre-vent buckling in the tension flange of a beam where flangesare not restrained by bracing or stiffeners and are subject toconcentrated loads. This failure mode may predominatewhen the compression flange is braced at closer intervals thanthe tension flange or when a monosymmetric section is usedwith the compression flange larger than the tension flange(e.g. wide flange beam with a cap channel). A maximum al-lowable concentrated load is used as the limiting criteria forthis buckling mode.

This criteria does not currently address beams subjectedto simultaneously applied multiple wheel loads.

For crane runway beams the following design procedureis recommended as both safe and reasonable where fatigue isnot a factor.

1. Compute the required moments of inertiato satisfy deflection control criteria.

L/600 to L/1000 for Vertical Deflection.L/400 for Lateral Deflection.

2. Position the crane to produce worst loading condi-tions. This can be accomplished using the equationsfound in the AISC Manual for cranes with two wheelend trucks on simple spans. For other wheel ar-rangements the maximum moment can be obtainedby locating the wheels so that the center of the spanis midway between the resultant of the loads and thenearest wheel to the resultant. The maximum mo-ment will occur at the wheel nearest to the Centerlineof the span. For continuous spans the maximum mo-ment determination is a trial and error procedure.Use of a computer for this process is recommended.

3. Calculate Bending Moments includingeffects of impact.

4. Select a section ignoring lateral load effectsfrom:

51


Administrator

Rectangle

where is obtained from AISC Equations in Chap-ter F.

5. Check this section by using:

= Section modulus of top half of section about y-axis. For rolled beams without channel caps,should be taken as 1/2 of the total of theshape, since only the top flange resists the lat-eral crane loads. For sections with channelcaps, is the section modulus of the channeland top flange area. Values of this parameterare provided in Appendix A, Table 1, for vari-ous W and C combinations. Table 1 also listsvalues for refer tobottom and top flange section moduli respec-tively, is the distance from the bottom flangeto the section centroid. Table 1 also gives themoment of inertia of the "top flange" of thecombined W and C sections.

6. Check the section with respect to sidesway webbuckling as described in Section K1.5 of the AISCSpecifications.

In selecting a trial rolled shape section, it may be helpfulto recognize that the following ratios exist for various Wshapes without channel caps:

W ShapeW8 through W16 3 to 8W16 through W24 5 to 10W24 through W36 7 to 12

Table 2 in Appendix A provides the radius of gyrationand for commonly used channel and wide flange combi-nations. In addition, for these combinations, the maximumspan (unbraced length) for which the allowable bending stresscan be taken as 0.6 is listed.

Where fatigue is a consideration, the above procedureshould be altered so that the live load stress range for the criti-cal case does not exceed fatigue allowables as per AISC Ap-pendix K.

An often overlooked aspect of crane girder design is thelocal longitudinal bending stresses in the top flange of the run-way girder due to the passage of the crane wheel. The fact thatsuch localized stresses exist was first determined from straingage measurements which showed that top flange stresseswere higher than calculated from conventional bending the-ory.

52

AISE Technical Report No. 13 provides the followingequation for use in calculating this localized stress:

EXAMPLE 18.1.2: Crane Runway Girder Design (ASD)

Crane Capacity = 20 Tons (40 kips)

Bridge Span = 70 Ft.

Type of Control - Cab Operated

Bridge Weight = 57.2 kips

Trolley Weight = 10.6 kips

Maximum Wheel Load (without impact) = 38.1 kips

where:

local longitudinal flange stress due tobending under wheel load, ksimaximum wheel load, kipsthickness of top flange, in.moment of inertia of rail cross-section, in.4

moment of inertia of top flange, in.4

clear depth between flangest, in.web thickness, in.

This stress is added to the normal bending stress in theflange. Thus, the stress on the top of the top flange will be in-creased by the value of and the stress on the bottom of thetop flange will be decreased by the value of Experiencehas shown the value of to be in the range of 1.0 to 4.0 ksi,thus, an increase in top flange compression can be anticipated.

EXAMPLE 18.1.1: Local Wheel Support Stresses.

Assume the following:

1. Maximum wheel load = 65 kips

2. W36x300 crane girder.

3. ASCE 85# rail.

Rev.3/1/03


f

h

t

Administrator

Rectangle

Wheel Spacing = 12'-0"

Runway Girder Span = 30' - 0"

Assume no reduction in allowable stress due to fatigue.

Use AISC criteria and A36 steel.

The critical wheel locations with regard to bending momentare:

Assume the girder weight =125 plf

(including impact)

Assuming

Using Tables 1 and 2 from Appendix A, try a W27x94 with aC15x33.9 cap channel.

for top flange and cap channel

Check bending about the x-axis:

From Table 2 of Appendix A,Span = 30' span limit therefore:

Check biaxial bending in the top flange.

= Maximum combined bending stress

Check sidesway web buckling.

Using Equation K1-7 from the AISC ASD Steel Specifica-tion,

53

The critical wheel locations with regard to deflection are:

At working loads,

Max. Vertical Load/Wheel

Max. Horizontal Load/Wheel

Using the vertical deflection criterion of L/600,

Using horizontal deflection criterion of L/400,


Administrator

Rectangle

= 31.6 kips.

Maximum wheel load with impact = 38.1 (1.25)

= 47.6 kips

47.6 > 31.6 kips N.G.

Section does not meet sidesway web buckling criteria as de-scribed in the AISC ASD Steel Specification. A wide flange,section with a larger flange width would be required. Calcu-lations show a W24x104 w/ MC18x42.7 (A36) or a planeW27x161 (A36) beam to be adequate. See the comments onsidesway web buckling at the end of Example 18.1.3.

Use a W24x104 w/MC18x42.7 (A361 or a W27x161 (A361.

EXAMPLE 18.1.3: Crane Runway Girder Design(LRFD)

Same criteria as used in Example 18.1.2 but design check isper the AISC LRFD Specification.

Calculate factored loads.

The load factors that are currently proposed for crane loadsare as follows:

Bridge weight: Load Factor = 1.2.

Trolley weight and lifted load: Load Factor = 1.6.

For the crane used in this example, the factored wheel loadsare calculated as follows:

= 4.05 kips/wheel.

The deflection criteria is based on working loads and there-fore is the same as calculated for Example 18.1.2.

54

Assuming the girder weight to be 125 plf, the factored mo-ments including impact are calculated as:

Investigate the same W27x94 w/C15x33.9 section reviewedin the ASD solution.

The calculation of the warping constant for a beam with acap channel is an involved problem. Therefore, rather thansolving for directly using Eqn. F1-6, is determined by atrial and error iteration using the following equation forfor singly symmetric members as provided in Table A-F1.1of the LRFD Specification.

Check bending about the x-axis.

For vertical loads,

For horizontal loads,

The iterative solution for involves assuming a value forin the above equation, calculating and then comparing

Rev.3/1/03


2.25[2(Iyc/Iy)-1]

Administrator

Rectangle

this calculated value of to as defined in the Specifica-tion. This process is iterated until is equal to Thevalue of that provides this equivalence is equal to

For this shape, the pertinent geometric properties are as fol-lows:

Substituting and solving:

= 7990 in.-kips.

Defining as the section modulus about the bottom (ten-sion) flange and as the section modulus about the top(compression) flange.

55

Check biaxial bending in the top flange.

As previously discussed, a stability or buckling check on thesection for this load case is not required. Since this problem isreally a case of combined bending and torsion on the compos-ite section, Equation H2-1 of the Specification can providethe appropriate check for this load case. This equation corre-lates well with the design check used in the ASD solution.

Equation H2-1 states that:

However, according to the Specification, "if the concentratedload is located at a point where the web flexural stress due tofactored load is below yielding, 24,000 may be used in lieu of12,000" in this equation. The previously discussed biaxialbending criteria will ensure that the web flexural stress be-neath the load is less than Therefore, is evaluated as:

From table values in the LRFD Manual of Steel Construction,

Since the channel cap is welded to top flange, use

For the second iteration, try

For this case,

Therefore,

Therefore,

Check sidesway web buckling.

Using Equation K1-7 from the AISC LRFD Steel Specifica-tion,

Rev.3/1/03


8493 in.-kips.

Administrator

Rectangle

Maximum factored wheel load w/ impact =55.2(1.25)

=69.0 kips.

69.0 < 94.8 o.k.

W27x94 w/C15x33.9 (A36) is adequate based on LRFDcheck.

It should be noted that the ASD Specification currently doesnot have a comparable increase in allowable concentratedload for the sidesway web buckling check when flexuralstress in the web is less than 0.6 Therefore, there is an in-consistency in the two Specifications (ASD and LRFD) withthe ASD Specification providing more conservative criteria.This explains why the W27x94 w/C15x33.9 section was ade-quate for the LRFD check but inadequate for the ASD check.Although it is generally not recommended that ASD andLRFD design criteria be mixed, since sidesway web bucklingis an independent failure mode, it seems reasonable that cranerunways designed using ASD procedures can be checked us-ing LRFD equations for sidesway web buckling.

18.2 Plate Girders

Plate girder runways can be designed in the same manneras rolled sections, but the following items become more im-portant to the design.

1. Plate girder runways are normally used in mill build-ings where many cycles of load occur. Since theyare built-up sections, fatigue considerations are ex-tremely important.

2. Welding stiffeners to the girder webs may produce afatigue condition which would require reduction instress range.(31) Thickening the girder web so thatstiffeners are not required (except for the bearingstiffeners which are located at points of low flexuralstress) may provide a more economical solution.However, in recent years, numerous cases of fatiguecracks at the junction of the top flange of the girderand the web have been noted. These cracks havebeen due to:

a. The rotation of the top flange when the cranerail was not directly centered over the web. (SeeFig. 18.2.1)

b. The presence of residual stresses from thewelding of the flange and stiffeners to the web.

c. Localized stresses under the concentratedwheel loads.

The presence or absence of stiffeners affects problems a.and c. If intermediate stiffeners are eliminated or reduced, theproblem of eccentric crane rail location becomes more severe.If intermediate stiffeners are provided, full penetration welds

should be used to connect the top of the stiffener to the under-side of the top flange. At the tension flange the stiffenersshould be terminated not closer than 4 times nor more than 6times the web thickness from the toe of the web-to-flangeweld.

Shown in Figs. 18.2.2 through 18.2.7 are details whichpertain to heavy crane runway installations. The difference inweld and stiffener detailing between older AISC publicationsand the stiffeners shown here are generally the result of re-vised detailing techniques for fatigue conditions.

Fig. 18.2.1 Rail Misalignment

Fig. 18.2.2 Girder Splice

56


Administrator

Rectangle

Fig. 18.2.3 Crane Runway Girder Detail

57


Administrator

Rectangle

Fig. 18.2.4 Detail at Ends of Crane Girders

Fig. 18.2.5 Sections A and C of Fig. 18.2.4

58


Administrator

Rectangle

Fig. 18.2.6 Section at Different Depth Crane Girders

59


Administrator

Rectangle

Fig. 18.2.7 Typical Crane Stop

18.3 Simple Span vs. Continuous Runways

The decision as to whether simple span or continuouscrane girders should be used has been debated for years. Fol-lowing is a brief list of advantages of each system. It is clearthat each can have an application.

1. Advantages of Simple Span:

a. Much easier to design for various combinationsof loads.

b. Generally unaffected by differential settlementof the supports.

c. More easily replaced if damaged.

d. More easily reinforced if the crane capacity isincreased.

2. Advantages of Continuous Girders:

a. Continuity reduces deflections which quiteoften control.

b. End rotations and movements are reduced.

60


Administrator

Rectangle

c. Result in lighter weight shapes and a savings insteel cost when fatigue considerations are not adetermining factor.

Continuous girders should not be used if differential set-tlement of the supports is of the magnitude that could causedamage to the continuous members. Also, when continuousgirders are subjected to fatigue loading and have welded at-tachments on the top flange (rail clips) the allowable stressrange is reduced considerably. Any advantage therefore maybe eliminated.

Shown in Fig. 18.3.1 are the results of several runway de-signs for spans from 20 to 30 feet. A36 and 50 ksi steel de-signs were made for a 4 wheel, 10T crane, with a 70' bridgefor continuous (two span) vs. simple span conditions. In theseexamples, deflection did not control. Fatigue was not consid-ered. The curves represent (in general) the trends for heaviercranes as well. In general, the use of two span continuouscrane girders could save about 18% in weight over simplysupported girders.

Fig. 18.3.1 Weights of Runways

18.4 Channel Caps

Use of channel caps is normally required to control lat-eral deflections and to control the stresses due to lateral loads.For light duty-lightweight cranes (less than 5T) channel capsmay not be required. Studies have found that a steel savingsof approximately 15 lbs/ft, is required to justify the cost ofwelding a cap to a structural shape.

18.5 Runway Bracing Concepts

An excellent paper on the subject of bracing of cranegirders is that of Mueller.(26) Several significant (and com-mon) considerations that need to be emphasized are:

1. As illustrated in Figure 2 in the Mueller paper (re-peated here as Fig. 18.5.1), improper detailing at theend bearing condition could lead to a web tear in theend of the crane girder. The detail shown in Figure18.5.2 has been used to eliminate this problem forlight crane systems. The details shown in Fig. 18.2.3and 18.2.4 would represent a similar detail for heavycranes. Use of this detail allows the end rotation andyet properly transfers the required lateral forces intothe column.

Fig. 18.5.1 Improper Girder Connection Detail

2. A common method of bracing the crane girder is toprovide a horizontal truss (lacing) or a horizontalplate to connect the crane girder top flange to an ad-jacent structural member as previously illustrated inFigures 17.2.1 and 17.2.2.

A critical consideration in the use of this system is tohave the lacing flexible in the vertical direction, ena-bling the crane girder to freely deflect relative to thestructural member to which it is attached. If the lac-ing is not flexible, stresses will be produced whichcould cause a fatigue failure of the lacing system,thereby losing the lateral support for the girder.

3. AISE Technical Report No. 13 requires that girdersmore than 36 feet in length must have the bottomflange braced by a horizontal truss system. Wherecompliance with this AISE Standard is not requiredmany engineers have used a bottom flange channelto brace the flange on long spans (perhaps 40 feet or

61


Administrator

Rectangle

Fig. 18.5.2 Proper Tie Back Detail

more). The origin of this requirement is not obvious;however, it appears that compliance with the AISCsidesway web buckling equations may analyticallysatisfy this requirement.

4. Occasionally two parallel crane girders are con-nected by a top plate to "mutually brace each other".This, of course, results in a very stiff girder in termsof lateral load. Also, the plate can be used as awalkway for maintenance purposes. When tied to-gether the loading and unloading of parallel girderscan cause a fatigue failure of the bracing plate unlessit is properly detailed. The interconnecting platemust be flexible to allow differential deflections be-tween two girders. Also if the horizontal plate isused it is likely that OSHA(30) will consider the platea footwalk, thus OSHA requirements must be met.

18.6 Crane Stops

The end section of a crane runway must be designed for alongitudinal force applied to the crane stops. For spring typebumper blocks the longitudinal crane stop force may be calcu-lated from the following formula.

where

W = total weight of crane exclusive of lifted load.V = specified crane velocity at moment of impact,

fps (required by AISE Technical Report No. 6 tobe 50% of full load rated speed.

= stroke of spring at point where the cranestopping energy is fully absorbed, ft.

F = total longitudinal inertia force acting at theelevation of the center of mass of the bridge andthe trolley. The force on each runway stop is themaximum bumper reaction from the inertia forceacting at such locations,

g = acceleration of gravity, 32.2 fps2

For bumper blocks of wood or rubber (commonly foundin older cranes) the above equation is not directly applicable.Manufacturers literature or experience must be used for suchinstallations. In the absence of specific data, it is recom-mended that the designer assume the bumper force to be thegreater of:

1. Twice the tractive force, or

2. Ten percent of the entire crane weight.

18.7 Crane Rail Attachments

There are four general types of anchoring devices used toattach crane rails to crane runway beams. These types arehook bolts, rail clips, rail clamps and patented clips. Detailsof hook bolts and rail clamps are shown in the AISC manual.

18.7.1 Hook Bolts

Hook bolts provide an adequate means of attachment forlight rails (40 lbs. - 60 lbs.) and light duty cranes (CMAAClasses A, Band C). The advantage of hook bolts are: l)theyare relatively inexpensive, 2) there is no need to provide holesin the runway beam flange and 3) it is easy to install and alignthe rail. They are not recommended for use with heavy dutycycle cranes (CMAA Classes D, E and F) or with heavycranes (greater than 20 ton lifting capacity), because hookbolts are known to loosen and/or elongate. It is generally rec-ommended that hook bolts should not be used in runway sys-tems which are longer than 500 feet because the bolts do notallow for longitudinal movement of the rail. Because hookbolts are known to loosen in certain applications, a program ofperiodic inspection and tightening should be instituted forrunway systems using hook bolts. Designers of hook bolt at-tachments should be aware that some manufacturers supplyhook bolts of smaller than specified diameter by the use of up-set threads.

62


Administrator

Rectangle

18.7.2 Rail Clips

Rail clips are forged or cast devices which are shaped tomatch specific rail profiles. They are usually bolted to therunway girder flange with one bolt or are sometimes welded.Rail clips have been used satisfactorily with all classes ofcranes. However, one drawback is that when a single bolt isused the clip can rotate in response to rail longitudinal move-ment. This clip rotation can cause a camming action thusforcing the rail out of alignment. Because of this limitationrail clips should only be used in crane systems subject to infre-quent use, and for runway systems less than 500 feet in length.

18.7.3 Rail Clamps

Rail clamps are a common method of attachment forheavy duty cycle cranes. Rail clamps are detailed to providetwo types: tight and floating. Each clamp consists of twoplates: an upper clamp plate and a lower filler plate.

The lower plate is flat and roughly matches the height ofthe toe of the rail flange. The upper plate covers the lowerplate and extends over the top of the lower rail flange. In thetight clamp the upper plate is detailed to fit tight to the lowerrail flange top, thus "clamping" it tightly in place when thefasteners are tightened. In the past, the tight clamp had beenillustrated with the filler plates fitted tightly against the railflange toe. This tight fit-up was rarely achieved in practiceand is not considered to be necessary to achieve a tight typeclamp. In the floating type clamp, the pieces are detailed toprovide a clearance both alongside the rail flange toe and be-low the upper plate. The floating type does not in realityclamp the rail but merely holds the rail within the limits of theclamp clearances. High strength bolts are recommended forboth clamp types.

Tight clamps are generally preferred and recommendedby crane manufacturers because they feel that the transverserail movement allowed in the floating type causes acceleratedwear on crane wheels and bearings.

Floating rail clamps may be required by crane runwayand building designers to allow for longitudinal movement ofthe rail thus preventing (or at least reducing) thermal forces inthe rail and runway system.

Because tight clamps prevent longitudinal rail move-ment, they should not be used in runways greater than 500 feetin length. Since floating rail clamps are frequently neededand crane manufacturers' concerns about transverse move-ment are valid, a modified floating clamp is required. In sucha clamp it is necessary to detail the lower plate to a closer tol-erance with respect to the rail flange toe. The gap betweenlower plate edge and flange toe can vary between snug and agap of 1/8". The 1/8" clearance allows a maximum of 1/4"float for the system. This should not be objectionable to crane

manufacturers since this amount of float is within normalCMAA tolerances for crane spans in the range of 50-100 feet,i.e. spans usually encountered in general construction. In or-der to provide field adjustment for variations in the rail width,runway beam alignment, beam sweep and runway bolt holelocation, the lower plate can be punched with its holes off cen-ter so that the plate can be flipped to provide the best fit. Analternative would be to use short slotted or oversize holes. Inthis case one must rely on bolt tightening to clamp the connec-tion so as to prevent filler plate movement.

Rail clamps are generally provided with two bolts perclamp. Two bolts are desirable in that they prevent the cam-ming action described in the section on forged or cast railclips. A two bolted design is especially recommended ifclamps of the longitudinal expansion type described aboveare used. Rails are sometimes installed with pads between therail and the runway beam. When this is done the lateral floatof the rail should not exceed 1/32", to reduce the possibility ofthe pads being worked out from under the rail.

18.7.4 Patented Rail Clips

This fourth type of anchoring device covers various pat-ented devices for crane rail attachment. Each manufacturer'sliterature presents in detail the desirable aspects of the variousdesigns. In general they are easier to install due to theirgreater range of adjustment while providing the proper limita-tions of lateral movement and allowance for longitudinalmovement. Patented rail clips should be considered as a vi-able alternative to conventional hook bolts, clips or clamps.Because of their desirable characteristics patented rail clipscan be used without restriction except as limited by the spe-cific manufacturer's recommendations. Installations usingpatented rail clips sometimes incorporate pads beneath therail. When this is done the lateral float of the rail should belimited as in the case of rail clamps.

18.7.5 Design Of Rail Attachments

The design of rail attachments is largely empirical. Theselection of the size and spacing of attachments is related torail size. This relation is reasonable in that rail size is relatedto load.

With regard to spacing and arrangement of the attach-ment the following recommendations are given. Hookedbolts should be installed in opposing pairs with three to fourinches between the bolts. The hook bolt pairs should not bespaced further than two feet apart. Rail clips and clampsshould be installed in opposing pairs. They should be spacedthree feet apart or less.

In addition to crane rail attachment, other attachments inthe form of clips, brackets, stiffeners, etc. are often attached tothe crane girder. These attachments are often added by plant

63


Administrator

Rectangle

engineering personnel. Welding should only be done after acareful engineering evaluation of its effects. Welding (in-cluding tack welding) can significantly shorten the fatiguelife. Therefore:

1. Never weld crane rail to girder.

2. Clamp, screw or bolt all attachments to crane girdersto avoid fatigue problems.

3. All modifications and repair work must be submit-ted to engineering for review and approval beforework is done.

18.8 Crane Rails and Crane Rail Joints

The selection of rail relates to crane considerations (basi-cally crane weight) and is generally made by the crane manu-facturer. Once this decision is made, the principal considera-tion is how the rail sections are to be joined. There are severalmethods to join rails but two predominate at the present time.

The bolted butt joint is the most commonly used railjoint. Butt joint alignment is created with bolted splice plates.These plates must be properly maintained (bolts kept tight). Ifsplice bars become loose and misaligned joints occur, a num-ber of potentially serious problems can result, including chip-ping of the rail, bolt fatigue, damage to crane wheels, and as aresult of impact loading, increased stresses in the girders.Girder web failures have been observed as a consequence ofthis problem.

The welded butt joint, when properly fabricated to pro-duce full strength, provides an excellent and potentially main-tenance free joint. However, if repairs are necessary to therails, the repair procedure and consequently the down time ofplant operations is generally longer than if bolted splices hadbeen used. The metallurgy of the rails must be checked to as-sure use of proper welding techniques, but if this is accom-plished the advantages can be significant. Principal amongthese is the elimination of joint impact stresses, existent innon welded rail construction, resulting in reduced cranewheel bearing wear.

Rail joints should be staggered so that the joints do notline up on opposite sides of the runway. The amount of stag-ger should not equal the spacing of the crane wheels and in nocase should the stagger be less than one foot.

Rail misalignment is the single most critical aspect of thedevelopment of high impact and lateral stresses in crane gird-ers. Proper use and maintenance of rail attachments is criticalin this regard. Rail attachments must be completely adjust-able and yet be capable of holding the rail in alignment. Be-cause the rails may become misaligned regular maintenanceis essential to correct the problem.

One aspect of crane rail design is the use of crane railpads. These are generally preformed fabric pads that workbest with welded rail joints. The resilient pads will reduce fa-tigue, vibration and noise problems. Reductions in concen-trated compression stresses in the web have been achievedwith the use of these pads. Significant reductions in wear tothe top of the girder flange have also been noted. With the ex-ception of a few patented systems, the pads are generally notcompatible with floating rail installations since they can worktheir way out from under the rail. Also prior to using a padsystem careful consideration to the cost benefits of the systemshould be evaluated.

19. CRANE RUNWAY FABRICATION &ERECTION TOLERANCES

Crane runway fabrication and erection tolerances shouldbe addressed in the project specifications because standardtolerances used in steel frameworks for buildings are not tightenough for buildings with cranes. Also, some of the requiredtolerances are not addressed in standard specifications.

Tolerances for structural shapes and plates are given inthe Standard Mill Practice section of the Manual of Steel Con-struction published by AISC. These tolerances cover the per-missible variations in geometrical properties and are takenfrom ASTM Specifications, AISI Steel Product Manuals andProducer's Catalogs. In addition to these Standards, the fol-lowing should be applied to crane runways.

a. Sweep: not to exceed 1/4 inch in a 50 foot beamlength.

b. Camber: not to vary from the camber given on thedrawing by plus or minus 1/4 inch in a 50 foot beamlength.

c. Squareness: within 18 inches of each girder end theflange shall be free of curvature and normal to thegirder web.

Columns, base plates and foundations should adhere to thefollowing tolerances.

a. Column anchor bolts shall not deviate from theirtheoretical location by 0.4 times the difference be-tween bolt diameter and hole diameter throughwhich the bolt passes.

b. Column base plates: Individual column base platesshall be within ± 1/16 inch of theoretical elevationand be level within ± 0.01 inches across the platelength or width. Paired base plates serving as a basefor double columns shall be at the same level and notvary in height from one to another by 1/16 inch.

Crane runway girders and crane rails shall be fabricated anderected for the following tolerances.

64


Administrator

Rectangle

a. Crane rails shall be centered on the Centerline of therunway girders. The maximum eccentricity of cen-ter of rail to Centerline of girder shall be three-quar-ters of the girder web thickness.

b. Crane rails and runway girders shall be installed tomaintain the following tolerances.

1. The horizontal distance between crane railsshall not exceed the theoretical dimension by ±1/4 inch measured at 68° F.

2. The longitudinal horizontal misalignment fromstraight of rails shall not exceed ± 1/4 inch in 50

feet with a maximum of ± 1/2 inch total devia-tion in the length of the runway.

3. The vertical longitudinal misalignment of cranerails from straight shall not exceed ± 1/4 inch in50 measured at the column centerlines with amaximum of ± 1/2 inch total deviation in thelength of the runway.

The foregoing tolerances were taken from AISE Techni-cal Report No. 13. The following Table shown in Fig. 19.1 istaken from MBMA's Low Rise Building Systems Manualand gives an alternative set of runway erection tolerances.

65


Administrator

Rectangle

Fig. 19.1 MBMA Crane Runway Erection Tolerances

66

Item Tolerance

MaximumRate ofChange

Span

Straightness

Elevation

Beam to BeamTop Running

Beam to BeamUnderhung

Adjacent Beams


Administrator

Rectangle

20. COLUMN DESIGN

No attempt will be made to give a complete treatise onthe design of steel columns. The reader is referred to a num-ber of excellent texts on this subject.(19)(35)

This section of the guide includes a discussion of themanner in which a crane column can be analyzed, how the de-tailing and construction of the building will affect the loadsthe crane column receives, and how shears and moments willbe distributed along its length. The guide also includes a de-tailed example of a crane column to illustrate certain aspectsof the design.

In most crane buildings, the crane columns are structur-ally indeterminate. Normally the column is "restrained" atthe bottom by some degree of base fixity. The degree of re-straint is to a large extent under the control of a designer, whomay require either a fixed base or a pinned base.

It is essential to understand that the proper design ofcrane columns can only be achieved when column momentsare realistically determined. This determination requires acomplete frame analysis in order to obtain reliable results.Even if a complete computer frame analysis is employed, cer-tain assumptions must still be made of the degree of restraintat the bottom of a column and the distribution of lateral loadsin the structure. Further, in many cases a preliminary designof these crane columns must be performed either to obtain ap-proximate sizes for input into a computer analysis or for pre-liminary cost and related feasibility studies. Simplifying as-sumptions are essential to accomplish these objectives.

20.1 Base Fixity and Load Sharing

Crane columns are constructed as bracketed, stepped,laced, or battened columns. (See Figure 20.1.1.) In each case,

Fig. 20.1.1 Column Types

the eccentric crane loads and lateral loads produce momentsin the columns. The distribution of column moments is oneprincipal consideration.

For a given loading condition, the moments in a cranecolumn are dependent on many parameters. Most parameters(e.g. geometry, nonprismatic conditions) are readily accom-

modated in the design process using standard procedures.However, two parameters which have a marked effect on col-umns moments are:

1. Base fixity.

2. Amount of load sharing with adjacent bents.

As an example, refer to Figure 20.1.2. The loading con-sists of 100T crane (vertical crane load = 310 kips, lateralcrane load to each side = 23 kips). A stepped column is used,but the same general principles apply to the other columntypes.

Fig. 20.1.2 Example Frame

1. Base Fixity: The effect of base fixity on column mo-ments was determined by a computer analysis for the framefor fixed and pinned base conditions. The results of the analy-sis shown in Figure 20.1.3 demonstrate that a simple base willresult in extremely large moments in the upper portion of thecolumn and the structure will be much more flexible as com-pared to a fixed base column. For fixed base columns the larg-est moment is carried to the base section of the column whereit can, in the case of the stepped column, be more easily car-ried by the larger section.

It is frequently argued that taking advantage of fullfixity cannot be achieved in any practical detail. However,the crane induced lateral loads on the crane column are ofshort duration, and for such short term loading an "essentiallyfixed" condition can normally be achieved through proper de-

67


Administrator

Rectangle

Fig. 20.1.3 Analysis Results

sign. The reduced column moments (22 percent in the previ-ous example) due to the fixed base condition provide goodeconomy without sacrificing stiffness.

There will be cases where subsoil conditions, existingconstruction restrictions, property line limitations, etc., willpreclude the development of base fixity and the hinged basemust be used in the analysis. Although the fixed base conceptas stated is deemed appropriate due to short term nature ofcrane loadings, for other long duration building loads the as-sumption of full fixity may be inappropriate. The reader is re-ferred to an excellent article by Galambos(18) which deals withthe effects of base fixity on the buckling strength of frames.

2. Load Sharing To Adjacent Bents: If a stiff system ofbracing is used (i.e., a horizontal bracing truss as shown inFigure 20.1.4) then the lateral crane forces and shears can bedistributed to adjacent bents thereby reducing column mo-ments. Note that such bracing does not reduce column mo-ments induced by wind, seismic or roof loads but only the sin-gular effects of crane loads. Figure 20.1.5 depicts the momentdiagram in the column from a frame analysis based on lateralcrane loads being shared by the two adjacent frames (i.e. two-thirds of the lateral sway force is distributed to other frames).The significant reductions in moment are obvious when com-pared to Figure 20.1.3. (Note the "two-thirds" is an arbitrarydistribution used at this point only to illustrate the concept andthe significant advantage to be gained. The following para-graphs describe in detail how load sharing actually occurs andhow it can be evaluated.

68

Fig. 20.1.5 Moment Diagram with Load Sharing

Consider a portion of a roof system consisting of fiveframes braced as shown in Figure 20.1.6. The lateral craneforce will result in a reactive force at the level of the lowerchord of the roof truss. (Figure 20.1.7.) The distribution ofthis reactive force to the adjacent frames can be obtained bystiffness methods. This is accomplished by analyzing thehorizontal bracing system as a truss on a series of elastic sup-ports. The supports are provided by the building frames andhave linear elastic spring constants equal to the reciprocal ofthe displacement of individual frames due to a unit lateral load(Figure 20.1.8). The model is depicted in Figure 20.1.9. Thesprings are imaginary members which provide the same de-flection resistance as the frames.

This procedure has been programmed and analyzed formany typical buildings. It is obvious that the degree of loadsharing varies, and is dependent upon the relative stiffness ofthe bracing to the frames; however, it was found that for usualhorizontal bracing systems a lateral load applied to a single in-terior frame will be shared almost equally by at least 5 frames.This is logical because bracing of reasonable proportions

Fig. 20.1.4 Horizontal Bracing


Administrator

Rectangle

Fig. 20.1.6 Roof Portion

Fig. 20.1.7 Reactive Force

Fig. 20.1.8 Unit Lateral Load

Fig. 20.1.9 Computer Model

made up of axially loaded members is many times as stiff asthe moment frames which are dependent upon the bendingstiffness of their components.

A building supporting a 100 ton crane is used to illus-trate the effect of load sharing. A roof system consisting offive frames X-braced as shown in Figure 20.1.6 was analyzedto determine the force in each frame due to a 20 kip force ap-plied to the center frame. This 20 kips represents the reactiveforce at the elevation of the bottom chord bracing due to ahorizontal crane thrust at the top of the crane girder as illus-trated in Figure 20.1.7. The final distribution is shown in Fig-ure 20.1.10.

Even though reasonable truss type bracing will distrib-ute a concentrated lateral force to at least 5 frames, it is recom-mended that load sharing be limited to 3 frames (the loadedframe plus the frame to either side). The reason for this con-servative recommendation is that unless pretensioned thehorizontal bracing truss members may tend to sag eventhough "draw" is provided. Thus, a certain amount of move-ment may occur before the truss "takes up" and becomes fullyeffective in distributing the load to adjacent frames.

The designer may conclude that if load sharing occursthat a simple method to handle the analysis is to design a givencolumn for one-third the lateral load. This is wrong and un-safe! Each individual crane column must be designed for thefull lateral force of the crane. It is only the reactive force ap-plied at the level of the bracing which is distributed to the ad-jacent frames. The results of this analysis must be added to orcompared with the results of other analyses which are unaf-fected by the load sharing, i.e., gravity, wind, and seismicloadings.

69


Administrator

Rectangle

Fig. 20.1.10 Final Force Distribution

To summarize, the most economical designs will resultwhen the following "assumptions" are designed into thestructure:

1. Fixed base columns.

2. Horizontal bracing truss (unless wind loads control)such that lateral crane loads can be distributed to ad-jacent columns.

3. When the roof frames are fabricated trusses the mosteconomical bracing truss location is at the elevationof the bottom chord where they are generally easierto erect. The bottom chord bracing system that is re-quired for uplift and slenderness ratio control mayalso be adequate for distributing concentrated lateralforces.

20.2 Preliminary Design Methods

Preliminary design procedures for crane columns are es-pecially helpful due to the complexity of design of thesemembers. Even with the widespread availability of comput-ers a good preliminary design can result in substantial gains inoverall efficiency. The preceding sections of this guide havepointed out the fact that in order to obtain meaningful columnmoments a frame analysis is required. A reliable hand calcu-lation method for preliminary design is not only helpful butessential in order to reduce final design calculation time.

The frame analysis which is required to obtain an "ex-act" solution accomplishes the following:

1. It accounts for sidesway.

2. It properly handles the restraint at the top and at thebase of the column.

3. It accounts for non-prismatic member geometry.

What is needed for a preliminary design procedure is amethod of analysis that will provide suitable column stiffnessestimates so that an "exact" indeterminate frame analysis pro-cedure need be conducted only once. The model given belowin Figure 20.2.1 has been found to give remarkably good re-sults for crane loadings, providing horizontal bracing is usedin the final design. It is a "no-sway" model, consisting of afixed base, and supports introduced at the two points wherethe truss chords intersect the column.

Fig. 20.2.1 No-Sway Computer Model

A moment diagram obtained from the "no-sway" modelfor the 100 ton crane column previously shown in Fig. 20.1.2is shown in Figure 20.2.2.

Comparing Figure 20.2.2 to Figure 20.1.5 it can be seenthat the general moment configuration is similar, and themagnitudes of moments are almost identical. For preliminarydesign purposes the two-support "no-sway" model is ade-quately accurate. The two support no-sway model is stati-cally indeterminate to the second degree. Thus, even a pre-liminary design requires a complex analysis and certain otherassumptions.

The preliminary design procedure for wind or seismicloadings can usually be made by assuming an inflection point

70

Rev.3/1/03


Administrator

Rectangle

Fig. 20.2.2 Results of No-Sway Model

and selecting preliminary column size to control sway underwind loads. An appropriate procedure is shown in the brack-eted crane column design example in the next section.

The sizes of bracketed columns are often controlled bywind; therefore, the design should first be made for wind andsubsequently checked for wind plus crane. AISE recom-mends that bracket vertical loads should be limited to 50 kips.

Stepped and laced or battened columns are another mat-ter. To obtain accurate values for moments, the effects of thenonuniform column properties must be included in the analy-sis. In doing a preliminary analysis of a stepped column an-other assumption is practical. The assumption involves thesubstitution of a single top hinge support to replace the twosupports in the two support no-sway mode. The singlehinged support is located at the intersection of the bottomchord and the column.

The simplified structure is depicted in Figure 20.2.3.Equations for the analysis of this member are given in Figure20.2.4.

In each case, the equation for the top shear force isgiven. For the single support assumption the indeterminacy iseliminated once this shear force is known. The moment dia-gram for the single hinge, no-sway column evaluated usingthe equations is given in Figure 20.2.5.

While the variation in moment along the length is not ingood agreement with that of the "exact" solution given in Fig-

Fig. 20.2.3 Simplified Structure

ure 20.1.5, the values and signs of the moments at critical sec-tions agree quite well.

There is one aspect of preliminary design that has notbeen discussed which is essential in the handling of thestepped and double column conditions. The non-prismaticnature of these column types requires input of the moment ofinertia of the upper and lower segments of the column which,of course, are not known initially. Therefore, some guidelinesand/or methods are required to obtain reasonable values forand

20.2.1 Obtaining Trial Moments of Inertia for SteppedColumns:

The upper segment of the stepped column may be sizedby choosing a column section based on the axial load actingon the upper column portion. Use the appropriate unsup-ported length of the column in its weak direction and deter-mine a suitable column from the column tables contained inthe AISC manual. Select a column about three sizes (byweight) larger to account for the bending in the upper shaft.

The size of the lower segment of the stepped columnmay be obtained by assuming that the gravity load from thecrane is a concentric load applied to one flange (or flange-channel combination). The preliminary selection may bemade by choosing a member such that P/A where Ais the area of one flange or flange plus channel combination.The depth of the lower shaft is normally determined by thecrane clearance requirements (see Figure 20.2.6).

71

Rev.3/1/03


Administrator

Rectangle

Fig. 20.2.4 Equations for Simplified Structure

Fig. 20.2.5 Column MomentsUsing Fig. 20.2.4 Equations

Fig. 20.2.6 Column Clearance Requirement

20.2.2 Obtaining Trial Moments of Inertia for DoubleColumns:

The building column portion of a double column canagain best be selected based on the applied axial load. Selectthe size of the crane column based on the crane gravity loadapplied to the "separate" crane column. The allowable stressof this portion will normally be based on the major axis of thecolumn assuming that the column is laced or battened to thebuilding column to provide support about the weak axis. Theactual sizes of the columns should be increased slightly to ac-count for the bending moments. The moment of inertia of thecombined sections can be calculated using standard formulasfor geometrical properties of built-up cross sections. If themoment of inertia of the combined sections is obtained by as-suming composite behavior, the lacing or batten plates con-necting the two column sections must be designed and de-tailed accordingly.

20.3 Final Design Procedures

After obtaining the final forces and moments in thecrane column, it can be designed. The design of a crane col-umn is unique in that the column has both a varying axial loadand a "concentrated" moment at the location of the bracket or"step" in the column.

The best design approach for prismatic bracketed col-umns is to design the upper and lower portions of the columns

72

Rev.3/1/03


Administrator

Rectangle

as individual segments with the top portion designed forand the associated upper column moments, and the lower por-tion designed for , and the lower column moments(Figure 20.3.1). The column can normally be considered to be

Fig. 20.3.1 Column Loads

laterally braced about the y-axis at the crane girder elevation.When considering the x-axis, and K should be calcu-lated based on the entire length of the column and the proper-ties of the cross section. can be assumed to be 0.85 sinceeach column segment is free to sway. A formal theoreticaltreatment of this procedure can be found in The Design ofSteel Beam-Columns, by Peter F. Adams, published by theCanadian Steel Industries Construction Council.(1) The bestreference for the design of crane columns is contained in theAISE Technical Report No. 13, Specifications for the Designand Construction of Mill Buildings. The AISE proceduresuggests that two equations be checked.

These equations are nearly identical with equationsH1-1 and H1-2 of the ASD AISC Specification for memberssubjected to both axial compression and bending stresses, ex-cept for the introduction of as different from In addition,the other terms are in some cases evaluated in a manneradapted especially to the stepped-column problem.

The terms in these equations are defined as follows:

- In the lower shaft where A is thearea of the lower shaft. In the upper shaft.

with A the area of the upper (building)shaft.

- In checking the lower shaft for bending about they-y axis, it is conservatively assumed that thecrane support segment resists all of the bendingintroduced by eccentricity of the crane girder re-actions. The amplifications of as a result ofdeflection is dependent on the average axialstress in the crane segment alone. The stress

is determined by adding (or subtracting) theaverage stress due to moment about the x-axis,calculated at the centroid of the crane segment,to (or from) the average stress of the entirelower shaft.

- The allowable axial stress under axial load. Itmay be determined for buckling of the entirestepped column about the x-x axis, based on theequivalent length , or by buckling about they-y axis for whatever column length is unsup-ported, in either the upper or lower shaft. It is tobe taken as the minimum of the two values ineach of the two sets pertinent to the upper andlower shafts, respectively. Exterior wall girtsare often assumed not to provide longitudinal(lateral) support to the columns in mill buildingsbecause building alterations may result in theirremoval. If support in the x direction is providedonly at locations A, B and C (Figure 20.3.2) theequivalent length KL for buckling about the y-yaxis should be taken as the full unsupportedlength AB in checking the upper shaft. In check-ing the lower shaft for the y-y axis, the equiva-lent length KL should be taken as 0.8 of lengthBC if the base is assumed to be fully fixed, or aslength BC if base fixity against rotation cannotbe assured.

- For bending about the x-x axis, use a value of0.85 when all bents are under simultaneous windload and sidesway is assumed to take place.When one bent is being considered, under maxi-mum crane loading, without wind (AISE Case 2loading) assume a value of 0.95 for

- Since the crane segment of the lower shaft is as-sumed to resist all of the bending about the y-yaxis, this term is applied to the lower shaft isassumed zero in the upper shaft). Assumingfixity at the base but no interaction with thebuilding column, half of the moment introducedat B as a result of unequal reactions from adja-cent girders will be carried down to the base, in

73


Administrator

Rectangle

Fig. 20.3.2 Typical Column

which = 0.4 (AISC Specification SectionH1). If base fixity cannot be assumed, take =0.6 (hinged condition at base), or, in intermedi-ate situations, interpolate between 0.4 and 0.6.

- Maximum stress due to bending about the x-xaxis, assuming an integral action of crane and

74

building column segments in the lower shaft,and the building column alone in the upper shaft.

- Maximum stress due to bending about the y-yaxis in the crane column segment of the lowershaft; usually zero in upper shaft.

- For compression on the crane column side of thelower shaft, is the permissible extreme fiberstress due to bending about the x-x axis, reducedif necessary below because of lack of lat-eral support. The reduced allowable stress maybe based on the permissible axial stress in thecrane column segment for buckling about they-y axis as shown in Figure 20.3.2. (The y-yaxis in this sketch would correspond to the x-xaxis of the individual wide flange segment in theAISC Manual.) The permissible column stress,so determined, should be multiplied by the ratio

as defined by Section B-B in Figure20.3.2. In no case is the allowable stress to begreater than

- Since this component of bending is about theweak axis of the combined crane and buildingcolumns, no reduction in permissible stress needbe made for lateral buckling. Also, since thebending resistance is assumed to be providedsolely by the crane segment of the lower shaft,the allowable stress for a compact section maybe used if the provisions of Section F2 of theAISC Specification are met.

- Since this stress is used as a basis for the determi-nation of the amplification of column deflectionin the plane of bending, it should be based on theequivalent length of the completed stepped col-umn, as in the case of for bending about thex-x axis.

- If the base may be assumed as fixed let K = 0.8for the crane column segment alone; otherwiseassume K = 1.0. The length in the determinationof KL is that of the column segment BC.

Example 20.3.1 will illustrate the procedure.

Contained in the AISE publication are effective lengthvalues for stepped columns, in terms of three parameters: theratio of the length of the reduced section to the total length ofthe column; B, the ratio of the maximum moment-of-inertiaof the combined column cross section to that of the reducedsection; and the ratio of the axial force in the upper seg-ment to the crane force in the lower segment. (See Figure20.3.2.)

The AISE tables do not address column end conditionsother than fixed or hinged and often times the ratios of


Administrator

Rectangle

fall outside the scope of the tables. Contained in Reference 2and reproduced in Appendix B are tables which address sevendifferent column end conditions, these include:

a) Pinned - Pinned

b) Fixed - Free

c) Fixed - Pinned

d) Fixed - Slider

e) Fixed - Fixed

f) Pinned - Fixed

g) Pinned - Slider

In addition, these tables include prismatic and nonpris-matic columns, and virtually all combinations of andload ratios.

EXAMPLE 20.3.1: Bracketed Crane Column Design

Design the column shown in Fig. 20.3.3:

Fig. 20.3.3 Example

Use AISE provisions and A36 steel.

1. Load Cases:

A frame analysis was performed using the AISE loadcombinations. The critical moment diagrams were ob-tained from Load Cases 2 and 3.

Case 2 = DL+LL+Crane (Lateral and Vertical)

Case 3 = (DL+Crane Vertical + Wind) .75

Fig. 20.3.4 Critical Moment Diagrams

Case 2 produces the most critical moment condition inthe lower portion of the column, and Case 3 in the upperportion.

2. Preliminary Design:

Since this structure is quite high it is very likely that lat-eral sway movement could control the column size.Thus, it is recommended that the preliminary design ofthe column be based on deflection considerations.

Base the allowable sway on:

Fig. 20.3.5 Sway Calculation

75


Administrator

Rectangle

Upper Column Check:Fig. 20.3.6 Example

76

Assuming is divided equally between both columns.

3. Stress Check:

The properties of the W16x77 are:

Lower Column Check:

Use the effective length charts contained in the AppendixB to determine Assume the column base is fixed andthe column top is a fixed roller.

From Case 2,

Note that is based on the midheight of the roof truss.

By interpolation from the tables: K2= 0.97

Checking the AISC interaction equations:

From Case 3,Use the effective length charts contained in the AppendixB to determine

By interpolation from the tables:

Checking the AISC interaction equations:

EXAMPLE 20.3.2: Stepped Crane Column Design

Design the column shown in Fig. 20.3.6:

Use a W16x77Deflection Controls.

Rev.3/1/03


24x30000x2.09.0

Administrator

Rectangle

Use AISE provisions and A36 steel.

1. Load Cases:

A frame analysis was performed using the AISEloading combinations. The critical moment diagram wasobtained from Case 2. See Fig. 20.3.7.

Case 2 = DL + LL + Crane (Lateral and Vertical)

Fig. 20.3.7 Critical Moment Diagram

2. Preliminary Design

Use the strength preliminary design procedures dis-cussed in this guide.

For the upper shaft, P=31K.Based on the AISC Manual column tables, try a W12x35section. For the lower shaft the crane load equals 50 kips.Estimate the flange area.

A W24 section is required for crane clearance.Try a W24x62,

As an approximation to the moment of inertia for thestepped column, use a weighted average of the moment of in-ertia for the upper and lower shafts.

Since this average is greater than 984 in.4 from the previ-ous example, the column should satisfy the L/240 deflectionrequirement. After a final stress check the deflection checkcan be verified by computer analysis.

3. Stress Check:

The properties of the W12x35 are:

77

For the W24x62:

Use the effective length charts contained in the AppendixB to determine values. Assume the column base is fixedand the column top is a fixed roller.

By interpolation from the tables:

Lower Column Check:

DetermineFrom the moment diagram for the lower shaft (Fig. 20.3.8):

Fig. 20.3.8 Critical Moment Diagram

AISC Eq. F1-6 and F1-8 apply.


Administrator

Rectangle

Checking the AISE interaction equations:

Repeating the above calculation for a W24x68 section yieldsthat a W24x68 is o.k.

Upper Column Check:

P = 31 kips, M = 53.31 ft-kips.The change in I1/I2 (change due to W24x68) causes aslight change in the effective length of the upper shaft.Iterating from the effective length tables yields K1=0.85

Use a W12x35 with a W24x68.

Fig. 20.4.1 "Clean" Column

2. Separate crane columns are economical for heavycranes. Fabricators favor tying the crane column tothe building column with short W shapes acting as adiaphragm as opposed to a lacing system using an-gles. (See Figure 20.4.2.)Lacing systems are economical as compared to thediaphragm system if the miscellaneous framingpieces are not required. For example, if the buildingcolumn flange width is equal to the crane columndepth, the columns can be laced economically usingfacing angles. (See Figure 20.4.3.)

3. Bracketed columns are generally most efficient upto bracket loads of 25 kips. Crane reactions between25 and 50 kips may best be handled by either abracket column or a stepped column.

4. If the area of one flange of a stepped column multi-plied by .5Fy is less than the crane load on the col-umn, a separate crane column should definitely beconsidered.

20.4 Economic Considerations

Although it is not possible to provide a clear-cut rule ofthumb as to the most economical application of the variouscrane columns, i.e., bracketed, stepped, or separate crane col-umn, due to differences in shop techniques; it is possible how-ever, to generalize to some degree.

1. The stepped column will be economical if "clean".In fact, for many jobs a "clean" stepped column canprove economical as compared to the bracketed col-umn even for light loads. By "clean" is meant thatthe column is fabricated without a face channel orextra welded attachments. (See Figure 20.4.1.)

78


Administrator

Rectangle

Fig. 20.4.3 Laced Column

21. OUTSIDE CRANES

Outside cranes are common in many factories for scraphandling, parts handling and numerous other operations.There are several important aspects of outside crane usagewhich are unique to that type of crane.

1. The exterior exposure in many climates requires thatextra attention be given to painting and generalmaintenance, material thickness, and the elimina-tion of pockets which would collect moisture.

2. Due to drive aisles, railways and other similar re-strictions, exterior cranes often require longer spansthan interior cranes. The outside crane has no build-ing columns from which to derive lateral support.Therefore, long, unbraced spans are more commonto these installations. Horizontal bracing trusses,wide truss columns or other bracing elements mustoften be employed to achieve stability.

3. Long spans may dictate that trusses, rather than plategirders or rolled sections, be used for the runwaybeams. This can have certain advantages includingimproved stiffness. The disadvantages are clearlythe increased depth plus joints which are highly sus-ceptible to fatigue problems. Secondary stressesmust be calculated and included in the fatigue analy-sis for trusses used as crane girders.

4. Another special girder that may be appropriate foruse in these long span applications is the trussedgirder. This "hybrid" involves the coupling of agirder (top flange) and a truss. The member can de-velop excellent stiffness characteristics and manytimes can temporarily support the crane weight evenif a truss member is damaged. As with the basictruss, the overall greater depth is a disadvantage.

5. Still another solution to the long span problem maylie in the use of "box" or "semi-box" girders. An ex-cellent reference on this subject was developed bySchlenker.(36) These girders have excellent lateraland torsional strength. In addition, the problem as-sociated with off center crane rails is eliminated.

6. Brittle fracture should be considered for cranes op-erating in low temperature environments.

22. UNDERHUNG CRANES

Underhung cranes in industrial buildings are very com-mon and quite often prove to be economical for special appli-cations. One of the distinct operational advantages that un-derhung cranes possess is that they can be arranged to providefor trolley transfer from one runway or aisle to another.

79

Fig. 20.4.2 Connections with W Shapes


Administrator

Rectangle

Proper provision in the design must be made for handling lat-eral and impact loads from underhung cranes. The conceptspresented in this guide (e.g. load transfer) are, in general, ap-plicable to underhung crane systems. Since these cranes aregenerally supported by roof members load is not transferreddirectly to the columns and therefore the column design doesnot involve the moment distribution problems of the top run-ning crane column. Pay particular attention to the method ofhanging the cranes. Fatigue problems with these connectionshave existed in the past and proper provisions must be madewith the hanging connection to guarantee adequate servicelife.

Hanger systems should provide for vertical adjustmentin order to properly adjust the elevation of the runway beam.After the runways are positioned vertically, a lateral antiswaybrace should be attached at each hanger location. The swaybrace prevents the hanger system from flexing perpendicularto the runway. Most hanger systems fatigue at a relativelylow stress level if they are allowed to sway. In addition to thelateral antisway braces, longitudinal braces should be in-stalled parallel to the runway beams to prevent sway along thelength of the runway. These braces should be placed at ap-proximately 100 foot intervals and at all turns in the runway.

Runway splices can be accomplished in many ways.The splice should allow for a smooth running crane as thewheels transfer from one beam to the next. A typical splicedetail is shown in Fig. 22.1.

Fig. 22.1 Underhung Crane Beam Splice

Many crane suppliers prefer to supply the runwaybeams. The building designer must carefully coordinatehanger locations and hanger reactions with the crane supplier.Many times the structure must be designed prior to the selec-tion of the crane system. Hanger locations and reactions mustbe estimated by the building designer. Hanger reactions can

80

be calculated from manufacturer's catalogs. Hangers shouldbe provided at a 15 to 20 foot spacing if possible. The deflec-tion limit for underhung crane runway beams due to wheelloads should be limited to span divided by 450.

23. MAINTENANCE AND REPAIR

As stated earlier in this manual, crane buildings requirean extra measure of maintenance. Crane rail alignment is es-pecially critical, wear on crane and rail, and potential fatigueproblems can result. Crane rails must also be inspected foruneven bearing, to minimize fatigue problems.

If fatigue cracks occur and must be repaired, the repairprocedure may create additional problems if proper proce-dures are not taken. Simple welding of doubler plates, stiffen-ers or other reinforcement may create a "notch effect" whichcould be more serious than the original problem. Engineersshould use common sense in detailing procedures for repair offatigue cracks. In particular they should not create a worse fa-tigue problem with the repair. Referral to Appendix K of theAISC Specifications is essential.

24. SUMMARY AND DESIGNPROCEDURES

Many concepts have been presented in this guide rela-tive to the design and analysis of structural frames for cranebuildings. In an effort to optimize design time, the followingprocedural outline has been developed for the designer.

1. Determine the best geometrical layout for the build-ing in question.

2. Design the crane girders and determine column andframe forces from the crane loadings.

3. Perform preliminary design of the crane columns.

4. Design the roof trusses or roof beams for dead loadsand live loads.

5. Determine all loading conditions for which the en-tire frame must be analyzed.

6. Analyze the frame in question for dead, live, windand seismic loadings. This analysis should be per-formed without load sharing from the adjacentframes. Also determine the lateral stiffness of theframe.

7. Analyze the frame (considering load sharing) forcrane loadings.

8. Combine moments and forces from the two analysesfor subsequent design.

9. Perform the final design of columns, trusses, bracesand details.


Administrator

Rectangle

81

ACKNOWLEDGMENTS

The authors wish to thank their colleagues, Mr. John A.Rolfes, P.E., Mr. Michael A. West, P.E., A.I.A., and Mr. NeilGlaser for their contributions to this guide. Special apprecia-tion is also given to Carol T. Williams for typing the manu-script.

The authors also thank the American Iron and Steel Institutefor their funding of this guide.


Administrator

Rectangle

1. Adams, Peter R, "The Design of Steel Beam-Col-umns", Canadian Steel Industries Construction Coun-cil, Willowdale, Ontario, 1974.

2. Agarawal, Krishna M., and Stafiej, Andrew P., "Calcu-lation of Effective Lengths of Stepped Columns",American Institute of Steel Construction, EngineeringJournal, Vol. 17, No. 4, 1980.

3. "Approval Guide, 1987 - Equipment, Materials, Serv-ice for Conservation of Property", Factory Mutual Re-search Corporation, Factory Mutual System, Norwood,MA, 1987.

4. "Building Code Requirements for Reinforced Concrete(ACI 318-89)", American Concrete Institute, Detroit,MI, 1989.

5. Cannon, R.W., Godfrey, D.A. and Moreadith, F.L.,"Guide to the Design of Anchor Bolts and Other SteelEmbedments", Concrete International, Vol. 3, No. 7,July, 1981.

6. "Code of Standard Practice for Steel Buildings andBridges", American Institute of Steel Construction,Chicago, IL, June 10, 1992.

7. "Code Requirements for Nuclear Safety Related Con-crete Structures (ACI 349-85)", American Concrete In-stitute, Detroit, MI, 1985.

8. "Column Base Plates", AISC Steel Design Guide SeriesNo. 1, American Institute of Steel Construction, Chi-cago, IL, 1990.

9. "Design Manual for Composite Decks, Form Decks andRoof Decks", Steel Deck Institute, Canton, OH, 1987.

10. "Detailing for Steel Construction", American Instituteof Steel Construction, Chicago, IL, 1983.

11. "Diaphragm Design Manual", Steel Deck Institute,Canton, OH, 1987.

12. "Energy Efficient Design of New Buildings ExceptLow-rise Residential Buildings", ASHRAE 90.1, At-lanta, GA, 1989.

13. "Engineering for Steel Construction", American Insti-tute of Steel Construction, Chicago, IL, 1984.

14. "Expansion Joints in Buildings", Federal ConstructionCouncil, Technical Report No. 65, National Academyof Sciences - National Research Council, Washington,D.C.

-82-

REFERENCES

15. Fisher, James M., "The Importance of Tension ChordBracing", American Institute of Steel Construction, En-gineering Journal, Vol. 20, No. 3, 1983.

16. Fisher, James M, "Structural Details in Industrial Build-ings", American Institute of Steel Construction Engi-neering Journal, Vol. 18, No. 3, 1981.

17. Fisher, James M. and West, Michael A., "ServiceabilityDesign Considerations for Low-Rise Buildings", SteelDesign Guide Series 3, American Institute of Steel Con-struction, Chicago, IL, 1990.

18. Galambos, Theodore, V., "Influence of Partial BaseFixity on Frame Stability", ASCE Structural DivisionJournal, Vol. 86, No. ST5, New York, 1960.

19. Gaylord, Gaylord and Stallmeyer,, "Design of SteelStructures", Third Edition, McGraw-Hill, New York,N.Y., 1992

20. "Guide for the Design and Construction of Mill Build-ings", AISE Technical Report No. 13, Association ofIron and Steel Engineers, Three Gateway Center, Suite2350, Pittsburgh, PA 15222-1097, 1979.

21. "Load and Resistance Factor Design Specification forCold-Formed Steel Structural Members", AmericanIron and Steel Institute, Washington, D.C., March 16,1991.

22. "Load and Resistance Factor Design Specification forStructural Steel Buildings and Commentary", AmericanInstitute of Steel Construction, Chicago, IL, September1, 1986.

23. "Loss Prevention Data for roofing Contractors", Fac-tory Mutual Research Corporation, Factory Mutual Sys-tem, Norwood, MA, various dates.

24. "Low Rise Building Systems Manual", Metal BuildingManufacturers Associations, Cleveland, OH, 1986.

25. Lutz, L.A., and Fisher, J.M., "A Unified Approach forStability Bracing Requirements", American Institute ofSteel Construction, Vol. 22, No. 4, 1985.

26. Mueller, John E., "Lessons from Crane Runways",American Institute of Steel Construction, EngineeringJournal, Vol. 2, No. 1, Chicago, IL, 1965.

27. Nair, R. Shankar, "Secondary Stresses in Trusses", En-gineering Journal, American Institute of Steel Construc-tion, Vol. 25, No. 4, 1988.

28. Nair, R. Shankar, "Simple Solutions to Stability Prob-lems in the Design Office", AISC, National Steel Con-struction Conference Proceedings, 1988.


Administrator

Rectangle

-83-

29. "N.R.C.A. Roofing and Waterproofing Manual", TheNational Roofing Contractors Association, Chicago, IL,1981.

30. OSHA "Construction Industry" Occupational Safetyand Health Administration, U.S. Department of Labor,U.S. Government Printing Office.

31. Reemsynder, "Fatigue Cracking in Welded Crane run-way Girders: Causes and Repair Procedures", Iron andSteel Engineer, April, 1978.

32. Ricker, David T., "Tips for Avoiding Crane RunwayProblems", American Institute of Steel Construction,Engineering Journal, Vol. 19, No. 4, 1982.

33. Ringo, Boyd C. and Anderson, Robert B., "DesigningFloor Slabs on Grade", The Aberdeen Group, 426 SD.Westgate St., Addison, Illinois 60101, 1992.

34. Robertson, G.W., and Kurt, C.E., "Behavior of NestedZ-shaped Purlins", Eighth International Specialty Con-ference on Cold-Formed Steel Structures, St. Louis,MO, 1986.

35. Salmon, Charles G. and Johnson, John E., "Steel Struc-tures Design and Behavior", Third Edition, Harper andRow, New York, NY, 1990.

36. Schlenker, Norman, "The Case for the Semi-BoxGirder", American Institute of Steel Construction Engi-neering Journal, Vol. 9, No. 1, Chicago, IL, 1972.

37. "Seismic Design for Buildings, TM 5-809-10", Depart-ments of the Army, Navy and the Air Force, 1982.

38. "Specification and Commentary for Steel Roof Deck",Steel Deck Institute, Canton, OH, 1987.

39. "Specifications for Electric Overhead TravelingCranes", CMAA Specification #70, Crane Manufactur-ers of America, Inc., Charlotte, NC, 1988.

40. "Standard Specifications, Load Tables And Weight Ta-bles for Steel Joists and Joist Girders", Steel Joist Insti-tute, Myrtle Beach, SC, 1992.

41. "Structural Design of Steel Joist Roofs to Resist Pond-ing Loads", TECHNICAL DIGEST #3, Steel Joist Insti-tute, Myrtle Beach, SC, 1971.

42. "Specification for Structural Steel Buildings, AllowableStress Design, Plastic Design with Commentary",American Institute of Steel Construction, Chicago, IL,June 1, 1989.

43. "Specification for the Design of Cold-Formed SteelMembers", American Iron and Steel Institute, Washing-ton, DC, August 19, 1986 with December 11, 1989 Ad-dendum.

44. "Structural Welding Code Sheet Steel" AWS D1.3-81,Second Edition, American Welding Society, 550 NWLe Jeune Road, Miami, FL, 1981


Administrator

Rectangle

BEAMSECTION

W14x22

W14x26

W14x30

W14x34

W14x38

W14x43

W14x48

W14x53

W14x61

W14x68

W14x74

W14x82

W16x26

W16x31

W16x36

W16x40

W16x45

CHANNELSECTION

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

TOTALWEIGHT

42.755.9

46.759.9

50.763.9

54.767.9

58.771.9

63.776.9

68.781.9

73.786.9

94.9103.7

101.9110.7

107.9116.7

115.9124.7

46.759.9

51.764.9

56.769.9

60.773.9

65.778.9

AXIS X-X

333.9372.5

394.7440.8

447.1498.7

507.1565.4

562.4627.1

601.0667.2

667.1740.0

732.9812.2

923.6968.8

1023.71073.4

1108.31161.8

1211.01269.0

492.5550.5

585.3654.4

670.3748.3

753.1839.5

834.6930.1

S1

33.4134.50

40.0941.34

46.7148.05

53.5054.97

59.9161.56

67.3768.97

75.3877.17

83.3185.30

99.43100.53

110.97112.19

120.66121.99

132.58134.07

44.2545.66

53.4455.07

62.8164.60

71.1273.05

79.7381.93

S2

82.92111.37

90.79120.82

98.25129.19

106.01138.06

112.59145.37

119.67152.15

127.73160.85

135.61169.26

184.71206.04

196.30218.10

205.84227.98

217.18239.65

101.74136.48

112.36148.84

122.53160.04

132.04170.69

140.39179.66

AXIS Y-Y

Y1

9.99410.795

9.84410.661

9.57110.379

9.47810.284

9.38610.186

8.9199.674

8.8499.589

8.7979.521

9.2899.637

9.2259.568

9.1859.523

9.1349.464

11.13012.055

10.95211.882

10.67111.584

10.58811.491

10.46711.352

It

132.48318.48

133.44319.44

138.77324.77

140.63326.63

142.31328.31

151.57337.57

154.67340.67

157.79343.79

368.66607.66

375.63614.63

381.80620.80

389.06628.06

133.78319.78

135.18321.18

141.21327.21

143.40329.40

145.39331.39

St

22.0842.46

22.2442.59

23.1243.30

23.4343.55

23.7143.77

25.2645.00

25.7745.42

26.2945.83

49.1567.51

50.0868.29

50.9068.97

51.8769.78

22.2942.63

22.5342.82

23.5343.62

23.9043.92

24.2344.18

APPENDIX A TABLE 1

-84-

Ix


Administrator

Rectangle

-85-

BEAMSECTION

W16x50

W16x57

W16x67

W16x77

W16x89

W16x100

W18x35

W18x40

W18x46

W18x50

W18x55

W18x60

W18x65

W18x71

W18x76

W18x86

W18x97

CHANNELSECTION

C12x20.7C15x33.9

C12x20.7C15x33.9

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

TOTALWEIGHT

70.783.9

77.790.9

100.9109.7

110.9119.7

122.9131.7

133.9142.7

55.768.9

60.773.9

66.779.9

70.783.9

75.788.9

80.793.9

85.798.9

91.7104.9

109.9118.7

119.9128.7

130.9139.7

AXIS X-X

918.61022.6

1033.11148.7

1363.61429.8

1543.41617.5

1766.71849.9

1980.82071.9

785.8880.5

908.61016.3

1026.81147.8

1120.41247.7

1224.31361.6

1331.61478.1

1431.61587.1

1543.11708.2

1860.71950.0

2088.92187.3

2341.22448.9

S1

88.4190.84

100.40103.22

126.53127.85

144.67146.20

167.06168.85

188.04190.04

65.5767.68

76.6478.97

87.7090.38

97.36100.04

107.25110.20

117.24120.41

126.77130.23

137.38141.17

157.73159.34

178.80180.62

201.79203.84

S2

149.31189.28

160.87201.47

229.06255.49

246.89273.88

268.71296.28

289.77317.90

131.01173.01

143.60187.15

154.78199.25

165.65210.82

175.48221.23

185.86232.27

195.05241.83

205.20252.32

273.10303.65

293.92325.01

316.89348.54

AXIS Y-Y

Y1

10.38911.257

10.28911.128

10.77611.183

10.66811.063

10.57510.956

10.53410.902

11.98313.010

11.85512.869

11.70712.699

11.50812.471

11.41512.355

11.35712.276

11.29212.186

11.23212.099

11.79612.238

11.68312.109

11.60112.013

It

147.55333.55

150.50336.50

374.41613.41

384.10623.10

396.19635.19

408.00647.00

136.65322.65

138.52324.52

140.21326.21

148.99334.99

151.41337.41

153.97339.97

156.32342.32

159.04345.04

391.14630.14

402.51641.51

415.36654.36

St

24.5944.47

25.0844.86

49.9268.15

51.2169.23

52.8270.57

54.4071.88

22.7743.02

23.0843.26

23.3643.49

24.8344.66

25.2344.98

25.6645.33

26.0545.64

26.5046.00

52.1570.01

53.6671.27

55.3872.70

APPENDIX A TABLE 1

Ix


Administrator

Rectangle

BEAMSECTION

W18x106

W18x119

W21x44

W21x50

W21x57

W21x62

W21x68

W21x73

W21x83

W21x93

W21x101

W21x111

W21x122

W21x132

W21x147

W24x55

W24x62

W24x68

W24x76

CHANNELSECTION

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

TOTALWEIGHT

139.9148.7

152.9161.7

64.777.9

70.783.9

77.790.0

82.795.9

88.7101.9

93.7106.9

103.7116.9

113.7126.9

143.7

153.7

164.7

174.7

189.7

75.788.9

82.795.9

88.7101.9

96.7109.9

AXIS X-X

2526.62641.2

2841.32966.1

1254.41408.5

1418.61589.3

1629.21818.8

1800.91999.9

1963.82175.9

2101.62324.5

2352.62593.9

2617.32876.4

3369.7

3662.5

3985.0

4280.5

4739.9

1919.42152.9

2157.52411.9

2443.72712.7

2746.93036.4

S1

218.89221.16

247.16249.72

92.9796.19

106.31109.87

123.11127.01

138.40142.27

151.83156.00

163.26167.71

184.10189.09

205.99211.56

245.48

268.75

294.20

317.40

353.08

128.89133.51

146.35151.39

168.21173.13

190.39195.62

S2

333.01364.91

360.85393.26

168.39219.51

182.62234.95

200.90254.71

218.02272.75

231.63287.01

242.97298.76

263.36319.74

284.60341.46

416.88

439.56

464.19

486.75

521.68

214.20274.41

232.48293.82

257.64320.58

281.03345.13

AXIS Y-Y

Y1

11.54211.942

11.49611.877

13.49214.643

13.34414.465

13.23214.319

13.01114.057

12.93413.948

12.87213.859

12.77813.717

12.70513.596

13.726

13.627

13.545

13.485

13.424

14.89116.124

14.74115.931

14.52715.668

14.42715.522

It

425.05664.05

441.27680.27

139.29325.29

141.41327.41

144.25330.25

157.67343.67

161.28347.28

164.19350.19

169.58355.58

175.26361.26

677.75

691.01

706.16

720.04

741.62

143.46329.46

146.15332.15

164.12350.12

170.17356.17

St

56.6773.78

58.8375.58

23.2143.37

23.5643.65

24.0444.03

26.2745.82

26.8846.30

27.3646.69

28.2647.41

29.2148.16

75.30

76.77

78.46

80.00

82.40

23.9143.92

24.3544.28

27.3546.68

28.3647.48

APPENDIX A TABLE 1

-86-

Ix


Administrator

Rectangle

APPENDIX A TABLE 1

-87-

BEAMSECTION

CHANNELSECTION

TOTALWEIGH

AXIS X-X AXIS Y-Y

W24x84

W24x94

W24x104

W24x117

W24x131

W24x146

W24x162

W27x84

W27x94

W27x102

W27x114

W27x146

W27x161

W27x178

W30x99

W30x108

W30x116

W30x124

W30x132

W30x173

C12x20.7C15x33.9

C12x20.7C15x33.9

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MX18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

104.7117.9

114.7127.9

146.7

158.7

173.7

188.7

204.7

117.9126.7

127.9136.7

135.9144.7

147.9156.7

188.7

203.7

220.7

132.9141.7

141.9150.7

149./9158.7

157.9166.7

165.9174.7

215.7

3026.93335.4

3390.23721.0

4319.7

4807.2

5366.1

5993.5

6642.1

4047.74255.4

4530.94756.2

4919.25157.8

5437.45693.9

7354.2

8069.9

8839.2

5542.85825.0

6067.36366.5

6593.06907.8

7053.77382.0

7496.47837.7

10427.2

210.91216.56

237.56243.72

280.24

314.87

354.13

397.54

442.14

237.11240.27

267.82271,24

292.38596.00

325.92329.93

440.34

486.07

535.00

299.50303.77

329.95334.45

360.23364.91

386.80391.69

412.67417.82

574.19

301.76366.59

328.47794.06

474.91

509.10

548.83

592.63

637.00

403.19450.32

435.55483.59

461.23509.91

494.02543.20

660.82

705.55

753.02

480.17533.21

512.36566.20

544.52599.11

571.88627.00

597.58653.05

819.10

14.35115.401

14.27015.267

15.414

15.267

15.152

15.076

15.022

17.07017.710

16.91717.534

16.82417.424

16.68317.257

16.701

16.602

16.521

18.50619.175

18.38819.035

18.30118.929

18.23518.846

18.16518.758

18.159

176.09362.09

183.31369.31

683.54

702.54

723.94

748.99

775.05

367.69606.69

376.89615.89

384.47623.47

394.13633.13

775.28

802.01

831.09

378.71617.71

387.79626.79

396.88635.88

405.10644.10

412.71651.71

852.63

29.3448.27

30.5549.24

75.94

78.06

80.43

83.22

86.11

49.0267.41

50.2568.43

51.2669.27

52.5570.34

86.14

89.11

92.34

50.4968.63

51.7069.64

52.9170.65

54.0171.56

55.0272.41

94.73

Rev.3/1/03


WEIGHT

Administrator

Rectangle

BEAMSECTION

W30x191

W30x211

W33x118

W33x130

W33x141

W33x152

W36x33.9

W36x150

W36x160

W36x170

W36x182

W36x194

CHANNELSECTION

MC18x42.7

MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

TOTALWEIGHT

233.7

253.7

151.9160.7

163.9172.7

174.9183.7

185.9194.7

168.9177.7

183.9192.7

193.9202.7

203.9212.7

215.9224.7

227.9236.7

AXIS X-X

Ix

11475.9

12640.7

7898.28280.0

8789.09194.7

9592.710018.8

10342.410786.5

10218.710696.0

11542.412050.6

12317.712843.2

13098.813641.6

13979.514541.0

14826.615405.1

S1

634.90

701.84

394.75400.13

442.11447.76

484.68490.60

524.31530.48

480.60487.19

545.73552.61

584.17591.27

623.01630.41

667.25675.03

709.36717.49

S2

879.05

944.80

596.00656.25

645.76707.01

689.70751.67

730.17792.73

695.73761.52

764.41831.45

803.80871.38

842.63910.60

885.95954.20

927.32995.85

AXIS Y-Y

Y1

18.075

18.010

20.00720.692

19.87920.534

19.79120.421

19.72520.333

21.26221.954

21.15021.806

21.08521.721

21.02421.639

20.95021.541

20.90121.470

It

889.95

931..66

408.29647.29

423.64662.64

437.78676.78

450.99689.99

427.34666.34

449.51688.51

461.88700.88

474.59713.59

488.12727.12

501.70740.70

St

98.88

118.35

54.4371.92

56.4873.62

58.3775.19

60.1376.66

56.9774.03

59.9376.50

61.5877.87

63.2779.28

65.0880.79

66.8982.30

-88-

APPENDIX A TABLE 1

BEAMSECTION

W14x22

W14x26

W14x30

W14x34

W14x38

W14x43

W14x48

W14x53

W14x61

W14x68

W14x74

W14x82

W14x90

W14x99

W14x109

W14x120

W14x132

W16x26

CHANNELSECTION

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C12x20.7C15x33.9

COMPOSITE SECTION

rT

4.1305.231

4.0335.144

3.9985.086

3.9185.006

3.8544.941

3.8314.876

3.7724.807

3.7184.743

4.7405.648

4.6755.567

4.6225.502

4.5705.435

5.664

5.612

5.558

5.510

5.460

4.0925.193

d/Af

1.801.21

1.731.18

1.621.13

1.551.10

1.501.07

1.350.99

1.290.96

1.240.93

0.870.75

0.840.73

0.810.71

0.780.69

0.63

0.60

0.58

0.56

0.54

1.991.35

MAXIMUM SPAN (FT.)FOR

Fb = 0.6Fy

Fy = 36 ksi

25.6338.09

26.7539.05

28.4540.80

29.7341.94

30.8242.93

34.2946.74

35.7548.08

37.1949.39

53.1561.49

55.0963.34

56.7664.93

58.6066.68

73.29

75.92

78.86

81.83

85.07

23.1534.11

Fy = 50 ksi

18.4527.42

19.2628.11

20.4929.37

21.4030.20

22.1930.91

24.6933.65

25.7434.62

26.7735.56

38.2744.27

39.6745.60

40.8746.75

42.1948.01

52.77

54.66

56.78

58.91

61.25

16.6624.56

APPENDIX A TABLE 2

-89-


Administrator

Rectangle

BEAMSECTION

W16x31

W16x36

W16x40

W16x45

W16x50

W16x57

W16x67

W16x77

W16x89

W16x100

W18x35

W18x40

W18x46

W18x50

W18x55

W18x60

W18x65

CHANNELSECTION

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C15x33.9MC18x42.7

Cl5x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

COMPOSITE SECTION

rT

3.9835.091

3.9405.024

3.8604.941

3.8004.876

3.7404.810

3.6684.728

4.7255.622

4.6475.523

4.5625.414

4.4915.319

3.9765.078

3.8704.973

3.7914.892

3.7914.851

3.7384.790

3.6844.727

3.6424.676

d/Af

1.891.31

1.771.25

1.691.21

1.631.18

1.561.15

1.491.11

0.990.86

0.950.83

0.900.79

0.850.76

2.081.44

1.961.39

1.881.35

1.761.29

1.691.25

1.631.22

1.581.19


Fb = 0.6Fy

Fy = 36 ksi

24.4035.23

26.0836.91

27.3438.06

28.3939.02

29.5040.05

30.9741.40

46.3953.54

48.6655.71

51.3658.32

53.9160.77

22.2431.99

23.5433.18

24.6234.17

26.2535.82

27.2736.77

28.3437.77

29.2738.64

Fy = 50 ksi

17.5725.37

18.7726.57

19.6827.40

20.4428.09

21.2428.83

22.3029.80

33.4038.55

35.0340.11

36.9841.99

38.8143.75

16.0123.03

16.9523.89

17.7324.60

18.9025.79

19.6326.47

20.4027.20

21.0727.82

APPENDIX A TABLE 2

-90-


Administrator

Rectangle

BEAMSECTION

W18x71

W18x76

W18x86

W18x97

W18x106

W18x119

W21x44

W21x50

W21x57

W21x62

W21x68

W21x73

W21x83

W21x93

W21x101

W21x111

W21x122

W21x132

W21x147

CHANNELSECTION

C12x20.7C15x33.9

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

COMPOSITE SECTION

rT

3.5994.623

4.7325.598

4.6645.508

4.5965.417

4.5545.358

4.4885.264

3.9305.024

3.8414.933

3.7334.819

3.7594.782

3.7044.714

3.6644.664

3.6024.582

3.5484.506

5.496

5.434

5.369

5.316

5.242

d/Af

1.521.16

1.060.92

1.010.89

0.960.85

0.930.82

0.880.79

2.321.63

2.201.57

2.061.50

1.901.42

1.821.37

1.761.34

1.661.28

1.571.23

0.97

0.93

0.90

0.87

0.83


Fb = 0.6Fy

Fy = 36 ksi

30.3039.60

43.4449.87

45.5851.94

47.9254.21

49.5855.82

52.3458.50

19.9228.32

21.0129.33

22.4530.67

24.2832.52

25.4133.59

26.3034.44

27.8635.91

29.4237.40

47.61

49.32

51.24

52.93

55.50

Fy = 50 ksi

21.8128.51

31.2835.91

32.8137.40

34.5039.03

35.6940.19

37.6842.12

14.7920.39

15.1321.12

16.1622.08

17.4823.41

18.2924.19

18.9324.79

20.0625.86

21.1826.93

34.28

35.51

36.89

38.11

39.96

APPENDIX A TABLE 2

-91-


Administrator

Rectangle

BEAMSECTION

W21x147

W24x55

W24x62

W24x68

W24x76

W24x84

W24x94

W24x104

W24x117

W24x131

W24x146

W24x162

W27x84

W27x94

W27x102

W27x114

W27x146

W27x161

W27x178

W30x99

CHANNELSECTION

MC18x42.7

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

C12x20.7C15x33.9

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C15x33.9MC18x42.7

COMPOSITE SECTION

5.242

3.8604.940

3.7774.851

3.8054.798

3.7344.707

3.6754.628

3.6154.543

5.553

5.470

5.387

5.300

5.223

4.7445.654

4.6535.543

4.5875.460

4.5165.368

5.438

5.376

5.320

4.7255.613

d/Af

0.83

2.471.77

2.341.71

2.111.58

1.981.51

1.871.44

1.751.38

1.10

1.05

0.99

0.94

0.89

1.651.43

1.561.36

1.501.31

1.431.26

1.06

1.01

0.96

1.771.53


Fb = 0.6Fy

Fy = 36 ksi

55.50

18.6826.06

19.7427.06

21.8529.17

23.3430.59

24.7531.94

26.3933.52

41.86

43.99

46.31

48.99

51.67

27.8932.34

29.4933.90

30.7735.15

32.3136.65

43.61

45.80

48.10

26.1330.14

Fy = 50 ksi

39.96

14.5218.77

14.2119.48

15.7321.00

16.8022.03

17.8223.00

19.0024.13

30.14

31.67

33.34

35.27

37.20

20.0823.28

21.2324.40

22.1525.31

23.2626.39

31.40

32.97

34.63

18.8121.70

TABLE 2

-92-

APPENDIX A

rT


Administrator

Rectangle

BEAMSECTION

W30x108

W30x116

W30x124

W30x132

W30x173

W30x191

W30x211

W33x118

W33x130

W33x141

W33x152

W36x135

W36x150

W36x150

W36x170

W36x182

W36x194

W36x210

CHANNELSECTION

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

MC18x42.7

MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

C15x33.9MC18x42.7

MC18x42.7

COMPOSITE SECTION

4.6515.521

4.5845.435

4.5305.364

4.4865.306

5.463

5.408

5.357

4.7035.539

4.6255.433

4.5625.346

4.5115.274

4.6935.498

4.6025.372

4.5615.311

4.5235.255

4.4905.204

4.4595.155

5.100

d/Af

1.681.47

1.611.41

1.541.36

1.491.32

1.08

1.02

0.96

1.801.57

1.691.49

1.601.42

1.521.36

1.851.63

1.701.52

1.641.46

1.571.41

1.511.36

1.461.32

1.27


Fb = 0.6Fy

Fy = 36 ksi

27.4431.43

28.7432.70

29.8933.83

30.9134.83

42.80

45.24

47.87

25.6829.31

27.3730.97

28.8932.47

30.2733.83

24.9828.34

27.0930.42

28.2231.54

29.3632.65

30.5133.79

31.6534.92

36.35

Fy = 50 ksi

19.7622.63

20.6923.55

21.5224.36

22.2525.08

30.81

32.57

34.47

18.4921.10

19.7022.30

20.8023.38

21.7924.35

17.9820.69

19.5021.90

20.3222.70

21.1423.51

21.9624.33

22.7925.14

26.17

APPENDIX A TABLE 2

-93-

rT


Administrator

Rectangle

Appendix BCALCULATION OF EFFECTIVE LENGTHSOF STEPPED COLUMNS

KRISHNA M. AGRAWAL AND ANDREW P. STAFIEJ

Travelling cranes are frequently used to move heavy loads inindustrial buildings. To accomplish a general movement, thecrane traverses a crane bridge, which in turn moves on railsalong the length of the building, supported by the mainbuilding columns.

Designers frequently choose stepped columns, with thewider lower section serving a dual purpose: (a) to support thecrane rail and (b) to provide the necessary strength to supportthe extra load from the crane. The design of the steppedcolumns is time-consuming and complicated. Effectivelengths, which must be calculated for each segment, dependupon the following: the end fixity types at the two ends, theratio of segment lengths the ratio of the segmentinertias and the ratio of the applied axial loads

applied at the top of the column and at the steppedlevels.

Various cases of end fixities are encountered in practice(Fig. 1). Anderson and Woodward1 have presented equationsfor five end-fixity types to be used in calculating effectivelengths. These types are: (1) Pin-Pin (2) Fix-Free, (3) Fix-Pin,(4) Fix-Slider, and (5) Fix-Fix. Two other cases which havenot been dealt with previously are: (6) Pin-Fix and (7) Pin-Slider (Fig. 1). Industrial building frames are often designedas pinned at the bottom, supporting a deep roof truss at thetop which provides for a fix or slider end condition.

The characteristic equation in Ref. 1 for case (5) whenP2 = 0 [Eq. (A-15) and FUNCTION FC5(x)] appears to be inerror. This technical note is intended to correct the equationfor the end-fixity case (5) and to extend the directory ofcharacteristic equations for end-fixities to include two addi-tional cases: (6) Pin-Fix and (7) Pin-Slider. The derivation ofthe equations is omitted in this paper, since the process hasbeen adequately described in Ref. 1. The nomenclature ofRef. 1 is used throughout to maintain continuity.

Krishna M. Agrawal is senior structural analyst, H. A. Simons(International) Ltd., Vancouver, B.C., Canada.Andrew P. Stafiej is Structural Engineer, H. A. Simons (Inter-national) Ltd., Vancouver, B.C., Canada.

Reprinted from Engineering Journal, Fourth Quarter, 1980.

-94-

Fig. 1. End condition types.

CHARACTERISTIC EQUATIONS

The parameters in the equations have the following definitions:

Finding the lowest root of the characteristic equationZ = ZRT allows the calculation of buckling load

Equating these critical loads to the Euler buckling formulaone obtains:

These effective lengths are used in the AISC interactionformula for designing beam columns.

The variable definitions are:Moments of inertia of upper and lower segments,respectivelyApplied axial loads at the top and at step levelTotal column axial loadLengths of upper and lower segments, respectivelyTotal column lengthEffective lengths of the upper and lower seg-ments for Euler buckling formula, respectivelyEffective length factors for upper and lower seg-ments, respectively, with the following definition:

Case 5—Fix-Fix [corrected characteristic equation to replaceEq. (A-15) in Ref. 1]:

c.

Case 6—Pinned-Fixed:

a.

b.

c.

Case 7—Pinned-Slider:

a.

b.(A-20)

c.

(A-21)

Reference 4 outlines a computer program similar to the onedescribed in Ref. 1. This program was developed to calculatethe roots of the characteristic equations. The solution routinewhich serves to find the lowest root was modified to improvethe speed of convergence to the root. Residual values werecalculated by spacing points at equal intervals until a signchange in the residual was observed. At this point, instead ofhalving the incremental value of Z, a new value for Z wascalculated by interpolating the two values of Z, which gaveresiduals of differing signs. The process was repeated retain-ing two values for Z, which produced the smallest residualsfor further interpolation. The last step was repeated severaltimes, producing a much faster convergence to the charac-teristic root.

The output from this program (Table 1) lists the slendernessratios for all seven end-fixity types (Fig. 1) for a wide selec-tion of segment inertia ratios, segment length ratios, and top-and step-level axial-load ratios. Any intermediate value canbe easily interpolated from the values presented.

Note that the axial load ratio variesfrom 0 to 1. A value of zero corresponds to anda value of 1 corresponds to and All other valuesof the ratio correspond to P1 and P2 both greater than zero.

ACKNOWLEDGMENT

Checks provided by Tony Katramadakis, Curt Shelton, andSteve Lang are gratefully acknowledged. Financial supportfor this project was provided by H. A. Simons (International)Ltd.

REFERENCES

1. Anderson, J. P. and J. H. Woodward, "Calculation of EffectiveLengths and Effective Slenderness Ratios of Stepped Col-umns," AISC Engineering Journal, Oct. 1972, pp. 157-166.

2. Dalai, S. T., "Some Non-Conventional Cases of ColumnDesign," AISC Engineering Journal, Jan. 1969, pp. 28-39.

3. Huang, H. C., "Determination of Slenderness Ratios forDesign of Heavy Mill Building Stepped Columns," Ironand Steel Engineering, Nov. 1968, pp. 123-134.

4. Agrawal, K. M., STEPCOL—Computer Program to Cal-culate Slenderness Ratio of Stepped Columns, H. A. Si-mons (International) Ltd. Engineering Computing Depart-ment, Vancouver, B.C., Canada. March 1980.

-95-


Administrator

Rectangle

-96-

Table 1. Equivalent Length Factors for Various End Conditions


Administrator

Rectangle


-97-


Administrator

Rectangle

-98-



Administrator

Rectangle


-99-


Administrator

Rectangle

-100-



Administrator

Rectangle

-101-



Administrator

Rectangle

-102-



Administrator

Rectangle

-103-



Administrator

Rectangle

DESIGN GUIDE SERIESAmerican Institute of Steel Construction, Inc.One East Wacker Drive, Suite 3100Chicago, Illinois 60601-2001

Pub. No. D807 (5M697)


Administrator

Rectangle

Documents

Steel Design Guide Series Industrial Buildings - صفحه اصلی · PDF fileSteel Design Guide Series Industrial Buildings Roofs to Column Anchorage Roofs to Column Anchorage James