Guidelines for the Use of Structural Precast Concrete in ...

Guidelines for the Use ofStructural Precast Concrete

in Buildings

NEW ZEALAND CONCRETE SOCIETY Centre for Advanced EngineeringUniversity of Canterbury, Christchurch, New Zealand

NEW ZEALAND SOCIETY FOREARTHQUAKE ENGINEERING

Centre for Advanced EngineeringUniversity of Canterbury Christchurch New Zealand


in BuildingsSecond Edition

Report of a Study Group of the

New Zealand Concrete Societyand the

New Zealand Society for Earthquake Engineering

All rights reserved. No part of this publication may be reproduced, stored in aretrieval system, transmitted, or otherwise disseminated, in any form or by anymeans, except for the purposes of research or private study, criticism or review,

without the prior permission of the Centre for Advanced Engineering.

Guidelines for the Use of StructuralPrecast Concrete in Buildings

— Second Edition —

ISBN 0-908993-20-X

Printing History

First published August 1991Reprinted June 1992

Second Edition December 1999

Copyright

© 1999 Centre for Advanced Engineering

PublisherCentre for Advanced EngineeringUniversity of CanterburyPrivate Bag 4800ChristchurchNew Zealand

Printing

Wickliffe Press, Christchurch

Cover Design

Hudson Design, Christchurch

Editorial Services, Graphics and Book Design

Charles Hendtlass and Úna O'Grady, Centre for Advanced Engineering

Disclaimer

The authorship of this report has been attributed to a number of individuals involved inits production. The report does not necessarily reflect the views of the New ZealandConcrete Society, the New Zealand Society for Earthquake Engineering, or the Centrefor Advanced Engineering as sponsoring organisations. Although the authors haveexercised due care in writing this report, no responsibility can be taken in its applicationby the authors, their employers, or the sponsoring organisations for the safety of anypersons or property in buildings designed using this report or the buildings themselves.Recommendations in the report need to be interpreted with care and judgement.

i

Contents

Acknowledgements ............................................................................................................................. iii

1 Introduction .................................................................................................................................... 11.1 General .......................................................................................................................................... 11.2 Formation of Study Group ............................................................................................................ 11.3 Scope of the Guidelines ................................................................................................................ 21.4 Summary ....................................................................................................................................... 21.5 Conclusions ................................................................................................................................... 31.6 References ..................................................................................................................................... 4

2 Floor Unit Support and Continuity .............................................................................................. 52.1 Introduction ................................................................................................................................... 52.2 Types of Support for Precast Concrete Floor Units ...................................................................... 52.3 Precast Floor Unit Seating ............................................................................................................ 72.4 Seating Details for Precast Concrete Hollow-core and Flat Slab Floor Units ............................ 112.5 Types of Support for Ribbed Units ............................................................................................. 142.6 Special Diaphragm Requirements for Hollow-core Floors ......................................................... 152.7 Overseas Practice ........................................................................................................................ 162.8 Related Considerations ............................................................................................................... 162.9 Recommendations ....................................................................................................................... 172.10 References ................................................................................................................................... 18

3 Frame Connections ...................................................................................................................... 213.1 Introduction ................................................................................................................................. 213.2 System 1 - Precast Beam Units between Columns ..................................................................... 213.3 System 2 - Precast Beam Units through Columns ...................................................................... 263.4 System 3 - Precast T or Cruciform Shaped Units ....................................................................... 303.5 Low Frames with Strong Beam-Weak Column Design.............................................................. 303.6 Mixed Precast Prestressed Concrete and Cast-in-place Reinforced Concrete

Moment-resisting Frames ........................................................................................................... 303.7 Pinned Joints ............................................................................................................................... 333.8 Composite Reinforced Concrete Moment-resisting Frames of Limited Ductility ...................... 343.9 Industrial Buildings ..................................................................................................................... 353.10 Frame Connections in North American and Japanese Practice .................................................. 353.11 Recommendations ....................................................................................................................... 353.12 References ................................................................................................................................... 38

4 Structural Wall Elements ............................................................................................................ 414.1 Introduction ................................................................................................................................. 414.2 Types of Precast Concrete Structural Wall Construction ........................................................... 414.3 Monolithic Wall Systems ............................................................................................................ 424.4 Jointed Wall Systems .................................................................................................................. 464.5 New Zealand Examples .............................................................................................................. 494.6 Weathering Details ...................................................................................................................... 564.7 Recommendations ....................................................................................................................... 564.8 References ................................................................................................................................... 56

5 Diaphragms ................................................................................................................................... 595.1 Introduction ................................................................................................................................. 595.2 Requirements of Diaphragms...................................................................................................... 595.3 Internal Restraint Actions in Diaphragms ................................................................................... 625.4 Analysis and Design of Diaphragms ........................................................................................... 625.5 Interaction of Plastic Hinge Zones and Diaphragms .................................................................. 67

ii

5.6 Overseas Practice ........................................................................................................................ 685.7 Recommendations ....................................................................................................................... 705.8 References ................................................................................................................................... 71

6 Connections between Precast Concrete Units by Grouted and Welded Bars ........................ 736.1 General ........................................................................................................................................ 736.2 Selection and Types of Grout ...................................................................................................... 736.3 Grouting Situations ..................................................................................................................... 746.4 Grouting Specific Connections ................................................................................................... 786.5 Quality Assurance ....................................................................................................................... 836.6 Grouting Workmanship and Construction Aspects .................................................................... 836.7 Connections between Precast Elements using Welded Reinforcing Bars .................................. 846.8 Mechanical Connectors For Splicing Reinforcing Bars ............................................................. 886.9 Recommendations ....................................................................................................................... 896.10 References ................................................................................................................................... 89

7 Embedded Steel Connectors ........................................................................................................ 917.1 Introduction ................................................................................................................................. 917.2 Types of Connectors ................................................................................................................... 917.3 Selection of Fixings .................................................................................................................... 917.4 Structural Actions on Steel Embedments .................................................................................... 927.5 Design Approach......................................................................................................................... 937.6 Design Methods .......................................................................................................................... 947.7 Detailing of Steel Embedments in Concrete ............................................................................... 977.8 Recommendations ....................................................................................................................... 987.9 References ................................................................................................................................... 98

8 Tolerances ................................................................................................................................... 1018.1 Introduction ............................................................................................................................... 1018.2 Product Tolerance ..................................................................................................................... 1018.3 Erection Tolerance .................................................................................................................... 1018.4 Interface Tolerance ................................................................................................................... 1048.5 Clearances ................................................................................................................................. 1048.6 Implications for Design ............................................................................................................. 1048.7 Recommendations ..................................................................................................................... 1048.8 References ................................................................................................................................. 105

9 Appendices .................................................................................................................................. 111A1 Beam Elongation due to Plastic Hinging .................................................................................. 111A2 Allowances for Effects of Spalling ........................................................................................... 115B1 Precast Concrete Frame Connection for System 1 Design Bending Moments ......................... 117B2 Experimental Tests on Precast Concrete Frame Midspan Joint Connections for System 2 ..... 119B3 Laboratory Tests on the Performance of Grouted Connections for System 2 .......................... 123C1 Assessing the Influence of Thermal Gradients, Creep and Shrinkage Strains in

Composite Concrete Members .................................................................................................. 125C2 Method for Locating the Position of Creep and Shrinkage Cracks and

Assessing Crack Widths ............................................................................................................ 127D1 Criteria for the Selection of Fixings .......................................................................................... 129D2 Embedments Subject to Corrosion ............................................................................................ 131E1 Example Calculation for Embedded Steel Connectors ............................................................. 133

10 Details of Organisations Involved ............................................................................................. 137New Zealand Society for Earthquake Engineering ............................................................................. 137New Zealand Concrete Society ........................................................................................................... 137Centre for Advanced Engineering ....................................................................................................... 138

Index ................................................................................................................................................... 139

iii

Acknowledgements

Contributors —Second Edition (1999)

D K Bull (Editor), Holmes Consulting Group Ltd, Christchurch

D P Barnard, retired (from Cement and Concrete Association of New Zealand)

B J Brown, Fraser Thomas Ltd, Auckland

R C Fenwick, University of Auckland

D C Hopkins, Sinclair Knight Merz Ltd, Wellington

L G McSaveney, Firth Industries Ltd, Auckland

R Park, University of Canterbury

J Restrepo, University of Canterbury

R G Wilkinson, Holmes Consulting Group Ltd, Christchurch

Study Group — First Edition (1991)

A W Charleson (Editor), Victoria University of Wellington

D C Hopkins (Chairman), Sinclair Knight Merz Ltd, Wellington

D P Barnard, Cement and Concrete Association of New Zealand

B J Brown, Fraser Thomas Ltd, Auckland

D K Bull, Cement and Concrete Association of New Zealand

R C Fenwick, University of Auckland

J L Lumsden, Centre for Advanced Engineering

L G McSaveney, Firth Industries Ltd, Auckland

B Newsome, Downer and Company Ltd

R Park, University of Canterbury

R G Wilkinson, Holmes Consulting Group Ltd, Christchurch

Financial Support and Editorial Services

Acknowledgement is made of financial support for the two editions of this publication from the Centre for AdvancedEngineering at the University of Canterbury, the New Zealand Concrete Society, the New Zealand Society forEarthquake Engineering, and the Cement and Concrete Association of New Zealand. Final compilation, editorialservices, graphics and book design have been carried out by the Centre for Advanced Engineering.

Photographic Material

Many photographs from the first edition have been reproduced in this edition and the publishers are grateful to thefollowing organisations for making photographic material available:

Firth Stresscrete Ltd, Holmes Consulting Group Ltd, Kerslake and Partners, Kingston Morrison Ltd, Murray-NorthLtd, University of Canterbury, Victoria University of Wellington, Wilkins and Davies Ltd and Works CentralLaboratories.

Introduction • 1

Chapter 1Introduction

1.1 General

This is a second edition of this book, which was firstpublished in 1991. The second edition came aboutthrough the need to incorporate relevant research un-dertaken during the first half of the 1990s, to bringsome of the technical aspects in to line with the 1995revision of the Standard for the Design of ConcreteStructures (NZS3101:1995), and to respond to thecontinuing demand for the Guidelines that outstrippedthe quantity published, including the second printing.

Since the introduction of seismic design requirementsin 1935, New Zealand has favoured reinforced con-crete as a building material. Developments in its usehave been related to successive changes in seismicdesign requirements, notably in 1965 and 1976. Thesechanges were summarised by Park [1.1]. Further sig-nificant changes occurred with the publication of acode of practice for general structural design anddesign loadings for buildings [1.2].

In New Zealand, general design provisions for rein-forced and prestressed concrete structures are con-tained in the code for general structural design anddesign loadings for buildings, and in a code for thedesign of concrete structures [1.3]. For buildings, thecodes contain comprehensive provisions for the seis-mic design of cast-in-place concrete structures, but donot have provisions covering all aspects of precastconcrete structures. Nevertheless, significant develop-ments in the use of precast concrete have been made inspite of the fact that some aspects of the seismic designof precast concrete building structures have not yetbeen formally codified. This reflects an on-goinginnovation by New Zealand practitioners that will nodoubt continue.

Since the early 1960s there has been a steady increasein the use of precast concrete for structural compo-nents. Precast concrete fabricators have developedskills to meet the increasing demand, using their expe-rience with increasingly popular “non-structural” clad-ding units. The use of precast concrete in flooringsystems very rapidly became commonplace, leavingcast-in-place floor construction generally less com-mon and uncompetitive. However, until the late 1970s

to early 1980s, the use of precast elements for seismicresistance in moment resisting frames and walls wasthe exception rather than the rule.

The boom years of the mid-1980s produced a signifi-cant increase in structural applications of precast con-crete, which had the advantages of familiar materialsand methods, high-quality factory made units, andspeed of construction. Time and resources availablefor experimental verification were significantly re-duced.

With high interest rates and pressure for new space, theadvantage of speed gave precast concrete a distinctcost advantage. Designers quickly responded, as didfabricators. Above all, the number of designers speci-fying precast grew rapidly, and the ranks of fabricatorsswelled to meet demand.

Contractors quickly adapted to demands of economi-cal precast assembly with enhanced craneage andconstruction techniques, and maximized off-site fabri-cation to compensate for a shortage of skilled labour.There was pressure to perform.

The result was a remarkable development in all aspectsof the structural use of precast concrete as all sectorsapplied their ingenuity to obtain a competitive edge.Whilst design aspects were generally carefully consid-ered, the solutions proposed normally assumed thatextrapolations from testing of cast-in-place specimenswould be valid. In some instances, especially in floorconstruction, the monolithic integrity of cast-in-placeconcrete was not fully replaced.

With this increase in the use of precast concrete struc-tural elements came an increasing concern that some ofthe design solutions being used should be more fullyresearched. Even if there was no reason to doubt thevalidity of extrapolating cast-in-place results, thenumber of major buildings employing precast concretefor seismic resistance demanded that more researchand testing be done to justify confidence in the struc-tural systems.

1.2 Formation of Study GroupIn February 1988, a seminar at the University ofCanterbury, attended by designers, researchers, fabri-

2 • Guidelines for the Use of Structural Precast Concrete

cators and constructors, highlighted a growing need toinvestigate and verify the performance of precast con-crete in structural members for seismic resistance.

Following the seminar a study group jointly-funded bythe New Zealand Concrete Society, the New ZealandSociety for Earthquake Engineering, and the Centre forAdvanced Engineering at the University of Canter-bury, was formed. Members were selected to representthe design, research, fabrication and construction as-pects. Its objectives, as with previous study groups,have been:

• to bring together and summarise existing data;

• to present the data in a form useful for New Zealandconditions;

• to identify any special concerns;

• to indicate recommended practices (and draw at-tention to practices that are not recommended); and

• to recommend topics requiring further research.

Principal areas to be covered were:

• precast beams (both shell and solid), precast col-umns and their jointing;

• beam-column joints, especially if cast-in-placebetween precast elements;

• support and continuity of floor slabs;

• jointing techniques and connectors, constructabilityand tolerances;

• diaphragm actions; and

• behaviour of precast concrete wall systems (subse-quently included).

Within each area, investigations focused on aspectsthat could possibly lead to different behaviour whencompared to cast-in-place construction. Special con-sideration was given to related matters, including fireresistance, beam elongation, robustness, integrity andworkmanship.

For this second edition, those responsible for contrib-uting to the first edition were asked to make necessarychanges reflecting developments since 1991. The origi-nal Study Group was not reconvened, but John Lumsdenof CAE co-ordinated the individual responses.

1.3 Scope of the GuidelinesThe scope of these guidelines follows the areas identi-fied for investigation listed above. Generally, the

emphasis is on building structures rather than civilengineering structures. Furthermore, only structuralelements are dealt with since architectural (non-struc-tural) precast concrete is not normally designed tocontribute to the overall structural integrity and re-quires a different set of design criteria. Although thefocus is on seismic aspects, many sections refer togravity load effects as well as volume changes such ascreep, shrinkage and thermal actions, since these ef-fects can result in a significant reduction in seismicperformance. Durability of precast concrete is notincluded in the scope. Readers are referred to reference1.3 for further information.

Each chapter contains references to overseas experi-ence, research and testing. Overseas research resultshave required careful interpretation to allow for NewZealand’s specific demands for ductility and capacitydesign, and for this reason much potentially valuablework could not be included with confidence. Interestin this subject is evident in work carried out in theUSA, Japan, China, Romania and other countries. Inaddition, the Prefabrication Commission of the FIPhas sets of recommendations which contain muchrelevant material [1.4, 1.5]. Where appropriate, over-seas material is referred to and referenced in thevarious sections.

1.4 Summary

Floor Unit Support and ContinuityThis chapter highlights the need for careful detailing toprovide adequate continuity and maintenance of loadpaths through the structures. Typical floor unit supportdetails are examined and their advantages and disad-vantages described. Many details are seen to be sensi-tive to fabrication and construction tolerances. Detailsthat reduce this sensitivity are indicated, notably theinclusion of continuity steel across beams.

The potential for beam elongation due to yielding ofsteel in plastic hinges is seen as a particularly importantconsideration for precast systems. Such lengtheningcould result in loss of support for precast floors. Guid-ance is given on this topic, utilising the considerableresearch that was undertaken in the early 1990s.

Frame ConnectionsCommon assembly systems for moment-resistingframes are examined in detail and their advantages anddisadvantages noted. Extensive reference is made tolaboratory tests and illustrations of recent projects.Most are “monolithic systems” that seek to reproducethe essential features of cast-in-place systems. Generalindications are that with proper care in detailing, and in

Introduction • 3

construction, adequate performance is achieved by allmethods.

Structural Wall ElementsThese are examined under two headings: “monolithic”and “jointed” wall systems. “Monolithic” systemsseek to reproduce a cast-in-place condition with strongjoints between elements. “Jointed” wall systems, onthe other hand, behave as a discrete number of precastelements with ductile connections.

Examples of details and applications of monolithicwall systems are given and a cautionary note includedregarding the use of jointed wall systems.

DiaphragmsCast-in-place diaphragms generally provide a com-forting degree of integrity and continuity. Precastconcrete diaphragms rely on connections between in-dividual elements, with typically only reinforced con-crete topping providing continuity. As a result, dia-phragms incorporating precast elements need specialattention to the functions they are required to perform.These are analysed and described and include trans-mission of shear, and resistance to volume changes.

Guidance is given on detailing for both the Serviceabil-ity and Ultimate Limit states.

Some potential dangers are described, especially inrelation to hollow-core floors where there is a reducedshear area available at mid-depth. Beam elongation ishighlighted as a further factor that can reduce theeffectiveness of connections between floor diaphragmsand supporting beams.

Grouted and Welded BarsTechniques of anchoring bars are vital to the success ofjoints between precast members. Detailed guidance isgiven in all common methods, with particular empha-sis on practical aspects. Advice is given on groutselection and on the large number of factors thatinfluence it. Techniques for successful grouting invarious common situations are described, includingvertical, horizontal and sloping bars. A special sectionis devoted to grouting in specific situations such asbeam-to-column joints and beam-to-beam joints.

Emphasis throughout is on achieving necessary qual-ity, with a separate section on quality assurance.

A comprehensive chapter giving guidance on the weld-ing of reinforcing bars is included. Welding, whenproperly controlled, is a practical method for joints inprecast members. Special care is needed to account fortolerances, varying materials, site conditions and cor-rosion protection.

Embedded Steel ConnectorsCommon types are described and their importance ofdetailing stressed. Many details are seen to be particu-larly sensitive to variation in component dimensionsand the proximity of other connectors. Attention to fireresistance and corrosion protection of such connec-tions is emphasised.

TolerancesClose control of dimensions has allowed successfuljointing of precast concrete members. Production,erection and interface tolerances are described andrecommended values given. It is important to recognisethat the tolerance values needed to achieve many of theprecast systems need to be much tighter than thosequoted. It is vital that designers make it clear whattolerances are required to meet design assumptions sothat fabricators and erectors are fully aware of theimplications. Close co-operation between designers,fabricators and erectors at an early stage is essential.

1.5 ConclusionsResearch carried out since 1991 has generally con-firmed the integrity of the procedure given in the firstedition. However, concern for the effects of tolerancesand possible beam elongation has heightened the needfor attention to these aspects.

The Study Group remains confident that precast con-crete can be used successfully in earthquake resistantstructures. However, it is essential that careful atten-tion is paid to:

• conceptual design;

• detailed design;

• fabrication;

• transport;

• erection;

• jointing;

• durability;

• fire protection;

• workmanship supervision; and

• overall quality assurance.

The quality of construction must justify the designassumptions. Equally, design must acknowledge thepractical constraints.

This publication is intended to assist in providing


consistently safe and economical applications of struc-tural precast concrete, and at the same time allowinnovation in design and construction to continue.

1.6 References

1.1 Park, R. “Review of Code Developments forEarthquake Resistant Design of ConcreteStructures in New Zealand”, Bulletin of NewZealand National Society for EarthquakeEngineering, Vol. 14, No. 4, December 1981,pp. 177-208.

1.2 Code of Practice for General Structural Designand Design Loadings for Buildings, NZS4203:1992, Standards Association of NewZealand, Wellington, 1992, 134 pp.

1.3 Concrete Structures Standards, NZS3101 Parts1 and 2, Standards Association of NewZealand, Wellington, 1995.

1.4 FIP. Guide to Good Practice — CompositeFloor Structures, Fib, May 1998, 58 pp.

1.5 FIP. Precast Prestressed Hollow Core Floors,Thomas Telford, London, 1988, 31 pp.

Floor Unit Support and Continuity • 5

Chapter 2Floor Unit Support and Continuity

2.1 IntroductionThe use of precast concrete flooring units is a popularand economical method of construction in New Zea-land. General design and construction requirementsfor these units are covered in the appropriate NewZealand codes of practice [2.1, 2.2, 2.3]. More detaileddesign requirements, which are often used by manu-facturers of the precast units, are available from over-seas sources [2.4, 2.5, 2.6, 2.7, 2.22].

New Zealand Standards (and some overseas standards)now recognise the widespread use of precast concreteconstruction, and include specific reference to precastconcrete support details. Research in this area is ex-panding however; new issues and newly recommendedsupport details continue to simplify the design andconstruction of precast concrete floor systems.

The emphasis with precast concrete design must al-ways be on constructability, with components detailedto reflect the sensitivity of supports to tolerances and tothe seismic performance of the primary structure. Thischapter seeks to provide guidance to designers, manu-facturers and constructors.

2.2 Types of Support forPrecast Concrete Floor Units

2.2.1 GeneralSupport for precast concrete floor units may be simpleor continuous. Both have their advantages in differingapplications.

Simple support suits long spans, or heavily loadedstructures, where it would be difficult and costly toprovide the required degree of negative moment re-straint at the supports.

Precast flooring support with moment fixity at the endssuits the more general commercial and residential typeof construction, but requires attention to support de-tails to ensure that the required degree of continuity canbe achieved.

Between true simple support and full continuity, adesigner may choose any degree of end continuity. In

making such a choice, however, the designer must beaware of the need to detail the end supports for crackcontrol, as partial end continuity relies on the ability ofthe top reinforcing steel to yield over the continuousends of the precast concrete floor unit. This yieldingmay occur at service loads and may lead to serviceabil-ity or durability problems in some applications.

2.2.2 Simple SupportMany designers and manufacturers assume precastconcrete floor units to be simply supported. While verypredictable in terms of serviceability criteria such ascamber, deflection and vibration, simply supportedspans have less redundancy and require more attentionto the cumulative effects of construction tolerances.Seating lengths, movement at expansion or controljoints, and support on ductile moment resisting framesare particularly critical for simply supported precastflooring units.

The New Zealand Concrete Structures Standard [2.1]requires minimum bearing lengths, as shown inFigure 2.1 (Fig C4.3 NZS 3101 Part 2) to maintain thestructural integrity of precast flooring systems. TheStandard also requires minimum longitudinal tensionreinforcement across the end supports of some precastfloor slabs, as discussed in Section 5.2.4.

Precastfloor member

Support

Unarmoured edge

Bearing length

15 mm minimum

L/180 ≥ 50 mm (slabs)

L/180 ≥ 75 mm beams or ribbed floor

Figure 2.1: Required bearing length at the sup-

port of a member in relation to its clear span

Failures have occurred in New Zealand and overseas inlarge-area car parking buildings. These structures,which are exposed to daily temperature cycles andgenerally constructed with long clear spans, requirecareful attention to support details. References [2.6]


and [2.9] provide guidance on the spacing of controljoints in large area structures. The primary structuremust be designed to ensure that thermal movements,together with creep and shrinkage shortening, are dis-tributed to all control joints and not concentrated at thefew that offer the least resistance to movement. (Referto Appendix C2).

Experience, both in New Zealand and overseas, showsthat split or double columns are more effective thansliding bearings in providing predictable control ofmovement in large-area exposed structures (see Fig-ures 2.2 and 2.3). Sliding bearings, where used, shouldbe detailed with movement-limiting linkage bars.

Bearing pads or mortar seating pads are required totake the concentrated reactions and end rotations of allsimply supported beams, or flooring units. For floorsystems subjected to daily thermal movements, such asoccur in the upper floor of parking structures, bearingstrips that allow for differential movement between theprecast floor unit and the support are essential.

Crack control in the cast-in-place topping concreterequires careful detailing if the surface will be visiblein the finished structure. The placement of construc-tion joints and saw cuts must follow the anticipatedpattern of cracking, and joints should be sealed forcorrosion protection where they will be exposed to theweather or to chemical attack.

2.2.3 Continuous SupportContinuity in precast concrete floors is used to:

• enhance fire resistance;

• limit deflection;

• control floor vibration;

• increase the load carrying capacity of the floor;

• provide cantilever support; and

• resist diaphragm forces.

The degree of continuity possible and its effectivenessvaries with the type of precast concrete flooring unitsand their method of support.

Flange-supported double tees, for example, cannoteasily achieve continuity, while to achieve the desiredeffects from continuity in other types of precast unitsrequires consideration of support details, constructiontolerances and creep and shrinkage movements in boththe precast units and the topping concrete.

Flexural continuity, achieved by means of reinforcingbars placed in the topping concrete at the ends of theprecast concrete flooring units, requires an adequatedepth of topping concrete in which to embed the barsand maintain the minimum cover. Experience showsthat 16 mm diameter bars can be effectively anchoredin 65 mm thick topping, but research is required todetermine safe design limits for bar size and toppingthickness.

Horizontal shear between the precast member and thecomposite topping concrete at continuous supports can

Figure 2.2: Split column detail for control joint in

large area exposed parking structure

Figure 2.3: Sliding support at a control joint that

has moved more than anticipated


be adequately resisted, under static loads, by meetingthe interface roughness requirements set out in theNew Zealand Concrete Structures Standard [2.1]. How-ever, recent work by Herlihy and Park [2.24] andOliver [2.25] has shown that a more ductile sheartransfer mechanism, or an alternative support mecha-nism, is required at the ends of precast floor slabssupported on beams that could be subject to seismiceffects as discussed in Appendix A.

2.3 Precast Floor Unit SeatingAdequate support of precast concrete floor units is oneof the most basic requirements for a safe structure.

The following factors must be considered when deter-mining required seating lengths.

2.3.1 TolerancesAs discussed in Chapter 8, in the design of seatinglengths allowance must be made for tolerances arisingfrom:

• the manufacturing process;

• the erection method; and

• the accuracy of other construction.

2.3.2 Construction MethodologyIf propping is to be avoided during construction, the

specified seating length must be increased to allow forthe cumulative effect of the various tolerances previ-ously mentioned. A seating length that is too short canlead to failure during construction, or a bearing failurein the completed structure. For systems that are proppedduring construction, the effect of an unintentionallyshort unit may not be serious, as cast-in-place concreteand additional support reinforcement can enable thegap to be bridged (see Figure 2.14).

2.3.3 Transverse Load DistributionSome ribbed systems, such as double tees, have lim-ited capacity to redistribute loads transversely in theevent of damage to support concrete under a rib. As aresult of this sensitivity, NZS 3101 [2.1] requiressupport lengths for ribbed floors to be increased by 25mm (see Figure 2.1).

2.3.4 Volume Changes and ThermalEffectsVolume changes as a result of concrete shrinkage,creep and temperature effects, may cause axial short-ening which reduces the actual seating lengths. Crack-ing may also result (see Figure 2.4). Reference 2.9 andAppendix C2 give appropriate methods for calculatingthe amount of this movement.

Thermal effects are most significant on a floor or roofexposed to the sun. Appreciable hogging due to differ-ential temperature effects can occur. If allowance is notmade for these displacements or induced actions, dam-

movement crack in topping

cast-in-place reinforced concrete topping

spalling at end of precast concrete unit

spalling of cover concrete

precast concrete beam

Figure 2.4: Damage of bearing seat due to movement


age can occur [2.11] as illustrated in Figures 2.5 and2.6.

Temperature gradients obtained from measurementsof bridge structures may be used in design [2.12]. InNew Zealand, the top levels of car parking buildingshave proved particularly vulnerable to differential tem-perature problems due to upward camber. Bearingpads that allow slip to occur without edge spalling arerecommended.

2.3.5 Seismic EffectsSeismic actions on a building can be expected toadversely affect the support conditions of precast con-crete floor units in the following two ways.

(i) Loss of Beam Cover Concrete In PlasticHinge Regions

Particular care needs to be given to rib and infill precastconcrete systems due to their limited ability to redis-tribute loads laterally if the support to one or more ribsis lost due to the spalling of support beam coverconcrete in “plastic hinge regions” (zones of plasticityin the beams). The reinforced concrete topping, whichrelies on tensile bond between the topping and theprecast unit, cannot be expected to transfer shear forcefrom the precast flooring to the supporting beam. Inthese beam regions it is recommended that concrete orsteel corbels, or top flange hung details that do not rely

Figure 2.6: Spalling at double tee support due to

differential temperature bowing of a parking

structure exposed to the sun

cast-in-place reinforcedconcrete topping

precast concretebeams

Elevation

Detail

back of beam spalls

Rotation induced by differentialtemperatures may split columnor bearing area of beam.

Figure 2.5: Structural damage due to differential temperature


on cover concrete, should be used unless it can beshown that transverse load shedding can occur to ribsor webs outside the plastic hinge region. Arch, ormembrane, action provides an effective mechanism totransfer load to adjacent ribs where there are stirrupsconnecting the precast rib to the topping.

(ii) Elongation of Ductile Moment ResistingBeams

Seismic actions can cause elongation of moment resist-ing beams due to plastic extension of the longitudinalreinforcing steel. Refer to Appendix A1 for details.

Both (i) and (ii) above relate to structural movementsassociated with significant seismic damage. It is rec-ommended that a load factor for gravity loads of notless than 1.1 be used for this design load case.

Alternatives to providing the necessary seating lengthsare:

• to provide hanger bars (as shown in Figures 2.7 and2.8) of sufficient ductility to accommodate theanticipated movement; and

• to investigate other load paths such as transverseload distribution or catenary action of toppingreinforcement if the precast floor units are ad-equately tied into the topping concrete diaphragmby reinforcing steel detailed to accommodate theanticipated displacement.

Recent research [2.23, 2.24, 2.25] has provided simplesolutions to this potential problem.

2.3.6 Appropriate SeatingIt is essential that floor systems do not collapse as aresult of any imposed movements that reduce seatinglengths or cause spalling of seating. Test results [2.13]indicate that top reinforcement in slabs cannot beexpected to provide an adequate load path for supportforces (see Figure 2.9 (a)).

It is therefore recommended that special reinforcementbe provided where the specified seating length (ateither end of the precast floor slab), minus tolerancesand allowances for volume changes and edge spalling,is less than the calculated increase in span due toelongation associated with plastic hinging of adjacentbeams, as may occur in a beam forming part of a ductileframe during a severe earthquake (Appendix A).

Special reinforcement required to provide an effective

200 600

R10 bars asrequired forvertical shear

continuityreinforcement

packerpacker

extent ofcast-in-place

concrete

cast-in-placeconcrete

hollow-coreunit

Fig 2.8: Remedial technique for lack of bearing length

with Type 2 support system

Figure 2.7: Support bars for precast concrete

floor units


30°

Vi

Vi

Vi = As fy sin30°

special reinforcing

precast concrete unit

beam

special reinforcing


beam

topping reinforcement cannot transfershear force in the event of lossof bearing


loss of bearing

beam

top bars can transfer shear forceif bearing to the precast unit is lostprovided the bars are enclosed byhanger stirrups located close to theend of the precast unit

(a) Inability of topping to transfer shear stress

(b) Alternative special reinforcing tosupport precast concrete floor unitsin the event of loss of bearing

(c) Kinking of the special reinforcing to provide aload path from the precast concrete floor unitto the beam

Figure 2.9: The use of special reinforcing steel to prevent collapse

of precast concrete floor systems


load path by kinking in the event of loss of bearing tothe precast concrete unit is illustrated in Figures 2.9 (b)and (c).

With concrete masonry construction, ribbed precastconcrete flooring units should be seated on the infillconcrete, not the shell of the block. Flat slab or hollow-core flooring may be seated on the block face thicknessbut temporary props are recommended to avoid failureduring construction, if the top of the block wall has notalready been fully grouted.

The use of packing or bearing strips between theprecast concrete member and its support is recom-mended when damage to either the precast concreteunit or the support is likely to occur. Examples ofwhere this recommendation should be followed are:

• leg or flange supported double tees;

• heavily loaded beams;

• precast units bearing on steel supports;

• uneven bearing surfaces of cast-in-place concreteor reinforced concrete masonry; and

• any precast flooring units subjected to daily tem-perature variations.

Bearing strips or packing may consist of the following.

• Sand cement mortar, suitably plasticised and ofsufficient consistency to adequately support theprecast unit without squeezing out of the joint.Shims may be required to maintain the correct jointthickness. Note that sand-cement mortar is not suit-able for floor slabs subjected to daily temperaturemovements (such as the top level of car parkingstructures).

• Neoprene rubber pads.

• Proprietary plastic shims or strips (recommendedfor slabs subjected to daily temperature move-ments).

• Epoxy mortar.

2.4 Seating Details for PrecastConcrete Hollow-core and FlatSlab Floor UnitsThe types of support for precast concrete hollow-core orflat slab flooring units seated on beams can be dividedinto the three groups as shown in Figure 2.10. Thedifferences between these types are the depth of thesupporting beam prior to the cast-in-place concrete beingpoured. Each type is examined in the following sections.

2.4.1 Type 1 Support (Figures 2.11 and2.12)

Advantages of the Type 1 Support

(i) As the top surface of the first stage of the beam isat the level of the soffit of the precast concrete floorunit, cast-in-place concrete can easily be placedbetween the floor units and into recesses at the endsof hollow-core units. This enhances the shear ca-pacity of hollow-core units.

cast-in-place concrete

Type 1

Type 2

Type 3



precast concretehollow-core

floor unit


floor unit


floor unit

Figure 2.10: Types of support of precast

concrete hollow-core floor units by precast

concrete beams


(ii) The presence of well compacted cast-in-place con-crete against the bottom of the precast concretefloor unit enables reliable negative moment conti-nuity to be developed.

(iii)Placement of reinforcement to provide ductilityand load capacity following loss of slab support dueto seismic effects is simplified (Figure 2.13).

(iv)With the Type 1 method of support for hollow-coreunits, problems of construction tolerances can beeasily overcome on site. The method illustrated inFigure 2.14 has been successfully used for lengthdiscrepancies of up to 100 mm with the detail asconstructed being checked by load tests as requiredby NZS 3101. Studies have confirmed the ad-equacy of this method [2.14, 2.21, 2.24]. Providedthat extra topping steel (saddle bars) plus steel“paper clip” bars in the broken-back voids are used,the shear strength is similar to the case with a 40mm seating length and no special detailing. In thesetests failure of the precast concrete units, whichwas initiated by debonding of the prestressing

strands at the ends of the units, occurred at anacceptable load level.

Disadvantages of the Type 1 Support

(i) Due to the reduced depth of beams at the stagewhen precast floor units are erected, more proppingis generally required than with other support types.If propping is not desirable, the adequacy of thestrength and stiffness of the beam during the con-struction phase must be considered by the designer.

(ii) Additional reinforcing steel may be required at thetop of precast beam units to provide negative mo-ment capacity over the supporting props duringconstruction.

(iii) Additional negative moment reinforcement will berequired to resist the dead load continuity momentsat the beam supports after the props have beenremoved (based on the full-depth composite beam).This additional beam strength may require an in-crease in the column strength to ensure that beamhinging is the ductile mechanism for seismic resist-ance.

2.4.2 Type 2 Support

Advantage of Type 2 Support

(i) With this form of support (See Figures 2.15 and2.16), the precast concrete beam depth may extendto the top of the floor unit. Because the beam depthis greater when the floor units are erected, lesspropping is required.

Disadvantage of Type 2 Support

(i) If the vertical gaps between the precast beam andfloor units are too small, there may be difficulty in

precast concrete hollow-core unit

bedding strip



Figure 2.11: Type 1 precast beam support system

Figure 2.12: Example of a Type 1 precast beam

support system


65

200

200 600

“Paperclip” seismictie reinforcement, twoper 1.2m wide slab

Support beam formingpart of a two way ductilemoment resisting frame structure

1.2m wide precasthollow-core floor

D12 orD16

Serviceability deflection andcrack control reinforcement

barrier or “dam”in the cell

barrier or “dam”in the cell

R10 plain round bar

Figure 2.13: Recommended final tie detail for hollowcore slabs in ductile moment-resisting frame

construction. Herlihy and Park [2.24]

Section

break back top ofcores as required

R10 hairpin bar, 2 per slab,or as required by shear frictioncalculations

Plan

PROP

200 600200 hollow-core unit

extent of cast-in-placeconcrete

D10

break back top ofeach core

R10 paperclip 2 per slab or asrequired by shear friction orcartenary calculations

≥ 0.75 x beam width

Barrier or“dam”

Figure 2.14: End support detail when seating is inadequate [2.26]


placing concrete both in the gap and in the recessesof hollow-core units. This can reduce the shearstrength of the precast flooring and prevent thedevelopment of negative bending moment actionsin the floor units resulting in a detrimental influ-ence on serviceability. If this system is to work asintended, the precast flooring units must be suit-ably detailed. It is recommended that some of thecores of hollow-core flooring units are broken outto enhance the shear strength.

(ii) There is a construction difficulty if the accumu-lated tolerances reduce the actual bearing length atthe ends of precast concrete floor units. The siteremedial method, which is suitable for Type 1, isnot possible. An alternative shown in Figure 2.8could be adopted. Adequate ductility in the tie bardetail is required for floor slabs seated on beamsthat could be forced apart by seismic actions asdescribed in Appendix A1 [2.23]. This detail is

similar to one that is recommended by the FIP [2.5].The bar size is designed for the ultimate shear load.

(iii)Beam stirrups are anchored in the thin, cast-in-placeconcrete topping. Horizontal shear stresses due toflexure must be transferred from the top of theprecast beam into the topping concrete. Some de-gree of anchorage of stirrups is also necessary. Arelationship between stirrup diameter and mini-mum topping thickness has been suggested [2.15].However, the recommendations were based onmonotonic loading tests and may not be conserva-tive for plastic hinge regions in seismic resistingbeams. There is need for further research.

2.4.3 Type 3 SupportFigures 2.17 and 2.18 show this support system, whichmay be used for perimeter beams or lift and stairwells.No edge formwork for slab topping concrete is re-quired. As for Type 2, it is recommended that the coresof hollow-core flooring units be broken out to ensureadequate compaction of cast-in-place concrete. Forthis reason, even more attention must be paid to provid-ing adequate seating. If, for example, a short unit is notto be rejected, some type of structural steel bracket,such as that shown in Figure 2.19, is required.

2.5 Types of Support forRibbed UnitsIn general, with the exception of cantilever support, itis unrealistic to assume that moment continuity forserviceability limit states can be achieved at the sup-ports of ribbed flooring units. Combined creep andshrinkage shortening in the highly prestressed rib can

continuityreinforcement

precast concrete hollow-core unit



Figure 2.15: Type 2 precast beam support system

Figure 2.16: Example of a Type 2 precast beam

support system


form a gap between the end of the rib and the cast-in-place beam concrete. Epoxy mortar and epoxy injec-tion have been attempted, but because the operationcannot be inspected properly such practice is not rec-ommended [2.16]. The same comments apply to theuse of dry pack mortar: it is difficult to use and hard toverify correct placement.

When the beam structural depth is to be minimised,precast concrete units may be supported by their flanges.Reference 2.6 provides two design methods. Designersshould appreciate the sensitivity of the strength offlange hung details to tolerances. Allowance for thisand other influences affecting precast concrete floorunit seating are discussed in Chapter 8.

2.6 Special DiaphragmRequirements for Hollow-coreFloorsThe New Zealand Concrete Structures Standard [ 2.1 ]allows reduced roughness for the top surface of ma-chine produced extruded hollow-core floor slabs ascompared to other precast concrete floor units.

There is convincing evidence [2.19] that a 5 mmamplitude roughness at the interface of precast ex-truded hollow-core slabs and cast-in-place toppingconcrete is unnecessarily conservative for flexuralshear transfer. This is because the dry concrete mixesand normal curing regimes used for the production of

precast concretebeam

600 200serviceabilityreinforcement

cast-in-place concrete topping

cores broken out

R10 hanger bars asrequired for vertical shearpacker

precast concretehollow core unit

mesh

end of coresto be filled in

extent of cast-in-placeconcrete

Fig 2.17: Type 3 precast concrete beam support system

Fig 2.18: Example of a Type 3 precast concrete

beam support systemFig 2.19 : Steel fixing providing support to

double-tee webs


extruded hollow-core slabs do not cause an accumula-tion of laitance on the top surface of the slabs. A lightbrushing (similar to the U4 class of finish shown inFigure 14 of NZS 3114: 1987 [2.20]) has proven to beadequate to ensure composite action under gravityloads. These comments apply to dry mix extrudedconcrete slabs, but not to conventionally cast flooringunits.

Recent research [2.24] has found that for hollow-corefloor slabs subjected to direct tension forces from beamelongation, as described in Appendix A, the bondbetween topping concrete and the top surface of pre-cast hollow-core floor slabs cannot be relied on totransmit diaphragm forces to the seismic resistingframes or walls. The details shown in Figures 2.20 and2.21 are recommended as a means of transferringdiaphragm forces.

Shear ties, when required to satisfy code requirements[2.1], can be placed in the shear keys between adjacenthollow-core units or in holes cut into selected cores.Where the hollow-core flooring units are supported onbeams that form part of a ductile moment-resistingframe, beam elongation (as described in Appendix A)may cause separation of the topping from some of thehollow-core units. Shear ties anchored in the slab coreswill provide a more dependable connection in this case.

Calculations on the horizontal shear capacity of toppedextruded hollow-core floor diaphragms usually ignorethe capacity of the hollow-core units and take all thediaphragm shear on the topping concrete. Further re-search on the shear capacity of topped and untoppedhollow-core floor diaphragms is required before a lessconservative approach can be taken.

2.7 Overseas PracticeThe Prestressed Concrete Institute (PCI) [2.4] hasproduced details of connections for hollow-core units.In most cases there is no structural topping shown in thePCI details, nor any flexural continuity provided at theends of units. There also appears to be a preference for

full-depth precast beams. Horizontal and draped rein-forcement is used to provide continuity for diaphragmaction and it is grouted in the gaps between units asshown in Figure 2.20.

European practice is similar [2.5, 2.22]. The steel isanchored either into the cores, which have been brokenout at the top, or the joints (see Figure 2.21).

In both the PCI and European details, anchorage ofcritical tie reinforcing bars in the joint between slabsmay create a potential problem in structural frames thatcould undergo significant beam elongation due to theformation of plastic hinges, as outlined in Appendix A.Tie bars anchored in the cores of hollow-core slabs(lapped with the pretensioning tendons in the slabs)appear to offer a more dependable load path [2.24, 2.25].

2.8 Related Considerations

2.8.1 Fire ResistanceManufacturers of precast concrete flooring componentscan produce units with a Restrained or an UnrestrainedFire Resistance rating [2.17]. Floor slab manufacturer’sproduct literature should indicate the method of achiev-ing the required fire resistance. In general terms (fornormal weight concrete), the lighter the self weight of aflooring system, the more it relies on restraint to achievefire resistance. The designer using a Restrained Fire-rated floor system must check that the structure canprovide the required degree of restraint.

Restraint may be provided by end moment continuity,resistance to thermal expansion, or a combination ofboth of these effects. Reference 2.17 sets out themethods of calculating the required degree of restraint.The ability of the structure, and the floor unit support,to provide this restraint must then be checked. Exteriorbays, with moment continuity only possible at one endand with limited resistance to thermal expansion, areparticularly critical.

The use of heavier precast concrete units with Unre-

topping if requiredreinforcement grouted inlongitudinal joints betweenfloor units

bearing strip

concrete beam

grout orcast-in-placeconcrete

topping if requiredreinforcement grouted inlongitudinal joints betweenfloor units

bearing strip

concrete beam

grout orcast-in-placeconcrete

Figure 2.20: Details of continuity steel (from Reference 2.4)


strained Fire Resistance ratings avoids the problems ofrestraint in the exterior bays of a structure. Flange-supported double tees must also be designed as unre-strained units (in fire engineering terms) as the ther-mally induced forces located above the centroid of theconcrete section can significantly reduce the fire resist-ance rating of the floor system.

2.8.2 Deflection and Vibration ControlThe use of structural continuity at the supports ofprecast concrete floor units requires judgment andcaution, at serviceability limit states. Highly pre-stressed narrow webbed units, such as leg supporteddouble tees and ribbed multi-piece floors, can undergoappreciable creep shortening that removes the endbearing essential for continuity at service loads. Thishas more effect on service load deflection than onvibration, possibly due to sliding friction at the ends ofthe unit providing damping.

Dry mix extruded floor slabs are not normally as highlystressed and so do not creep or shrink as much asconventionally cast ribbed units. Moment continuity istherefore more effective for the control of deflection atservice loads in extruded hollow-core slabs.

Vibration in precast concrete floor systems has notbeen a serious problem in New Zealand. As spans areextended however, and as manufacturers respond to

designers’ needs for lighter weight floors, vibrationcharacteristics need to be checked.

The critical parameters for human perception are fre-quency, amplitude and damping. Reference 2.18 pro-vides guidance on recommended values. Typical of-fice construction without partitions can be expected toprovide 3% to 4% of critical damping.

For commercial use, floor frequencies greater than6 Hz are recommended while for gymnasiums or areasused for aerobic exercises frequencies of 2.5 Hz ormultiples of 2.5 Hz should be avoided.

2.9 Recommendations• Precast concrete floor unit support details must

reflect practical achievable tolerances and the an-ticipated seismic performance of the supportingstructure (Sections 2.3.1 and 2.3.6).

• Simply supported units require specifically de-tailed design at movement control joints (Section2.2.2).

• Moment continuity support requires considerationof creep and shrinkage. Topping thickness must beadequate to provide reinforcement embedment forcontinuity (Section 2.2).

slit forconcreting

tie bar

tie bar

tie bar

tie bar tie bar

tie beam

SectionSection

PlanPlan

Figure 2.21: Provision of continuity or tie steel (from Reference 2.5)


• Precast concrete floor unit seating should followthe manufacturer’s recommendations. If reducedseating lengths are required, the manufacturers ofthe precast components and the constructor of thebuilding must take special precautions to ensuresafety during construction. Reduced seating mayalso require additional reinforcement to preventcollapse in the event of gross seismic damage to theprimary structure (Section 2.3).

• Suitable bearing material is required to preventconcrete spalling where precast concrete units seaton rigid supports (Section 2.3.7) or where dailytemperature movements occur (Sections 2.2.2 and2.3.4).

• Type 2 or Type 3 supported hollow-core or flat slabfloors require special attention to tolerance to en-sure adequate end support and moment continuity.

• Support for ribbed units requiring moment conti-nuity must be detailed to allow for constructiontolerances, creep and shrinkage effects, and toensure the ease of placement of well consolidatedconcrete in critical parts of the support.

• If moment continuity at supports is used to limitdeflection or to reduce the human perception ofvibration, special calculations are required (Sec-tions 2.8.1 and 2.8.2).

2.10 References2.1 Concrete Structures Standard, NZS 3101 Parts

1 and 2, Standards Association of NewZealand, Wellington, 1996.

2.2 Specification for Concrete Construction, NZS3109, Standards New Zealand, Wellington,1997.

2.3 Code of Practice for General Structural Designand Design Loadings for Buildings, NZS4203, Parts 1 and 2, Standards New Zealand,Wellington, 1992.

2.4 PCI Manual for the Design of Hollow-CoreSlabs, Prestressed Concrete Institute, Chi-cago, 1985.

2.5 FIP. Precast Prestressed Hollow Core Floors,Thomas Telford, London, 1988.

2.6 PCI Design Handbook - Precast and Pre-stressed Concrete, Prestressed ConcreteInstitute, Chicago, 3rd Edition, 1985.

2.7 Manual for Quality Control for Plants andProduction of Precast and Prestressed

Concrete, Prestressed Concrete Institute,Chicago, 1985.

2.8 “Proprietary concrete floor systems - interpreta-tion of New Zealand Code Requirements”,Technical Report No. 1, New ZealandConcrete Society, August 1981.

2.9 Metric Design Manual: Precast and PrestressedConcrete, Canadian Prestressed ConcreteInstitute, Ottawa, 1987.

2.10 Structural Use of Concrete, BS 8110 Parts 1and 2, British Standards Institution, MiltonKeynes, 1985.

2.11 Fintel, M and Ghosh, S K. “Distress due to suncamber in a long-span roof of a parkinggarage,” Concrete International, July 1988,pp. 42-50.

2.12 Transit New Zealand. Bridge Manual, TransitNew Zealand, Wellington, 1994 and Amend-ments.

2.13 Hawkins, N M and Mitchell, D. “Progressivecollapse of flat plate structures”, ACI Jour-nal, July 1979, pp. 775-807.

2.14 Yap K K. Shear Tests on Proprietary Pre-stressed Voided Slabs using Various EndSupport Conditions, Report 5-85/3, CentralLaboratories, Ministry of Works and Devel-opment, Lower Hutt, 1985.

2.15 Mattock A H. “Anchorage of stirrups in a thincast-in-place topping”, PCI Journal, Nov-Dec. 1987, pp. 70-85.

2.16 Additional Notes for Specification of StahltonFloors, Stahlton Flooring (NZ), Auckland,March 1988.

2.17 Design for Fire Resistance of Precast Con-crete, Prestressed Concrete Institute, Chi-cago, 1977.

2.18 Allen, D E, Rainer, J H and Pernica, E.“Vibration criteria for long-span concretefloors”, ACI Special Publication SP-60:Vibrations in Concrete.

2.19 Composite Systems without Roughness,Concrete Technology Associates TechnicalReport 74 B6, Tacoma, Washington.

2.20 Specification for Concrete Surface Finishes,NZS 3114:1987, Standards Association ofNZ, Wellington 1987.

2.21 Proprietary Prestressed Voided Slabs Using


Various End Support Conditions, Report 5-85/3, Central Laboratories, Ministry ofWorks and Development, Lower Hutt, 1985.

2.22 FIP Prefabrication Commission. Planning andDesign Handbook on Precast BuildingStructures, SETO, London, 1994.

2.23 Mejia-McMaster, J C and Park, R. “Tests onspecial reinforcement for the end support ofhollow-core precast concrete floor units”,PCI Journal, Vol. 39, No 5, pp. 90-105,1994.

2.24 Herlihy, M D and Park, R. “Detailing precastflooring systems to survive loss of support”,

New Zealand Concrete Society ConferenceProceedings, October 1996, pp 130-139.

2.25 Oliver, S J. The Performance of ConcreteTopped Precast Concrete HollowcoreFlooring Systems Reinforced with andwithout Dramix Steel Fibres under SimulatedSeismic Loading, Master of EngineeringProject, Department of Civil Engineering,University of Canterbury, 1998.

2.26 Blades, P S, Jacks, D H and Beattie, G J.Investigation of the Influence of the EndSupport Condition on the Shear Strength ofPrestressed Voided Slabs, Central Laborato-ries, Lower Hutt, 1990.

Frame Connections • 21

Chapter 3Frame Connections

3.1 IntroductionExperience of earthquakes, and extensive laboratorytesting, have shown that well-designed, detailed andconstructed cast-in-place continuous reinforced con-crete frames perform very well during severe earth-quakes. Moment-resisting frames incorporating pre-cast concrete members, designed to be ductile andproviding the primary earthquake resistance, have nothad the same extensive laboratory testing. The use ofprecast concrete in moment-resisting frames wasshunned for many years in New Zealand, due mainly tothe observation of poor performance of connectiondetails between the precast elements during majorearthquakes in many overseas countries. However,moment-resisting frames incorporating precast con-crete members have become widely used in NewZealand since the 1980s [3.1].

Confidence in the use of precast concrete elements inmoment-resisting frames in New Zealand has requiredthe use of capacity design to ensure that yielding duringa major earthquake occurs only in the preferred ductileregions of the frame. Also, moment-resisting framescontaining precast concrete elements have been de-signed and constructed so as to possess stiffness, strengthand ductility similar to that of cast-in-place concretemonolithic construction. In other words, monolithicconstruction is emulated [3.1].

The basic challenge in the design of building structuresincorporating precast concrete elements for earthquakeresistance is in finding an economical and practicalmethod for connecting the precast elements together sothat the seismic performance will be as for a monolithicstructure. If the connections between the precast ele-ments are placed in critical regions, such as in potentialplastic hinge zones, the design approach is to ensurethat the behaviour of the connection region approachesthat of a monolithic cast-in-place structure. Possiblebrittle connections between members should be madeoverstrong and placed away from the critical regions.Reinforcing details and structural configurations shouldbe arranged to ensure that potential plastic hinge re-gions are away from the jointing faces of precastmembers if possible [3.1].

The general trend in New Zealand for reinforced con-crete framed buildings incorporating precast concreteis to design the perimeter frames with sufficient stiff-

ness and strength to resist most, if not all, of the seismicloading. The interior columns of the building thencarry mainly gravity loading and can be more widelyspaced. References 3.2 to 3.8 give details of severalbuildings designed in New Zealand since the 1980s,which incorporate significant quantities of precastconcrete in their frames and floors.

The New Zealand standards for concrete design cur-rent in the 1980s [3.9, 3.10], like the design standardsof many countries, contained comprehensive designprovisions for the seismic design of cast-in-place con-crete structures, but did not have seismic provisionscovering all aspects of precast concrete structures. Therevisions of these standards published in the 1990s[3.11, 3.12] contain more design provisions for theseismic design of structures incorporating precast con-crete as a result of the significant research and develop-ment conducted in New Zealand during that decade.

A number of possible arrangements of precast concretemembers and cast-in-place concrete forming ductilemoment-resisting multi-storey reinforced concreteframes, commonly used for strong column-weak beamdesigns, have been identified [3.1] and are shown inFigure 3.1. The three systems illustrated are describedbelow. These systems can also be used in a modifiedform for one- or two-storey frames where strong beam-weak column design is permitted. The aim has been todesign the systems so to achieve behaviour as for amonolithic structure. Ductile frames are designed us-ing the capacity design procedure and according to theprovisions for totally cast-in-place concrete structures,or alternatively the frames can be designed using thelimited ductility procedure [3.11, 3.12].

This chapter discusses the possible arrangements ofprecast members in moment-resisting frames, com-ments on some design aspects and test results, andconcludes with recommendations.

3.2 System 1 - Precast BeamUnits between Columns

3.2.1 Construction DetailsAn arrangement involving the use of precast membersto form the lower part of the beams is shown in Figure3.1(a). The precast beam elements are placed between


Figure 3.1: Arrangements of precast members and cast-in-place concrete for constructing

moment-resisting reinforced concrete frames [3.1]

cast-in-place concreteand steel in column

cast-in-place concreteand top steel in beam

precast beam unitprecast beamunit

MIDSPAN

cast-in-placejoint

precast beam unit

cast-in-place concreteand top steel in beam

MIDSPAN

precast orcast-in-placecolumn unit

precast orcast-in-placecolumn unit

mortar or grout joint

(b) System 2 - Precast Beam Units Through Columns

(a) System 1 - Precast Beam Units Between Columns

vertical leg ofprecast T-unit

mortar or grout joint

MIDSPAN

cast-in-placejoint

precast T- unit

(c) System 3 - Precast T-Units

Notes:

1 Shading denotes precast concrete

2 Reinforcement in the precast concrete is not shown

3 Capacity design is used to ensure that the flexural strengthof the columns is suitably greater than the flexural strengthof the beams, so that in the event of a severe earthquakeplastic hinging occurs in the beams rather than in the columns.


columns and seated on the cover concrete of the previ-ously cast-in-place reinforced concrete column belowand/or propped adjacent to the columns. In some casesthere may be two precast beam elements per span,requiring additional props, with a cast-in-place joint atmidspan where longitudinal beam bars are spliced. Aprecast concrete floor system is placed, seated on thetop of the precast beam elements and spanning be-tween them. Reinforcement is then placed in the top ofthe beams, over the precast floor and in the beam-column joint cores. The topping slab over the floorsystem and the beam-column joint cores are cast. Thenext storey height of columns is then prepared.

3.2.2 Some Design Aspects

Anchorage of Bottom Longitudinal Bars

A possible difficulty with this connection detail is thatthe bottom longitudinal bars of the beams, protrudingfrom the precast beam elements, need to be anchored inthe joint cores. Hence the column dimensions need tobe reasonably large to accommodate the required de-velopment length and to reduce the congestion causedby the hooked anchorages.

The previous concrete design standard [3.10] requiredthat beam bars that are terminated at an interior columnshould be passed right through the core of the columnand be terminated with a standard hook immediatelyoutside the ties around the perimeter of the columncore.

However, full-scale laboratory cyclic load tests in NewZealand [3.13 - 3.16], showed that anchoring all thebottom bars within the joint core with a hooked lap did

not affect the seismic performance of the joint. In thespecimens tested all the bottom flexural steel wasterminated in standard 90˚ hooks in the far side of thejoint core.

Hence in the 1995 revision of the concrete designstandard NZS3101:1995 [3.12], this design provisionwas amended to permit the anchorage detail within thejoint core of interior columns shown in Figure 3.2.

The anchorage of beam bars within the joint core inSystem 1 can then be designed using the same coderules as for anchorage in an exterior column. Theanchorage is considered to commence at one-half ofthe depth of the column or 8db from the face at whichthe bar enters the column, whichever is less, where dbis the bar diameter.

Location of the Cast-in-place ConcretePrecast-concrete Cold Joint

A further possible problem is that the critical section ofthe potential plastic hinge region in the beam occurs atthe vertical cold joint at the column face between thecast-in-place concrete of the joint core and the precastbeam. Figures 3.3 and 3.4 show examples of this typeof construction using System 1 (in Figure 3.1).

In some of the full-scale laboratory tests [3.10, 3.11], itwas found that by the end of the tests vertical slidingshear displacements were occurring at the cold joints atthe column faces. In another test [3.15], no movementsoccurred at the cold faces at the ends if the precast beamswere seated on 30 mm of cover concrete of the columnbelow. Tolerances in normal construction may meanthat this seating is reduced. Hence it is recommended

≥

top bars slidinto place

column

cast-in-placejoint concrete

precast beam

hc

dh + 8db or dh + 0.5hcwhichever is less but not less than 0.75 hc

(hooks to terminate at the far side of the joint core)

similarly for the “dotted” bar

Figure 3.2: Hooked lap of bottom bars within joint core for System 1


Figure 3.3: Frame incorporating precast elements in the beams between columns (System 1)

Figure 3.4: A detail of the joint region for a System 1 frame [3.15]


[3.12] that vertical shear should be transferred acrossthese interfaces by shear friction or by mechanical keys.That is, the end of the precast beam should be clean, freeof laitance and intentionally roughened to a full ampli-tude of not less than 5 mm. Alternatively, the key couldtake the form of either reinforced concrete projectionsfrom the end of the precast beam unit into the cast-in-place concrete of the joint core, or otherwise recesses inthe end of the beam into which the cast-in-place jointcore concrete could project.

The option of using such a mechanical key would makethis aspect of New Zealand practice similar to overseaspractice. For example, Figure 3.5 shows joint detailsfrom China, Japan and Romania where, in each case,shear keys have been provided at the vertical interfacesbetween the precast and cast-in-place concrete. Never-theless, intentional roughness as described above shouldalso provide adequate shear transfer.

Horizontal Interface between the PrecastBeam Unit and the Topping Concrete

It is recommended that the top surface of the precastbeam unit should be clean, free of laitance and inten-tionally roughened to a full amplitude of not less than5 mm or mechanically keyed.

Reduction in Design Negative Moments

When using composite beams jointed at column loca-tions, advantage can be taken of the presence of deadload during construction to reduce demand for nega-tive moment reinforcement. Significant savings can bemade by this design process especially in relation tojoint core shear reinforcement. A description of thistechnique, design issues and details are given inAppendix B1.

mortar bed


precast concretebeam unit

precast concretecolumn

Detail From Japan

precast column

precast concretebeam unit

Detail From China

cast - in placeconcrete

precast concreteteam unit

cast - in placeconcrete column

Detail From Romania

precastcolumn

Figure 3.5: Mechanical keys used at the vertical interface between the precast

and cast-in-place concrete


3.3 System 2 - Precast BeamUnits through Columns

3.3.1 Construction DetailsAn arrangement that makes more extensive use ofprecast concrete, and avoids the placing of cast-in-place concrete in the congested beam-column jointcore regions, is shown in Figure 3.1(b). The success ofthis system depends on tighter than normal tolerances.The reinforced concrete columns can be either precastor cast-in-place to occupy the clear height betweenbeams. The precast portions of the beams extend fromnear midspan to midspan, and hence include within theprecast element over the columns the complex arrange-ment of joint core hoop reinforcement, which is fabri-cated in the precast factory. The precast portions of thebeams are seated on the concrete column below with asuitable jointing material between and propped forconstruction stability.

Protruding longitudinal column bars from the rein-forced concrete column below pass through preformedvertical holes in the precast beam element and extendabove the top surface of the element. The holes in theprecast beam elements are preformed using corrugated

steel ducting and are grouted with the horizontal inter-face gap, as discussed in Section 3.6, after the columnbars have been passed through.

Protruding bottom bars of the precast beam elementsare lapped in the cast-in-place joint at midspan or,alternatively, they can be connected by welding to steelplates that are bolted together in the cast-in-place joint.A precast floor system is seated on top of the precastbeam elements and spanning between them. Rein-forcement is then placed in the top of the beam andtopping slab, and cast-in-place concrete is poured.

One variation on this system is for the top steel of thebeam to be cast within the precast beam section. Thisis particularly suitable for perimeter beams as no edgeformwork for the topping is then required. Columns ofthe next storey are then positioned above the beamsusing grouted steel sleeves to connect the vertical barsif columns are precast, or using normal reinforcedconcrete details if columns are cast-in-place.

Figure 3.6(a) shows a 22-storey building under con-struction using this system. The structure consists ofmoment-resisting perimeter frames with interior framescarrying mainly gravity loading. Some constructiondetails are shown in Figures 3.6(b), (c) and (d).

Figure 3.6(b): Precast concrete beam corner unit

being lowered into place using temporary plastic

tubes as guides

Figure 3.6(a): Structure of building frame incor-

porating precast concrete beam elements pass-

ing through the columns


Figure 3.6(c): Column bars after being grouted in the joint core of a precast concrete beam unit

3.3.2 Some Design Aspects

General

An advantage of System 2 is that the beam-columnjoint core reinforcement can be incorporated in the

precast concrete beam element and the potential plastichinge regions in the beam occur within the precastelements away from the vertical cold joints betweenprecast elements and cast-in-place concrete. Also, it ispossible to incorporate in precast concrete beam ele-

Figure 3.6(d): Preparing a cast-in-place concrete midspan joint of a perimeter frame


ments reinforcing details to relocate the potential plas-tic hinge regions away from the column faces if neces-sary (see Figure 3.7).

Splicing at Midspan of Beams

An aspect of the previous concrete design code [3.10]that caused problems in design was the requirementthat, when the critical section of a potential plastichinge region of a beam is located at a column face, nopart of the splice of the longitudinal reinforcementwas to occur within 2d of the column face, where d isthe effective beam depth. To satisfy this requirementthe clear span of the beam had to be at least 4d + lswhere ls is the splice length. This code requirementmade it difficult to use beams of relatively short span.However, as reported in Appendix B2, a number oftests on midspan joints have now been conducted[3.14, 3.15, 3.16] and the 1995 edition of the concretedesign standard [3.12] modified this requirement bypermitting the splice to commence at distance d fromthe column face. Hence conventional straight barsplices are now possible at midspan of short spanbeams.

Some details for midspan connections in beams whichhave been used are illustrated in Figure 3.8. The con-ventional straight bar lap of Figure 3.8(a) can be short-ened by using hooked laps as shown in Figure 3.8(b)and (c). The double hooked lap (see Figure 3.8(c)),involving the use of hooked “drop in” bars, is the mostconvenient hooked lap to construct. These details, in

some cases with slight modification, have all shownexcellent performance in laboratory tests (see labora-tory tests [3.14, 3.15, 3.16] described in Appendix B2)and hence can be recommended as suitable for use.

Diagonal reinforcement has been used where shearforces in the beams are high (see Figure 3.8(d)). Thedesign and detailing of the welded connection detailsrequire extreme care. Significant vertical ties are re-quired between the bends in the diagonal reinforce-ment to resist the vertical component of the force in thediagonal bars. Also, it should be checked that bearingfailure of the concrete cannot occur under the bends ofthe diagonal reinforcement (see the laboratory tests[3.15, 3.16] described in Appendix B2). The jointsshould be capacity designed and any eccentricities ofplate and reinforcing bars be minimized and providedfor by basketing reinforcement. Grinding back of thereinforcing bars to provide 45˚ double-V butt weldsmay be required to provide the necessary high standardwelded connection.

Strength of Grouted Connections

Other aspects of System 2 frames which have been ofconcern are the performance of grouted column barsand horizontal joints between the columns and theprecast beams, and the performance of column tiesaround the grout-filled ducts which are oversized toprovide construction tolerances (see Figures 3.6(c) and3.9). Laboratory testing in New Zealand [3.14, 3.15,3.16], summarised in Appendix B2, has indicated that

(a) Reinforcement for conventionalplastic hinge positions

A B

A B

(b) Two reinforced arrangementsfor relocated plastic hinges

Figure 3.7: Relocated plastic hinge design for moment-resisting frames dominated by seismic loading


Figure 3.8: Some details for midspan connections for beams which have been used in New Zealand

column

column

cast-in-placejoint

column

precast beam

column

precast beam

(a) Conventional Straight Bar Lap

cast-in-placejoint

column

precast beam

≥ d ≥ s ≥ d

n ≥ 2d +

≥ d ≥ dh ≥ d

d

d

d

(b) Hooked lap

(c) Double hooked lap

column

precast beam

cast-in-placejoint

precast beam

s

≥ d ≥ 2 dh ≥ d

column

precast beam

overstrength stub

steel platesbolted together

cast-in-placejoint

diagonal bars weldedto steel plate

column

precast beam

(d) Diagonal Beam Reinforcement

Note: Transverse reinforcement is not shown


provided design and construction are adequate, theseaspects are satisfactory. The performance of this sys-tem was shown to be similar to that of a conventionalcast-in-place joint.

3.4 System 3 - Precast T orCruciform Shaped UnitsA further possible structural arrangement incorporat-ing T-shaped precast concrete elements is shown inFigure 3.1(c). The vertical column bars in the precastT units are connected using grouted steel sleeves. Atthe midspan of the beams, bottom bars can be splicedin a cast-in-place concrete joint or connected by weld-ing to steel plates which are bolted together. An alter-native to the T-shaped units is to use cruciform-shapedprecast concrete units with joints between columnsoccurring at mid-height of the storeys. Precast floorsystems can be used as with the other systems.

An advantage of System 3 is the extensive use ofprecast concrete possible, and the elimination of thefabrication of complex reinforcing details on the build-ing site. A possible constraint is that the precast ele-ments are heavy and crane capacity may be an impor-tant consideration.

Figure 3.10 shows the details of a precast concrete T-shaped unit used in the perimeter frame of a building inwhich plastic hinging was designed to occur in thediagonally reinforced beam regions away from columnfaces. Grouted steel sleeves were used to splice col-umn bars at interfaces between units above the jointcore, and high strength friction grip bolts were used toconnect beam bars which were welded to steel chan-nels at midspan. As for Figure 3.8(d), extreme care isrequired for the welded connection detail and the detailat the bends of the diagonal bars.

Figure 3.11 shows a perimeter frame of a buildingconstructed using precast concrete cruciform-shapedunits two storeys in height, with grouted steel sleeves

connecting column bars at mid-storey height and hookedsplices connecting beam bars in cast-in-place concretejoints at midspan.

3.5 Low Frames with StrongBeam-Weak Column DesignSystems 1, 2 and 3 described in Sections 3.2, 3.3 and3.4 have typically been used for strong column-weakbeam design.

For one- or two-storey frames a strong beam-weakcolumn design concept is permitted [3.11, 3.12]. Hencefor such low frames the post-elastic deformations in amajor earthquake can be designed to occur by plastichinges forming at the column ends. For a strong beam-weak column design, System 1 of Figure 3.1(a) wouldbe suitable. Also suitable would be System 3 but withcruciform-shaped units as in Figure 3.11, with connec-tions between precast elements at mid-height of col-umns and at midspan of beams, rather than T-shapedunits, in order to keep column splices out of potentialplastic hinge regions.

A further possible arrangement of precast elements forsuch a design is shown in Figure 3.12. For this building,two-storey columns are precast in one length for the fullheight of the frame and the beams are precast to occupyabout the middle 60% of each span. Top and bottombeam bars protruding from precast beam elements arespliced in end regions of beams in cast-in-place con-crete. These bars are spliced with lengths of beam barwhich are cast passing through the precast columns.

3.6 Mixed Precast PrestressedConcrete and Cast-in-placeReinforced Concrete Moment-resisting FramesA further building system which has become popular

precast concrete column

column tie

preformed metal duct

grout

grouted-in bar

Figure 3.9: Part column section with grouted-in bars


typical internal panel typical corner panel

(both symmetrical about )

1415 1300 1300 120512

25

240

520

240

2155

87 125 225 225226 251 87

250

R6 hairpins4-D204-D20R6 spirals

2-D12

2-D16 2-D24 R6 hairpins

D20

2-H20

4-D32

NMB splice sleeves

R10 ties at 70

R10 stirrups at 100

R10 stirrups at 210

R6 spirals

connected to next unitby high strength frictiongrip bolts

elevation

70160

70

1100

section A-A

As assembled structure

Figure 3.10: Details of precast concrete T-shaped units of the perimeter frame of a

reinforced concrete building

in recent years involves the use of precast concretebeam shells as permanent formwork for beams. Theprecast beam shells are typically pretensioned pre-stressed concrete U-beams and are left permanently inposition after the cast-in-place reinforced concretecore has been cast. The precast U-beams support theself weight and construction loads and act compositelywith the reinforced concrete core when subjected toother loading in the completed structure.

Precast U-beams are generally not connected by steel tothe cast-in-place concrete of the beam or column. Reli-ance is normally placed on the bond between the rough-ened inner surface of the precast U-beam and the cast-in-place concrete core to achieve composite action. Occa-sionally, protruding stirrups or ties from the U-beamshave been used to improve the interface shear strength.

The typical structural organisation of a building floorand moment-resisting frame system incorporating pre-cast pretensioned U-beam units is shown in Figure3.13. Figure 3.14 shows a ductile frame under con-struction using precast U-beams.

This form of composite beam construction has beenused in multi-storey moment-resisting framed struc-tures. In this application, the composite beams will berequired to develop ductile plastic hinges during majorearthquakes.

Doubts have been expressed by some designers andchecking authorities concerning the ability of this formof composite construction to be able to perform asductile moment-resisting frames. It had been felt thatcracking may concentrate in the beam at the column


Figure 3.12: Two-storey frame with columns precast in one length for the full height and cast-in-place

spliced joints at the ends of precast beams

Figure 3.11: Perimeter frame of a reinforced concrete building constructed using two-storey

high precast concrete cruciform units, Unisys House, Wellington


face at the discontinuity caused by the end of theprecast U-beam. However, tests have demonstrated[3.17] that during severe seismic loading there is atendency for the plastic hinging to spread along thecast-in-place reinforced concrete core within the pre-cast U-beam due to breakdown of bond. Hence plastichinge rotation does not concentrate in the beam at thecolumn face and no undesirable concentration of cur-vature results. Seismic design recommendations forsuch construction are available [3.17]. It is consideredthat this type of construction is suitable for ductilemoment-resisting frames.

It is very important to ensure during construction thatthe inside surface of the shell beams is clean when thecast-in-place concrete is placed, otherwise sufficientbond between the shell and core cannot develop. A sitefailure of a beam due to lack of bond because of a dirtyinterface has been observed.

3.7 Pinned JointsPinned joints can be used to connect secondary beamsto primary beams and sometimes at beam-columnjoints to reduce the moment input from the beam to thecolumn. Examples of pinned joints at secondary beamto main beam connections are shown in Figure 3.15.The typical 20 mm tolerance gap, between the end ofthe secondary beam and the side of the main beam,implies that the precaster and contractor may be re-quired to work to more stringent tolerances than thosespecified in NZS 3109 [3.18]. Also, the welding ofreinforcing bars to RHS, or bars to the seating angle, is

reinforced concretecolumn

pretensioned precastconcrete U-beam

cast-in-placeconcrete beamcore reinforcement

proprietary floor systemand cast-in-placereinforced concretetopping Note: not all

reinforcementis shown

Figure 3.13: Construction details of a structural system involving precast concrete beam

shells and cast-in-place reinforced concrete

Figure 3.14: Construction of a ductile frame

using precast concrete shell beams

critical. Weld throat thicknesses need to be carefullymonitored.

In many instances beams that are supported each end


20 gaptypical

additional stirrupsrequired at changein beam depth

cast-in-placeconcrete topping

Lap bars

precast concretefloor ribs

steel flats buttwelded to corbelweldplate

extra stirrups atcorbel location

precast concretemain beam

fabricated steelcorbel withgussels

bearing/seatingangle

precast concretesecondary beam

beam reinforcing barsplug welded to seatingangle

(a) Using Steel Corbel Angle Bracket

(b) Using RHS Hanger Brackets

pair of RHS hangerbrackets withU-bars welded on

cast-in-placeconcrete topping

top continuityreinforcement

precast concretefloor ribs

bottomreinforcement(occasionaly used)

bearing angle

precast concretemain beam20 gap

typical

precast concretesecondary beam

endplate weldedon RHS

Figure 3.15: Example of secondary beam to main beam connections

on pairs of RHS hangers or on wide corbel anglebrackets will only bear on the two diagonally oppositeRHS hangers or on one side of the wide corbels. Thisis because the various component parts are not cast orfabricated perfectly square and true. The bearing sur-faces should be shimmed as necessary to ensure evenbearing on all surfaces. However, designers shouldtake into account in the design of the components thiscommon miss-fit scenario.

Bottom reinforcement by tying the secondary beam tothe main beam is occasionally used to improve theoverall robustness of the “Pin Joint” details.

3.8 Composite ReinforcedConcrete Moment-resistingFrames of Limited DuctilityCast-in-place ductile reinforced concrete structuralwalls in a building can be designed to resist almost allof the seismic loading acting on the building, if they arevery stiff compared with the frames in the building.Then the frames in the building are present mainly tocarry the gravity loading. Such moment-resisting framescan be designed for limited ductility, using less trans-verse reinforcement than for ductile frames, providing


it can be shown that when the walls have deformedinelastically to the required displacement ductilityfactor during severe loading, the ductility demand onthe frames is not large. This is possible when the framesare much more flexible than the walls.

A building so designed is illustrated in Figure 3.16.The central cast-in-place reinforced concrete wallsforming the service core of the building were designedto resist the seismic loading. The perimeter frame ofprecast concrete beams, and shell columns infilledwith cast-in-place concrete, was designed mainly forgravity loading.

3.9 Industrial BuildingsTotally precast concrete frames have proved to besuitable for industrial buildings with a large degree ofrepetition, or enclosing processes giving off corrosivevapour (for example, pulp and paper processing).Typically the frame would consist of precastpretensioned, prestressed concrete members. Nor-mally the roof beam is simply supported at its ends oncantilever columns, and the seismic loading is resistedby the cantilever columns.

Figure. 3.17 shows photographs of some examples ofsingle-storey industrial buildings which have beenconstructed in New Zealand.

3.10 Frame Connections inNorth American and JapanesePracticeThe New Zealand design approach for moment-resist-ing frames has been to use cast-in-place concrete con-nections between precast concrete elements and to seekstructural behaviour as if of monolithic construction.

Moment-resisting frames incorporating precast con-crete elements have had very little use in the UnitedStates. United States practice has been mainly to use“dry connections” formed by welding, dry packing andgrouting. Dry connections do not always behave as ifpart of monolithic construction. The Prestressed Con-crete Institute of the United States has sponsoredresearch on moment-resisting connections. The em-phasis of the early research work has been on dryconnections. A summary of some of this work ispresented in References 3.19 and 3.20. In most caseswhere welding or bolting was used, there was a lack ofductility. It is considered that as long as welding andbolting is used in the critical region it will be extremelydifficult to ensure ductile behaviour in primary seismicload-resisting elements.

In Japan, structural steel or steel-reinforced concretehas traditionally been used for tall building structures.However, in the 1980s the technical feasibility andeconomics of reinforced concrete for tall buildings wasrealised and now reinforced concrete moment-resistingframes typically 20- to 30-storeys in height are com-monly-used in Japan for apartment buildings. Manyincorporate precast concrete elements. The types ofmoment-resisting frame connections between precastmembers used in Japan are similar to those in use inNew Zealand. Examples of Japanese practice are givenby Kurose et al [3.21] and Kanoh [3.22]. It is noted thata considerable amount of laboratory testing is beingconducted in Japan by construction companies.

3.11 Recommendations

3.11.1 Arrangements of PrecastConcrete Members in Moment-resistingFramesSeveral possible arrangements of precast concrete mem-bers and cast-in-place concrete forming ductile mo-ment-resisting frames have been used in New Zealand(see Sections 3.2 to 3.6). The design aim has been to

Figure 3.16: Precast concrete perimeter frame in

building with seismic loading resisted mainly by

interior core of cast-in-place walls


Figure 3.17(b): Aluminium smelter, Tiwai point (cantilever columns and pinned rafters

built in the early 1970s)

Figure 3.17(a): Hamilton showgrounds building (bolts at tops of cantilever columns pass

through Y-shaped rafters - built in the late 1960s)


Figure 3.17(c): Caxton paper mills, Kawerau (built in mid-1980s)

connect the precast concrete elements together so thatthe seismic performance will have the essential integ-rity of a monolithic cast-in-place concrete structure.

3.11.2 Frames Formed from PrecastBeams Spanning between Columns withCast-in-place Beam-column JointsThe column dimensions need to be large enough toaccommodate the required development lengths of thebeam longitudinal reinforcement and to avoid conges-tion of the hooked bottom bars in the beam-columnjoints as far as possible. The ends of the precast beamsshould be roughened or keyed to assist shear transferacross the vertical cold joints at the column faces (seeSection 3.2).

3.11.3 Frames Formed from PrecastBeams Extending from Midspan toMidspan with the Longitudinal ColumnBars Grouted in the Beam-Column JointsThe splicing of the longitudinal beam bars in a cast-in-place joint in the midspan regions can be achieved byeither straight or hooked laps, or by diagonal reinforce-ment welded to plates, which are bolted together.Splices in longitudinal beam bars should not com-mence within d of the column face, where d is theeffective depth of the beam. The grouting of horizontaljoints between the columns and the soffits of theprecast beams, and the grouting of the longitudinalcolumn bars passing through precast beams, when

carried out satisfactorily, will result in performance ofthe joint similar to that of a cast-in-place concrete joint.However, strict grouting procedures need to be fol-lowed (see Section 3.3).

3.11.4 Frames Formed from T-Shapedand Cruciform Shaped Precast UnitsAdequate splicing of the beams of precast units in themid-span regions, and the columns at the beam-col-umn joints or at the mid-height regions of storeys, canbe achieved (see Section 3.4).

3.11.5 Frames Formed from MixedPrecast Prestressed and Cast-in-PlaceConcreteMoment-resisting frames constructed from precastconcrete beam shells and cast-in-place concrete can bedesigned to perform satisfactorily. Special attention indesign needs to be paid to the performance of theprecast concrete shell and its cast-in-place concretecore (see Section 3.6).

3.11.6 Low Frames With Strong Beam-Weak Column DesignThe above structural arrangements have been usedmainly for strong column-weak beam designs. Forone- or two-storey frames where a strong beam-weakcolumn design is permitted, only the arrangement withgrouted column bars may be unsatisfactory (see Sec-tion 3.5).


3.11.7 Industrial BuildingsPrecast concrete has been successfully incorporated inframes used for industrial buildings in seismic areas.

3.12 References3.1 Park, R. “Seismic design considerations for

precast concrete construction in seismiczones”, Seminar on Precast ConcreteConstruction in Seismic Zones, Japan Societyfor the Promotion of Science - United StatesNational Science Foundation, Tokyo, Vol. 1,1986, pp. 1-38.

3.2 O’Leary, A J, Monastra, D P and Mason, J E.“A precast concrete moment-resistingframing system”, Proceedings of PacificConcrete Conference, Vol. 1, Auckland, NewZealand, November 1988, pp. 287-298.

3.3 Billings, I J and Thom, G W. “NZI Centre -design of multi-storey towers”, Proceedingsof Pacific Concrete Conference, Vol. 1,Auckland, New Zealand, November 1988,pp. 309-318.

3.4 Poole, R A and Clendon, J E. “Mid CityTowers - an efficient precast concrete framedbuilding”, Proceedings of Pacific ConcreteConference, Vol. 1, Auckland, New Zealand,November 1988, pp. 319-332.

3.5 Silvester, D B and Dickson, A R. “FanshaweStreet building - a precast concrete study”,Proceedings of Pacific Concrete Conference,Vol. 1, Auckland, New Zealand, November1988, pp. 333-344.

3.6 O’Grady, C R. “Precast cruciform columns, Hframes and precast concrete shear walls inbuilding construction”, Proceedings ofPacific Concrete Conference, Vol. 1, Auck-land, New Zealand, November 1988, pp. 345-354.

3.7 Raymond, W. “Efficient use of structuralprecast concrete in high rise buildings — acase study”, Transactions of Institution ofProfessional Engineers New Zealand, Vol.19, No. 1/CE, November 1992, pp 21-27.

3.8 Park, R. “A perspective on the seismic design ofprecast concrete structures in New Zealand”,Journal of the Prestressed/Precast ConcreteInstitute, Vol. 40, No. 3, May-June 1995, pp40-60.

3.9 “Code of Practice for General Structural Design

and Design Loadings for Buildings, NZS4203:1984”, Standards Association of NewZealand, Wellington, 1984, 100 pp.

3.10 Code of Practice for the Design of ConcreteStructures, NZS 3101:1982, StandardsAssociation of New Zealand, Wellington,1982, 127 pp.

3.11 Code of Practice for General StructuralDesign and Design Loadings for Buildings,NZS4203:1992, Standards Association ofNew Zealand, Wellington, 1992, 134pp.

3.12 Design of Concrete Structures, NZS3101:1995,Standards New Zealand, Wellington, 1995,256pp.

3.13 Stevenson, R B and Beattie, G J. “Cyclic loadtesting of a beam column cruciform incorpo-rating precast beam elements”, Report 88-B5204/1, Works Central Laboratories, LowerHutt, 1988, 66 pp.

3.14 Beattie, G J. “Recent testing of precast struc-tural components at Central Laboratories”,Proceedings of the New Zealand ConcreteSociety Conference, Wairakei, New Zealand,October 1989, pp 111-118.

3.15 Restrepo, J I, Park, R and Buchanan, A H.“Tests on connections of earthquake resistingprecast reinforced concrete perimeter framesof buildings”, Journal of the Prestressed/Precast Concrete Institute, Vol. 40, No. 4,July-August 1995, pp 44-61.

3.16 Restrepo, J I, Park, R and Buchanan, A H.“Design of connections of earthquakeresisting precast reinforced concrete perim-eter frames of buildings”, Journal of thePrestressed/Precast Concrete Institute, Vol.40, No. 5, September-October 1995, pp 68-77.

3.17 Park, R and Bull, D K. “Seismic resistance offrames incorporating precast prestressedconcrete beam shells”, Journal of PrestressedConcrete Institute, Vol. 31, No. 4, July/August 1986, pp. 54-93.

3.18 Specifications for Concrete Construction,NZS3109:1997, Standards Association ofNew Zealand, Wellington, 1997, 67pp.

3.19 Stanton, J F, Anderson, RG, Dolan, C W andMcCleary, D E. “Moment resistant connec-tions and simple connections”, ResearchProject No. 1/4, Prestressed Concrete Insti-tute, Chicago, 1986, 436 pp.


3.20 Dolan, C W, Stanton, J F and Anderson, R G.“Connections and simple connections”,Journal of Prestressed Concrete Institute,March/April 1987, pp. 62-74.

3.21 Kurose, Y et al, “Reinforced concrete beam-column joints in construction practice inJapan”, US-NZ-Japan Seminar on Earth-quake Resistant Design of Reinforced

Concrete Beam-Column Joints, Honolulu,Hawaii, May 24-26, 1989, 32 pp.

3.22 Kanoh, K. “Review of Japanese precastconcrete frame systems used as buildingstructures”, Seminar on Precast ConcreteConstruction in Seismic Zones, Japan Societyfor the Promotion of Science- United StatesNational Science Foundation, Tokyo, Vol. 2,1986, pp 35-54.

Structural Wall Elements • 41

Chapter 4Structural Wall Elements

4.1 IntroductionThe usefulness of structural concrete walls in buildingshas long been recognised in New Zealand as they havebeen found to be efficient in resisting lateral forces dueto wind and earthquakes. Their large inherent stiffnessmeans that drift is limited during a severe earthquake,thus providing a high degree of protection against non-structural damage. Comprehensive seismic design pro-visions for cast-in-place reinforced concrete walls aregiven in the New Zealand Concrete Structures Stand-ard [4.1].

Tests [4.2] have shown that well-detailed walls givethe required strength and ductility. It is recognised thatwell-proportioned, ductile, cast-in-place reinforced con-crete, coupled cantilever walls, probably form the bestearthquake resistant structural system available in re-inforced concrete [4.3]. In these, the coupling beamscan be designed so that most of the hysteretic energydissipation occurs in them, thereby limiting the dam-age that is sustained by the vertical load-resisting wallelements.

Although most seismic resistant structural concretewalls for multi-storey buildings in New Zealand are ofcast-in-place construction, there has been some con-struction utilizing precast concrete wall panels. Thischapter examines and makes recommendations on theuse of precast concrete in structural wall elements.

4.2 Types of Precast ConcreteStructural Wall ConstructionPrecast concrete structural wall construction usuallyfalls into two broad categories [4.4]:

• “monolithic” wall systems; or

• “jointed” wall systems.

The distinction between these two types of construc-tion is based on the design of the connections betweenindividual precast concrete panels, which when jointedtogether form the structural wall. For “monolithic”wall systems, connections are designed as “strong”connections, so their elastic limit is not exceeded insatisfying the building’s ductility demands. Alterna-tively, in the case of “jointed” wall systems, connec-

tions may be designed as “ductile”, with energy dissi-pation occurring in the connection, thereby contribut-ing to the building’s overall ductility.

It is possible to use a combination of monolithic andjointed details. For example, multi-storey residentialconstruction in Japan, America and Yugoslavia oftenutilises “monolithic” vertical joints and “jointed” hori-zontal joints [4.5].

Some examples of monolithic and jointed wall connec-tions used in Yugoslavia, Japan and USA are shown inFigures 4.1 to 4.8.

The design of the connections between precast con-crete wall elements largely depends on the type ofconstruction. However, factors such as capacity de-sign principles, cartage, craneage capacity, erectionprocedures, volume changes from creep, shrinkageand temperature, need to be taken into consideration.

A common precast concrete wall construction methodfor single and two-storey buildings in New Zealanduses the tilt-up wall construction technique. With tilt-up construction, relatively large wall panels are casthorizontally on top of concrete floor slabs or castingbeds adjacent to final wall panel positions. When theconcrete used has gained sufficient strength for thewall panels to remain uncracked during lifting opera-tions, the walls are tilted up and lifted into their perma-nent positions. Generally tilt-up walls are secured tothe adjacent structural elements with jointed connec-tions comprising various combinations of concreteinserts, bolts, weldplates, angle brackets and lappedreinforcement splices within cast-in-place joining strips.Reference 4.6 provides full details for tilt-up walldesign and construction for New Zealand conditions.Reference 4.21 discusses recent research into the seis-mic performance of a wide range of panel-to-panelconnections and floor-to-panel connections.

Reference 4.22, a report by the PCI Ad Hoc Committeeon Precast Walls, discusses recent experience, re-search, and design regulations employed in NorthAmerica for the use of structural (shear) precast con-crete walls. Included in this comprehensive report, arereferences to a number of papers describing the per-formance of precast walls during major seismic eventsover the last 25 years. This report and references within


will be of considerable interest to designers dealingwith the seismic design of structural wall elements.

4.3 Monolithic Wall Systems

4.3.1 GeneralIn monolithic construction, precast elements are joined

continuous vertical reinforcement ofvertical joints and adjacent to panelends within full height metal ducts

cast-in-place joint concrete

prefinished hollow corefloor planks.

n bars from top and bottomwall panels.

precast concretewall panels

floor ties at thejunction of each plank

Figure 4.1: Nearly monolithic precast wall construction horizontal joint of the

SCT System, Yugoslavia (Reference 4.12)

horizontal hairpin bars.By altering the size andspacing of these bars theperformance of the jointcan vary from “jointedbehaviour to “monolithic”behaviour

continuous verticalreinforcement adjacentto panel ends within fullheight metal ducts


continuous verticalreinforcement withinthe vertical joints

cast-in-placeconcrete joint

Figure 4.2: Nearly monolithic precast wall construction vertical joint of the

SCT System, Yugoslavia (Reference 4.12)

by reinforced concrete connections possessing stiff-ness, strength, and ductility believed to be comparableto cast-in-place concrete. Monolithic precast wall sys-tems are often used as a cost-effective alternative tocast-in-place ductile cantilever shear walls or ductilecoupled shear walls. Designers expect these to becapable of sustaining the significant inelastic deforma-tions required of a ductile structure.


a proportion of verticalbars are lap weldedtogether

cast-in-placeconcrete fill

precast concretefloor planks


Figure 4.3: Jointed precast wall construction example of horizontal joint for

2-storey houses in Japan (Reference 4.9)

cast-in-placeconcrete column

precastconcretewall panels

the joint tends to behave“Monolithically” if all horizontalreinforcement extends into thejoint and is lap welded. Ifonly a proportion of horizontalreinforcement extends into thejoint then the joint will behavein a “Jointed” manner.

Figure 4.4: Jointed or monolithic precast wall construction example of vertical joint for

2-storey houses in Japan (Reference 4.9)

The satisfactory performance of well-detailed cast-in-place concrete walls has been established by exhaus-tive laboratory testing and observation of their behav-iour during earthquakes. If designers of precast con-crete wall systems detail the connections between thecomponents to possess stiffness, strength and ductilitycomparable to cast-in-place concrete, then the com-pleted walls can be expected to perform satisfactorily.

Care is needed, however, in design and construction asindicated below.

4.3.2 Design of Monolithic WallsIn the absence of a New Zealand code of practicewritten specifically for precast concrete construction,designers usually design monolithic precast structuralwalls to the requirements of NZS 3101 [4.1].


continuous reinforcementat ends of panels only

cast-in-placeconcrete infill


Figure 4.5: Jointed precast wall construction horizontal joint of the

“Rad” System, Yugoslavia (Reference 4.15)


cast-in-placeconcrete infill

by varying the size and spacingof the hairpins the joint canbehave in either a “Monolithic”or “Jointed” manner

Figure 4.6: Jointed or monolithic wall construction vertical joint of the

“Rad” System, Yugoslavia (Reference 4.15)

Since NZS 3101 has been developed principally forcast-in-place concrete construction, there are somedifficulties in applying this standard to precast con-crete wall designs. Care is needed when applying thedimensional limitation rules. In NZS 3101 it is as-sumed that structural walls have full flexural continu-ity about their minor axis at each level of lateralrestraint (floor level). With precast concrete wallsthere may be a major discontinuity in the flexural

strength about their minor axis at each horizontal jointlocation. Designers are therefore encouraged to adopta conservative approach to the ratio of the clear storeyheight to the wall thickness.

NZS 3101 requires two layers of longitudinal (vertical)reinforcement when structural walls are more than 200mm thick. This can be easily achieved over the heightof the wall panels, but having two layers of reinforce-


Note that the verticalcontinuity reinforcementoccurs at the end of eachwall panel only (USAexample [4.10, 4.20])

grouted floor/walljoints

prefinished hollowcore floor planks

bearing strip

grout placed within1 hour of stressing

2 post tensioningDywidag bars ateach end of wallpanels precast concrete

wall panels

grouting duct

cast-in stressing head

stressing platestressing nut

paper or plastic dam

floor ties at 1200 crsin each floor slab joint

dry pack for 150 eachside of stressing bar

alternative locationof coupler sleeve

600 mm unbonded inplastic sheath

coupler sleeve

Figure 4.7: Jointed precast wall construction horizontal joint — platform type construction

vertical continuityreinforcement ofends of wallpanels only (weldedor bolted connections)

cast-in-place infillsat ends of panels

precast concretefloor panels


Figure 4.8: Jointed precast wall construction elevation of horizontal joint

(Reference 4.8)

ment at horizontal panel joints is usually impractical.Furthermore, NZS 3101 has a restriction that the diam-eter of bars used in any part of a wall shall not exceedone-tenth of the thickness of that wall. This limitationcan often be a major constraint when using centralvertical lap bars at horizontal joints that lap with pairsof smaller bars placed either side of the central bar (seeFigure 4.10, Section A-A).

NZS 3101 typically requires staggering of the splicesof the principal vertical flexural tension reinforcementwithin potential plastic hinge zones. This is verydifficult to achieve with practical details. Usually, lapsin the cast-in-place concrete bandage joints and thecast-in-place end wall thickenings are staggered inlevel from those in the precast wall sections. Yieldingof the bars splicing the precast wall sections together is


usually avoided by selecting a lap bar size and/or steelgrade that is slightly stronger than the vertical bars thatlap with it, i.e. one D24 (452 mm2) could lap with twoD16 bars (402 mm2). This helps to minimise flexuralcracking over the lapping region.

Research is therefore needed to check the sensitivity ofmonolithic precast concrete wall design and construc-tion to the difficulties which have been identified.

4.3.3 Vertical JointsVertical joints between wall panels are typically cast-in-place “bandage” type joints. Horizontal reinforcingfrom a precast panel projects into the joint zone and islapped with horizontal reinforcing from an adjacentpanel. The amount and spacing of the horizontal shearreinforcement is established using capacity designprinciples and concrete code requirements [4.1]. Thewidths of the cast-in-place vertical joints are deter-mined by concrete code requirements for lap lengths ofthe horizontal reinforcement. Typical details of mono-lithic vertical joints are shown in Figure 4.9.

A wide range of details are in common use in NewZealand. The design of the lap or splice regions shouldcomply with the requirements of NZS 3101, or alterna-tively be laboratory tested. In many instances, the endsof monolithic precast walls have cast-in-place columnswhich improve the wall stability where flexural rein-forcement and transverse stirrups and ties are concen-trated.

4.3.4 Horizontal JointsThe vertical reinforcement in precast walls is usuallylapped at horizontal joints. Proprietary grouted steelsleeve splices may be used for this purpose, or alterna-tively the lap can be formed by grouting a bar extendingfrom one unit into a metal duct in the matching unit.The spacing of steel sleeve splices or metal ducts isusually at no more than 450 mm centres to comply withthe maximum spacing provisions of NZS 3101. Sometypical details of horizontal joints in monolithic con-struction are shown in Figure 4.10. Some steel sleevesplices have been cyclic load tested and comply withthe concrete code requirements as high strength me-chanical connectors suitable for use in a plastic hingezone, provided that the grout used complies with thesleeve manufacturer’s specification [4.7].

When corrugated metal preformed pipe ducts are used,starter bars that project into the ducts are usuallydesigned for a full lap length as defined in NZS 3101.Generally central starter bars are lapped with pairs ofsmaller bars, one on each face of the precast concretewall section or, alternatively, all of the main flexuralreinforcement is lapped on the precast concrete wall

centreline and some additional basketing cover rein-forcement is provided.

The horizontal joint between precast concrete panels isusually scabbled or retarded and cement paste removedto provide appropriate surface roughness to avoid asliding shear failure. The joint is then grouted using themethod outlined in Chapter 6. When using steel sleevesplices, these are usually grouted individually. Foamplastic rings are used to seal the base of the steel sleevesplices to allow the separate grouting of the sleeves andthe horizontal joint between the precast concrete panels.

4.4 Jointed Wall Systems

4.4.1 GeneralJointed construction describes the connection of pre-cast components, which result in planes of signifi-cantly reduced stiffness, strength, and ductility at theinterface between adjacent precast concrete members.This type of precast concrete wall construction is notcommon for high rise construction in New Zealand but,because of potential economy in construction, requiresconsideration.

Design and construction of buildings up to eight sto-reys high utilising jointed precast wall systems iscommonplace in countries such as Japan, America,Yugoslavia, USSR, Romania and Bulgaria [4.5]. Typi-cally, this form of construction has been used for highdensity residential developments. Various forms ofjointed precast wall systems are occasionally used inNew Zealand for medium height commercial or indus-trial structures.

4.4.2 AnalysisAnalysis assumptions commonly used for monolithicor cast-in-place wall structures are usually not suitablefor jointed precast concrete wall systems. For theultimate limit state, in which member strengths areassigned according to the relative elastic stiffness ofthe undamaged structure, there is a reliance on inelasticload redistribution to adjust for differences betweenactual and computed load paths. Jointed precast con-crete structures that rely on discrete connectors ofsomewhat limited ductility have a relatively low de-gree of redundancy. Consequently, their load redis-tributing capacity may be insufficient to accommodateinaccuracies of the simplified analyses usually per-formed for monolithic wall structures.

Changes in the relative wall stiffness of jointed wallsystems that occur as joints open and close during asevere earthquake response are often neglected. The


Note 1. Vertical joints shown as Types D, E and F need to be detailedwith extreme care. Once the lapping bars have been overlapped theability for lowering the wall panels over starter bars is very restricted.These details will typically work only when grouted steel splice sleevesare used to splice the vertical flexural reinforcement and when thelaps of the vertical bars in the “bandage” joints are made below floorlevel.

Joint D is not preferred because joint reinforcement and concreteinfill cannot be inspected.

cast-in- place concreteor grout-filled joint

sealant

sides of jointroughened


Type E

all horizontal shearreinforcementhairpin spliced

vertical reinforcementtypically lap splicedabove or belowfloor level(see Note 1)

cast-in-place concrete“bandage” joint


for horiz. barsall horizontal shearreinforcement lapspliced

vertical reinforcementtypically lap splicedabove floor level

sides of wall panelskeyed and roughened

Type A

d


all horizontal shearreinforcement splicedwith 90 standardhooks i.e.


sides of wall panelskeyed and roughed


Type B

dh



sides of wall panelskeyed and roughened


all horizontal shearreinforcement splicedwith 90° standardhook lap barsi.e.

Type C

dh

cast-in-place concrete“bandage” joint.


all horizontal shearreinforcement is lappedwith stirrups of samebar diameter i.e.

sides of jointroughened and keyed

vertical reinforcementtypically lap splicedabove floor level.

Type G


cast-in-place concrete“bandage” joint.

sides of jointroughened and keyed



Type F

sealant



sides of jointroughened

cast-in-place concreteor grout-filled joint


Type D

Figure 4.9 : Monolithic precast concrete wall construction vertical joints


Figure 4.10: Monolithic precast concrete wall construction

A A

A A

d (o

f lap

bar

) +

Ad

(of l

ap b

ar)

+ A

(a) Monolithic precastconcrete wall constructionhorizontal joint - type A

steel splice sleeve

grout outlet vent

grout inlet tube

sealing strips

cast-in-place toppingover hollow core floorplanks

lap bar

precast wall panels

sealant

horizontal jointgrouted

foam plasticring

A A

A A

d (o

f lap

bar

) +

Ad

(of l

ap b

ar)

+ A

(b) Monolithic precastconcrete wall constructionhorizontal joint - type B

grout outlet vent

preformed metal ductfilled with expansivegrout

grout inlet tube

sealing strips

cast-in-place toppingover hollow core floorplanks

lap bar

precast wall panels

sealant

horizontal jointgrouted

lap bars to lap withpairs of vertical bars

Lap bars in preformed metalducts or grouted steel splicesleeve. Area of lap bar > 2 xArea of verticalbars

lap bars lap withpairs of vertical bars

horizontal shearreinforcement

reinforcement same asAlternative 1 excepthorizontal shearreinforcement is tieddirectly to the preformedmetal ducts or groutedsplice sleeves

full length bars lap within full height preformedmetal ducts or full length lap bars spliced withgrouted steel splice sleeves

vertical basketingreinforcement

horizontal shearreinforcement

Section A-AAlternative 1



assumption that foundations and floor diaphragms arerigid may cause further inaccuracies [4.4]. Designersshould be cautious in their approach to the design ofjointed precast concrete panel systems. Inelastic ac-tions of the joints should be considered and a rationalapproach to the design of the connections formulated.

4.4.3 Design and DetailingThere has been considerable research and prototypetesting of jointed panel systems overseas and a widerange of jointing details have been developed. Manypapers have been written and readers are directed toreferences such as 4.8 to 4.18.


Some of the usual features of jointed panel systems,sometimes referred to as “large panel building sys-tems”, are:

• they are normally associated with medium riseresidential construction;

• the floor spans are typically less than 6 m;

• the floors are either of precast concrete slabs withcast-in-place toppings or large (room size)prefinished precast concrete floor slabs;

• the vertical joints often incorporate insitu “band-ages” with non-standard laps of nominal amountsof horizontal reinforcement;

• the horizontal joints usually have only two discretewelded connections per panel, above or below floorlevel;

• the horizontal joints are usually dry packed formost of their length;

• the panels are often rebated at the base in thevicinity of the connections to allow for the con-struction of cast-in-place concrete shear keys;

• the design shear stresses and flexural demands onpanels are usually low (mainly because there isusually a large number of shear walls of adequatelength resisting lateral loads in both principal direc-tions); and

• the inelastic demand on the shear walls is usuallyaccommodated by panel rocking and/or slidingshear mechanisms.

4.5 New Zealand Examples

4.5.1 Police Training CollegeResidential BuildingsThese three-storey buildings have external and closecentred internal walls forming a jointed wall system ofprecast concrete. The walls are of “I” or channel crosssection with each being made up of three precast panels(two flanges and one web) per storey height (Figure4.11). The wall sections were placed one above theother up the height of the building, separated by cast-in-place concrete floor slabs, and connected togetherso as to form vertical cantilevers to resist seismic loads.

A

A

D16

200

80D12

D12

R24

50

stop offD16 bar

230x150x12plate

5012

0

1510

030

100

A A

(a) Detail of welded connection in wall flange

(b) Typical wall arrangement

wallconnection

insitufloor slab

insitujoint

Figure 4.11: Precast reinforced concrete walls

(Police Training College residential buildings, Porirua)


Figure 4.11 (continued): Precast reinforced concrete walls

(Police Training College residential buildings, Porirua)

(c) General views

of precast walls of

first story

(d) View of top of

precast walls

before placing

cast-in-place floor

(e) A welded

connection in wall

flange above

cast-in-place floor


The panels also provide vertical support for gravityload from the floors. A cast-in-place concrete founda-tion beam system gives support to the walls and pro-vides resistance against overturning.

A relatively high seismic design load was chosen toreduce the ductility demand during a severe earthquakeand prevent damage during moderate earthquakes. Ahorizontal dry-pack mortar joint was used between thefloor slab and wall element above. Connections be-tween vertical wall elements at the ends of the flangesand webs were made by welding the protruding rein-forcing bars to steel plates (Figure 4.11(a)).

These connections were designed to transfer twice thedesign tensile force in the panel reinforcing and theassociated shear, to ensure that any yielding wouldoccur away from the connection. A cast-in-place con-crete connection was made between the web and flangewalls. Tests conducted in a laboratory on two panelunits under cyclic loading in the yield range confirmedthat a capacity design approach, in which the connec-tions were deliberately made stronger than the ele-ments connected, was essential if a reliable assessmentof performance was to be made [4.19].

4.5.2 Rotorua District Council CivicCentreThis two-storey building utilizes full height precastconcrete wall panels to provide support for the precastconcrete floor system and the roof. The precast con-crete walls, which provide the lateral load resistance ofthe building, are designed as cantilevered shear wallsof limited ductility to the requirements of NZS 4203[4.3]. The connection detail between walls and founda-tions was designed to withstand larger forces corre-sponding to an elastic response. Vertical joints be-tween panels, shown during construction in Figure4.12 consist of cast-in-place concrete joining horizon-tal overlapping reinforcing steel hair pins which projectfrom each of the wall panels, creating a monolithicjoint. A vertical steel bar was placed in the spacebetween the ends of the hair pins prior to concreting.Details of the connection between the wall and thefoundation are shown in Figures 4.13 and 4.14. Holeswere formed in the bases of the panels so that horizon-tal reinforcing bars could be placed through to rein-force the joint. Due to the squat shape of the walls thisdetail was adequate to resist design horizontal shearforces and tension forces resulting from overturningmoments.

4.5.3 Salvation Army Citadel, VivianStreet, WellingtonFour-storey high precast concrete walls occur on two

parallel sides of this building to provide seismic lateralload resistance in the north-south direction. Thesewalls are simple cantilevered shear walls designed forlimited ductility.

The walls are approximately 13 m high x 20 m long x200 mm thick. The precast concrete wall panels usedwere two storeys high and 2.7 m long. The horizontaljoints between the precast sections have lap bars groutedinto metal ducting similar to the detail shown in Figure4.10(b). These bars lap with pairs of vertical flexuralbars in the precast panels. The vertical joints betweenpanels as shown in Figures 4.15 and 4.18 have a halfthickness cast-in-place concrete bandage joint with 90˚hooked drop in lap bars.

A pair of large perforated tilt panel walls with a cou-pling “beam” tilt panel between, provide seismic loadresistance in the east-west direction. This wall-beamsystem as shown in Figures 4.16 and 4.17 has beendesigned for limited ductility. The walls at 5.1 m wideby 7.5 m high with four large window openings behavemore like deep membered frames than cantileveredshear walls. These tilt walls sit on a one-storey highcast-in-place concrete wall with a base connectiondetail similar to that shown in Figure 4.10(b). The

Figure 4.12: Precast concrete panels during

erection (Rotorua District Council Civic Centre)


coupling “beam” tilt panel was connected to the adja-cent panels by full strength butt welding 3 -100 x 16 mmmild steel flats at each end of the beams. All the tiltpanels were “match” cast to ensure a high degree ofaccuracy for the welded connections.

A simplified detail of the welded connection is shownin Figure 4.19.

4.5.4 Sheraton Hotel - AucklandThe main block of the Sheraton Hotel consists of a ten-storey building. Two cast-in-place walls are combined

with precast concrete walls to resist seismic loads inone direction while at right angles the total load iscarried by precast concrete walls only.

The walls have been designed as ductile cantilevershear walls complying with NZS 4203:1976. TheStructural Type Factor, S, used in the design was 1.0and the Structural Material Factor, M, used was 1.0. Atypical cross section of a precast wall unit is shown inFigure 4.20. The critical end regions of the precastwalls are confined by stirrup cages as shown in thatfigure.

precast concrete wall

D16 bar per holein precast panels

D12 at 200

D12 at 200

4-H32

4-H20

4-H20

Figure 4.13: Detail at the wall/foundation junction

(Rotorua District Council Civic Centre)

Figure 4.14: Connection between base of wall panel and foundation

(Rotorua District Council Civic Centre)


Figure 4.15: Interior view of part of 4-storey precast walls

(Salvation Army Citadel, Wellington)

Figure 4.16: Interior view of perforated tilt panel walls showing location of weld connections (A) 2 storey

perforated tilt panel walls, (B) Coupling “beam” tilt panel (Salvation Army Citadel, Wellington)

The floors consist of precast untopped units, supportedon the walls as illustrated in Figure 4.21. Continuity isestablished between the floor units by hairpin bars thatoverlap and are linked by a longitudinal bar in the jointplaced above the precast wall unit.

The typical precast concrete wall units have 65 mmdiameter ducts at 150 mm centres at the ends of the walland at 300 mm centres in the middle regions. Deformed

reinforcing bars were placed in each duct as shown inFigure 4.21. These bars extend a lap length above thesurface of the floor slab ready to receive the nextprecast wall unit. The vertical reinforcement placed inthese ducts consists of 28 mm diameter bars in the endregions and 12 mm diameter bars in the mid regions.The duct together with the gap between the floor unitsabove the precast concrete walls was fully grouted.


Figure 4.17: Perforated tilt panel walls (Salvation Army Citadel, Wellington)

of joint

lap bars shapedto match reinforcing projectingfrom precast concrete panels

horizontal reinforcementshapes

sealant

200

Figure 4.18: Vertical joint between precast concrete panels (Salvation Army Citadel, Wellington)

200

200 x 200 x 110 deep pocketat each fixing location

cut stirrups and F.S.B.W. inreplacement bars afterpanel erection

full strength butt weld

grout fill over full depth ofbeam section of panel

backing plate3-100 x 16 m.s. flats 850 long and700 longwith ex. 20 x 20 x 100 crs.welded on each side

Figure 4.19: Connection between tilt walls and coupling beams

(Salvation Army Citadel, Wellington)


150

125 150 275 300 300

3800 not to scale

to

D28 bars in end regions

65 mm dia. ducts

D12 bars in mid region

DH12 at 350each face

3-D16 each face andR6 stirrups at 150 centres

Figure 4.20: Horizontal section through typical precast wall panel

(Sheraton Hotel, Auckland)

reinforcement lappedwithin duct

grout

125

150

65 mm dia. duct at 300 mmcrs. in mid regions and 150 mmcrs. in end regions

precast concrete untoppedfloor units

longitudinal bar between hairpins

Figure 4.21: Vertical section through precast wall - floor junction

(Sheraton Hotel, Auckland)


4.6 Weathering DetailsWhen precast concrete structural walls are used as partof the external envelope of a building, special attentionto the joints is usually required to ensure weather-tightconstruction. During the life of a building smallmovements will often occur at wall panel joints throughthe actions of concrete shrinkage, creep, thermal move-ments, settlement or seismic movements. For exteriorwalls it is usually considered prudent to seal all junc-tions between adjacent precast concrete wall elementsand all junctions between precast concrete wall ele-ments and adjacent cast-in-place bandage joints withflexible sealants. Rebates are usually required at thejunction between precast panels and cast-in-place band-age joints to accommodate the weatherproofing seal-ants. Often designers detail stepped or sloping hori-zontal wall panel joints to assist with weather proofing.These stepped or sloping horizontal joints are usuallysealed with flexible sealants.

4.7 Recommendations

4.7.1 Monolithic Wall SystemsIf the connections between precast concrete panelshave been designed and detailed to possess stiffness,strength and ductility comparable to cast-in-place con-crete construction, there is every reason to believe thata precast concrete wall would perform as satisfactorilyas a cast-in-place wall. Care is needed with the detail-ing of both horizontal and vertical joints to ensure thesatisfactory behaviour of the wall, particularly over thezone of yielding at the base of the walls (plastic hingezone).

4.7.2 Jointed SystemsIt is recommended that a very cautious approach betaken to the design of jointed precast concrete panelsystems. Their seismic performance is clearly verydifferent from that on which New Zealand codes ofpractice are based. A design approach based on bothexperimental test data verifying the seismic behaviourand detailed theoretical analysis is recommended [4.4].

4.7.3 Precast Concrete Wall Designsand NZS 3101It is recommended that research is needed to check thesensitivity of monolithic precast concrete wall designconstruction to:

• the discontinuity at horizontal joint locations (Sec-tion 4.3.2);

• single-layer lap bar reinforcement at horizontaljoints (Section 4.3.2);

• lap bars with diameters greater than one tenth of thewall thickness (Section 4.3.2);

• not complying with the full requirements for stag-gering laps in plastic hinge zones (Section 4.3.2);and

• concrete compression zones, in plastic hinge zones,of walls with single layers of reinforcement and notransverse reinforcement (confinement) at the endsof these wall regions.

4.8 References4.1 Concrete Structures Standard Parts 1 and 2:

NZS3101:1995, Standards New Zealand,Wellington, 1995.

4.2 Park, R and Paulay, T. Reinforced ConcreteStructures, John Wiley and Sons, New York,1975, 760 pp.

4.3 Code of Practice for General Structural Designand Design Loadings for Buildings, NZS4203:1992, Standards Association of NewZealand, Wellington, 1992, 45 pp.

4.4 Clough, D P. “Development of seismic designcriteria for connections in jointed precastconcrete structures”, Proceedings of theEighth World Conference on EarthquakeEngineering, San Francisco, 1984, Vol. V,pp. 645 - 652.

4.5 “Design and construction of prefabricatedreinforced concrete building systems”,Building Construction Under SeismicConditions in the Balkan Regions, UnitedNations Industrial Development Organiza-tion, Vienna, Vol. 2, 1985.

4.6 Tilt-up Technical Manual: TM34, Cement &Concrete Association of New Zealand, 1991,32 pp.

4.7 Yong, P M F. The Performance of NMB Splicesfor Grade 380 Bars Under Cyclic Loading,Report 5-85/9, Central Laboratories, Ministryof Works and Development, Lower Hutt,1985, 60 pp.

4.8 Velkov, M. “Large panel systems in Yugosla-via. Design, construction and research forimprovement of practice and elaboration ofcodes”, ATC - 8 Proceedings of a Workshopon Design of Prefabricated Concrete Build-ings for Earthquake Loads, Applied Technol-ogy Council, Berkeley, 1981, pp. 87-120.


4.9 Watabe, M and Hiraishi, H. “On the currentdevelopments in earthquake resistant designprocedures for prefabricated concrete build-ings in Japan”, ATC - 8 Proceedings of aWorkshop on Design of PrefabricatedConcrete Buildings for Earthquake Loads,Applied Technology Council, Berkeley,1981, pp. 61-85.

4.10 Fintel, M and Ghosh, S K. “The seismic designof large panel coupled wall structures”, ATC -8 Proceedings of a Workshop on Design ofPrefabricated Concrete Buildings for Earth-quake Loads, Applied Technology Council,Berkeley, 1981, pp. 383-401.

4.11 Oliva, Clough, Velkov and Gavrilovic. Corre-lation of Analytical and ExperimentalResponses of Large-Panel Precast BuildingSystem, Earthquake Engineering ResearchCentre, Report UCB/EERC-83-20, 1988.

4.12 Fischinger, M, Fajfar, P and Capuder, F.“Earthquake resistance of the “SCT” largepanel building system”, Bulletin of the NewZealand National Society for EarthquakeEngineering, Vol. 20, No. 4, December1987, pp. 281-289.

4.13 Harris, H G and Caccese, V. “Seismic behav-iour of precast concrete large panel buildingsusing a small shaking table”, Proceedings ofthe Eighth World Conference on EarthquakeEngineering, San Francisco, 1984, Vol. VI,pp. 757-764.

4.14 Cauvin, A and Zanon, P. “Test results on keyjoints of large panel prefabricated buildingssubject to cycliced reversed actions”, Pro-ceedings of the Eighth World Conference onEarthquake Engineering, San Francisco,1984, Vol. VI, pp. 765-771.

4.15 Velkov, M, Ivkovich, M and Perishich, A.“Experimental and analytical investigation ofprefabricated large panel systems to be

constructed in seismic regions”, Proceedingsof the Eighth World Conference on Earth-quake Engineering, San Francisco, 1984,Vol. VI, pp. 773-780.

4.16 Youlin, C, Yinsheng, L, Rui, C, Qixun, G andGuang, S. “Experimental study on theseismic behaviour of multi-storey precastlarge panel residential buildings”, Proceed-ings of the Eighth World Conference onEarthquake Engineering, San Francisco,1984, Vol. VI, pp. 781-788.

4.17 Clough, R W, Malhas, F and Oliva, M G.“Seismic behaviour of large panel precastconcrete walls: analysis and experiment”,Journal of Prestressed Concrete Institute,Vol. 34 No. 5, 1989, pp. 42-46.

4.18 Scanlon, A and Kianoush, M R. “Behaviour oflarge panel precast coupled wall systemssubjected to earthquake loading”, Journal ofPrestressed Concrete Institute, Vol. 33, No.5, 1988, pp. 124-151.

4.19 Edmonds, F D. Testing of Two ReinforcedConcrete Panel Units, Report 5-75/10,Central Laboratories, Ministry of Works andDevelopment, Lower Hutt, 1975, 92 pp.

4.20 Clough, D P. Design of Connections forPrecast Prestressed Concrete Buildings forthe Effects of Earthquake, PrestressedConcrete Institute Technical Report No. 5,1985.

4.21 Restrepo, J I, Crisafulli, F J and Park, R, 1996.“Earthquake resistance of structures: thedesign and construction of tilt-up reinforcedconcrete buildings”, Research Report 96-11,Department of Civil Engineering, Universityof Canterbury, September 1996, 182 pp.

4.22 PCI. “Design for lateral force resistance withprecast concrete shear walls”, PCI Ad HocCommittee on Precast Walls, PCI Journal,September-October 1997, pp 44-64.

Diaphragms • 59

Chapter 5Diaphragms

5.1 Introduction

5.1.1 DefinitionA diaphragm in the context of this chapter is a horizon-tal or near horizontal element, such as a floor or a roof,which links the lateral force-resisting members in abuilding. In addition to distributing horizontal forcesto the lateral force-resisting elements, it ties the struc-ture together. Two forms of diaphragm have beenidentified [5.1]. The first of these is the simple dia-phragm, which distributes forces that are applied di-rectly to it to the lateral force-resisting system. Theseforces may arise from seismic inertial effects, wind orsoil pressures. The second of these is the transferdiaphragm, which transfers shears arising from windor earthquake actions between the different lateralforce-resisting elements, in addition to functioning asa simple diaphragm.

5.1.2 GeneralThe continuity of reinforcement, which is achieved intypical cast-in-place floor slabs, provides diaphragmswith a high degree of integrity. With precast construc-tion incorporating a cast-in-place topping, the level ofintegrity is reduced as only some of the reinforcementis continuous. The New Zealand Concrete StructuresStandard gives requirements for the interface betweencast-in-place concrete and precast components [5.2].Where this interface is required to sustain cyclic forcesarising for bars, which are subjected to yielding intension and compression, caution is required. Testswith dry extruded members (hollow-core type compo-nents) have indicated that bond failure can occur be-tween the precast and cast-in-place concrete at lowshear stress levels [5.3]. This is not such a problem withunits made by normal casting methods as the surfacecan be roughened and treated to remove laitiance.

In the case of precast concrete flooring units withoutstructural topping, continuity is generally provided byhaving relatively few connections between the precastunits. This results in a loss of redundancy and greatervulnerability to damage from concrete volume changesand seismic actions. In this case particular care isrequired in assessing the forces and displacementsimposed on the connections and ensuring these haveadequate ductility to avoid premature brittle failure.

This chapter addresses the question of what character-istics precast concrete diaphragms require to performadequately. The actions arising on these are discussedand recommendations for their design are made. Inresearching this topic a major difficulty was apparent.With many structural elements extensive testing hasbeen carried out to check theoretical predictions anddevise design criteria. However, with diaphragms thebackground of structural testing is very limited due tothe difficulty of modelling the situation realisticallyand applying the high forces involved. As a result ofthe lack of this basic research, designers are urged totake a conservative approach to the design of floors thatare required to act as diaphragms.

5.2 Requirements ofDiaphragms

5.2.1 GeneralMost diaphragms serve a dual purpose in that, as wellas tying the structure together and distributing lateralforces arising from wind and seismic actions, they arecalled upon to also resist gravity loads. In this chapteronly the diaphragm actions are discussed, but thesimultaneous gravity load actions must also be consid-ered.

In the design of diaphragms the following require-ments need to be considered:

• robustness of the structure;

• serviceability limit state; and

• ultimate limit state.

5.2.2 Serviceability Limit StateCombinations of actions, which may be expected toarise several times during the life of the structure, needto be considered to ensure it is serviceable. Furtherdetails are given in Section 5.4.3.

5.2.3 Ultimate Limit StateThe design strength [the design strength is equal to thenominal strength (theoretical strength) times the strengthreduction factor] of the structure is required to be equal


to or greater than the maximum design action (thedesign action is equal to the sum of the load factoredactions) determined from the specified load combina-tions. As yielding of the reinforcement is permitted inthis limit state, redistribution of actions due to plasticbehaviour may be assumed and the forces (but notcracks) arising from internal restraint due to factorssuch as creep, shrinkage or temperature change, maybe neglected.

For structures that are designed to be ductile in severeearthquakes there is a further requirement related tocapacity design. This requires the nominal strength ofa diaphragm to be sufficient to sustain the structuralactions associated with over-strength actions acting inthe potential plastic hinges of the chosen energy dissi-pation mechanism. Further details on the ultimate limitstate requirements are given in Section 5.4.3.

5.2.4 RobustnessAn important aspect in the design of precast construc-tion is to ensure that the structure maintains its integrityin the event of premature failure of a connection ormember. Following a progressive collapse in thecorner of a multi-storey precast concrete building atRonan Point in London in the 1960s, specific designrequirements were introduced into the British code ofpractice for the structural use of concrete. These wereextended and incorporated in the British code, BS8110[5.4], where they are referred to as “robustness require-ments”. In the 1995 revision of the New ZealandConcrete Structures Standard [5.2] the term “structuralintegrity” is used instead of robustness.

Premature failure of an element, such as a column or abeam, may arise for a reason not specifically consideredin the design. Reasons for such failures may include: anintense localised fire, impact of vehicles, poor work-manship, explosions due to gas, flour milling or chemi-cals, or seismic effects not allowed for in the design,such as the impact with adjacent structures or fallingmaterial. In the event of such a failure it is importantthat a progressive collapse is prevented. The robustnessrequirements of the British code are summarised be-low. For details, the reader is referred to reference 5.4.In developing these criteria, seismic considerations didnot receive the detailed attention that would be appro-priate for seismically active regions.

• The key load bearing elements in a structure, that isthose members whose removal or failure could leadto a general collapse, must be identified and spe-cifically designed, or otherwise protected, to pre-vent removal or failure by accident (gas explosion,impact with vehicle, etc.).

• In the design, the removal in turn of each verticalload carrying element, excluding the key elements

identified above, is considered. Under this condi-tion, collapse of a significant part of the structuremust not result.

• All buildings should be capable of resisting a no-tional horizontal force, applied at each floor simul-taneously, of 1.5% of the characteristic dead weightof the structure at that level.

• All diaphragms are to be provided with effectivehorizontal ties at each level:

(a) around the periphery;

(b) internally; and

(c) to all columns and walls.

• Each column, or other vertical load carrying mem-ber, is to be tied to each floor so that the tie can resista force not less than:

(a) a value related to the loading and tributary areasupported by each column or wall; and

(b) 3% of the design vertical load carried by themember.

In the 1995 edition of the New Zealand ConcreteStructures Standard [5.2] a number of structural integ-rity requirements, which relate to the design of dia-phragms, were introduced. These are contained in twogroups of clauses, namely 4.3.6 and 13.4.3. In section4.3.6 (see NZS3101, 1998 amendment) the designer isrequired to provide a rational load path for forcesacting on a diaphragm. In addition, for the cases wherethe building has three or more storeys and it is sup-ported on precast walls there are three requirements fornominal reinforcement as shown in Figure 5.1 anddescribed in the following paragraphs.

• Reinforcement placed parallel to the precast unitsis required to tie units together above internalsupports and to tie the units into the external walls.This reinforcement is to be proportioned to resist aforce of not less than 22 kN/m and the spacing is notto exceed 3 m.

• Reinforcement transverse to the span of the precastunits is also required to resist a force of 22 kN/mtimes the length of the span(s) attributed to thesupporting wall, based on simply-supported spans.For example, an exterior wall, supporting a 10.0 mspan of precast floor, requires (10 /2) x 22 kN/m =110 kN capacity, in the line of the supporting wall.In this case the reinforcement may be located eitherin the topping, or the supporting wall elementswithin 600 mm of the plane of the floor, or it maybe divided between these zones. The tension ca-pacity has to be continuous over the full width of

Diaphragms • 61

T

T

Greater ofP or T

Greater ofP or T

Greater ofP or T

T = TransverseL = LongitudinalV = VerticalP = Peripheral

Figure 5.1: Typical locations for tying reinforcement in large panel structures [5.2]

the structure. There is no maximum spacing ofreinforcement specified for this case.

• Reinforcement is required to provide a continuoustension capacity of not less than 70 kN right roundthe perimeter of the building. This reinforcement isto be located within 1.2 m of the building perimeterand it may be located within the perimeter walls,precast diaphragm units or topping.

The American Concrete Institute code (ACI 318-95)has very similar requirements as does the CEB-FIPmodel code [5.5], though in the latter code the require-ments are very sensibly not limited to structures sup-ported on precast walls. It is suggested that the NZS3101-95 provisions could well be applied to structuressupported on other components, particularly wheremasonry walls are used.

Robustness is an important characteristic for structuresdesigned to behave in a ductile manner during severeearthquakes. Many actions that may occur in such anevent tend to be neglected. These include the effects ofvertical accelerations, which may increase or reducegravity actions, differential vertical deflections due tosurface waves and the elongation of beams, columnsand walls due to the formation of plastic hinges. Thedetailed consideration of all possible effects and theircombinations is impractical. To prevent possible col-lapse due to these actions robustness is required.

The elongation of members, such as beams, due to theformation of plastic hinge zones has received little

attention in the literature. However, as discussed inChapter 2 and Appendix A1, it could, in extremesituations, lead to loss of support for precast concretefloor or stair systems. Numerous tests have indicatedthat, at a displacement ductility of 6, an elongation of2 percent or more of the overall depth of the beam canbe expected at each plastic hinge, see Appendix A1.This raises the possibility of one floor falling ontoanother and setting off a progressive collapse. Thissituation occurred in some parking buildings in theNorthridge earthquake [5.6]. Provisions to prevent thisoccurrence are detailed in sections 2.3 and 2.4.

Additional potential elongation effects in ductile mo-ment resisting frame structures, associated with theformation of plastic hinges in the beams, are outlinedin the following paragraphs.

• Additional lateral deflection of the external col-umns occurs as indicated in Figure 5.2. This resultsin increased rotations of the plastic hinges at thebase of the columns, and the formation of addi-tional plastic hinges in the external columns near oradjacent to the level 1 beams. The increasedcolumn shear and plastic hinge rotations should beconsidered in design. This behaviour has beenobserved in a frame test [5.7].

• With precast floors beam elongation may be ex-pected to generate wide cracks in the diaphragms atthe sections which are weak in tension. These areat the support positions of the precast units, at thejunction between any two precast units and be-


tie force requiredfor equilibriumdue to localisedP-delta actions inexternal column

beam elongation causes displacementof some columns — additional plastichinging may be induced

Figure 5.2: Localised P-delta action and additional hinging in columns due to beam elongation

tween a precast unit and a beam which is spanningparallel to the unit. In these locations the strengthdepends on the reinforcement in the cast-in-placeconcrete. Wide cracks result in a major reductionin both strength and stiffness of diaphragms. Thisis considered in sections 5.4.3 and 5.5.

• The high flexural and shear deformations in theplastic hinge zones impose high deformations onthe precast units in their immediate vicinity. Thismay lead to local failure of non-ductile precastfloor components, as discussed in section 5.5.

• The elongation of the plastic hinge zones can resultin tension being induced in the precast units that areclose to the plastic hinges. This is particularlycritical if the plastic hinges are adjacent to the mid-span regions of the precast unit. The compressionforce balancing the additional tension force in theslab acts on the beam and it can significantlyincrease the resultant flexural strength. This hasimportant implications for capacity design, as thedesign actions in the beam-column joints, shear inthe beams and the strength of the columns arederived from the over-strength of the beam plastichinge zones. This aspect is considered in moredetail in section 5.5.

It should be noted that current methods of analysis,including time-history methods, do not predict theincreased lateral displacements, column shears, plastichinge rotations and additional column plastic hingeformation, which arise due to elongation.

5.3 Internal Restraint Actionsin Diaphragms

5.3.1 Volume ChangesThese arise due to creep, shrinkage and thermal effectsin the concrete. Their main significance lies in their

influence on the serviceability of the structure. Differ-ential creep and shrinkage effects in concrete memberscan lead to significant out-of-plane deflections. Floorslabs and roofs exposed to the sun can sustain appreci-able deformation due to differential temperature con-ditions. As described in section 2.3.4 these can causedamage if adequate allowance is not made for them.Appendix C1 indicates how these actions can be as-sessed.

Volume change strains resulting from creep, shrinkageand temperature change (seasonal variation) can resultin crack formation in the concrete topping. In somesituations the formation of a crack, at say a control jointin the slab, may suppress the opening up of othercontrol joints, leading to a few unacceptably widecracks forming in the structure. A method of assessingpotential crack locations and widths is outlined inAppendix C2.

5.3.2 Minimising Volume ChangeRestraint ForcesWhere practical, the lateral force-resisting elements ina structure should be arranged to minimise volumechange restraint actions. For example, in the buildingshown in Figure 5.3, the arrangement of walls in (a)allows the volume changes to occur with a minimum ofrestraint, while the arrangement in (b) is likely to causeeither extensive cracking in the suspended floors, ordiagonal cracking in the walls.

5.4 Analysis and Design ofDiaphragms

5.4.1 Diaphragm FlexibilityIn New Zealand many simple diaphragms consist ofprecast concrete units with cast-in-place concrete top-ping. These are commonly assumed to be rigid in thestructural analysis for seismic actions. However, in the

Diaphragms • 63

transfer diaphragm case, where the level of shearsustained is high, it may be necessary to allow fordiaphragm flexibility.

Some common structural analysis programs, such asolder versions of ETABS, assume diaphragms arerigid. Where these programs are used, some allowancefor the flexibility of diaphragms can, in some in-stances, be incorporated into the analysis by reducingthe stiffness of the wall or frame elements to which thelateral forces are being transferred.

The importance of flexibility in a diaphragm on theforce distribution in a structure changes with the limit-state being considered. In general terms it is muchmore significant in the serviceability than in the ulti-mate limit-state, where inelastic behaviour and conse-quential increase in the displacements of the lateralforce-resisting elements occurs.

The New Zealand Concrete Structures Standard [5.2]contains guidance on when a diaphragm may be as-sumed to be rigid for the purposes of analysis. Thecommentary to clause 13.3.4 indicates the rigid dia-phragm assumption may be made where the lateraldeformation of the diaphragm is less than twice theaverage inter-storey drift in the relevant storey found inan elastic analysis for the design seismic forces at theultimate limit state. This provision is similar to thecorresponding recommendations in the 1997 UBCcode and the proposed International Building code2000 [5.8, 5.9].

For design purposes, a diaphragm may be assumed toact as a deep beam. Cracking in the topping concretedue to volume changes and beam elongation can re-duce both the effective flexural and shear stiffnesses of

the diaphragm. To assess the stiffness appropriate foran analysis it should be noted that cracking is likely toexist in the cast-in-place concrete between precastconcrete units and at the support positions. The stiff-ness can be expected to vary with the direction of theaction. In most situations the analysis is not sensitiveto the diaphragm stiffness and only in a relatively fewcases is it necessary to assess this.

Where the “strut and tie” approach is used to design adiaphragm the basic truss may be used to assess thestiffness. In such calculations allowance should bemade for tension stiffening of the concrete.

5.4.2 Influence of Cracking on StrengthTests on the transfer of forces across cracks by aggre-gate interlock action have been made for crack widthsup to and including 0.5 mm. The stiffness has beenshown to decrease with increasing crack width [5.10,5.11 and 5.12] and at some stage the interlock actionmust become ineffective. This limit has not beenestablished. Tentative recommendations are made inthe following paragraph.

For wide cracks (where aggregate interlock action isineffective) some shear force can be resisted by kinkingof the reinforcement. Hawkins and Mitchell found thatlongitudinal reinforcement, which was placed in thebottom of slabs to control punching shear failures,could work at an angle of 30o [5.13]. If this mechanismis assumed for design two points should be considered,namely:

(i) the diameter and spacing of the reinforcementshould be such that the concrete is not split, asillustrated in Figure 5.4; and

Wall Wall

Plan Plan

(a) Restraint to volume changeStrains minimised

(b) Walls restrain movements dueto volume change (not preferred)

This arrangement may lead to theformation of wide cracks in theslab or diagonal cracking in thewalls

Figure 5.3: Structural arrangement to minimise volume change restraint actions


30° max

shear displacement A

A

Kinking of bar across a crackMax shear force transferred bybar = Asfy/2

localised highbearing stressesdue to sheartransfer by bar

lines of force due tobar bearing on concrete

bursting tension maylead to cracking

Figure 5.4: Shear transfer across a crack by kinking of bar

(ii) strain in the reinforcement should not exceed thelevel corresponding to the maximum stress valuefor the reinforcement. If this value is exceeded itmust be assumed that the bar has failed in tension.Plain bars should be used where wide cracks areanticipated as the poorer bond allows yielding todevelop over a longer length.

5.4.3 Design Limit StatesTo ensure adequate performance a diaphragm needs tobe proportioned so that it can satisfy the different limitstates.

Serviceability Limit State

Under serviceability actions diaphragms should beproportioned so that crack control and deflection crite-ria are satisfied and the members remain in the elasticrange, with the possible exception of allowing limitedyielding of reinforcement at expansion joints or atsupports where continuity moments have developed.With this limit state restraint forces arising from creep,shrinkage and thermal effects should be included [5.2].

Ultimate Limit State

There are two sets of requirements that arise from thislimit state:

• The design strength of the diaphragm must besufficient to sustain the combinations of loadfactored actions associated with gravity loads andlateral forces (arising from seismic effects, wind orsoil forces). Yielding of the reinforcement dissi-pates forces arising from restraint to volume changesin the concrete. Consequently, these actions do not

need to be considered for this limit state. However,the possible effects on the load resistance due tocracks, which may have been opened up by theseactions, should be considered.

• The diaphragm must be designed to satisfy capac-ity design requirements. These ensure that theductile primary energy dissipating mechanism canbe sustained in a severe earthquake. For a ductilestructure the capacity design principles must beused, with the nominal strength of the diaphragmbeing equal to or greater than the actions sustainedwhen over-strength actions act in the potentialplastic hinge zones.

The diaphragm chapter of the New Zealand ConcreteStandard (Chapter 13 of Reference 5.2) also requiresthat the diaphragms should generally remain elasticunder the application of the capacity design forces.This condition is specified as it is difficult to detailthese elements to behave in a ductile manner. How-ever, it should be noted that inelastic deformation isinevitable where the supporting beams contain poten-tial plastic hinge zones. Once plastic hinging occurs,elongation ensures extensive yielding develops in thediaphragm. Where diaphragms are supported by precastwalls the Standard requires minimum levels of nomi-nal reinforcement to be used (see section 5.2.4).

In the situation where high shear forces are sustainedby a diaphragm, which has beams embedded in it thatare expected to form plastic hinges in a major earth-quake, consideration needs to be given to the change inbehaviour associated with beam elongation. The widecracks that may be imposed on a diaphragm in thissituation, can cause gross changes in both its stiffness

Diaphragms • 65

and strength. Consequently, the stiff behaviour of thediaphragm assumed in the elastic analysis of the struc-ture could cease to be appropriate. In extreme cases,the diaphragm may start to distribute the seismic shearsarising from loads it supports in a tributary mode.Consideration may need to be given to the change in theprimary energy dissipating mechanism associated withthe loss of stiffness and strength. Further research isrequired on the influence of beam elongation on theperformance of diaphragms. However, an essentialfeature is to ensure that the vertical loading carryingcapacity is not destroyed by elongation (see sections2.3 and 2.4). Redistribution of seismic actions due tothe loss of diaphragm stiffness can be accepted in theultimate limit-state provided this does not lead toexcessive inelastic deformations being imposed on thelateral force-resisting elements.

5.4.4 Strength of Diaphragms

Strut and tie method

This approach provides a valuable tool for the designof diaphragms [5.14, 5.15, 5.2 and Commentary to5.2]. Intermediate beams in the slab can act in a similarmanner to stirrups in a beam, with diagonal compres-sion forces being resisted by the concrete topping andthe precast units where these are continuous. A strutand tie model for a transfer diaphragm is illustrated inFigure 5.5. The need to provide reinforcement totransfer the forces to the lateral force resisting elements(such as wall or frames) should be noted. This isillustrated in the figure. The strut and tie method isparticularly useful for complex details, such as may becaused by large openings in the floor [5.14]. Thereinforcement in the topping may also act as tie or shearreinforcement. However, where there is poor bondbetween the precast units and the cast-in-place con-crete, particular care is required to prevent delamination(see Delamination ). This is likely to arise where aconcentrated band of reinforcement in the topping isrequired to sustain high tensile stresses or where directtension forces are introduced by reinforcement into thein situ concrete.

Delamination

Recent research has shown that considerable care isrequired where reinforcement located in the topping isused to transmit tension forces into a precast dia-phragm. This is particularly a problem where theprecast units are made from extruded dry concrete. Theproblem is accentuated where wide cracks develop dueto elongation of the beams. Tests have shown that insome conditions delamination may occur at this inter-face. This arises as a result of either high bond forcesgenerated by highly stressed reinforcement entering the

topping, or buckling of the reinforcement which hasfirst yielded in tension and then subjected to compres-sion, or a combination of these actions [5.3, 5.6].

The application of a tension force to the topping on aprecast slab is illustrated in Figure 5.6, together withthe failure mechanisms that have been observed. Inthese tests precast prestressed hollow-core units wereused. These are manufactured from a dry concrete mixusing an extrusion process. The units were eccentri-cally prestressed and they contained no vertical tie orstirrup reinforcement in the concrete between the cores.As shown in the figure, the tension force induced inthe continuity reinforcement in the topping above thesupport, together with the balancing compression force,induced a couple bending moment in the concretesection at the end of the beam. Initially, a tension crackformed in the topping concrete close to the terminationpoint of the starter or continuity bars. This was fol-lowed by a brittle failure of the wire mesh in the toppingas it had a low ductility, and either a delaminationfailure or a failure in the reduced concrete thickness inthe level of the cores.

The situation illustrated in Figure 5.6, where a tensionforce is transmitted to the topping over a wide crack, isimproved by using plain bars. The reduced bondresistance allows yielding to extend further into theconcrete and there is a reduction in the intensity of thedelamination shear stresses. In addition there is lessdisruption of the topping concrete due to localisedcracking associated with the high bond stresses ofdeformed reinforcement. The use of plain bars isparticularly recommended where this reinforcement isexpected to bridge wide cracks, such as may be inducedby elongation of beams when plastic hinges are formed(see Chapter 2 and Appendix A1).

Where hollow-core units are used, the reliance on thebond between the in situ and precast concretes can beeliminated by breaking out the flange above two ormore cores at the ends of each unit and adding rein-forcement and cast-in-place concrete in the voids (seesection 2.4.2).

Other factors

In designing diaphragms a number of other factorsshould be considered in connection with strut and tieanalyses.

• High diagonal compression stresses in the concretein diaphragms with hollow core units may causehorizontal splitting cracks to form as illustrated inFigure 5.7. Such cracking could allow the bottomhalf of a unit to drop out, leaving the top half without appreciable flexural reinforcement to resist thegravity loads.


• There is a potential problem with diagonal com-pression buckling, which is particularly severewhere there is poor bond between the precast and insitu concretes. This situation is made more criticalwhen the in situ concrete contains significant rein-forcement that has been yielded in tension andsubsequently subjected to compression [5.3].

• Pretensioned reinforcement in the precast units

gives the diaphragm very different stiffness andstrength characteristics in different directions. Thisaspect needs to be considered when the direction ofstruts and ties in strut and tie models is devised. Theprestress inhibits the formation of cracking acrossthe line of prestress, and this can have a direct effecton the formation of the actual struts and ties andhence the structural performance.

< > <>

<> < >

Earthquake

Forces

level oftransfer

diaphragm

Section A—A Section B—B

compression tension

wall

wall

wall wall

A B

AB

precastunits

Plan on transfer diaphragm

<>

<> <>

<>

<><><><><>

<>

<>

<>

<>

<>

<>

wallwall

<>

KEY

compression

tension

<>

<>

<>

<>

< >

< >

< > < > < >

< >

< >< >

< >

< >

<>

<>

<>

<>

Figure 5.5: Action in a diaphragm — with strut and tie model

Diaphragms • 67

crack

A

A

mesh

delamination failurefloor starter orcontinuity bar

Supportcompressionin concrete

prestressforce Section A-A

bar

prestressingstrand

mesh

compression force actson end of beam

Figure 5.6: Potential failure mechanisms at the ends of hollow-core members with highly

stressed bars in the topping

5.5 Interaction of Plastic HingeZones and DiaphragmsNumerous tests of beams and beam column sub assem-blies have indicated that at design levels of ductilityplastic hinge zones sustain both high elongation andhigh shear deformation in addition to flexural rotation.A typical average shear strain over a length equal to thebeam depth is 0.03, while a typical elongation is 0.02times the beam depth [see references A1 to A5 inAppendix A].

Where precast floor units are either placed adjacent toa plastic hinge, or supported by a beam containing aplastic hinge, high deformations may be imposed onthe units. Where these are brittle, and particularlywhere they do not contain any shear reinforcement, theimposed deformation is likely to result in shear failure,as indicated in Figure 5.8. Research is required toestablish the extent of this potential problem.

Elongation of beams, due to plastic hinge formation,may in part be restrained by diaphragm action. How-ever, tests on beams with composite reinforced con-crete slabs have indicated that these slabs provide nosignificant restraint. The reason for this is that oncereinforcement in the slab yields, on load reversal it actsto prop open the cracks, and hence increase elongation[5.16 and 5.17]. Tests have indicated that very highaxial forces were required to prevent elongation fromoccurring. If these restraint forces can be sustained theflexural strength of the beams is greatly enhanced[5.17]. For the purposes of capacity design it isimportant to assess the likely strength enhancement ofthe beams due to restraint from diaphragms. An underassessment could in some cases lead to a non-ductilefailure occurring, possibly due to shear failure in beamsor plastic hinges forming in the columns leading to acolumn sway failure mode.

Figure 5.9 shows a part of a perimeter frame for a

cast-in-placeconcrete

compression inmember

possible horizontal crack dueto inclined componentof compression force

Figure 5.7: Possible horizontal cracking in hollow-core slabs due to compression


building together with an associated diaphragm, madefrom precast pretensioned units and cast-in-situ con-crete topping. With the arrangement shown in thefigure the pretensioned units span directly past twoperimeter columns, which are on gridlines 5 and 6.Elongation of the beams, due to plastic hinge formationat the faces of these columns, would be partiallyrestrained by these units. There are a number ofpossible outcomes of this interaction between the beamsand the diaphragm as indicated in the following para-graphs.

• The restraint provided by the diaphragm induceshigh shears at the interface with the beam, asillustrated in Figure 5.9 (b). This might lead to acomplete shear failure at this interface or it mightlimit the shear transfer at this section. A situationsimilar to the one illustrated has been examinedanalytically in the literature [5.18]. The shear thatcould be transmitted across the critical interfacewas assessed with a strut and tie analysis (or byshear friction). It was found that this action in-creased beam negative moment flexural strengthby 130 percent. There was no significant influenceon the positive moment strength. However, theincrease in bending moment resistance is sufficientto significantly raise the shear forces in the beams,and the moments and shears in the columns, whichputs at risk the intended ductile failure mode.

• The resultant shear transmitted across the criticalinterface applies an eccentric axial tension force tothe precast units. As illustrated in Figure 5.9(c) thismay break the back of one or more units leaving itopen to potential collapse when the tension forcereduces.

• The restraint provided by the diaphragm is suchthat the axial compression force in the beams leadsto a primary flexural compression failure. Such

failures would be associated with high strengthgain in the beam.

A further source of strength increase due to diaphragmrestraint to beam elongation needs to be considered.The transverse beam at column 4 supports the precastunits. Elongation of the perimeter frame beams in thevicinity of this column could be expected to generatewide cracks in the topping on each side of the trans-verse beam. The situation is shown in Figure 5.9(d).Forces are transmitted across these cracks by rein-forcement placed in the in situ concrete. However, inaddition it can be seen that the two halves of the floorare acting as deep beams, with bending moments andshears being sustained at the end of the cantilever likemembers, see Figure 5.9(d)). This bending action,together with the force transmitted across the cracksalong the transverse beam on line 4, applies an axialforce to the beams in the vicinity of column 4, and thismay substantially increase the flexural strength of thebeams [5.18].

5.6 Overseas PracticeIn the USA, with its wide range of seismicity, there area wide variety of diaphragm construction practices.Generally there appears to be a preference for untoppedprecast concrete floor units to be used to form dia-phragms. For example, the PCI Manual for the Designof Hollow-Core Slabs [5.19] gives information on thedesign of untopped diaphragms. Shear forces areresisted by shear friction along the longitudinal jointswith the clamping force provided by transverse barsplaced at the ends of the units. Limited full-scaletesting [5.20] has shown that within the limits of theshear capacity available using this method, the systemexhibited reasonable strength and ductility.

Untopped diaphragms usually contain embedded steel

A A

B

B

plastic hingein beam

cracks open up dueto elongation inplastic hinges

Plan on support zone forprecast floor units

Section A—A

potential shearfailure

Section B—B

Figure 5.8: Deformations imposed on precast units adjacent to plastic hinges

Diaphragms • 69

plates, that are later connected by welding additionalsteel elements to these plates. The robustness of theseconnections is in doubt where they are called upon toresist seismic actions. Some testing of these connec-tions has been carried out [5.21] and it was found that,provided that the loose plates, which were welded tosteel embedments in the precast concrete units wereslotted or narrow (to avoid brittle behaviour as theresult of volume changes), the behaviour was satisfac-tory. Unfortunately only monotonic loading was ap-plied so the results cannot be immediately adopted fordiaphragms designed to resist seismic forces.

Concern has been expressed by several US engineersas to the adequacy of mechanical connections fordiaphragms in seismic conditions [5.22, 5.23]. Cloughhas stated [5.25] that:

Untopped diaphragms in which inter-element connection is made by groutingor mechanical connectors, have rela-tively low in-plane shear strength andductility and are most suitable whenseismic equilibrium and compatibilityforces are small. In zones of high seis-

portion of flooracting as a

cantilever beam

precastunits

plastic hinges

moment and shearin diaphragm

>compression in beams

forcestransmitted

across crackby reinforcementin insitu concrete

(d) Portion of precast floor acts as deep beam

><

shear applied toprecast unit mesh fails

pretension strands

(c) Shear transfer to precast unitcauses tensile failure

>< >< ><

<>tension in units

compression in beams

shear transfer between beamsand diaphragm

(b) Shear at interface betweenbeams and diaphragm

1 2 3 4 5 6 7

potentialplastic hingezones

precastunits

(a) Plan on perimeter frame and floor diaphragm

Figure 5.9: Interaction of beam elongation due to plastic hinges and diaphragm


mic intensity, or with structural configu-rations which impose large in-planecompatibility forces under lateral load,diaphragms joined by cast-in-place re-inforced concrete, either as pour stripor as a topping, usually are more satis-factory.

From this brief survey of US practice some majorconcerns emerge regarding the use of untopped dia-phragms in seismic situations.

European practice, at least as represented by the FIP[5.25], suggests an approach similar to that recom-mended by the PCI. The use of structural toppings inEurope is not common. However, in another FIPpublication [5.26] structural toppings are acknowl-edged as improving diaphragm action in seismic con-ditions.

A review of Japanese practice for multi-storey framebuildings [5.27] indicates that cast-in-place concretefloor units tend to be used for more cellular construc-tion such as apartment buildings [5.28]. The use of newsystems in Japan is preceded by extensive testingprogrammes [5.27].

5.7 Recommendations

5.7.1 GeneralIn this chapter a number of recommendations are maderegarding the design of precast diaphragms togetherwith areas that require further research.

5.7.2 RobustnessSatisfactory behaviour in extreme events, such as gasexplosions, impact by vehicles or severe earthquakes,requires the structure to be robust. For this purposeappreciable structural redundancy is desirable so thatthe failure of one element does not lead to a progressivecollapse. Where this is not practical the critical ele-ments should be designed conservatively. In additionit is important that the different elements in a structureare tied together. In this regard attention is drawn to the“robustness requirements” in the British code and the“structural integrity requirements” in the New ZealandConcrete Standard (see section 5.2.4).

5.7.3 Internal restraint actionsStresses and cracking may be induced due to internalrestraint to movements induced by thermal effects,shrinkage and creep of the concrete. As shown inSection 2.3.4 considerable damage can occur if inad-equate allowance is made for these effects. Methods ofassessing thermal, shrinkage and creep actions and

associated crack widths, are given in Appendices C1and C2. These effects are important for the serviceabil-ity limit state but of less significance for the ultimatelimit state (see sections 5.3 and 5.4.3).

5.7.4 Analysis and DesignThe influence of the flexibility of transfer diaphragmsshould be checked for the serviceability limit state.Criteria are given that indicate when allowance needsto be made for diaphragm flexibility and when thediaphragms may be considered to be rigid (see section5.4.1).

Section 5.4.4 illustrates how the “strut and tie” methodof analysis may be used to design a diaphragm. Theimportance of providing tie or drag bars to transfer theseismic force in the diaphragm into the lateral forceresisting frames or walls is indicated. Caution is re-quired where reinforcement is used to transmit tensiondirect into the in situ topping, as tests have shown thatdelamination can occur with some precast units. Thisis a particular problem at the support positions forprecast units, where elongation due to the formation ofplastic hinges in beams may generate wide cracks.

Where forces need to be transmitted across wide cracks(due to beam elongation) the situation is improved byusing plain bars. The lower bond strengths of thesebars reduces the delamination shear stresses and allowsyielding to occur over a longer length giving an in-crease in ductility. With hollow-core units the tensionforce may be introduced into the ends of the units bylocally breaking out the top flanges and placing rein-forcement and in situ concrete in the voids (see sections5.4.4, 2.3.5 and 2.4).

5.7.5 Deformation in Plastic HingesWide cracks, associated with plastic hinge formationin beams embedded in a diaphragm, may greatly re-duce the strength and stiffness of a diaphragm, chang-ing the way it distributes forces to the lateral forceresisting elements. The influence that this may have onthe seismic performance of a building should be as-sessed. Where this may occur, designers should ensurethat an adequate load path remains for the seismicforces (see sections 5.4.2 and 5.4.4).

The elongation in plastic hinges, in some situations,may be partially restrained by diaphragms, particularlywhere these contain prestressed units. This restraintcan impose appreciable axial compression on the beamplastic hinges and result in very significant increases inflexural strength. Two situations where this may occurin perimeter frame beams are outlined. It is importantto recognise and allow for this action as the unexpectedover-strength could result in the formation of a non-

Diaphragms • 71

ductile failure mode, leading to premature collapse ofthe building, see section 5.5.

The high flexural and shear strains sustained by plastichinges in beams embedded in a diaphragm may imposehigh deformations on precast units. As some of thecommonly used units are relatively brittle in terms ofthe shear and flexural deformations, the possibilityexists of local failures occurring in these units in theneighbourhood of the plastic hinge zones (see section5.5). Research is required to establish the extent of thisproblem.

5.7.5 Untopped DiaphragmsIn view of overseas findings untopped precast concretediaphragms should not be used in a seismic resistingstructure unless a special study is undertaken whichincludes assessing the effects of concrete volumechanges. Particular care is required where this form ofconstruction is used in frame structures, in which thebeam may be subjected to elongation in a severeearthquake due to plastic hinge formation. Specialdetailing is warranted to ensure that ductile perform-ance can be obtained (see section 5.6).

5.8 References5.1 Kolston, D and Buchanan, B W. “Diaphragms

in Seismic Resistant Buildings”, Bulletin ofthe New Zealand National Society forEarthquake Engineering, Vol. 13, No. 2, June1980, pp 162-170.

5.2 Code of Practice for the Design of ConcreteStructures, NZS 3101:1995 Parts 1 and 2,Standards Association of New Zealand,Wellington.

5.3 Herlihy, M D, Park, R and Bull, D K. “Theneed to consider capacity design principlesfor reinforcement concrete diaphragmssubjected to seismic actions”, ProceedingsNZNSEE Technical Conference, March 1996,p12-717.

5.4 BS 8110: Parts 1 and 2, 1985 - Structural Useof Concrete, British Standards Institute,Milton Keynes.

5.5 Comite Euro-international du Beton. CEB-FIPModel Code 1990, Thomas Telford, London,1993, 437pp.

5.6 Iverson, J K and Hawkins, N M. “Performance ofprecast/prestressed concrete buildings duringthe Northridge Earthquake”, PCI Journal,Vol.39, No. 2., March 1994, pp 38-55.

5.7 Fenwick, R C, Ingham, J M and Wuu, J Y.“Performance of ductile R.C. frames underseismic loading”, Proceedings NZNSEETechnical Conference, Mar 1996, pp 20-26.

5.8 International Conference of Building Officials,Uniform Building Code - UBC-1997, Vol. 2,Conference of Building Officials, Whittier,California.

5.9 International Building Code 2000 – Final Draft,International Code Council, July 1998.

5.10 Paulay, T and Loeber, P J. “Shear Transfer byAggregate Interlock”, Shear in ReinforcedConcrete , Vol. 1, ACI Publication, SP42,1974, pp 1-16.

5.11 Fenwick, R C and Paulay, T. “Mechanisms ofshear resistance of concrete beams”, Struc-tural Journal, ASCE, No ST10, Vol. 94, Oct1968, pp 2325-2350.

5.12 Houde, J. Study of Force-displacement Rela-tionships for the Finite Element Analysis ofReinforced Concrete, PhD Thesis, StructuresLaboratory, McGill University, Montreal,Canada, Dec 1973, 326pp.

5.13 Hawkins, N M and Mitchell, D. “Progressivecollapse of flat plate structures”, Journal ofAmerican Concrete Institute, Vol. 76, No. 7,July 1979, pp 775-808.

5.14 Schlaich, J, Schafer, K and Jennewein, M.“”Towards a consistent design of structuralconcrete”, PCI Journal, Vol. 32, No. 3,May-June 1987, pp 74-151.

5.15 Collins, M P and Mitchell, D. PrestressedConcrete Basics, Canadian PrestressedConcrete Institute, Ottawa, 1987, pp 386-427.

5.16 Fenwick, R C, Davidson, B J and McBride, A.“The influence of slabs on elongation inductile seismic resistant frames”, TechnicalProceedings NZNSEE Conference, Apr.1995, pp36-43.

5.17 Fenwick, R C and Megget, L M. “Elongationand load deflection characteristics of rein-forced concrete members containing plastichinges”, NZNSEE Bulletin, Vol. 26, No. 1,Mar. 1993, pp28-41.

5.18 Fenwick, R C, Davidson, B J and Lau, D.“Strength enhancement of beams in ductileseismic resistant frames due to prestressedcomponenets im floor slabs”, SESOC Jour-nal, Vol.12, No. 1, 1999, pp 35-40.


5.19 PCI. Manual for the Design of Hollow-CoreSlabs, Prestressed Concrete Institute, Chi-cago, 1985.

5.20 Moustafa, S E. “Effectiveness of shear frictionreinforcement in shear diaphragm capacity ofhollow-core slabs”, PCI Journal, Vol. 26,No. 1, Jan-Feb 1981, pp 118-132.

5.21 Stanton, J F, Anderson, R G, et al. Momentresisting Connections and Simple Connec-tions, Research Project No 1/4, PrestressedConcrete Institute, Chicago, 1986.

5.22 Gates, W E. “Seismic design considerations foruntopped precast concrete floor and roofdiaphragms”, Proceedings of a Workshop onthe Seismic Design of Prefabricated Buildings,Applied Technology Council, p 517-535.

5.22.1 Design of Prefabricated Buildings, AppliedTechnology Council, p 517-535.

5.23 Elsesser, E. “Analytical Modelling Problemsof Precast Diaphragms”, Proceedings of aWorkshop on the Seismic Design of Prefabri-cated Buildings, Applied TechnologyCouncil, pp 463 - 480.

5.24 Clough, D P. “Considerations in the design ofprecast concrete for earthquake loads”,Journal of Prestressed Concrete Institute,Vol. 27, No. 2, Mar-Apr, 1982, pp 618-639.

5.25 FIP. Recommendations of Precast PrestressedHollow Core Floors, Thomas Telford,London, 1988, 31pp.

5.26 FIP. Guide to Good Practice - HorizontalComposite Structures, Second Draft, FIP,Paris, 1998.

5.27 Kanoh, Y. “Review of Japanese precastconcrete frame systems used as buildingstructures”, Seminar on Precast ConcreteConstruction in Seismic Zones, JapanSociety for the Promotion of Science,Tokyo, October 29-31, 1986, Vol. 2, pp35-54.

5.28 Suenaga, Y. “Design guidelines for precastconcrete structures in Japan”, Seminar onPrecast Concrete Construction in SeismicZones, Japan Society for the Promotion ofScience, Tokyo, October 29-31, 1986,Volume 2, pp 1 - 34.

Connections between Precast Concrete Units by Grouted and Welded Bars • 73

Chapter 6Connections between Precast Concrete

Units by Grouted and Welded Bars

6.1 GeneralThe grouting of bars and interfaces, and welding ofreinforcing steel at connections between precast con-crete structural elements, is common in the New Zea-land precast concrete industry. Many types of connec-tion, such as between columns and footings, whichformerly would have always been of cast-in-placeconstruction, now frequently rely on grouted connec-tions for their structural integrity.

The basic requirement of any grouted or welded con-nection is that it satisfactorily meets all the strength,stiffness and other performance criteria for the service-ability and ultimate limit states. In most cases thismeans that connections should be at least as satisfac-tory in all respects as cast-in-place connections.

The design approach to grouted and welded connec-tions is similar to that of cast-in-place connections.However, a higher level of workmanship and qualityassurance is required. The procedures for makinggrouted and welded connections are far more complex,more difficult (sometimes impossible) to inspect, andit may not be possible to make adjustments to theconnection after it has been made. High quality work-manship is essential to the satisfactory achievement ofthese connections.

In this chapter, guidance and recommendations aregiven on the different stages of designing and con-structing grouted and welded connections.

6.2 Selection and Types ofGrout

6.2.1 General PrinciplesEvery grouting situation requires careful considerationby the specifier as to the most appropriate grout for thepurpose.

There are a large number of factors that may influencethe selection of a grout. These include the rate ofstrength gain, required final strength, shrinkage (orvolume change) characteristics, long-term appearance,durability (under chemical and water attack), place-

ment conditions (presence of moisture or dust, hot orcold ambient temperatures, vertical or horizontal place-ment, etc.), viscosity, and “wetting” characteristics.Viscosity and “wetting” determine how effectively airis displaced and how well grout bonds to componentsof the assembly. These factors also dictate what clear-ance is needed between interfaces of assemblies orbetween duct walls and reinforcing bars. Also affectedby grout type is the diameter of the tremie or injectiontube used to place the grout.

6.2.2 Types of Grouts

Epoxy and Polyester Grouts

Epoxies and polyesters are two-part mixtures. Theygenerally develop very high compressive strengths ina relative short time (40 MPa in 6 hours) and requireshorter embedment lengths for grouted-in bars etc.,than for cement-based grouts. However, the designermust exercise caution to ensure that even though thebond between grout, bar and surrounding concrete issufficient, the failure mechanism does not become aconcrete cone pull-out (see Chapter 7). Bar yieldingfollowed by bar ultimate tensile failure is the preferredfailure mechanism.

These grouts, which come in a wide range of viscositiesand strengths, are sensitive to construction conditions.In particular, water and dust in a void can seriouslyreduce the strength as well as the bonding capability ofthe grout. Water can have a detrimental influence onthe long-term properties of some polyesters.

The chemical reaction of these grouts is usually fastand generates significant amounts of heat. The resinshould not be present in such a large volume that theheat of reaction causes the resin to boil. If this occursthe grout loses strength and becomes a granularunbonded mass.

The manufacturers’ instructions should be consulted forfull details on appropriate usage of such special grouts.

Cement-based Grouts

Cement-based grouts, which are usually cheaper thanresin grouts, are often favoured when it is not necessary


to use the rapid strength-gain of resin grouts. Thetypical strength development of cement grouts is 15MPa at 1 day, reaching 60 MPa at 28 days.

Although in some situations neat cement grout can beused for grouting, high-strength, non-shrink cement-based grouts are usually specified. High strength groutis specified, firstly to ensure that forces in the bars canbe transferred to the surrounding concrete and, sec-ondly, to minimize the time before the assembly canwithstand loading.

It is recommended that the minimum grout strength be10 MPa greater than that of the surrounding concrete.Non-shrink or expanding grout is specified to ensurethat the bond and load carrying ability of the grout is notdiminished by the presence of shrinkage cracks orentrapped air [6.1].

The manufacturers’ instructions should be followed indetail when using proprietary grouts.

Occasionally, specifiers/contractors may choose toproduce their own cement-based grout, accepting theresponsibility for failure, rather than use a proprietarygrout. In these circumstances there should be enhancedquality control of production, placement and testing(for verification of design performance) when com-pared to that required for proprietary systems.

6.3 Grouting Situations

6.3.1 GeneralIn any grouting situation the following fundamentalprinciple applies:

• Replace air in the void completely by grout, so thatthe full dependable load carrying ability and stiff-ness of the connection are achieved.

Situations where bars and interfaces are grouted gener-ally fall into two groups, namely vertical and horizon-tal void grouting. A combination of these two situa-tions within a particular connection detail is not un-common. Techniques for successful grouting varyaccordingly.

Bars and holes must be free of grease, paint, dust, rustand mill scale for the concrete/grout/steel bond to befully effective. It is essential that surfaces are not oilcontaminated from blowing out with compressed aircontaining oil (Section 6.6.2).

Loss of Concrete Cover due to Gap SealingMaterials

Perimeter gaps or ducts are sealed to prevent grout loss

during grouting by flexible spongy strips, epoxy putty,or cement mortars. Designers must allow for thepossible reduction of cover to steel reinforcementcaused by the use of sealing materials. Cover reductionmay promote premature corrosion, and in these casesadditional cover to such steel elements should beprovided. A further consideration of the reduction ofcover concrete is the shortening of the internal leverarm ‘d’ between the tension reinforcement and theconcrete compression zone. The reduction of ‘d’ willdecrease the available bending moment capacity at thatsection.

6.3.2 Vertical Void Grouting

Vertical Starter Bars

Starter bars can be grouted into vertical holes that can beeither drilled, formed with formers left in place, or informed holes with the formers removed (see Figure 6.1).

T

grout

hole formedwith corrugated

metallicduct

e = development length of the reinforcing bar or test values

Figure 6.1: Preferred detail for formed voids

Masonry drills usually provide a desirable rough wallsurface for good mechanical bond at the grout/concreteinterface. However, care must be taken that the drillingmethod does not produce cracking in the concretearound the drilled hole which, although not readilyapparent, could reduce the grout/concrete bond strength.Diamond drills will also generally produce a suitablesurface for bonding. However, a polished surface,occasionally produced by diamond drills, may be det-rimental to bonding capability and additional surfaceroughening can be required. This roughening may bedone by mechanical scabbling of the wall of the hole.Some broken-in diamond drills produce a “rifling”effect that should produce adequate roughening. Alldust and debris that can reduce the bond between thegrout and concrete must be removed.


Formed voids should have some sort of roughening orcorrugations to assist the shear stress transfer betweenthe grout and concrete as illustrated in Figure 6.1.Where formers remain, experience and tests have shownthat corrugated metallic ducts or conduits should beused. PVC, rubber or polystyrene formers do notperform satisfactorily when left in place.

As noted previously, the size of spaces or clearancesbetween bars and holes is dependent on the viscosityand “wetting” ability of the grout. Low-viscosity,high-wetting grouts such as epoxies usually requireonly small clearances.

Placement of grout can be either before (pre-grouting)or after (post-grouting) the bar is located in the void.When the position of the bar is considered critical, thebar is usually located prior to grouting.

When pre-grouting, the hole is filled with sufficientgrout so that it fills the hole completely as it is displacedby the bar. When clearances between the bar and sidesof the hole are small, say less than 3 mm, it is recom-mended that a less viscous epoxy-type grout be used.Pre-grouting eliminates the critical workmanship phaseof post-grouting and is easier and less time consuming.

For post-grouting, larger hole diameters are requiredthan for pre-grouting. After the bar has been located inthe hole, two grouting methods are suggested.

“Tremie Tube”

The tremie tube method, shown in Figure 6.2, isrecommended. This technique is the most likely toproduce a fully-grouted connection.

“Decanting”

If the clearances between bar and hole sides aresufficient then the grout may be poured against anddown the bar to the bottom of the void. This methodis susceptible to air-locks which may lead to im-paired load transfer performance and corrosion.Air-locks result from either too high a rate of groutplacement, a grout that is too viscous, or a combi-nation of these effects.

Vertical Voids or Spaces

Figure 6.3 shows two typical situations of verticalvoids to be grouted. The steel section and the precastconcrete beam are grouted in the same manner. Thewidth of the space to be grouted is usually set byconstruction tolerances and typically ranges from20 mm to 50 mm. The sides and bottom of the spacecan be formed either by formwork (usually timber), orby sealing with a putty-like epoxy or mortar.

The grout used for such spaces is usually cement-based. The grout may be poured in from the top of thevoid provided the space is wide enough to prevent air-locks from developing. Experience and manufactur-er’s recommendations will guide that decision. If thespace is narrow and deep then it should be treated as ahole receiving a vertical starter bar (see Section 6.3.2.1)and precautions to prevent air-locks forming need to betaken. If the grout is expected to transfer loads itsminimum strength should be specified. Where loadsare to be transferred across the grouted connection,attention to details such as clean interfaces and appro-priate roughening are essential.

6.3.3 Horizontal Void Grouting

Horizontal Starter Bars

A typical situation is shown in Figure 6.4. This type ofdetail is used where the presence of protruding rein-forcing steel will hinder the installation or removal offormwork such as structural wall climbing forms. Bycasting in ducts at starter bar locations, the forms canbe left free of starter bars which are grouted-in later.

Voids can be drilled or formed as for vertical starter bars.The “horizontal” nature of the voids creates the potentialfor air-locks at the tops of the spaces. Inclining the ductor hole from the horizontal allows the grout to displacethe air as illustrated in Figure 6.5. However an allowancefor additional embedment length must be made to ac-count for the incomplete filling of the hole or duct(Figure 6.5(a)), unless it can be fully filled by sealing theend and grout pumped in so that it overflows out the top(Figure 6.5(b)). Any air gap (Figure 6.5(a)) should befilled with mortar or epoxy (putty consistency).

grout

grout

tremie tube

Tube may bewithdrawn asgrout level risesproviding the endof the tuberemains in the grout.Start at the bottom

Figure 6.2: Tremie tube method of grouting


Inclined holes, formed by ducts or drilling, tend toclash with other structural reinforcement. Care needsto be exercised with these details. Designers need toconsider the bursting force generated at the bend in thereinforcing bar under tension (Figure 6.5(c)).

Where horizontal ducts are used, pressure grouting,similar to post-tensioning grouting techniques used inprestressing, can be employed to minimise the residualair as shown in Figure 6.6.

In detailing grouted bars, attention needs to be given tolocating the bar in the centre of the hole. In particular,the bar must not end up in the top portion of the ductbecause of the potential for developing an air-pocket inthis location (Figure 6.7). Bar locators should be usedto place the bar in the centre of the duct and allow anunobstructed flow of grout throughout the duct.

When using horizontally grouted bars, designers shouldincrease the embedment/lap length of the bar to allowfor incomplete contact of the grout with the sides of the

hole/duct. A 20% increase in length is currently beingspecified by a number of designers. Notwithstandingthis, every care should be taken in construction to avoidair entrapment and, as far as possible, achieve com-plete grout-to-hole/duct contact.

As with vertical starter bars, the size of the ductdepends on restrictions imposed by the structure and itsreinforcement as well as the type of grout to be used.Larger voids allow a greater tolerance in placing bars,but more attention must be given to rigidly supportingthe bars during and after the grouting operation.

Horizontal Voids or Spaces

Unless special care is taken when grouting horizontalvoids there is a tendency to trap air. Vent holes throughthe base plates of steel columns or machinery assist inrelieving the trapped air. However, vents are usuallynot feasible where precast concrete units are locatedabove a void.

steel section(RHS,RSJ, etc.)

20-50mm

grout

bolt

temporary formsor mortar sealing

grout

precast concreteunit

20-50mm

shims and sealing

(a) (b)

Figure 6.3: Vertical voids requiring grouting

duct or hole

grout

barand duct

primary structure


HRC mesh or mat of reinforcing bars


Figure 6.4: Typical horizontal starters


transfer forces

(c)

burstingforce

applied force

e

duct or hole

bar

fill air gap withmortar or epoxy

duct or

bar

= development length of the

reinforcing bar or test values

(a)

additonal

length required

e

sealed duct or hole

groutand air

grout

(b)

Figure 6.5: Inclined starters

grout

groutand air

groutand air

grout

(b)(a)

duct or hole

sealant

bar andduct

duct or hole

Figure 6.6: Grouting of horizontal starters

To overcome this problem, experience of designersand grouting specialists has shown that the grout (epoxyor cement-based) should be injected under pressurefrom one point or side (Figure 6.8). In the case of longwalls (Figure 6.8(a)), timber bunds may be used toprovide a head for the grout. For grouting under platesor precast units, the sides of voids are usually sealed by

a cement or epoxy-based mortar with the consistencyof putty, which can be colour matched if necessary. Asan alternative, compressible proprietary sealing stripscan be used around the perimeter. Any sealing strip orcompound will occupy some of the area of the jointwhich will reduce the bearing area or cross-sectionalarea of the joint. The designer will need to check that


this reduction is not critical for the expected loadingsand that cover to reinforcing steel is still adequate(Section 6.3.1.1). Outlet ports or tubes (Figure 6.8(b))should be located so that air can escape from the void.When grout, free of air bubbles, is flowing from anoutlet, the port should be plugged.

When the underside surface of the top unit has beenroughened for shear transfer requirements, grout selec-tion is particularly important. There is concern amongsome designers that fine air bubbles might get trappedin the texture of the roughening, reducing the sheartransfer capacity. It is recommended that a low viscos-ity (very pourable) grout with good “wetting” charac-teristics be selected.

6.4 Grouting SpecificConnections

6.4.1 Precast Concrete Beam to ColumnJointFigure 6.9 illustrates a typical arrangement of a precastbeam placed on a cast-in-place or precast column.Column bars pass through corrugated metal ducts cast inthe precast beam. The choice of duct size is influencedby the likely size and location of the column bars passingup through the beam-column joint. The column bars, inan ideal situation, will be centrally located in the ducts.However, in reality, the column bars will vary from the

A

A

air lock

seal

grout

barduct

duct or hole

air lock

grout

Section A A

Figure 6.7: Bond loss in an air-lock

grout

Outlet

Outlet

Outlet

inlet

airair

airgrout

plug

groutthenplug

groutplug

0t

air

air

grout

groutthenplug

grout

plug

grout

air

groutthenplug

0t t 3

0t t20t t1

(b) Plan Views: Plate on Horizontal Surface(at different times)

precast concretewall unit

grout

timber

foundation

intermittent shims

grout

(a) Wall Section

Figure 6.8: Horizontal voids or spaces


ideal positions, but should be within certain contractualor specified tolerances. A duct size should be chosen toaccommodate these tolerances, plus a recommendedadditional 10 mm clearance to the bar to allow grout toflow between the duct wall and bar (Figure 6.10).Typically, duct diameters range from two to three timesthe nominal diameter of the grouted bar. (The CanadianPrestressed Concrete Institute [6.2] recommends threebar diameters for the internal duct size.)

Since column bars are ideally located at the centre ofducts, their cover will be larger than normal. This isbecause covers include the distance to stirrups or ductwall plus the clearance of the bar from the duct wall.

The two principal methods of grouting a precast con-crete beam to column joint are as follows.

• Method 1

After sealing the gaps as for Section 6.3.3.2, grout(typically non-shrink cement-based) is pumped in

at one inlet port (or tube) to displace air progres-sively across the interface (Figure 6.8). If the grouthas a relatively high viscosity, it may start to flowup the open ducts, starting at the duct closest to thegrout inlet (Figure 6.11).

It is recommended that outlet ports be provided onthe other three corners. These should be progres-sively plugged once grout without air bubbles flowsout. Once the outlets are stopped, pumping of groutwill continue to fill the ducts in the precast beamupwards from the bottom.

The duct nearest the inlet should fill first while theone furthest away (opposite corner) may requiretopping up by use of a tremie tube, or by pouring-in from a dispenser in such a way that grout runsdown in contact with the reinforcing bar to avoidair-locks. The inlet tube or port is plugged onceinjection is completed.

column bars

note: cover to columnmain reinforcing will belarger than normal

precast concrete beamlowered over column bars

ducts cast intobeam unit

15 - 30 mm gap

cast-in-place orprecast concrete column

Figure 6.9: Schematic beam-column assembly

Section

relocated bar with 10 mmminimum clearance

duct

bar in centreof duct

Figure 6.10: Column bar and duct clearance


• Method 2

The horizontal gaps are first sealed, then providingthe ducts are sufficiently large, grout may be pouredin from a dispenser down one corner duct. Progres-sively, the grout will flow across the column-beamsoffit interface and up the remaining ducts. It isrecommended that, as with Method 1, outlet portsbe used to confirm the progress of the interfacegrouting. Topping off for ducts remote from thefilling position may be necessary. Again, care mustbe exercised so that no air is trapped in the ducts.

In both Methods 1 and 2, outlet ports have beenomitted. This produces a degree of uncertainty as to theextent of grout along the interface between the top ofthe column and the underside of the beam. Almostcertainly there will be a small air-lock beside the ductat the furthest corner from the input duct or port. Thispractice is therefore not recommended.

Caution is required when sealing perimeter gaps. Seal-ing materials can reduce covers and may promotecorrosion of reinforcement as discussed in Section6.3.1.1.

6.4.2 Column SplicesFigure 6.12 shows typical grouted column splice con-figurations.

Some connections are made by bars protruding down-wards from the precast element above, with the barslocated in pre-grouted ducts or proprietary steel sleeves(Figure 6.12(b)(ii)). Care must be taken to ensure that:

• The sleeves or ducts, receiving bars from above, areclear of any debris or contaminants likely to reducethe bond or obstruct the placement of these bars.

• In the case of grout being placed in the receivingducts prior to the bars being inserted, the bars mustbe placed before the grout has started to set. If thisis not done, poor bond and partial bar penetrationwill result, leading to an inferior connection.

Other configurations consist of connecting bars whichprotrude upwards.

The precast units have ducts or proprietary sleeves to fitover the bars that extend upwards from the adjacentunits. If required, the units may be correctly levelled onshims then grouted with a horizontal interface and ver-tical duct grouting, operating similarly to Method 1(Section 6.4.1). Units may be seated on shims and a bedof cement or epoxy-based mortar consistent in strengthand colour with the precast units. Shims should notobstruct the flow of grout and, if metal, the shims shouldhave sufficient edge cover to prevent corrosion fromoccurring. If the mortar bed approach is used with thepost-grout method, care is needed to ensure that the ductsand inlet ports are not blocked by intruding mortar.

Two types of ducts are typically used in column splices:

• proprietary steel sleeves; and

• corrugated metal ducts (Figure 6.13).

For grouting either type of duct configuration it isnecessary to have an inlet tube at the base of each duct

column steel

groutair & grout

sealant

column

grout ‘top off’as necessary


(b)

(a)


column

grout grout

grout

sealant

corrugatedducts

grout levelrises

Figure 6.11: Grouting of a beam-column joint


(c) ‘T’ precast unit Beam + column cast together

(b) (ii) Cruciform Unit Mid height column joint Downward sleeving

(a) Full height column


(b) (i) Cruciform Unit Mid height column joint Upward sleeving

Figure 6.12: Typical column splice configurations

and an outlet tube at the top for bleeding air and grout(Figure 6.13). In practice, a significant amount of groutinjection can be done from one interface inlet port ortube, and topping up as necessary for the individualducts or sleeves through their lower inlet tubes. Thegrout may be injected under pressure that is main-tained, as in grouting of post-tensioned cables in pre-stressed concrete construction.

A steel sleeve splice type system may require tightercontrol over tolerances both in precasting and siteerection and a higher degree of supervision by thedesigner and construction staff may also be required.

Any proprietary splice system must be laboratorytested to show compliance with the criteria in the NewZealand Concrete Structures Standard [6.3]. For an


example of a test report, readers are referred to Refer-ence 6.4.

6.4.3 Beam SplicesFigure 6.14 shows the junction of two precast concretebeams linked by steel sleeve splices (other mechanicaltype couplers can be similarly installed). The installa-tion procedure follows this general approach:

• Align the beam units so that the sleeve connectors(which are slid completely along the bars of oneunit prior to alignment) can be slid over the bars ofthe second unit.

• Having slid the sleeves on to the bars of the secondunit, ensure that the correct amount of each bar isoccupying the internal space of the sleeve. Markson the bars should be provided to assist with this.

bar (upper)

duct

bar (lower)

25 mm minimum andcode spacing requirements

Section AA

bar

precast concrete member

high strength grout

splice sleeve

bedding mortar

bar (upper)(lapped beside lower bar)

grout and air

precast concrete member

duct

grout

shims

bar (lower)

beddingmortar

A A

Figure 6.13: Steel sleeve splice and duct splice detail

precast concrete units

steel sleeve splices

Figure 6.14: Proprietary reinforcing bar splices between precast concrete beams


• Seal sleeve ends and fill with the specified grout inaccordance with the manufacturers’/specifiers’ in-structions. Ensure the sleeves are not disturbeduntil the grout has adequate strength or else theconnection will be weakened.

• Place any other reinforcement and complete con-struction details to be cast-in in this zone.

• Place the required form-work for the final opera-tion.

• Complete the connection of the two precast units byfilling the beam form-work with cast-in-place con-crete (the specified 28-day compressive strength ofwhich should be at least equal to that of the precastelements).

Each proprietary coupler system will have its ownparticular method for installation, which should bestrictly followed.

6.5 Quality Assurance

6.5.1 GeneralThe designer should indicate the quality assurancerequirements for any grouting operation. These re-quirements should be part of the contract documents,available at time of tender. If all parties are aware ofquality assurance requirements at tender stage thereshould be fewer difficulties in implementing themduring the contract.

Quality assurance is essential for the success of anygrouting operation. Bar size and length, hole size,depth and cleanliness, grout type and quantity, pressu-rising procedures for the grout (if any), protectionagainst disturbing the immature joint, and avoidance ofairlocks all need careful attention and confirmation ofthe correct procedure.

In any detail where there are stages of partial comple-tion, checks are appropriate. For example, whenducts are cast into units, checks should be made ontype of ducts, location, length, etc. Similarly, for barsto be grouted-in, correct bar strength, size, embed-ment length and location within the duct should bechecked.

At pre-definable stages the contractor should sign offeach “milestone”. This approach is currently beingused in the precast industry quite successfully.

The designer, particularly during the initial stages,should co-operate with the contractor to ensure that thedetails are being correctly interpreted and built, and tomonitor the quality assurance programme.

6.5.2 Assurance of Complete GroutingFrom experience, it appears that the only practicalmethod of establishing whether units are fully groutedis by grout volume measurement. The volume of thevoids is calculated and this should equal the volume ofgrout that was inputted, minus the volume of groutcollected from outlet or bleed ports. The degree ofaccuracy is subject to judgement and should be estab-lished prior to commencement of grouting.

6.6 Grouting Workmanship andConstruction Aspects

6.6.1 WorkmanshipThe construction of grouted details and the groutingoperation are specialised practices requiring high stand-ards of workmanship. For example, details such aslocation of ducts to match others in adjacent precastunits and the location and bar length requirements forgrouted steel sleeve splices [6.5] require tighter toler-ances than those for conventional reinforced concreteconstruction. Grouting should always be done by expe-rienced applicators, either sub-contractors or trainedpersonnel from the contractor’s staff. It must not beconsidered as an operation to be done at some later,more convenient stage, by any available staff (typi-cally lacking the necessary skills).

At tender stage, the contractor should be made aware ofall grouting operations and any complexities so as toallow for the high standard of workmanship required.

6.6.2 Construction AspectsExperience has shown that attention should be paid tothe following matters during a grouting operation:

• Use of compressed air or a vacuum to remove dustfrom voids or spaces may not adequately removedust adhering to the sides of the void, particularlyif the concrete is of an early age or in a dampenvironment. Brushing with a stiff brush willnormally be necessary in addition to the use of airor water for flushing out the void.

• If air is used to blow out dust and debris from ductsor voids, it should be ensured that either an oil filterhas been installed, or no oil-bottle is or has beenused on that line as part of any air-driven equip-ment. Any oil will damage the bond between thesurfaces and grout.

• A major problem with grouting voids or gaps is thefailure of sealing compounds or forms around theperimeter of the voids. Leakage through the seal-


ants not only complicates the measurement of groutvolumes, but also causes unwanted delays. Surfacefinishes may also be damaged.

• Any excess grout which could reduce the bond tothat surface should be cleaned off.

• Under no circumstances should bars be moved orknocked before the grout has gained sufficientstrength. Such movement has a detrimental effecton the bond strength between the bars and grout.

• After cement-based grouting has been completed,grouted ducts should be protected from prematuremoisture loss.

6.6.3 Correct Usage of GroutsThe grout specifier and installer must ensure that themanufacturer’s instructions for the product are fol-lowed exactly. If more detailed information is re-quired, then the supplier/manufacturer should be ap-proached for technical advice.

Concern has been expressed regarding the relianceplaced on manufacturer’s specifications. Dependingon the integrity required for the grouted detail, thedesigner should be convinced that the manufacturer’sdetails are appropriate. Further tests could be requiredfrom the manufacturer, or as part of the contract inde-pendent laboratory testing, and trial grouting of pro-posed details could be specified. Due allowances needto be included at time of tender.

6.6.4 On-site TestingOn-site pull-out tests may also be specified as a meansof verifying correct workmanship and function of theconnections. Where load tests are to be conducted ongrouted-in bars, the test load may be limited to 85% ofthe tensile yield strength of the bar so as not to damageit for usage. This level of load is normally sufficient toidentify improperly installed dowels. Usually anyslippage or pull-out of the bar at this level indicatesincorrect installation. The designer must, however,realise that there is still a degree of uncertainty that thegrouted detail will reach its full design strength.

In order to verify the strength capacity of grouted bars,it may be necessary to pull embedded items to failure.“Failure” is either tensile rupture of the bar, concretecone pull out, or bar slippage beyond a pre-specifiedlimit. Such tests should be done on items that will notbe part of the final structure.

In either the non-destructive method or the full failuremethod, the testing should not interfere with the gen-eral construction programme of the site.

6.7 Connections betweenPrecast Elements using WeldedReinforcing Bars

6.7.1 GeneralThe welding of reinforcement bars is a practical methodof developing force transfer in many connections,provided that the weldability of the bars and adherenceto sound welding practice can be assured. The threesituations of welding reinforcing bars are as follows:

• To each other;

• Through splice members (such as angles, reinforc-ing bars, or flat plates);

• To structural steel members anchoring the bars.

Some of the common welds used with reinforcing barsare shown in Figure 6.16 and two typical examples ofsplicing reinforcing bars are shown in Figure 6.17.Figure 6.18 shows two welded steel details used to joinprecast concrete beams.

These guidelines have considered the welding of twonew “seismic” grades of reinforcement, designated as“E” grade in the joint Australian/New Zealand Stand-ard (currently being produced) for the production ofreinforcing bars. Welding of bars that are not classed as“E” grade should be undertaken in accordance withrequirements of the joint Australian/New ZealandWelding Standard AS/NZS 1554.3.

6.7.2 Welded Reinforcement SplicesA new joint Australian/New Zealand Standard for theproduction of reinforcing bars, which is to supersedeNZS 3402 [6.7], lists two grades of reinforcing bar thatmeet seismic requirements, and hold the “E” (forEarthquake) designation mark. These two seismicgrades 300E and 500E are high ductility grades in the300 MPa and 500 MPa lower characteristic yieldstrength category.

The New Zealand Concrete Structures Standard [6.3]defines the requirements of welded splices and me-chanical connections when used in given structurallocations. The designer must ensure that weld detailsbeing proposed comply with this code and to anyadditional requirements of the joint Australian/NewZealand Welding Standard AS/NZS 1554.3 [6.9], whichis replacing NZS 4702[6.6]. At the time of writing thisguideline, AS/NZS 1554.3 was still in the draft stageand the current NZ Standard was NZS 4702. Howeverit is noted that NZS 4702 does not cover the new NewZealand produced micro alloyed seismic grades 300Eand 500E.


Parameters controlling ductility such as minimum andmaximum yield strength, tensile to yield strength ratioand uniform elongation are set tighter than for the lowductility (L) and normal (N) ductility reinforcementbars. When joining these bars by welding, theseproperties need to be maintained ensuring that the weldhas sufficient tensile strength for it not to be theweakest link. However the weld should also contributeto the deformation requirements and therefore the weldshould provide yielding close to the yield of the parentbar combined with acceptable elongation.

The current Grade 300E and Grade 500E manufac-tured in New Zealand are of the micro alloyed typeproduced by hot rolling. The “micro alloyed” methoduses small, closely controlled additions of alloyingelements such as vanadium, which increase strength,overall ductility and assist in maintaining goodweldability. All grades manufactured to the antici-pated joint Australian/New Zealand standard for theproduction of reinforcing will carry specific mark-upsto identify the bars and to recognise the specific weld-ing requirements. Figure 6.15 shows the marking to thenew reinforcement production standard for the 300Eand 500E reinforcement bars.

An extensive study by NZ HERA [6.11, 6.12] on theweldability of the micro alloyed Grades 300E and500E produced by Pacific Steel proved their excellentweldability and mechanical properties, provided thecorrect welding procedures are followed. The studyalso showed that for seismic applications the appropri-ate choice of joint type is of importance.

6.7.3 Selecting Joint Types for SeismicApplicationsUsing “seismic” grade reinforcing bars, the correctwelding consumable, and following qualified weldingprocedures guarantees that well executed welded jointswill satisfy the mechanical strength and ductility re-quirements, for seismic loads in the plastic rangewithin the design expectation. This is provided thejoints are symmetric to the central axis of the applied

loads. Therefore butt joint configurations or bar toplate welds as shown in Figure 6.16 will perform wellif loaded past the yield point in the inelastic range.

However, non-symmetric joints such as the standardlap joints will develop a moment of rotation, which canlead to brittle fracture at the weld ends if stressed pastyield. This effect particularly shows at the higherstrength 500E bar and in larger diameter. The 300EGrade performed with satisfactory ductility in the testsup to and including the 32 mm bar diameter. If splicejoints are jointed by indirect means like using anglebacking or flat backing (see Figure 6.16), the dimen-sions used for the angle or the backing plate willdetermine the degree of bar rotation and if prematurefracture will occur. Therefore careful considerationfor plate and angle dimensions needs to be given.Alternatively, the indirect butt splice joint alternative(two bar pieces in a symmetrical lap splice) as shownin Figure 6.16 can be chosen. Due to its symmetricnature ductility performance is excellent.

Extending the overlapping length of the lap joint is alsoan option and provided the overlapped length is suffi-ciently long the resulting moment (resulting from theeccentricity of the bars) is sufficiently low to notsustain any brittle fracture even in the heavier thick-ness higher strength bars.

Similar to the non-symmetric lap joints, the precastconcrete detail with the bent and welded reinforcingbar ends (Figure 6.18) should be avoided for seismicapplications especially in the higher strength Grade500E and larger diameter bar alternatives. Symmetricalternatives such as shown in Figure 6.18(a) are pre-ferred to avoid brittle fracture if stressed past yield.

When specifying a joint detail, the designer must beaware that for some joints it is difficult for the welderto readily achieve the required weld quality. Joints thatprovide for good access of the welding arc to the rootof the weld are preferred. For example, the bar to platejoint configuration shown in Figure 6.16, the singleside fillet weld (detail on the left) or the double side

Grade 300 E Grade 500 E

Figure 6.15 : Identification Marking of Grade 300E and 500E


Bar-to-plate joint configurations

Single-sided external fillet weld Double-sided external fillet welds Single V butt weld

fillet weld

plate

filletwelds

preferred detail

plugweld

Critical detailbecause ofroot qualityproblems

Indirect butt splice joint configuration

Butt joint configurations

G

θ

θ

G

Double lap splice

Bar-to-bar joint configuration

90˚ bar to bar

lap length

effective length

W45˚

G 45˚ W

S

Indirect butt splice with flat backing Indirect butt splice with angle backing

S

G

G

Indirect butt splice using two reinforcing bar straps (symmetrical)

S

W

Figure 6.16 : Typical reinforcing bar welds


fillet weld (detail in the centre), is much preferred to thesingle V butt weld (detail on the right). In the newlyrevised AS/NZS 1554.3 [6.9], this “prequalified” jointalternative considers the difficulty of weld access byrequiring an additional 2 mm of weld throat, plusconsideration of the maximum possible gap betweenthe bar and the corresponding plate hole. In the singleV-butt weld detail considerable care is needed toobtain a satisfactory weld in the root area of the jointdue to the poor access for the welding arc. Verification

of the cross sectional area of the weld achieved can alsobe costly because weld inspection is difficult. For thefillet weld alternatives, not only is access for thewelder easily achieved, the weld quality can be readilyconfirmed by visual inspection.

The 90˚ bar-to-bar weld detail as shown on Figure 6.16is not a prequalified joint and, like the double lap splice,it should not be chosen if full plastic performance isrequired. Welding procedures for the 90˚ bar to bardetail need to be fully qualified by the contractor.

fill pocketwith grout

steel angle

Fillet welds

Section

(a) Panel: Vertical bar spliceusing welded angle

Section

Fillet welds

(b) Beam: Splice using weldedangle

note: special stirrups and ties are necessary

Figure 6.17 : Examples of splicing reinforcing bars

Note: Special stirrup detailsto control bursting at thebar ends are required

bolted splice plate

rebar welded toconnector plates

Note: NZS 3101 restrictswelding adjacentto bends in rebar



plate thickness ≥ bar diameter

slotted plate

bolted splice plate

rebar welded toconnector plates

(a) preferred(b) traditional

Figure 6.18: Bolted splice connection between precast concrete beams


6.7.4 Weld Specifications

As previously stated, welding of reinforcing bars inNew Zealand is covered in AS/NZS 1554.3 [6.9]. Theprinciple of this standard is that satisfactory weldquality is achieved by the use of qualified materials(bars and consumables), using qualified welders fol-lowing qualified procedures under adequate supervi-sion. Welding inspection as defined in the standardsupports this.

While the standard should be consulted on all aspectsof welding, some general comments of interest to thespecifier follow:

Preheating

Generally no preheat requirement applies for theGrade 300E and 500E manufactured in New Zealandby Pacific Steel. Exceptions may be called for in AS/NZS 1554.3 [6.9] for specific joints of heavier thick-ness bars (≥32 mm) where the joints configurationcombined with low energy input may lead to fastcooling and preheat may be necessary. However, it isgood practice, especially in cold and damp condi-tions, to consider slight preheat (30˚C) to dry off anymoisture around the weld site, reducing the risk forhydrogen induced cracking, lack of fusion and weldporosity.

Welding Process

The choice of the welding process should be left to thefabricator. Manual Metal Arc Welding (MMAW), GasMetal Arc Welding (GMAW), Flux Cord Arc Welding(FCAW) gas shielded and FCAW self shielded aresuitable. Well suited to on site welding are the MMAWand the self shielded FCAW process. In the workshopthe GMAW and the gas shielded FCAW process alter-natives are best suited and more economic. While forthe MMAW and FCAW of the 300 Grade non-low-hydrogen consumables are acceptable, for the 500Egrade only low hydrogen consumables are recom-mended.

Welding Procedure Specification

AS/NZS 1554.3 [6.9] requires the contractor to estab-lish welding procedures. These should be made avail-able to the welder and on request to the principle’srepresentative who is usually the inspector or theengineer. The welding procedure details all essentialvariables including joint preparation and welding set-tings which determine repeatable weld quality. As thewelding procedure qualification is based on testing, ithas been verified that these parameters produce asatisfactory joint by an adequately experienced andqualified welder.

Qualification of Welding Personnel

AS/NZS 1554.3 [6.9] requires that a suitably qualifiedwelder under the supervision of a suitably qualifiedwelding supervisor shall carry out all welding. It isimportant to note that the welder needs to be qualifiedin the both the welding process and for the weldingposition used. Welder qualification can be specific tothe on the particular procedure applied, or in the formof wider qualification such as NZS 4711 [6.10].

Inspection

AS/NZS 1554.3 requires that all welding shall besubject to the examination of an inspector appointed bythe Principal. This inspector shall be suitably qualifiedand shall inspect the work prior and after welding to therequirements of the standard.

Welding near bends in reinforcing bars

An example of where this welding detail occurs can beseen in Figure 6.18(b). NZS 3101 [6.3] states thatwelding can not be undertaken any closer to a bend(including that of a restraightened bar) than three timesthe diameter of the bar being welded.

6.7.5 TolerancesParticular care is necessary when fabricating precastconcrete units joined by welding. Bars to be joined bybutt welds require accurate alignment. Welded spliceswith steel angles in most cases require less accuracy.The designer must be aware of the achievable accuracyof alignment for welded details and check that this doesnot exceed the maximum permitted by NS/NZS 1554.3.

6.7.6 Welding Information for Grade300E and Grade 500E ReinforcementBar Manufactured by Pacific SteelShould any queries outside AS/NZS 1554.3 [6.9] arise,it is recommended that they be directed to HERA’sNew Zealand Welding Centre. A large number ofwelding procedures covering MMAW and GMAW inthe most common joint alternatives and welding posi-tions have been developed and qualified, and copies ofthe procedures can be made available to product users.

6.8 Mechanical Connectors forSplicing Reinforcing BarsThe performance of mechanical connectors must com-ply with the New Zealand Concrete Structures Stand-ard [6.3]. Various forms of these connectors are avail-able in New Zealand. The principal method used isbutt splicing.


Types of mechanical connectors are as follows.

• Threaded bars

— Ends of bars are threaded to accept a threadedsleeve coupler.

— Threaded sleeves that are swaged onto the endsof bars and a threaded coupler used to jointhem.

— Some bar types have the deformations for bondroled as threads, where proprietary sleeveswith matching internal thread to those defor-mations are used to make the splice.

• Cadweld

A proprietary system consisting of a sleeve intowhich a molten metallic filler is poured. Once thefiller has solidified, the splice is complete.

• Swaged sleeves

The splice is achieved by the swaging of a propri-etary sleeve on to the ends of butted bars.

In all cases these splicing systems, as with butt welds,require a high degree of accuracy for alignment of bars.These systems are perhaps most effectively used for aconnection where there is at least one “free” bar whichcan be adjusted on site.

Most proprietary systems offer a range of strengths fortheir couplers. Some couplers are designed to reach100% or 125% of the yield strength of the bars. Thedesigner must ensure that the system nominated meetsthe design and code requirements.

As with all proprietary systems, the manufacturer’sinstructions for usage should be fully adhered to.

6.9 Recommendations

6.9.1 Connections made by Grouting ofPrecast Concrete ComponentsIt is recommended that:

• When cement-based grouts are used they should behigh strength and shrinkage compensating. Theminimum grout compressive strength should be 10MPa greater than that of the surrounding concrete.(Section 6.2.2.2.)

• For all grouting situations, extreme thoroughnesswith respect to cleanliness and the following ofmanufacturer’s instructions is required. (Section6.3.1.)

• When bars are grouted in horizontal or inclined

holes, bar locaters should be used to keep the barsin the centre of the holes. (Section 6.3.3.1.)

• When precast concrete beam-to-column joints arebeing grouted, one of the two methods in Section6.4.1 should be used, together with a low viscositygrout with good “wetting” characteristics.

• Designers should communicate to contractors attender stage all the specialised requirements for thegrouting operations, including the need for experi-enced operators and a satisfactory quality assur-ance programme. (Sections 6.5 and 6.6.)

• The grout volume method should be used to deter-mine whether or not the units have been fullygrouted. (Section 6.5.2.)

• Before compressed air is used to blow out dust fromholes, it should be tested for oil contamination.(Section 6.6.2.)

6.9.2 Connections made by MechanicalSplicesIt is recommended that when using mechanical splicesdesigners should ensure the strength of the splice meetsdesign and code requirements, and ensure that themanufacturer’s instructions for using the splices arefollowed in full.

6.10 References6.1 Thurston, S J. Precast Grouted Beam-Column

Joint, Report 5-86/9, Ministry of Works andDevelopment, Central Laboratories,Gracefield, September 1986.

6.2 Metric Design Manual: Precast and PrestressedConcrete, 2nd Ed., Canadian PrestressedConcrete Institute, 1987, Canada.

6.3 Concrete Structures Standard Part 1 and 2:NZS 3101:1995, Standards New Zealand,Wellington, 1995.

6.4 Yong, P M F. The Performance of NMB Splicesfor Grade 380 Bars under Cyclic Loading,Report 5-85/9, Ministry of Works andDevelopment, Central Laboratories,Gracefield, 1985, 60 pp.

6.5 NMB Splice Sleeve System User’s Manual,NISSCO Splice Sleeve Japan, Ltd.

6.6 Metal Arc Welding of Grade 275 ReinforcingBar, NZS 4702: 1982, Standards Associationof New Zealand, Wellington, 1982, 23 pp.


6.7 Steel Bars for the Reinforcement of Con-crete, NZS 3402:1989, Standards Asso-ciation of New Zealand, Wellington,1989, 16 pp.

6.8 General Information Sheet: G001: ReinforcingSteel Bar Welding Procedure, Pacific SteelLtd, Otahuhu, New Zealand, 4 pp.

6.9 Structural Steel Welding - Welding of Reinforc-ing Steel, AS/NZS 1554.3 (under revision),Joint Standard, Standards Australia andStandards New Zealand.

6.10 Qualification Tests for Metal-arc Welders,NZS4711:1984, Standards Association ofNew Zealand, Wellington, 1984, 31 pp.

6.11 Scholz, W and Roberts, B. “Welding newlydeveloped, high strength, seismic gradereinforcing bars”, 12th World Conference onEarthquake Engineering, Auckland, NewZealand, February 2000.

6.12 Scholz, W et al. Welding Pacific Steel Rein-forcement Bars Phase 1 + 2, New ZealandWelding Centre Tech. Rep. TR9807, HERA,1998, 1999.

Embedded Steel Connectors • 91

Chapter 7Embedded Steel Connectors

7.1 IntroductionEmbedded steel connectors are used extensively by theprecast concrete industry. Although not used for con-nections between primary precast concrete seismicresisting members, embedded steel connectors may besubject to significant gravity and earthquake inducedforces resulting from:

• inertia effects, i.e. parts and portions provision ofthe New Zealand Loadings Code NZS 4203:1992[7.1]; and

• imposed deformations, as secondary structural(Group 2) elements, under seismic loading, asrequired by the New Zealand Concrete StructuresStandard, NZS 3101:1995 C1. 4.3.5, 4.4.13 and4.4.14 [7.2].

Steel embedments may also be required to accommo-date considerable localised movements resulting fromconcrete volume changes, member elongation underseismic loading, or temperature-induced structural de-formations in fires.

In many cases the applied force or deformation calcu-lated is at best an estimate, and the design of steelembedments should always be carried out conserva-tively, bearing in mind the consequences of failure.The deformation capacity of embedded connectors istherefore important, as the dependable strength may bereached and deformations may well exceed those cal-culated.

This chapter gives information on, and recommenda-tions for, the design and detailing of steel connectorsembedded in concrete, and interprets relevant provi-sions of the New Zealand Concrete Structures Standard.

7.2 Types of ConnectorsThe types of embedded connectors considered areshown in Figure 7.1, (a) through (e). These include:

(a) and (b) weld plates;

(c) cast-in steel sections;

(d) post-drilled fixings, i.e. resin or expansion type; and

(e) cast-in concrete inserts.

A comparison of the advantages and disadvantages ofeach type of fixing is set out in Table 7.1.

7.2.1 Weld PlatesThis type of connection (see Figure 7.1(a) and (b)) iscast into concrete members to allow for later attach-ment of similarly-equipped concrete members, or asteel member, by welding to the exposed steel surface.

7.2.2 Cast-in Steel SectionsChannel, angle, flat, hollow or universal sections maybe cast into precast concrete to provide a joint toanother member (Figure 7.1(c)).

7.2.3 Post-fixing to Precast ConcreteConnections can be formed by drilling into hardenedconcrete and inserting expanding anchors or resin-typeanchor studs (see Figure 7.1(d)). Expanding anchorsrely on friction or mechanical interlock (particularlywhere using undercut installation) to transfer axialtension and may need to incorporate low strengthreduction factors (high factors of safety) to coverperformance sensitivity to installation uncertainties.Note that in situations where the vertical gap betweena precast concrete unit and connecting member is to befilled with cast-in-place concrete or grout, a preferabledetail is to make use of an inverted steel angle. Thisavoids prying of the embedded fixing.

7.2.4 Cast-in Steel InsertsPrecast members may utilise cast-in steel inserts, whichsubsequently receive a bolt or threaded bar for con-necting other structural elements (see Figure 7.1(e)).

7.3 Selection of FixingsBefore detailed design commences, a careful choice offixing type should be made. There are a large number offactors to be considered. The checklist provided in Ap-pendix D1 may be used when making a selection for agiven situation. Other sources of information on mechani-cal fasteners in concrete are given in References 7.3 and7.4.


Specific questions requiring consideration are :

• What loads are to be carried by the fixing assembly?

• What deformation is the fixing assembly requiredto withstand?

• What load/deformation characteristic is requiredunder overload conditions?

• What kind of fixings are available for this applica-tion?

• What are their performance characteristics?

• Are these all the factors that influence the perform-ance of the fixing assembly?

• What about corrosion?

• What has to be considered in the structural designof the fixing assembly?

• How sensitive is the fixing assembly to a shift in theposition of applied load, or dimensional changes,etc?

• Are there any potential problems with the methodof assembly, or installation?

• How does the designer convey performance re-quirements to the user?

• Is any load verification testing required?

• Are there any other matters to be considered?

7.4 Structural Actions on SteelEmbedments

7.4.1 LoadsThe most common load types that may act on steel

insert with head

bar coupler type

rebar

actions onfixing

(a) Weldplate anchored by headed studs

(c) Cast-in steel section

(e) Concrete inserts

floor unit

(d) Seating angleFixed by epoxied stud,expanding anchor or groutedholding-down bolts(angle may be inverted)

(b) Weldplate anchored by reinforcing bars

Figure 7.1: Types of steel embedments


embedments are identified, together with appropriatecombinations in the loadings code [7.1].

7.4.2 Actions Resulting from Movementsof Structural ElementsAs discussed in Chapters 2 and 5, buildings incorporat-ing precast concrete elements are subject to concretevolume changes due to temperature effects, shrinkageand creep. In buildings where lateral seismic loads areresisted by ductile seismic frames, movements of fargreater magnitude may occur due to the elongation ofbeam plastic hinges. This phenomenon is described inAppendix A.1.

In most cases steel embedments do not have sufficientstrength to prevent such movements from occurring, sothe design and detailing must provide deformationcapacity through either flexibility or ductility. It is

therefore important that joint axial deformations androtations be correctly assessed, with some reserve toallow for inaccuracies in the global analysis. Thechange in position of reaction points as deformationoccurs needs to be assessed. Friction loads should beconsidered when an element moves in response tovolume changes, as illustrated in Figure 7.2.

7.5 Design ApproachThe strength of a steel embedment is governed byeither steel strength, the concrete resisting mechanismin epoxy type post-tensioned anchors, or also by bondfailure. The concrete resisting mechanism is itselfdependent on a number of factors. These includeconcrete strength, embedment length, edge distances,anchor spacings and other less frequently consideredfactors such as the presence of cracks, potential loss of

Anchor Type Advantages Disadvantages OverloadCharacteristics

Weld Plates

Figure 7.1(a)7.1(b)

Reliable installation

Reliable strength prediction

No load test required

Positioning accuracy (smalltolerances)

Reliable when anchoragedesign based on ductileyield in steel

Cast-in steelsections

Figure 7.1(c)



Difficulty of concretecompaction around steel


Reliable when anchoragedesign based on ductileyield in steel

Post-fixedanchor

Epoxy type

Figure 7.1(d)

Undercutanchors

Positioning accuracy easilyachieved


Load performancesensitive to installationmethod

Creep effects imposelimitation on magnitude oflong term loadings

Verification load testsrequired

See cast-in inserts

Reliable when anchoragedesign based on ductileyield in steel and verifiedby load tests

Post-fixedanchor

Expanding type

Figure 7.1(d)

Positioning accuracy easilyachieved

Proprietary components

Ease of removal

Load performancesensitive to installationmethod

Proprietary types havelimited embedmentdepth, i.e. difficult toanchor into confinedconcrete

Overload strength limitedby concretetension/anchor slip

Ductility achievable byusing specialistcomponents, e.g. barsplicers

Cast-in inserts

Figure 7.1(e)



Proprietary components


Proprietary types havelimited embedmentdepth, i.e. difficult toanchor into confinedconcrete

Overload strength limitedby concretetension/anchor slip

Ductility achievable byusing specialistcomponents, e.g. barsplicers

Table 7.1: Comparative advantages of types of embedded connectors


concrete cover and the type of loading (monotonic orcyclic) and the method of installation.

In an overload situation the steel component of theembedment is generally ductile whereas the concreteresisting mechanism is brittle. Brittle failures in thesteel component may still occur in bolts or in shortconnections, where the ductility of the connector overa small length results in a very limited deformationcapacity. Also, brittle failures may arise in boltedconnectors subjected to cyclic loads in shear or com-bined tension and shear. Capacity design, where thestrength of the steel embedment is required to be lessthan the dependable strength of the concrete resistingmechanism, is the recommended design approach.This approach should be used not only for seismicloads, but also for any other type or combination ofloads, so that in all situations brittle failure of theconcrete resisting mechanism is prevented. Capacitydesign is particularly relevant when the effects ofconcrete movements are considered since, as previ-ously mentioned, the forces that are generated mayreach the strength of the connection.

7.6 Design Methods

7.6.1 GeneralMany methods have been proposed for design withembedded steel connectors. Designers should be cau-tious when choosing a design method, since the limita-tions of the method may not always be stated, or theequations may have been calibrated to obtain meanvalues rather than the 5% lower characteristic strengths.Some methods underestimate the detrimental effectson the concrete resisting mechanism caused by edgesin the anchoring element or by overlapping of thepotential surface failures. A recent state-of-the-artreport [7.4] highlights the limitations of many designmethods available.

The design philosophy in the Concrete StructuresStandard [7.2], Cl. 4.3.5(a) and (c) is rational. The

principle objective of Cl. 4.3.5(c) is to demonstrate thatyield or slip occurs in connectors in order to precludea brittle concrete pull-out failure. This implies that,while the connector is sized for the design loads, theembedment length in the concrete is proportioned toensure the development of yielding in tension in theconnector itself, particularly in statically indetermi-nate structures or components.

Note that not all proprietary embedments, which quoteultimate limit state loadings, satisfy the above criteriaand these cannot therefore be “prequalified” for use inseismic resistant applications.

To calculate the embedment length, it is important toincorporate the statistical variation inherent in thefactors contributing to the capacity, or deformationdemand, in the connector.

The main parameters affecting these are:

• the actual tensile or yielding strength of the connec-tor surrounding the embedment;

• accuracy in establishing the embedment length ofthe connector;

• scatter between observed and predicted capacity —design equations are based on “best-fitting” ofexperimental results over a range of values;

• long-term performance of the embedment, givenits reliance on friction forces (which may decreaseas a result of creep or cracking in the surroundingconcrete) or sensitivity of its capacity to installa-tion procedures;

• likely overstrength that can develop in steel con-nectors, where extensive yield strains are likely,and the effects of high material strengths and strainhardening contribute to the overstrength designactions applied; and

• consequences of failure of the embedment.

The consequences of failure should be accounted for inthe design of an embedment. In lieu of a particularstudy, the component safety index should be β = 3.5, avalue considered appropriate for a near brittle pull-outfailure. This concept is illustrated in Figure 7.3. InReference 7.5, the second moment probabilistic meth-ods are used to derive such factors. The variation inmaterial strengths, embedment length and curing of theconcrete were considered in the study. With referenceto Figure 7.3, it was proposed that λ = 1.21 and φc = 0.51be used to modify the 5% lower characteristic strengthof the connector and the mean value predicted for theconcrete resisting mechanism respectively (λ = mate-rial factor for steel and φc = strength reduction factor

movement fromvolume change

movement

shift in reactionposition

friction force

Figure 7.2: Friction load acting on a steel

embedment


associated with the concrete cone pull-out failure). Thedesign procedure and factors proposed in Section 7.6.2are consistent with the values proposed in Reference 7.5.

7.6.2 Weld Plates with Headed StudsIt is recommended that the design of a connection usingweld plates with headed studs be carried out in twostages. In the first stage the dimensions of the studs aredetermined using a strength reduction factor for ten-sion, shear or combined tension and shear equal to φ =0.85 when capacity design is not used, and φ = 1 whenthe actions have been derived using capacity design. Inthe second stage the embedment length of the studs orgroup of studs is then determined using capacity designto preclude a brittle failure of the concrete resistingmechanism. In this second stage, it is explicitly as-sumed that a connector with a well-defined yieldplateau will develop the yield strength in tension, andthat connectors without a yield plateau will develop theultimate tensile strength. The dependable strength ofthe concrete resisting mechanism should be based onequations derived to predict the 5% lower characteris-tic strength multiplied by a strength reduction factorequal to φ = 0.85 when capacity design is not used andwith φ = 1 when the actions have been derived usingcapacity design.

The recommended design procedure generally followsthat outlined in Reference 7.4 with the following pro-posed modifications:

• the procedure for designing steel embedments inlight-weight concrete should be modified to ac-count for the brittleness behaviour of this concretetype;

• effects of spalling of cover concrete should beallowed for, in accordance with Section 7.6.7; and

• the cover for ductility and fire resistance shall be inaccordance with Sections 5 and 6 of Reference 7.2,respectively.

The recommended design steps, incorporating the abovemodifications, are as follows:

• establish weld plate and stud geometry, and thematerial properties;

• proportion the studs to carry the design actions;

• calculate the embedment length of the studs:

— where shear load is applied towards a free edge,check the shear edge distance or provide hairpinreinforcement to carry the entire shear force;

— check minimum tensile edge distance againstactual edge distance available;

• check service load stresses in studs and in theconcrete and, where cyclic loading is applied, afatigue check is also required; and

• check effects of other factors, e.g. spalling of coverconcrete on load resistance.

For cyclic loading, laboratory cyclic shear load tests[7.4, 7.6] indicate that, provided the stud embedment isadequate, the steel will fail due to the effects of lowcycle fatigue.

Under cyclic loads such as might occur on an embed-ment supporting lift machinery, the design stress shouldbe reduced as recommended in Reference 7.7 to allowfor fatigue.

Two worked examples of embedment design are pre-sented in Appendix E1.

7.6.3 Weld Plates Anchored by HookedReinforcing BarsWeld plates anchored by hooked reinforcing barspresent similar design considerations to weld plateswith headed studs and design can follow the procedure

Load or strength

Pro

babi

lity

dens

ityAsfy

λ = 1.21 and φc = 0.51

Connectoryield strength Probability of failure

Concrete pulloutstrength

Tc

λAsfy =φcTc

Figure 7.3: Derivation of the load factors for embedments


described in Section 7.6.2. The embedment length ofhooked bars is slightly larger than for studs [7.5]. Inelements where a reinforcing steel cage is present anda concrete pull-out failure is prevented from occurringdue to the presence of well-anchored transverse rein-forcement, the embedment length of hooked reinforc-ing bars can be obtained from the requirements of theNew Zealand Concrete Structures Standard [7.2].

7.6.4 Concrete Inserts and HookedStarter Reinforcing BarsHeaded steel inserts and hooked starter reinforcingbars are similar to headed studs and design can followSection 7.6.2.

The load resistance of shallow concrete inserts orhooked starter reinforcing bars can be adversely af-fected by spalling of cover concrete. In situationswhere loss of cover concrete may occur, the inserts orhooked starter reinforcing bars should not rely on thetensile strength of and bearing on cover concrete.Transfer of the design load to the member should thenbe achieved by use of an additional U-shaped reinforc-ing bar passing through the eye-hole in the base of theinsert and anchored in the core of the section, or the useof threaded bar couplers also anchored within the core(see Figure 7.1 (e)).

The use of inserts without bars through the eye-holes isnot recommended, unless the embedment length of theinsert is such that a brittle failure of the concrete resist-ing mechanism is prevented. Note that some proprietaryinserts include eye-holes whose diameter is less thanthat required to accommodate the bar size necessary toresist the ultimate pullout load on the insert. Theseinserts should not be used when subject to seismicloading. The effect of adverse influences discussed inSections 7.6.7 and 7.6.8 should always be considered.

7.6.5 Post-drilled FixingsPost-drilled fixings, involving the installation of theanchor with epoxy, or non- shrink cement grout, can bedesigned using References 7.4 or 7.8. Post-drilledfixings, involving the use of expansion anchors, can bedesigned using a similar approach to that shown inReference 7.4.

In general, the effect of concrete cracking on expansionanchors is significant (see Section 7.6.7) and completeloss of preload can occur if cracking coincides with theanchor location.

Proprietary post-drilled fixings, exhibiting the follow-ing features, are known to provide improved loadresistance over conventional expanding anchor types:

• ability to pre-load the anchor against a load indicat-

ing washer to a level that ensures the externallyapplied load never exceeds the long-term preloadcapacity; and

• ability to under-ream (or “bell”) the anchor hole, topermit the expanded base of the anchor to mechani-cally engage the surrounding concrete under a pre-load condition.

Notwithstanding the above, it is recommended thatexpansion anchors should not be used in areas wherecracking of the surrounding concrete is likely to occur.

In cases where expanding anchors are loaded dynami-cally or in static tension, it is recommended that theloading not exceed a proportion of the first-slip load,typically 65%. The first-slip load is commonly definedas the load at 0.1 mm slip. These particular uses forexpanding anchors should be discussed, prior to use, withthe manufacturer/supplier of the proprietary anchor.

7.6.6 Cast-in Steel SectionsThe resistance of cast-in sections to shear and bendingcan be calculated assuming the formation of concretecompression stress blocks to resist the actions, andwith due regard to the rigidity of the section as shownin Figure 7.4.

Tension forces should be resisted by anchor lugs thatwill mobilise the strength of a cone of concrete. Designwill then be in terms of the principles previouslyoutlined.

A recommended design approach is included in Exam-ple 2 of Appendix E1.

7.6.7 Influence of CrackingThe performance of embedments is adversely affectedby cracking [7.4]. It is recommended that if an anchor islocated in an area where cracking is likely to occur (fromany cause) extra precautions should be taken, such asadditional reinforcement to transfer the loads on theembedment back to an uncracked region of concrete.

It is not advisable to locate embedments in potentialplastic hinge zones where extensive cracking can beexpected and from where cover concrete may be lost.If relocation is not possible, the embedment should bedesigned neglecting the tensile and compressionstrength of the concrete. The load transfer should relyon truss action using transverse reinforcement to trans-fer tension where required, noting that the core con-crete can also become badly cracked.

The recommended approach is to embed the anchors asdeeply as possible into the core, and to ensure thepotential failure surface engages the transverse rein-


forcement in that region. It is recommended that post-drilled anchors should not be used in potential plastichinge regions.

Fixings anchored in zones of flexural cracks shouldhave their failure cones modified to account for thestrength reduction of the embedment caused by thecracks. Reference 7.4 recommends that the resistanceof the concrete mechanism be reduced in proportion tothe crack widths. Refer to Section 7.6.8 for the influ-ence of cracking on the design of welded plates withheaded studs.

7.6.8 Detailing for the Effects of SpalledConcrete CoverThe design of embedments will normally include thecontribution of all concrete surrounding the anchor-age, including that in the cover zones.

Where embedments are unavoidably located in zoneswhere spalling of cover concrete is likely, additionalprecautions are required as follows:

• The type of embedment selected must still permitthe applied loads to be resisted by the assembly(albeit at reduced load factors), without the coverconcrete forming part of the load transfer mecha-nism. This could be by flexure and tension in studsanchored into confined concrete, rather than bybearing on concrete.

• Anchorage beyond the cover zone should includeallowance for concrete cracking. This requirementmay preclude the use of headed studs, for example,in favour of longer reinforcing bars with hookedanchorage into the confined concrete.

• The type of embedment specified must exhibitload-deformation characteristics that are inherentlyductile, i.e. not highly sensitive to assembly defor-mation, or changes in position of applied loadswhich may occur concurrently with the spalling ofcover concrete.

• Special attention is required in the welding ofgalvanised and structural steel, to ensure that pre-mature (and brittle) failure arising from weld zoneembrittlement is avoided.

7.6.9 Testing of Embedded SteelConnectionsThe strength of embedded steel connectors may bedetermined by means of load testing no less than 10(ten) specimens to failure. The test loads shouldsimulate the most unfavourable combination of loadsand forces likely to be applied to the embedded steelconnector when in service. The design is the load atwhich no specimens fail. Less than 10 (ten) specimenscan be tested if full account is taken of the effects ofstatistical variation.

7.7 Detailing of SteelEmbedments in Concrete

7.7.1 GeneralIn the case of overload, the embedments must yield orslip before failure, thus having sufficient ductility orflexibility to preclude a brittle concrete pull-out fail-ure, or brittle fracture at or near the welds. The embed-

45˚R1

R2

R1 + V = R2

potential pull outfailure surface undertension load

assumed concrete bearingstress distribution

V

T

maximum bearingstress only nearweb

Figure 7.4: Actions on an embedded length of RHS


ment must also yield or slip in the attached connectinghardware before the component of the fixing anchoredinto the concrete can fail in a non-ductile manner.Where failure of the concrete resisting mechanismmight occur, embedments should be attached to, orhooked around, reinforcing steel to effectively transferforces to the reinforcing and surrounding concrete (seeFigure 7.5).

7.7.2 Detailing for Limited EdgeDistanceDetails for this situation are provided in Reference 7.6.The design approach is to transfer the force from theembedment to reinforcing steel, which is anchored intothe concrete core.

Tests on the shear resistance of anchor bolts underreversed cyclic loading [7.6], indicate that for smalledge distances the fixing’s performance can be ad-equate with the provision of two reinforcing steelhairpins or U-bars (Figure 7.5).

While the Type 1 detail (Figure 7.5(a)) provided betterprotection against concrete spalling at lower loads, itwas observed that cyclic loading could cause concretebetween the hairpin and the bolt to spall away. It istherefore recommended that for seismic loads the Type2 detail be used.

7.7.3 Welding to Embedded PlatesWhen welding anchors or reinforcing bars to plates(Figures 7.1 (a) and (b)), plate laminar tearing must beprevented. It is recommended that a “pass-through andplug weld” detail be used (Figure 7.6).

To avoid concrete spalling at the edges of the platewhen heat from on-site welding causes expansion, agap between plate and concrete should be formed. Theplate thickness should also be sufficiently thick tominimise welding distortions.

7.7.4 Corrosion ProtectionSteel embedments are often vulnerable to corrosion,and are normally specified either hot-dipped galva-nised, or with stainless steel components.

The following points should be noted when specifyingcorrosion protection:

• hot-dipped galvanising of mild steel can engenderembrittlement in cold-worked sections, reducingductility; and

• stainless steel, where attached to mild steel bywelding, can promote crevice corrosion, or gal-vanic corrosion, in the contact areas.

Both of these phenomena are discussed in Appendix D2.

7.8 Recommendations

7.8.1 Steel EmbedmentsSteel embedments in all situations should be designedand detailed to suppress concrete pull-out failure,thereby ensuring a ductile mechanism.

7.8.2 Design and Detailing ofEmbedmentsSteel embedments selected for a specific applicationshould be evaluated on their ability to:

• withstand all imposed deformations, and appliedloads; and

• accommodate structural damage, e.g. by spallingof cover concrete without loss of strength below anacceptable level.

7.8.3 Design MethodsDesigners should check that design method assump-tions, e.g. in source documents, are valid for the in-stalled condition of the steel embedment. Specificreferences are recommended for use in particular ap-plications.

7.8.4 CorrosionDesigners should specify embedments with a corro-sion resistance appropriate to the design life of thestructure and the envisaged frequency of maintenanceinspections.

7.8.5 Use of Overseas DesignInformation in New ZealandNew Zealand designers should review and modify thedesign information contained in the listed references,to ensure that the approach used to determine structuralstrength is consistent with that adopted in other NewZealand standards.

7.9 References7.1 Code of Practice for General Structural Design

and Design Loadings for Buildings, NZS4203:1992, Standards New Zealand, Wel-lington, 1992, 45 pp.

7.2 Concrete Structures Standard, NZS 3101:1995,Standards New Zealand, Wellington, 1995,256 pp.


7.3 “Mechanical fasteners for concrete”, Informa-tion Bulletin IB31, Cement and ConcreteAssociation of New Zealand, Porirua, 10 pp.

7.4 Fastenings to Concrete and Masonry Structures- State of the Art Report, Comité Euro-International du Béton, Thomas Telford,1994, 249 pp.

7.5 Restrepo-Posada J I and Park, R. “Tensilecapacity of steel connectors with shortembedment lengths in concrete”, ResearchReport 93-6, Department of Civil Engineer-ing, University of Canterbury, August 1993,51 pp.

7.6 Klingner R E, Mendonca, J A and Malik, J B.“Effect of reinforcing details on the shearresistance of anchor bolts under reversedcyclic loading”, ACI Journal, Vol 79 No 1,Jan-Feb 1982, pp. 3-12.

7.7 Hsu T T C. ”Fatigue of plain concrete”, ACIJournal, Vol 78 No.4, July-August 1981, pp.292-305.

7.8 Brown, B J, Vautier E W and Shepherd, D A.“Grouting of anchors and reinforcing starterbars into concrete”, Research Project No.108905 (Draft Report), Transit NZ, Welling-ton, May 1990, 69 pp.

100

20

50

200

20

anchor bolt

concrete surface

hairpin

secondary hairpinfor cyclic loadsconcrete edge

~ L/2

L(a) Type 1 detail

50-100

20

anchor bolt

concrete surface

hairpin

secondary hairpinfor cyclic loadsconcrete edge

~ L/2

200

50

L(b) Type 2 detail

Figure 7.5: Hairpin reinforcement to embedded bolt with small edge distance


rebar plate

gap to allow forwelding heat generatedexpansion 45˚

1 - 2 mmclearance weld as per 6.7

2 - 3 mm

Weld Detail

Figure 7.6: Connection of reinforcing bars to plate

Tolerances • 101

Chapter 8Tolerances

8.1 IntroductionSuccessful precast concrete construction relies on afull understanding of the need for adequate tolerancesand the full implications of variations in dimensions.This understanding must be developed by designers,fabricators and constructors.

The requirements regarding tolerances for precast con-crete construction are contained in the New ZealandStandard NZS3109 1997 [8.1].

The 1997 version of this Standard was derived fromstudying Australian and American sources, coupledwith the discussions raised in Chapter 8 of the 1992edition of this publication. The revised tolerances arevery similar to the provisions contained in the Pre-stressed Concrete Institute (PCI) Committee on Toler-ances [8.2] and the subsequent incorporation of thismaterial into a more general publication [8.3].

In the preparation of a national code such as NZS3109,there remains always the difficulty of selecting appro-priate information for all eventualities. This can oftenlead to significantly over-complicated documentation.Given that PCI examples of tolerances provided sig-nificant illustration of the combination of tolerancesfrom manufacture and erection, the PCI examples ofFigures 8.1 (a) and (b), 8.2, 8.3 and 8.4 have beenretained. To help the designer, however, tolerancesfrom the New Zealand Standard 3109 Tables 5.1,“Tolerances for precast components” and 5.2 ,“Toler-ances for in situ construction” have been substitutedfor the PCI values where similar definitions occur.

It is essential that the designer and precaster under-stand the tolerance requirements of NZS3109 because,by reference, all work designed to NZS3101 is deemedto be constructed in accordance with NZS3109.

The approach in this section is to begin by consideringtolerances as three different types (namely product,erection, and interface tolerances) and then to look atthe question of clearances.

8.2 Product ToleranceProduct tolerances relate to the dimensions of an indi-vidual component. They are set by the designer to

control production in order to achieve the structuraland architectural requirements. They should be aslarge as possible, but commensurate with the natureand type of structure being built. Close tolerancesincrease costs, especially where limited productionnumbers are required.

Product tolerances for New Zealand are given in Table5.1 of NZS3109. An example of PCI, American prod-uct tolerances [8.2], is given in Figures 8.1 (a) and (b).The comparative New Zealand tolerance figures, whereavailable, have been added to the table. These toler-ances are generally accepted by precast concrete manu-facturers and may be used. However, designers shouldalways check the appropriateness of a given tolerancefor each particular situation. In some cases the recom-mended tolerance may not be suitable.

8.3 Erection ToleranceErection tolerance is the allowance between actualcomponent location and primary control surfaces suchas grids, datum levels, etc. Some examples of PCIerection tolerances [8.2], which are generally achiev-able by the New Zealand construction industry, aregiven in Figures 8.2 to 8.5. If smaller tolerances arerequired, increased building costs can be expected.

However, reference must be made to NZS3109 Table5.2, “Profile tolerances”, which deal with positionwithin a structure and plumbness of the structure.Tolerance differences between the two references aresmall, sometimes lower, sometimes higher, e.g. Planposition PCI Structural +13 mm, Architectural ~l0 mmNZS3109 ~l0 mm. Where precast concrete is to befitted to a structural steel frame, then the erectiontolerances from NZS3404 Clause 15.3 need to be takeninto account.

The use of “primary control surfaces” [8.2] are consid-ered a useful technique for specifying erection toler-ances. It is the positional dimensions of these surfacesthat are controlled during erection. The positionaldimensions of all other parts of the component will begoverned by the erection and product tolerances. Forexample, consider the beam-column junction in Fig-ure 8.3. The beam bearing surface for the precast


Figure 8.1(a): Product tolerances for building beams and spandrels.

(From Reference 8.2)

(This, and other figures from Reference 8.2 are included with permission of PCI).

b b

r 2

c

i

e

e1

d

e1

d

c

b

Cross Sections

e

3000s

q

o

o

q

k

f

p

k

Plan

l

n

r1

m

g j

a

h

end bulkhead angle

Elevation

Tolerances • 103

Figure 8.1 (b): Product tolerances for building beams and spandrels


f== Varies with length 3 to

L/1000 or 15 max±6 mmSweep (variation from straight line parallel to centerline of member)Up to 12m member length12 to 18m member lengthGreater than 18m member length

±13 mm±16 mm

g

h*

i

=

=

=

Variation from specified endsquareness or skewCamber variation fromdesign camberPosition of tendons

±3 mm per 300 mm depth, ±13 mm max

±3 mm per 3 m, ±19 mm max

Varies with length ±5 to ±15

Varies with length ±5 to ±15

±5IndividualBundled

±6 mm±13 mm

e1±6 mm

ab

==

Length

cde

Width (overall)Depth (overall)Depth (ledge)Stem width

===

Ledge width

±19 mm±6 mm±6 mm±6 mm±6 mm

Varies with length ±8 to ±20Varies with length ±5 to ±10Varies with length ±5 to ±10±5Varies ±5 to ±10±5

j

klmnop

=

======

q

r

=

=

s =

Longitudinal position from design locationof deflection points for deflected strandMember length 9m or lessMember length greater than 9mPosition of platesPosition of bearing platesTipping and flushness of platesTipping and flushness of bearing platesPosition of sleeves both horizontal and vertical planePositions of inserts forstructural connectionsPosition of handling devicesParallel to lengthTransverse to lengthPosition of stirrupsr1 longitudinal spacingr2 projection above surface of beamLocal smoothness any surfaceDoes not apply to top surface leftrough to receive a topping or tovisually concealed surfaces

±150 mm±300 mm±25 mm±13 mm±6 mm±3 mm

±25 mm

±13 mm

±300 mm±13 mm

±50 mm±6 mm - 13 mm

(6 mm in 3 m)

±5±5±12

±12

±12±12

±20Inserts ±12

See Clause 3.9 NZS3109

See NZS3114

†NZS3109 from Table 5.1mm

*

†

For members with a span to depth ratio approaching or exceeding 30, the stated camber tolerance may notapply. If the application requires control of camber to this tolerance in beams of such span-to-depth ratios,special premium production measures may be required. This requirement should be discussed in detail withthe producer.

Table 5.1 also has further information for precast slabs and additional fitments in the precast units.

concrete floor units is a control surface, not the top ofthe column.

The accuracy achieved in constructing the foundationsof buildings has a major bearing on the success orotherwise of the assembly of precast concrete units.The contractor must therefore establish appropriate

surveying controls. Before erection commences, aftererection, and before other trades start work, the accu-racy of erection should be verified. Failure to use thesecontrols can, and frequently does, lead to costly andtime consuming alterations and arguments.

Like product tolerances, erection tolerances should be


as liberal as possible for the efficient matching of theunits. It is recommended that, wherever possible, thedesigner review the proposed tolerances with themanufacturers and erectors prior to deciding on thefinal project tolerances.

8.4 Interface ToleranceInterface tolerance refers to the allowance needed forthe jointing or attaching materials in contact with theprecast units. However, the immediate area of interestis the interfacing between precast units themselves andbetween precast units and cast-in-place concrete. Thisinterfacing has often caused problems in the past.

One common example is the clashing of reinforcingbars at a beam-column junction. Often, the position ofthe projecting column steel is such as to displace thebeam from its true position. A possible consequence ofthis is the loss of bearing of the precast floor unitssupported on one side of the beam, and too muchbearing on the other side. Displacement of the beam bythis type of interfacing problem also has a cumulativeeffect, especially where long beams span two or threebays of the frame. Frequently, the lapping of reinforc-ing bars at midspan beam joints is only achieved withthe aid of gas sets or hydraulic jacks.

8.5 ClearancesThe provision of clearances is recognition of the needfor interface tolerances. Clearance should provide abuffer area where combined erection and productionvariations can be absorbed, and the actual clearanceprovided should reflect all the previously specifiedtolerances.

If allowance is made for the worst absolute algebraiccombination of permissible tolerances, namely prod-uct tolerance + erection tolerance (including the toler-ance of cast-in-place concrete) + interface tolerance,then it can become virtually impossible to detail inter-face connections. For example, consider the seating ofprecast concrete floor units on a precast beam, asshown in Figure 8.6(a). If all the maximum tolerancesfrom NZS 3109 are used, the cumulative tolerance forthe seating is + 36 mm. This would require a nominaldesign seating of approximately 101 mm, which is farmore than currently used in practice. This exampleindicates a need to review the current requirements ofNZS 3109. The reality is that many contractors havesuccessfully built structures to far more stringent toler-ances.

The recommended method to cope with this problem isfor the designer to indicate in the contract documents

(preferably the working drawings) where tighter toler-ances than those specified in NZS 3109 are required.The provided clearances should also be indicated so thecontractor can make appropriate allowances during bothtendering and construction. In many instances the con-tractor may be required to use special templates to ensureaccurate starter locations for reinforcing bars, etc.

Two approaches have been suggested to combinepermissible tolerances to calculate clearances. Refer-ence 8.2 emphasises the need for engineering judge-ment to assess the likelihood of maximum producttolerances occurring in one location. The second ap-proach [8.4] uses a statistical method in which thecombined tolerance is the square root of the sum of theindividual tolerances squared. This method is not onlysuitable where there is insufficient previous experi-ence as a basis for sound judgement, but it will also givedesigners a realistic value for the required clearance.An example is given in Figure 8.6(b).

Designers are referred to Reference 8.2 which providesexamples of calculation of clearances for several dif-ferent construction situations.

8.6 Implications for DesignWhen designing connection details involving precastmembers, it is important that designers allow for in-creased structural actions due to the occurrence ofunfavourable construction tolerances. The effect oftolerances will usually be negligible on the design ofstructural elements such as precast concrete floor units.However, when the design of a connection is largelyinfluenced by the value of a small dimension which issensitive to variation caused by tolerances, then toler-ances should be allowed for in the design. An exampleis shown in Figure 8.7, where a steel angle is designedto support a precast concrete floor unit. The designclearance between the end of the bolt and the unit is20 mm, but when cumulative permissible deviationsare considered, the actual gap is increased by 20 mm,resulting in significantly increased stresses in the boltand steel angle.

It is vital when designing connections where tolerancesare significant, that cumulative tolerances be used todetermine the worst design case.

8.7 Recommendations• Tolerances from NZS3109 from Table 5.1 and 5.2

should be used as the primary reference points, butreferences 8.2 and 8.3 should be used as recom-mended resource documents where special addi-tional concerns may need to be addressed.

Tolerances • 105

Figure 8.2: Erection tolerances for columns (Reference 8.2)

a = Plan location from building grid datumStructural applicationsArchitectural applications

b

c

d

ef

=

=

=

==

Top elevation from nominal top elevationMaximum lowMaximum highBearing haunch elevation from nominal elevationMaximum lowMaximum highMaximum plumb variation over height of element(element in structure of maximum height 30 m)Plumb in any (3 m) of element heightMaximum jog in alignment of matching edgesArchitectural exposed edgesVisually non-critical edges

±13 mm

±13 mm

±13 mm

±10 mm

±6 mm

± 25 mm

±6 mm

±6 mm

±6 mm±13 mm

b

c

precast concretecolumn

aa

a add

f

f

ee

a

3m

splice area

Building X grid datumor Y grid datum

Elevation datumElevation datum

Building Y grid datum

Building X grid datum

Where precast concrete is to be attached to a struc-tural steel frame, reference is also needed toNZS3404, Clause 15.3.

• Designers should liaise as closely as possible withcontractors when specifying tolerances so appro-priate allowances can be made, thereby reducingconstruction difficulties (Section 8.5).

• It is recommended that when designing connec-tions that are sensitive to tolerances, cumulative

maximum permissible tolerances be used to definethe worst design case (Section 8.6).

8.8 References

8.1 Specification for Concrete Construction, NZS3109:1997, Standards Association of NewZealand, Wellington, 1997, 67 pp.

8.2 PCI Committee on Tolerances, “Tolerances for


Figure 8.3: Erection tolerances for beams and spandrels, precast element to precast concrete,

cast-in-place concrete, masonry or structural steel (From Reference 8.2)

aa1b

c

d

e

fg

==

=

=

=

=

==

Plan location from building grid datumPlan location from centreline of steel*Bearing elevation** from nominal elevation at supportMaximum lowMaximum highMaximum plumb variation over height of elementPer 300 mm heightMaximumMaximum jog in alignment of matching edgesArchitectural exposed edgesVisually non-critical edgesJoint widthArchitectural exposed edgesHidden jointsExposed structural joint not visually criticalBearing length*** (span direction)Bearing width***

* For precast elements on a steel frame, this tolerance takes precedence over tolerance on dimension“a”.

Or member top elevation where member is part of a frame without bearings.**

*** This is a setting tolerance and should not be confused with structural performance reqiurementsset by the architect/engineer. the nominal bearing dimensions and the allowable variations in thebearing length and width should be specified by the engineer and shown on the erection drawings.

±25 mm±25 mm

13 mm6 mm

3 mm13 mm

6 mm13 mm

±6 mm±19 mm±13 mm±19 mm±13 mm

g

f

e

d

a

bearing area

a

b

b

d

c c

precast or cast-in-placecolumn


ElevationPlan

Column grid line

Column grid line

Column grid line

Design elevation

precast and prestressed concrete”, PCIJournal, January-February 1985,pp. 26-112.

8.3 Recommended Practice for Erection of Precast

Concrete, Prestressed Concrete Institute,Chicago, 1985, 87 pp.

8.4 Code of Practice for Accuracy in Building BS5606:1978, British Standards Institution,London, 59 pp.

Tolerances • 107

Figure 8.4: Erection tolerances for floor and roof members (Reference 8.2)

aa1b

c

d

e

fg

==

=

=

=

=

==

h =

Plan location from building grid datumPlan location from centreline of steel*

Top elevation from nominal top elevation at member endsCovered with toppingUntopped floorUntopped roofMaximum jog in alignment of matching edges(both topped and untopped construction)Joint widthMember length 0 - 12 m

12.1 - 1.8 m18.1 m plus

Differential top elevation as erectedCovered with toppingUntopped floorUntopped roof**Bearing length*** (span direction)Bearing width***Differential bottom elevation of exposed hollow-core slabs****

*

**

***

****

For precast concrete erected on a steel frame building, this tolerance takes precedence

It may be necessary to feather the edges ±6mm to properly apply some roof membranes.

This is a setting tolerance and should not be confused with structural performance requirementsset by the architect/engineer. The nominal bearing dimensions and the allowable variations inthe bearing length and width should be specified by the engineer and shown on the erectiondrawings.

Untopped installations will require a larger tolerance.

over tolerance on dimension “a”.

±25 mm±25 mm

±25 mm

±25 mm±19 mm

±19 mm

±19 mm

±19 mm

±19 mm±19 mm±19 mm

±13 mm

±6 mm

±6 mm

±6 mm

Plan Plan

a

a

g

d

c

bearing

precast concrete flooror roof members

a

g

g

d

c

f

a

steel structure


bf

e

precast or cast-in-placeconcrete support member

clearance


Elevation Elevation

a1

f

b

e


steel support structure

of steelstructure





Building elevation datumBuilding elevation datum


Figure 8.5: Erection tolerances for structural wall panels (a) precast element to precast or cast-in-place

concrete or masonry, (b) precast element to structural steel (Reference 8.2)

c = Bearing elevation from nominal elevationMaximum lowMinimum high

d = Maximum plumb variation over height of structureor 30 m whichever is less**

e = Plumb in any 3 m of element heightf = Maximum jog in alignmane of matching edgesg = Joint width (governs over joint taper)h = Joint taper over length of panel

h3 = Joint taper over 3 m lengthi = Maximum jog in alignment of matching faces

ExposedNonexposed

j = Differential bowing, as erected, between adjacentmembers of the same design

a = Plan location from building grid datum*a1 = Plan location from centreline of steel**

b = Top elevation from nominal top elevationExposed individual panelNonexposed individual panelExposed relative to adjacent panelNonexposed relative to adjacent panel

* For precast buildings in excess of 30m height, tolerances “a” and “d” can increase at a rate of3 mm per storey to a maximun of 50 mm.

** For precast concrete erected on a steel frame building, this tolerance takes precedence over toleranceon dimension “a”.

13 mm6 mm

25 mm6 mm13 mm±10 mm

13 mm

10 mm

10 mm19 mm

13 mm

±25 mm±25 mm

±13 mm±19 mm±13 mm±19 mm


Building X grid datumBuildingelevationdatum

a

ia

precast concretepanel

cast-in-place orprecast concrete

e ec

b

3m

d d

3m

g f h

h10

nominal joint widthcast-in-place foundationor precast concrete supportPlan Elevation

a

ia

a1

steelstructure

of steel structure

e e

c

b

3m

d d

steel supportbeam

nominal joint width

g

3m

h3

hfg

Plan Elevation

Buildingelevationdatum


Building Xgrid datum

(a)

(b)

Tolerances • 109

Figure 8.6: NZS3109 tolerances applied to a precast beam-floor unit interface using two

combination members

gridline

A

precast concretefloor unit

precast concretefloor unit

(b) Combination by square root ofthe sum of the squares

(a) Combination by absolutealgebriac addition

Tolerances for point A

Tolerances for floor unit

Maximum setting width required

= 82.2 mm

(102 + 2.52 + 82)0.5 = 13.0

(102 + 52)0.5 = 11.2

= 65 + (132 + 11.22)0.5

Tolerances for point A

Tolerances for floor unit

length (L > 12) (50% ± 20)squareness

± 10± 5

total ± 15

Maximum setting width required= 65 (design seating) + 20.5 + 15

= 100.5 mm

position in plan Table 5.2cross-sectional dimension (50% ± 5)straightness (L = 8 m)

± 10± 2.5± 8

total ± 20.5

Figure 8.7: Precast concrete floor unit - structural steel seating detail

20 20

load bearing wall

actual final position

assumed location of endof precast concrete floor unit

precast concretehollow-core floor unit

Appendix A1: Beam Elongation due to Plastic Hinging • 111

Appendix A1: Beam Elongation due toPlastic Hinging

Elongation occurs with the formation of a plastic hingein a member. This arises as the rotation occurs pre-dominantly due to tensile yielding of the reinforcementrather than the crushing of the concrete. Longitudinalextensions in beams of the order of 2 to 4 percent of thebeam depth per plastic hinge have been observed intests in which expansion was free to occur [A1.1 toA1.5]. In a test of a combined frame-wall structure, theelongation of the wall was found to result in a redistri-bution of internal action and a major increase in lateralstrength of the system [A1.6].

In a uni-directional plastic hinge in a beam, that is ahinge in which inelastic deformation occurs in onlyone direction in an earthquake (see Figure A1.1 (a)), orin the first inelastic displacement in a reversing plastichinge, the compressive strains in the compressionreinforcement are small and for practical purposes maybe neglected. On this basis the growth in length in eachhinge zone may be calculated from the expression:

extension(d d )

2=

− ′θ(A1.1)

where θ is the rotation of the plastic hinge zone andd - d′ is the distance between the centroids of top andbottom reinforcement in the beam [A1.3].

With reversing plastic hinge zones it has been foundthat cracks in the compression zone do not close unlessaxial forces appreciably greater than the maximumshear force are applied to the member [A1.4 and A1.5].In the absence of a restraining force, the membercontinues to elongate as cyclic inelastic deformationsare applied. Test results have indicated that the elon-gation in a typical reversing plastic hinge in a beam,which have been subjected to two complete cycles of±2 and ±4 displacement ductilities and then taken to thefirst displacement ductility of 6, is approximately twicethe value given by equation A1.1. For this situation, asillustrated in Fig A1.1, the plastic hinge rotation can beassessed from the expression:

θ =Δhk

(A1.2)

where Δ is the interstorey deflection, h is the interstoreyheight and k is the ratio of the distance between thehinge zones in a beam to the column centrelines.

On this basis the elongation in the length of a structure,which is not restrained against expansion and which

forms reversing hinges in a major earthquake, is ap-proximately given by:

extensionk

nh

d d= − ′2 Δ

( ) (A1.3)

where n is the number of bays.

For the case where uni-directional plastic hinges formin a severe earthquake, as illustrated in Fig 1.1(b), theinelastic rotations accumulate with the passage of theearthquake. The resultant magnitude of plastic hingerotation depends upon the duration of strong groundshaking. Several series of analyses have been madewith earthquake records with intensities and durationssimilar to that envisaged in the loadings code [A1.2,A1.8]. From these, it was found that the plastic hingerotation was typically 2 to 4 times the value for areversing plastic hinge as given by Eq.A1.2.

The elongation caused by tensile yielding of reinforce-ment associated with plastic hinge formation can bereadily visualised by drawing the plastic hinge zones asconcentrated areas as illustrated in Figures A1.1 andA1.2. As indicated in Figure A1.2, particularly severeeffects can arise where the seismic resistance is pro-vided by a perimeter frame. In this situation, the accu-mulated elongation from the beams in several bays maybe imposed on a single bay of a precast floor system.

In calculating the maximum interstorey deflectionduring a severe earthquake, allowance should be madefor the inelastic deformation that develops. For struc-tures of limited ductility, when the structural ductilityfactor is 2 or less, the values of the maximum interstoreydeflection can be estimated from the lateral deflectionsin an elastic analysis scaled by the structural ductilityfactor. However, in ductile high-rise moment-resistingframe structures, the critical interstorey deflection istypically 1.7 times as great as estimated by scalingelastic-based deflections by the structural ductilityfactor. This aspect is recognised by the code [A1.7],which gives different interstorey deflection limits de-pending upon whether an elastic-based analysis, or aninelastic-based time history analysis is used [A1.9,A1.10].

Further research is required to establish the magnitudesof member elongation that may occur in severe earth-quakes. The recommendations in this appendix shouldbe treated as tentative until this work has been carriedout.


ReferencesA1.1 Fenwick, R C and Fong, A. “The behaviour of

reinforced concrete beams under cyclicloading”, Bulletin of the NZ National Societyfor Earthquake Engineering, Vol. 12, No. 2,June 1979, pp 158-167.

A1.2 Restrepo-Posada, J I. Seismic Behaviour ofConnections Between Precast ConcreteElements, Report 93-3, Dept of Civil Engi-neering, University of Canterbury, pp 339-347.

A1.3 Megget, L M and Fenwick, R C. “Seismicbehaviour of a reinforced concrete portalframe sustaining gravity loads.” Bulletin ofNZ National Society for Earthquake Engi-neering, Vol 22, No. 1, Nov 1989, pp 39-49.

A1.4 Fenwick, R C, Tankat, A J and Thom, C W.The Deformation of Reinforced ConcreteBeams Subjected to Inelastic Cyclic Loading- Experimental Results, University of Auck-land, School of Engineering Report No 268,Oct 1981, pp 72.

A1.5 Fenwick, R C and Davidson, B J. “Elongationin ductile seismic resistant RC frames”,Recent Developments in Lateral ForceTransfer in Buildings, American Concrete

Institute Special Publication SP 157, 1995, pp143-170.

A1.6 Wight, J K (editor). Earthquake Effects onReinforced Concrete Structures, AmericanConcrete Institute, Special Publication SP84,1985, 428pp.

A1.7 Code of Practice for General Design andDesign Loadings for Buildings, NZS 4203-1992, Standards New Zealand, Wellington,1992.

A1.8 Fenwick, R C, Dely, R and Davidson, B J.“Ductility demand for unidirectional andreversing plastic hinges in ductile momentresisting frames”, Bulletin of New ZealandSociety for Earthquake Engineering, Vol. 32,No. 1, March 1999, pp 1-12.

A1.9 Paulay, T. “A consideration of p-delta effectsin ductile reinforced concrete frames”,Bulletin of NZ National Society for Earth-quake Engineering, Vol 11, No.3., Sept 1978,pp 151-160.

A1.10 Moss, P G and Carr, A J. “The effects oflarge displacement on the earthquake re-sponse to tall frames structures”, Bulletin ofNZ National Society for Earthquake Engi-neering, Vol 13, No.4, Dec 1980, pp 317-328.

Appendix A1: Beam Elongation due to Plastic Hinging • 113

θ3θ2

θθ14

hΔ

d d’

NOTE:

from the column face

of rotation may be taken

θ

θ

d d´

The hinge zones have beendrawn as concentrated rotationsat a section.To allow for the spread ofcolumn face hinges the centre

4

(d-d′)

(a) Uni-directional hinges in a bay

(b) Reversing hinges in a bay

Figure A1.1: Reversing and uni-directional plastic hinges in a beam


hΔ

growth in 3 bays imposed on one set ofprecast concrete floor units

precast concrete flooring

perimeter frame

Plan

Elevation on perimeter with interstorey drift

Figure A1.2: Growth in perimeter frame imposed on precast concrete flooring

Diaphragms • 115

Appendix A2: Allowances for Effectsof Spalling

Material of Support Distance Assumed Ineffective

Steel 0

15 mm

25 mm

25 mm

Concrete Grade 30 or over, plain orreinforced (in general)

*

Brickwork or masonry*

Concrete below Grade 30, plain orreinforced (in general)

*

Reinforced concrete less than 300 mmdeep at outer edge

Reinforced concrete where vertical-loopreinforcement exceeds 12 mm diameter

not less than nominal cover toreinforcement on outer face of support

Nominal cover plus inner radius of bend

Where unusual spalling characteristics are known to apply when particular constituentmaterials are being used, adjustment should be made to the distances recommended

*

Table A2.1: Allowances for effects of spalling at supports

(From Reference 2.5, with permission)

Reinforcement at Bearing ofSupported Member

Distance Assumed Ineffective

Straight bars, horizontal loops or verticalloops not exceeding 12 mm in diameter,close to end of member

Tendons or straight bars exposed atend of member

Vertical loop reinforcement of bar sizeexceeding 12 mm in diameter

10 mm or end cover, whichever isthe greater

0

End cover plus inner radius of bar

Table A2.2: Allowances for effects of spalling at supported members


Appendix B2: Experimental Tests — System 2 • 117

Appendix B1: Precast Concrete FrameConnection for System 1 Design

Bending Moments

With System 1 type connections, the precast beamsmay be supported at each column face so that the deadand construction live loads are initially carried by thebeam acting as a simply supported member. When thecast-in-place concrete hardens, the initial simple sup-ports for the beam become moment-resisting connec-tions, which sustain bending moments in response toany rotation subsequently applied to them. The struc-tural form has been changed. In its initial condition theinitial dead load acting on the simple span causesrotations to occur at the supports. Due to creep andshrinkage in the concrete these rotations continue toincrease with time. However, the hardening of thecast-in-place concrete and the resultant change in thestructural form leads to a redistribution of that portionof the bending moment arising from the dead loadapplied to the member before the structural form waschanged. Dead loads applied after the structure hasbeen modified are not subject to redistribution.

With the redistribution of bending moments describedabove, the values change from those sustained in theinitial state towards those that would be sustained if theload in question had been applied to the structure in itsfinal condition. This is illustrated in Figure B1.1. Howfar this redistribution goes depends upon the creep andshrinkage characteristics of the concrete, the timebetween the initial condition and when the change instructural form is made, and the arrangement of rein-forcement in the beam. Making a rational allowancefor these factors is complex. Readers wishing to pursuethis are referred to references B1.1 and B1.2. Anapproach by which the extent of redistribution can beassessed may be based on the method used in theconcrete design standard [B1.3] for finding long termdeflections. The long term deflection is given as Kcptimes the short term deflection, where the factor Kcp isgiven by the equation:

KA

Acps

s

= −′

>2 1 2 0 6. .

This value corresponds to an effective creep coeffi-cient for the beam as a whole, and it may be used withthe effective modulus method to indicate the propor-

tions of the initial dead load resisted by the structure inits initial and final forms when creep has ceased. Onthis basis the proportion of the load carried by the initialform, pin, is given by:

pKin

cp

=+

1

1

with the remainder of the load K

Kcp

cp1+

⎡

⎣⎢⎢

⎤

⎦⎥⎥ being resisted

by the structure in its final form.

In checking the performance of the beam for service-ability, it should be noted that this redistribution ofbending moments occurs with time. Consequently, itmay be necessary to check for both the initial and finaldistribution of bending moment.

In the strength limit state further redistribution mayoccur with the formation of plastic hinges in the beam.

The design and construction of beams propped only atthe ends also raises the issue of the shear strength of theprecast beam in its temporary construction state, asshown in Figure B1.2, since it could be considered thatthe stirrups are inadequately anchored. This issue hasbeen already discussed in Section 3.2. However, noshear problems have been observed in practice duringconstruction using this system.

ReferencesB1.1 ACI Committee 209, Designing for Creep and

Shrinkage in Concrete Structures, AmericanConcrete Institute, Special Publication, SP76,1982.

B1.2 Bryant, A H, Wood, J A and Fenwick, R C.“Creep and shrinkage in concrete bridges”,RRU Bulletin 70, National Roads Board,Wellington, 1984.

B1.3 Design of Concrete Structures: NZ3101:1995,Standards New Zealand, Wellington, 1995,256pp.


dead load applied to beambefore continuity was achievedwith the columns

the change in bendingmoments that occurswith time

initial distribution of bendingmoments with structure in itsinitial state

beam propped at the columnface during construction

the final distribution of bendingmoments when creep andshrinkage have ceased

distribution of bending momentsobtained if the dead load wasapplied to the structure in itsfinal form

Figure B1.1: Redistribution of bending moments associated with creep and shrinkage where a load is

applied to a structure which is subsequently changed in form

props atends only precast concrete

part beam

stirrups not fully anchoreduntil cast-in-place concretehardens

place stirrup anchors in the bottomof the beam where they will notinterfere with placing of the top bars

Section

Figure B1.2: Uncertain anchorage of stirrups providing shear strength of a precast beam during place-

ment of cast-in-place concrete


Tests on midspan joints of the System 2 precast con-crete beam-column configuration have been conductedat Works Central Laboratories [B2.1] and the Univer-sity of Canterbury [B2.2, B2.3].

The Central Laboratories’ tests joint details and test rigare shown in Figure B2.1. The test specimens were H-shaped. During cyclic loading there was no apparentdegradation in the three cast-in-place midspan joints.In these tests, however, the test frame applied lateralrestraint to beam ends and hence restrained beamgrowth caused by plastic hinging at the beam ends. Theresulting axial compression in the beams would havecaused a more favourable performance of the midspanjoints than had there been no axial load.

The University of Canterbury tests involved four H-shaped test specimens that were subjected to cyclichorizontal loading at the column tops with the columnbases pinned (see Figure B2.2(a)). The base support ofone column was also on a sliding bearing, thus avoid-ing as far as possible the presence of axial forces in thebeam due to beam growth during testing. Figure B2.2(b)shows the details of the four test specimens.

Unit 1 had a midspan connection with hooked lapswhich performed extremely well during testing. Athigh ductility factors there was only very minor crack-ing, namely one diagonal tension crack and some veryfine shrinkage cracks. Figure B2.2(c) shows the con-dition of Unit 1 after completion of the test. Theundamaged midspan region contrasts with the plastichinge damage. It is to be noted that once diagonaltension cracking commences in the beam, the “tensionshift” effect means that longitudinal reinforcement atthe midspan connection will have substantial tensilestresses, although the bending moment diagram forthis test loading has zero bending moment at midspan.The maximum stress measured in the longitudinalbeam steel at midspan in Unit 1 was 42% of the yieldstrength. Unit 2 incorporated the double-hooked “dropin” splice bars shown in Figure B2.2(b). Once again,the performance of the connection during testing wasexcellent. Unit 3 had a midspan connection with con-ventional straight bar splices and performed extremelywell during testing.

These tests on Units 1, 2 and 3 demonstrated that alength of beam of at least one effective beam depth d ateach end adjacent to the column face is sufficient to

achieve adequate plastic hinge behaviour as required inductile frames. Therefore, the provision in the 1995concrete design standard that allows splices to com-mence at a distance d from the column face is justified.This design provision gives designers considerableflexibility when detailing midspan connections.

The performance of Unit 4 was less ductile than that ofUnits 1, 2 and 3. For Unit 4 there was a drop in lateralload carrying capacity at a drift of 1.5% when thedisplacement ductility factor was 4. It was observedthat crushing of the concrete had occurred at the insideof the bends of the diagonal bars of Unit 4 due to radialpressure there from the diagonal bars in tension and thetie force from the diagonal bars in compression. Whenthis crushing occurred in the test it meant that the trussmechanism of Fig. B2.3, which was used in the designof the beam, could not develop effectively. It is evidentthat caution is required when designing the detailshown for Unit 4 [B2.3].

An improved method for designing Unit 4 would be toprovide transverse bars at the inside of the bends of thediagonal bars to improve the bearing strength there.Also, ties are required around the ends of the barsterminated at the ends of the strong region to permitproper transfer of forces. Reference B2.3 gives arecommended design procedure.

ReferencesB2.1 Beattie, G J. “Recent testing of precast

structural components at central laborato-ries”, Proceedings of the New ZealandConcrete Society Conference, Wairakei,October 1989, pp 111-118.

B2.2 Restrepo, J I, Park, R and Buchanan, A H.“Tests on connections of earthquake resistingprecast reinforced concrete perimeter framesof buildings”, Journal of the Prestressed/Precast Concrete Institute, Vol. 40, No. 4,July-August 1995, pp 44-61.

B2.3 Restrepo, J I, Park, R and Buchanan, A H.“Design of connections of earthquake resistingprecast reinforced concrete perimeter framesof buildings”, Journal of the Prestressed/Precast Concrete Institute, Vol. 40, No. 5,September-October 1995, pp 68-77.

Appendix B2: Experimental Tests onPrecast Concrete Frame Midspan Joint

Connections for System 2


(b) Details of midspan test specimens

D5 D6D2 D4

D3D1

3656

Test Beam

3656 - Beam 11828 - Beam 22285 - Beam 3

Reaction frame

Tie rods

Load cellRam

Pivot pin restrained against East West movementNorth South load measured with reaction load cell

North

Fixed pivot pin

D4 - Displacement Gauge 4

“Columns” rolledon load skatespositioned approx1m from loadingframe

Loading frame straingauged to determinelink force

NOTE: Ram and loadcellmoved to dotted positionfor reversed loading

(a) Loading rig

A

A

25 25

800 11781178

R6 tie sets at 75 crs.

8mm filletweld

Face of “column” face of “column”

3 D16T & B

2 D16T & B 2 DH16

8 D20, 200 long

400

Precastwith “column” Cast-in-place Precast

with “column”

Elevation Beam Unit 1

260

400

5 D16

5 D16

R6 tie

Cover to mainsteel - 35 mm

A-A

B-B

260

450 R6 tie

4 D16 U bars& D16 main bars

4 D16 U bars& D16 main bars

Cover to mainsteel - 35 mm

Face of “column”

220 220

75 75

R6 tie sets at 75 crs.

1328 cast-in-place

B

450

4 D16 U bars

8 D20 200 long

4 D16


260

375

2 D

20

2 D

20

3 D162 D12

2 D123 D16

R6 tie Cover to mainsteel - 35 mm

C-C240

773 773

370 C

375

R6 tie sets at 75 crs. R6 tie sets at 75 crs.

Precast with“column”

Precast with“column”

Cast-in-place

Face of “column”

3 D16 T & B

2 D12 T & B

2 D24,200 long

2 D20


Figure B2.1: Details of the Works Central Laboratories loading rig and midspan

joint test specimens [B2.1]


1905 2667 190519

2011

20

800

Precastunit

Cast-in-place

Slidingpin

32 mm brackets

2 150 UC 23

102 mm pin

Load cell

Ram (328 kN capacity)

51 mm pin

(a) Loading rig

(b) Details of midspan test specimens

2100

700

836

5 110

35 mm cover

4 D16, butt welded

4 D283 D28, 1680 long

5-R10 stirrups

at 110 = 4404-R10 stirrups

at 110 = 330

Cast-in-place

Construction joint

3 D28, 1680 long

Additional stirrup

Unit 1 - Overlapping 180˚ hooks

2100

700

780

7-R10 stirrupsat 102 = 612

4-R12 stirrupsat 102 = 306

30Cast-in-place

3 D28

3 D28

Construction joint

2 D28

D28 rods

Unit 2 - Double 90˚ hooked drop in bars

700

Construction jointInsituwelding

6 D20

2100

Cast-in-place

Unit 3 - non-contact straight lap splices

2100

700

645

4 4R10 stirrupsat 127 = 381

93 61

195 4 2R10 stirrupsat 200 = 600

4 D24

4 D28

Construction joint

12 M22 HFGB at 56

2 D28

40 mm cover

2 R16, 700 long

2 R16

2 10x30 m.s. straps3 R6 stirrups at 80

4 D24, 100 mm fillet welded

Unit 4 - Strong end regions and diagonal barswelded to bolted plates

Figure B2.2: Details of the University of Canterbury loading rig and midspan

joint test specimens [B2.2]


(c) Unit 1 at end of test

(d) Beam of unit 4 at end of test

Figure B2.2 (continued): Details of the University of Canterbury loading rig and

midspan joint test specimens [B2.2]

M˚

M˚

V˚V˚ V˚2

V˚2

Cd Td

27.4˚

Steel strap Crushing of concrete due tobearing occurred here

Crushing of concrete due tobearing occurred here

Steel strap

Figure B2.3: Truss model assumed for design of Unit 4 [B2.3]


Test of ComponentsTests were carried out at Works Central Laboratories[B3.1, B3.2, B3.3] and at the University of Canterbury[B3.4] to investigate the behaviour of grouted columnbars. The purposes and the conclusions of these testswere as follows.

(i) To measure the bond strength of the main columnsteel within grouted sleeves, and to compare theresults with the bond strength of bars in cast-in-place construction.

The test results indicated that grouted bars had aslightly improved pull-out strength when com-pared with bars in cast-in-place construction. Theimproved performance was attributed to the strengthof the grout being greater than that of the concrete.Failure mechanism did not involve the bond be-tween the bars and grout, but occurred due tobreakdown of bond between the precast concreteand ducting. For both the grouted and cast-in-placebars, very high bond stresses were developed, withsome bars reaching yield strength even though thedevelopment length was well below recommendedcode values.

(ii) To assess the effectiveness of column ties passingaround grout sleeves.

Monotonic and cyclic pull-out tests were performedon R12 ties which passed around grouted ductscontaining longitudinal bars and around longitudi-nal bars in cast-in-place construction. In all casesthe R12 ties yielded. There was little difference inthe stiffnesses measured in cyclic and monotonicloading tests. The R12 tie was slightly stiffer whenbent around bars in cast-in-place construction thanwhen bent around grouted ducts containing bars.This detail, therefore, involving the use of groutedcorrugated metal tubing can be used with confi-dence. In these tests the compressive strength of thegrout was 51 MPa and the concrete, 35 MPa. Untilfurther laboratory test results are available, as isrecommended in Chapter 6, the grout should be atleast 10 MPa stronger than the specified compre-hensive strength of the precast concrete.

(iii)To inspect the completeness of the grout fill ofsleeves.

The tests involved pumping grout into the horizon-tal joint between the beam soffit and column top,and up through the vertical ducts in the beams.Cementitious grouts with an expanding agent wereused. Inspection of the grout, after saw cuts hadbeen made, confirmed that grout had completelyfilled the horizontal joints and vertical ducting. Anumber of recommendations were made regardingthe grout design and grouting operation which aresummarized in Chapter 6.

(iv)To evaluate the use of a performance test unit,envisaged as a specimen which is clamped on to thestructure during the grouting operation.

Concern has been expressed about the difficulty ofassessing the quality of grouting after the operationhas been completed. This test indicated that a groutspecimen which is clamped on to the structure andlater removed for strength testing, was not practical.This means that greater emphasis must be placed oninspection during the grouting procedure.

Test of Beam-column JointsFull-scale tests on two System 2 moment-resistingjoints were performed at Works Central Laboratories[B3.2, B3.3] and at the University of Canterbury [B3.4].As is typical for a System 2 joint, longitudinal columnreinforcing bars passed up through corrugated metaltubes located in the precast beam. A proprietarycementitious grout was used. The grouting operationis shown in progress in Figure B3.1. The beam-columnjoints were subjected to cyclic loading to investigatetheir stiffness and strength degradation. Test resultsshowed that behaviour of these joints was similar tothat of a conventional cast-in-place beam-column joint.

There was no indication of bond failure between thecolumn bars and the precast unit. The overall jointperformance was satisfactory up to beam deflectionlevels of displacement ductility factor 10 for the WorksCentral Laboratories test, and to at least 6 for theUniversity of Canterbury test, at which stage the dam-age at the beam plastic hinge was sufficient to cause areduction in strength. These results indicate that Sys-tem 2 provides a very satisfactory beam-column jointproviding that the design and workmanship are ad-equate.

Appendix B3: Laboratory Tests on thePerformance of Grouted Connections

for System 2


Figure B3.1: Grout being pumped into the ducts

References

B3.1 Thurston, S J. Precast Grouted Beam-ColumnJoint, Report S-86/9, Ministry of Works andDevelopment Central Laboratories, LowerHutt, 1986, 49 pp.

B3.2 Beattie, G J. “Recent testing of precaststructural components at Central Laborato-ries”, Proceedings of the New ZealandConcrete Society Conference, Wairakei,October 1989, pp. 111-118.

B3.3 Stevenson, R B and Beattie, G J. Cyclic LoadTesting of a Beam-Column CruciformIncorporating a Precast Joint Zone andColumn Bars Grouted in Drossback Ducts,Report 89-B5204/2, Central Laboratories,Works and Development Service Corporation(NZ) Ltd, Lower Hutt, 56 pp.

B3.4 Restrepo, J I, Park, R and Buchanan, A H.“Tests on connections of earthquake resistingprecast reinforced concrete perimeter framesof buildings”, Journal of the Prestressed/Precast Concrete Institute, Vol. 40, No. 4,July-August 1995, pp 44-61.

Appendix A1 • 125

Appendix C1: Assessing the Influenceof Thermal Gradients, Creep andShrinkage Strains in Composite

Concrete MembersThe sun acting on a concrete surface, in low windconditions, can generate a significant thermal gradientin the member. Measurements made on bridges indi-cate that temperatures on deck surfaces may be ex-pected to rise by 25˚ C to 30˚ C above ambient one ortwo times a year. Similar conditions occur on concreteroofs and exposed suspended slabs. The deflectionsand support rotations that develop as results of theseactions have in the past caused appreciable damage(see Section 2.3.4). Thermal gradients used for bridgedesign may be used to assess the appropriate tempera-ture profiles for roofs and other elements exposed tothe sun, see reference C1. A method of assessing thestructural actions due to thermal gradients is outlinedlater in this appendix.

Creep, shrinkage in precast prestressed concrete mem-bers can result in significant deflections in certainsituations. This can occur in two ways:

• If the unit is made up of different thicknesses, thethinner portions generally shrink more than thickerelements causing distortion to occur.

• Reinforcement, normal or prestressed, restrainsshrinkage movements. If this is asymmetrical inthe section, the restraint leads to curvature.

When concrete topping is placed on precast unitsadditional deformation generally occurs due to differ-ential shrinkage between the precast and cast-in-placeconcrete. In addition, some redistribution of dead loadand prestress can occur between the precast memberand the composite section with creep of the concrete.

Where long spans are used, or the deflection of thefloor is important, a number of different methods ofanalysis are available to predict the deflected shape.The simplest of these, which is the modified modulusmethod [C.2 and C.3], is described in the later part ofthis appendix. A number of other different approachesare described in references C.4 and C.5.

With any method of analysis the concrete propertieshave to be known. Tests on New Zealand concrete[C.5] have indicated that the creep values can bepredicted with reasonable accuracy from the CEB-FIPrecommendations [C.6], but the shrinkage values in

this document are on the low side. The free shrinkagestrain and creep coefficient for a typical 65 mm thick 25MPa concrete topping are of the order of 650 x 10-6 and3.5 respectively. For typical precast concrete unitswith a maturity of about 3.5 weeks the free shrinkagestrain and creep coefficient are reduced to about halfthe values for the cast-in-place topping concrete. Precastunits such as “Dycore” (hollowcore), which are madefrom low water content concrete, have a free shrinkageof the order of 350 x 10-6 and a creep coefficient ofabout 1.8 at the end of its initial curing period (twodays). These reduce respectively to about 100 x 10-6

and 1.2 after a further period of three weeks. It shouldbe noted that the basic creep and shrinkage valuescannot be predicted accurately without testing andappreciable variation in these values may be expectedwith different aggregate types, cement contents andcuring conditions. The thermal coefficient of concreteis also subject to variation with aggregate type and mixconstituents, but a typical value is 12 x 10-6 per C˚.

AnalysisThe methods of analysis for thermal, creep and shrink-age induced actions in concrete members all followvery similar steps, as outlined below.

1 Find the elastic (or equivalent elastic) transformedsection properties of the members.

• For thermal gradients and other situations wherethe induced stresses are not sustained for longperiods of time, section properties based onelastic transformed sections are found. Allow-ance should be made for differences in themodulus of elasticity, Ec, of insitu (topping) andprecast concretes. Transformed propertiesshould generally include both the passive andprestressed reinforcement.

• Where actions arise from shrinkage of the con-crete allowance must be made for the relaxationin stresses due to creep. For shrinkage, whichdevelops gradually over a long period of time,the effective modulus of the concrete(s) shouldbe taken as Ec/(1+0.6φ), where φ is the creepfactor for the concrete and Ec is the initial elasticmodulus.


• For these cases where loads are sustained forlong periods of time, such as dead load orprestress, transformed properties are determinedusing an effective elastic modulus of theconcrete(s) of Ec/(1+φ).

2 Buttresses are assumed to hold the structure so thatno strain can develop in the member. The stressesin the concrete are found and the forces that thebuttresses apply to the members are determined.There is no deflection or deformation of the struc-ture associated with this step.

• For the differential temperature case the stressesin the concrete in the buttress held condition areequal to (T x α x Ec), where T is the temperaturerise of the fibre being considered and α is thecoefficient of thermal expansion of the con-crete.

• For the differential shrinkage case the buttressheld stress in the concrete are equal to the valueSd Ec/(1 +0.6φ), where Sd is the differentialshrinkage of the concrete assuming this valuewas free to develop.

3 Forces of equal but opposite sign to the buttressesforces found in step 2 are applied to the structure tocancel out the buttress forces. The effective elasticmodulii defined in step 1 must be used. Thedeflection is associated with this step.

4 The resultant stresses and deflections are now foundby adding the stresses and deflections found insteps 2 and 3.

If the form of a structure is changed after some load isapplied redistribution occurs as the concrete creeps.

This situation arises, for example, when insitu concreteis added to a precast unit. The dead load of the insituconcrete and the self weight of the unit are initiallyresisted by the precast unit alone. However, as creepdevelops some of this load is redistributed to thecomposite section. A method of assessing the resultantdeflection using the modified effective modulus ap-proach is given in references C.2 and C.3.

References

C.1 Transit New Zealand. Bridge Manual: Designand Evaluation, Wellington, New Zealand,1991 draft, p 131.

C.2 Sritharan, S and Fenwick, R C. “The influenceof creep and shrinkage on the behaviour ofprestressed concrete beams”, Proceedings ofNZ Concrete Society Technical Conference,1989, pp 17-25.

C.3 Sritharan, S and Fenwick, R C. “Creep andshrinkage effects in prestressed beams”,Magazine of Concrete Research, Vol. 74, No.170, March 1995, pp 45-55.

C.4 American Concrete Institute. Designing forCreep and Shrinkage in Concrete Structures,ACI Special Publication SP76, AmericanConcrete Institute, 1982, pp 484.

C.5 Bryant, A H , Wood, J A and Fenwick, R C.“Creep and shrinkage in concrete bridges”,RRU Bulletin 70, National Roads Board,1984, 106pp.

C.6 Comite Euro-international du Beton. CEB-FIPModel Code 1990, Thomas Telford, London,1993, 437pp.

Appendix A1 • 127

Appendix C2: Method for Locating thePosition of Creep and Shrinkage

Cracks and Assessing Crack Widths

In a precast concrete floor with cast-in-place topping itis possible to estimate over what length of floor theformation of a crack relieves volume change strainsand from this information the maximum possible crackwidth can be assessed. In the example shown in FigureC1(a) the crack causes the stresses in the shaded area tobe reduced. The maximum crack width c, at a pointsuch as C, is approximately equal to the volume changestrain x L/2, where L is the length over which thestresses are reduced. The formation of the wide crackat B could have been avoided if a concentrated band ofreinforcement had been run from the wall DB into thetopping concrete in the slab, as shown in Figure C1(b).With this steel, provided a sufficient quantity is used,the crack width is controlled at B, and the length over

BD

A

30˚

30˚

precast concrete floor unitswith topping

concrete block wallrestrains slab

stresses reduced by crackin shaded area

CL

Figure C1a: Examples of crack locations in slabs - crack in a floor slab partially

restrained by block walls

which the stresses are relieved when the crack formsfrom A to B is greatly reduced. As illustrated, furthercracks may now be expected to open up and absorbsome of the shrinkage and thermal strains. The 30˚angle dividing the restrained and unrestrained floorregions due to the presence of a wall, or some otherrelatively rigid element, is a simple working approxi-mation that is based on the results of a series of finiteelement analyses [C.5].

ReferenceC.5 Bryant, A H , Wood, J A and Fenwick, R C.

“Creep and shrinkage in concrete bridges”,RRU Bulletin 70, National Roads Board,1984, 106pp.


BD

A

30˚

L

reinforcing from wall totopping concrete restrainsthe slab at B

stresses reduced by crackin shaded area

Figure C1b: Examples of crack locations in slabs — an additional band of

reinforced cement helps to control the crack widths

Appendix A1 • 129

Appendix D1: Criteria for the Selectionof Fixings

Table D1: Checklist for selection of fixings

QUESTION ACTIVITY DETAILS TO BE CONSIDERED

What loads are to be carried bythe fixing assembly?

Nature of loading has beenascertained previously by thedesigner

Dead Sustained Tension

Live Intermittent Compression

Temporary Cyclic Shear

(Construction) Dynamic Torsion

What kind of fixings are availablefor this application?

Description of fixing types Cast-in fixings

Expanding fixings

Bonded fixings

What are their performancecharacteristics?

Test data Short-term tensile loading

Sustained axial loading

Cyclic loadings, shear loading

Load/deformation characteristics, e.g. ductility

Elevated and low temperatures, resin capsule fixings

Behaviour and vulnerability under fire conditions

Are these all the factors thatinfluence performance?

Limitations of use of particulartypes of fixing related to basematerial characteristics

Nature of base material

Bonded fixings

Sensitivity to tolerance in fabrication, and distortion.

What about corrosion? Durability Types of corrosion

Corrosion prevention

What has to be considered in thestructural design of a fixingassembly?

Design criteria Principles

Safety margins

Spacing of fixings

Embedment depth in base material

How does the designer conveyhis requirement to the user?

Specification and drawings Who is responsible for doing what?

Consult with supplier and user

Specification

Anything else? Activities not covered above Consider the advantage or disadvantages of casting-in (inconcrete) or using a fixing in a drilled hole

Is the fixture permanently attached or will it be removable? This may influence the detail of the connection (e.g.accessibility)

Consider method of attachment direct by bolting or indirectby use of bracket or corbels which themselves may besecured with a fixing

Allow for possible differential movement between the basematerial and the fixture

Appearance: hidden or seen. Staining of fixture or basematerial

Total cost: materials plus labour

Appendix A1 • 131

Appendix D2: Embedments Subject toCorrosion

The following clauses indicate areas of concern thatshould be addressed by designers. The clauses are fromPW 81/10/1:1985, Guidelines for the Seismic Designof Public Buildings, Civil Engineering DirectoratePublication, Ministry of Works and Development.This publication is no longer operational as a standardciting document, but the technical content providessound engineering advice in this area. Works andDevelopment Services Corporation (NZ) Ltd’s per-mission to publish these clauses is acknowledged.

12.4.6.2 Fixings exposed to weather orcorrosive conditions

Fixings exposed to weather or corrosive condi-tions, shall be of stainless steel suitable for theparticular application except that where theyare visible and can be inspected and main-tained they may be galvanised and painted mildsteel.

In the case of stainless steel, care shall be takento avoid creating narrow crevices which mayentrap water. If present these are to be filledwith a chromated mastic (see Clause 12.4.7.2).

Comment: Metal fixings exposed to weathercreate corrosion and maintenance problems.Whenever possible the building facade shouldbe designed in such a way that such fixings arenot required. Choice between galvanised mildsteel and stainless steel should take account oflikely maintenance costs including scaffolding,etc., required to gain access to them. Galva-nised steel exposed to weather has a limited lifeand should always be painted. Repainting inter-vals will depend on exposure conditions.

12.4.6.3 Dissimilar metals

Corrosion may result through the galvanic ac-tion of dissimilar metals in contact. Stainlesssteel and mild steel should not be welded to-gether.

Comment: Welding together of mild and stain-less steels is technically feasible but can intro-duce embrittlement and corrosion problems.

12.4.6.4 Stainless steel types

The following stainless steel types shall beused:

AISI 304 (304 L or 321 if to be welded): Generalpurposes.

AISI 316 (361 L if to be welded): For corrosiveconditions or marine atmosphere salt laden air.

Comment: Special applications may call forother types and expert advice should be ob-tained in these cases. Stainless Steel Alloys Ltd,Box 31193, Lower Hutt, publish useful refer-ences.

12.4.7.2 Stainless steel and bolts

The threaded portion of stainless steel boltsshall be coated with an approved chromatedmastic or Dulux Stag A red jointing paste toprevent corrosion due to oxygen exclusion.Nuts are not available in type 316 alloy, andheads of type 316 bolts shall therefore be ex-posed where corrosive conditions exist.

Comment: Chromates are among the mosteffective corrosion inhibitors, but if used inadequate concentration they then act as a corro-sion accelerant. The manufacturer’s instruc-tions should be followed.

12.4.8 Admixtures containing chlorides

Admixtures containing chloride ions shall notbe used in grouts, mortars or concrete in con-tact with fixings.

Comment: Admixtures containing chloride ionsmay have a corrosive effect on fixings and otherembedded metals. Also, if dissimilar metals arepresent, chlorides may accelerate any galvanicaction.

12.4.9 Cold worked steel

Cold worked steel which is to be galvanised orhas the cold worked area subsequently heatedby welding, shall be heat treated unless testevidence shows that embrittlement will not oc-cur.

Comment: Galvanising of cold worked steelcan result in serious embrittlement in the coldworked areas. Refer to ASTM A123 and A143for guidance. There is also a seriousembrittlement risk when steel is cold workedand subsequently heated by adjacent welding.


Strain ageing embrittlement is accelerated byhigher temperatures such as those encounteredduring galvanising and welding, and results in

loss of ductility and even in brittle fracture atbends during the ordinary process of handlingand erection.

Appendix E1: Example Calculation for Embedded Steel Connectors • 133

Appendix E1: Example Calculation forEmbedded Steel Connectors

NOTE: Refer to Chapter 7 “Embedded Steel Connec-tors” for background theory to these examples.

E1.1 Example 1 — Weld Platewith StudsDesign a weld plate with studs to be cast into a 400 mmsquare reinforced concrete column. The concretecompressive strength of the column is ′ =f MPac 30 .A 360 UB is to be fixed to the weld plate. The columnhoops are widely spaced and cannot be relied on fortransferring the forces in the connection. The follow-ing design actions have been derived from the struc-tural analysis:

V* = 100 kN

M* = 24 kNm

N* = 20 kN (tension)

Figure E1.1 shows an elevation of the weld plateconnection detail. Tension and compression forces arefound in equilibrium. The shear force, Vb, transferredby the bearing of the plate on to the concrete will beignored. Hence, the shear force needs to be transferredby friction at the compressive block of stresses, Vf , and

by direct shear through the top and bottom studs, Vdtand Vdb, respectively.

1 Design of the Steel Embedment

1.1 Design for Shear

Ensure that the dependable shear strength of the con-nector is equal or greater than the design shear force

φV Vn ≥ * (E1.1)

where the strength reduction factor for a steel connec-tor is φ = 0.85. Try using two rows of 2-12.7 mmdiameter Nelson studs (see Figure E1.1).

The nominal shear strength, Vn, is the sum of the shearforce transferred by friction, Vf, plus the force trans-ferred by direct shear through the top and bottom studs,Vdt and Vdb, respectively.

V V V Vn f dt db= + + (E1.2)

The shear force transferred by friction is computedfrom a shear friction coefficient μf = 0.7 recommendedby the New Zealand Concrete Structures Standard,NZS 3101:1995 [E1]. Thus

V x kNf = =0 7 58 6 41 0. . .

he

Vdt

VfVdb

Vb

T = 78.6

C = 58.6

350

100 24

20

100 200 100

Column

Weldplate

Stud

gap

gap

Figure E1.1: Elevation of weld plate connection and design actions


The force transferred by direct shear through thebottom studs is computed ignoring the presence ofaxial force. The ultimate shear strength under pureshear stress conditions is equal to 0.6 fut. Therefore, theshear force transferred by these studs is

V f Adb ut sb= 0 6.

Where Asb is the cross-section area of the bottom studs.

Assume two bottom studs. From Figure E1.2 the shearforce carried by the bottom studs is

V x x x x kNdb = =−0 6 415 2 127 10 63 23. .

The force required to be carried by direct shear throughthe top studs is found by substituting Equation E1.2 into Equation E1.1 and rearranging for Vdt

VV

V V

kN

dt f db= − +( )

= − +( ) =

*

.. . .

φ

100

0 8541 0 63 2 13 4

The required area of studs to transfer Vdt is

A13.4 x 100.6 x 415

54 mmstv

32= =

1.2 Design for Flexure

The nominal tensile strength of the studs, Tn, is givenby

T A fn st ut=

where Ast is the cross-section area of the studs and futis the ultimate tensile strength of the stud when a well-defined yield plateau is not apparent.

Now ensure that the dependable tensile strength of theconnector is equal to or greater than the design action

φT Tn ≥ (E1.3)

The area of top studs required for transferring tensiononly, At, is from Equation E1.3

AT

f

xmmst

t

ut

= = =* .

..

φ1 78 6 10

0 85

1

415222 8

32

1.3 Combined Flexure and Shear

The top studs are required for transferring tensionforce, resulting from flexure and axial tension, andshear force simultaneously.

The area of studs, Ast, required for transferring thecombined action is

A A A

54 222.8 229 mm

st stv 2

stt 2

2 2 2

12

12

= ( ) + ( )[ ]= +[ ] =

The area provided by 2-12.7 mm diameter studs is 2 x127 = 254 mm2, which is greater than 225 mm2. Hencethe design is satisfactory.

2 Embedment LengthUse capacity design to ensure that the dependablestrength of the concrete resisting mechanism, φTc, isequal to or greater than the probable strength of thestuds, λTs

φ λT Tc s≥ (E1.4)

where

φ = strength reduction factor for concrete inflexure and axial force

Tc = 5 percentile lower characteristic strengthof the concrete resisting mechanism

λ = ratio between probable and 5 percentilelower characteristic tensile strength forstuds (no yield plateau)

Ts = tension force in the studs, based on the 5percentile lower characteristic tensilestrength of a stud.

The Ψ-method given in references E2 and E3 recom-mend for Tc

T h f Nc R e c= ′10 7 3 2. ( )/ξ

ddh

12.7 25 127

15.9 32 198

19.1 38 285

d (mm) dh (mm) As (mm2)

Figure E1.2: Properties of Nelson studs

Appendix E1: Example Calculation for Embedded Steel Connectors • 135

where ξR is a factor that considers edge, concretecracking and group effects, and he is the embedmentlength at the stud, measured as shown in Figure E1.1.

In lieu of data supplied by the manufacturer, useλ = 1.21 as recommended in Reference E3. Thus

0 85 10 7 30 1 21 127 4153 2. . ./x h x xR eξ ≥

h mmeR

≥( )

1182 3

ξ/ (E1.5)

Evaluation of factor ξR [E3]

ξR CR CX SX CY SY= Ψ Ψ Ψ Ψ Ψ (E1.6)

For this example and with reference to Figures E1.1and E1.3

Ψ Ψ ΨCR SY CX= = = 1

Group effect factor ΨSX

nx = 2scr = 3hes = 200 mm

assume he = 260 mm and then check. Hence scr = 780mm.

ΨSX x= + −{ } = ≤1 2 1 200 780 2 0 63 1( ) / / .

Edge effect factor, Ψcy

cx = 100 mm

ΨCY x= +{ } = ≤0 3 0 7 100 390 0 48 1. ( . / ) .

Substituting ΨCR, Ψsy, Ψcx, Ψsx and Ψcy in EquationE1.6 results for ξR

ξR x x x x= =1 1 0 63 0 48 1 0 30. . .

Now substituting ξR in Equation E1.5

h mme ≥ =118 0 30 2612 3/( . ) /

Note that the value initially assumed for he is very closeto the value obtained. Consequently, there is no need tocarry a second iteration.

E1.2 Example 2 — Cast-in RHSStudDesign the cast-in RHS stud for the force shown inFigure E2.1.

102 76 6 3 225

35

x x RHS f MPa

f MPa

y

c

. =

′ =

A cast-in RHS stud is to support a load of

V kN* = 100 (E1.7)

Use the Steel Structures Standard [E17]. Check RHSfor shear,

V A F

x x x x

kN

n w yw=

=

=

0 55

0 55 6 3 102 2 225

159

.

. .

φV kN V satisfactoryn = >135 *,

Check RHS for flexure

M S F

x

kNm

n y=

=

=

64 225

14 4.

M x kNm

M kNm kNm satisfactoryn

* .

. ,

= =

= >

100 0 1 10

12 2 10φ

The dependable bearing stress fb on the concrete is

0 85 19 3. .φ ′ =f MPac

where φ = 0.65 (NZS 3101, Clause 3.4.2.2(f)) and′ =f MPac 35

500 100

102 x 76 x 6.3 RHS

100 kN

Figure E2.1: Elevation of cast-in RHS stud


The maximum bearing stress that the RHS can resist islimited by the flexural strength of the RHS wall span-ning between the webs.

At collapse

Mwl

p =2

16 (hogging and sagging plastic hinges)

Dependable

M S F x x x

Nmm mm

y= =

=

2251

41 6 3

2233

2.

/

Maximum stress = 2233 x 16/702 = 7.3 MPa

Hence allow 19.3 MPa under web and 7.3 MPa acrossflange of RHS, giving

12 6 19 3 76 12 6 7 3 706. . . . /x x N mm+ −( ) =

The depth of stress blocks is 85% of the distance to thepoint of zero strain, and stress block lengths are a andb as shown in Figure E2.2)

V 706a 706b

M 706a x 1.176b 0.676a 706b2

For V 100 kN

2

*

= − +

= +( ) −

=

a

125

b

268

706 N/mm

706 N/mmzero strain

stress blockstrain

0.176 (a + b)

Figure E2.2: Equivalent load distribution in the embedded cast-in RHS stud

a b mm

x b x

b b b

= −( )= −( )

+ −( )[ ] −

142 6

10 10 706 142 6

1 176 0 676 142 6 353

6

2

.

.

. . .

giving

b = 268 mm

a = 125 mm

l = 462 mm (minimum), hence 500 mmembedment is OK.

If, in addition, there was a tensile force in the RHS of20 kN, an end cap plate should be welded to the end ofthe stud and transverse reinforcement should be pro-portioned to cross the potential failure cone and carrythe tensile force.

ReferencesE1 Concrete Structures Standard, NZS 3101:1995, Stand-

ards New Zealand, Wellington, 1995, 256 pp.

E2 Fastenings to Concrete and Masonry Structures -State of the Art Report, Comité Euro-Interna-tional du Béton, Thomas Telford, 1994, 249 pp.

E3 Restrepo-Posada J I and Park, R. “Tensile capacityof steel connectors with short embedmentlengths in concrete”, Research Report 93-6,Department of Civil Engineering, University ofCanterbury, August 1993, 51 pp.

Diaphragms • 137

The objects of the society, one of the earliest specialistgroups of IPENZ, are to further those of the Interna-tional Association for Earthquake Engineering as ap-plicable to New Zealand, and to foster the advance-ment of the science and practice of earthquake engi-neering.

The Society has strong international affiliations, andhas run a number of major international conferences: in1965 the 3rd World Conference on Earthquake Engi-neering was held in New Zealand and the 12th WorldConference on Earthquake Engineering is to be held inAuckland, New Zealand, from 30 January to 4 Febru-ary 2000. The group has also been host to the SouthPacific Regional Conference, and the Pacific Confer-ences of 1987, 1991 and jointly with the AustralianSociety, in 1995.

A principal activity is the maintenance of a pool of

New Zealand Concrete Society

experts in diverse disciplines able to travel at shortnotice to the sites of damaging earthquakes in countriescomparable to New Zealand to learn first hand of theresponse of building structures and lifelines.

For further information please contact:

Andrew King (Chairman)BRANZPrivate Bag 50-908Poriruaphone 04-235-7604e-mail: [email protected]

Graham Quirke (Secretary)PO Box 48-046Upper Huttphone 04-528-4906e-mail: [email protected]

New Zealand Society for Earthquake Engineering

Details of OrganisationsInvolved

The New Zealand Concrete Society is a learned Soci-ety. The objectives of the Society are to “encourage agreater knowledge and understanding of all aspects ofstructural and architectural concrete and to supporttheir development and use where appropriate”.

Membership of the Society encourages closer contactwith many of the leading firms and individuals in theconcrete industry — designers, manufacturers, con-structors and material suppliers — to both give andtake freely of ideas and experiences on a wide range ofaspects of one of the most versatile and widely usedbuilding materials — concrete.

Additional membership benefits include entitlement to

the ‘New Zealand Concrete’ and ‘ƒib News’ publica-tions. Reduced prices are available to members onpublications, attendance at conference and most tech-nical seminars and participation in Concrete and Pre-stressed Concrete Awards.

For further information please contact:

New Zealand Concrete SocietyPO Box 12BeachlandsAucklandPhone: 09 536 5410, Fax: 09 536 5442E-mail: [email protected]


CAE, the Centre for Advanced Engineering, was es-tablished in May 1987, as a centre for the promotion ofinnovation and excellence in engineering and technol-ogy, to commemorate the centenary of the School ofEngineering at the University of Canterbury.

Vision Statement

To benefit New Zealand through promoting and en-couraging the application of advanced engineering andtechnology.

Objective

CAE aims to enhance engineering knowledge withinNew Zealand by technology transfer and the applica-tion of New Zealand and overseas research to engi-neering-related issues of national importance.

Key Activities

• CAE undertakes major projects that bring togetherselected groups of practising and research engi-neers and other experts from industry, researchorganisations, local and central government, terti-ary institutions, and the engineering profession.

• CAE also carries out smaller projects and organisesseminars, workshops and conferences as opportu-nities arise.

CAE projects aim to:

• Be of national importance with wide public appealand with tangible results.

• Facilitate technological co-operation amongst com-mercial and government organisations, tertiary in-stitutions and the engineering profession.

• Identify deficiencies in New Zealand’s technologi-cal capability and take action to promote the ad-dressing of these deficiencies.

• Undertake technology transfer rather than originalresearch.

Funding

CAE is an independent, non-profit organisation, fi-nanced mainly from the earnings of its trust fund. Thisfund, which currently stands at approximately $2.3million, consists of monies donated initially by 150corporate donors and 750 individual donors during the1987 Centennial Appeal, and more recently supple-mented by further donations during the 10th Anniver-

sary Appeal. Other income is derived from sponsor-ship for specific projects, book sales and seminars. TheUniversity of Canterbury continues to make a majorcontribution to CAE by providing accommodation andfinancial services.

Administration

CAE is controlled by a Board of Directors comprisingrepresentatives from industry and commerce (includ-ing government and consulting engineers), the Univer-sity of Canterbury and other tertiary educational insti-tutions. The present Chairman is Dr Francis Small ofWellington. The Trust Fund is currently administeredby the University of Canterbury under the direction offour trustees. CAE has two executive staff and threeother staff engaged on publications and secretarialduties. Executive Director John Blakeley has overallresponsibility for CAE activities and Projects DirectorJohn Lumsden co-ordinates CAE projects.

Principal Corporate Donors

Founders:

BHP New Zealand Steel Limited

Earthquake Commission

Electricity Corporation of New Zealand

Fisher & Paykel Industries Limited

Mainzeal Property and Construction Limited

McDonnell Dowell Constructors Limited

Opus International Consultants Limited

TransAlta New Zealand Limited

10th Anniversary Appeal:

Transpower New Zealand Limited

Contact:

CAEUniversity of CanterburyPrivate Bag 4800Christchurch, New Zealand

Street Address:39 Creyke RoadChristchurch 8004

Telephone: +64-3-364-2478Fax: +64-3-364-2069

e-mail: [email protected]://www.cae.canterbury.ac.nz

Executive Director: John P BlakeleyProjects Director: John L LumsdenPublications Editor: Charles A Hendtlass

Centre for Advanced Engineering

Index • 139

Index

Aadmixtures 131aggregate interlock 63analysis and design of diaphragms 62

design limit states 64flexibility 62strength of diaphragms 65

Bbeam column sub-assemblies 67beam elongation 16, 65, 111beam splices 82beam-column joint 26, 27, 33, 62

bearing surfaces 34cold joints 27core reinforcement 27hangers 34joint core hoop reinforcement 26

beam-column junction 104beams 60, 61, 62, 67

elongation of members 61plastic hinge zones 61

bearing pads 6bearing strips 11bond 65, 66bond failure 59

ductility 59bond strength 123

Ccantilever columns 35capacity design 21, 60, 64, 134cast-in-place jointcementitious grouts 123clearances see tolerancescold worked steel 131

galvanising 131column 60, 61, 68

plastic hinge rotations 61plastic hinges 61shear 61

column splice 80combined flexure and shear 134composite beam construction 31composite concrete members 125composite reinforced concrete moment-resisting

frames 34compression buckling 66concrete beam to column joint 78connecting precast elements using welded reinforcing

bars 84connection details 21connector types 91

cast-in steel sections 91concrete inserts 91post drilled fixings 91weld plates 91

construction joints 6corrosion 131

embedments 131coupled cantilever walls 41

coupling beams 41cover 95crack control 6crack widths, assessing 127cracking 63creep 117, 127creep and shrinkage 41creep and shrinkage cracks, locating 127creep and shrinkage strains 125creep and shrinkage values 125

creep coefficient 125free shrinkage 125free shrinkage strain 125thermal coefficient 125

cruciform-shaped precast concrete units 30plastic hinging 30splice column bars 30welded connection detail 30

Ddead load 117deep beams 68design approach, embedded steel connectors 93, 94detailing, steel embedments in concrete 97development lengths 37diagonal tension cracking 119diaphragm 3, 59-71

axial compression 68axial forces 67axial tension 68beam elongation 65bond strengths 70capacity design 62, 64, 67construction practices 68continuity 59creep 62critical interface 68deformation in plastic hinges 70design strength 59diagonal compression stresses 65effective flexural shear stiffness 63elongation 62, 64, 70elongation in plastic hinges 70flexibility 62flexural deformations 71flexural strength 62, 68, 70force distribution 63gravity loads 59horizontal ties 60in-plane shear strength 69interface 59, 68lateral forces 59load path 60local failure 62loss of support 61plastic behaviour 59plastic hinge zones 62plastic hinges 61, 71redistribution of actions 59reinforcement 60, 61restraint 68robustness 61


serviceability limit state 59, 64shear deformations 62shear failure 68shear forces 64, 68shear friction 68shear stresses 70shrinkage 62simple diaphragm 59simply supported 60splitting cracks 65stiffness 62, 64, 66strength 62, 65, 66strength increase 68strut and tie 65, 66, 68tension 61, 62thermal effects 62transfer diaphragm 59, 63ultimate limit state 59, 64untopped 68, 69, 71vertical accelerations 61volume changes 62, 69

diaphragm flexibility 62diaphragm forces 16diaphragm stiffness 65diaphragms, overseas practice 68differential shrinkage 125, 126differential temperature 126double columns 6ductile 21ductile frames 34ductile moment-resisting frames 35ductility 41

Eelongation of beams 67, 68

cracks 68embedded steel connectors 3, 91-100embedded steel connectors, worked examples 133

capacity design 134combined flexure and shear 134embedment length 134flexure and axial tension 134

embedment length 134embedments, subject to corrosion 131erection tolerances see tolerancesexplosions 60

Ffire 60fixings exposed to weather 131fixings, selection 91flexural rotation 67flexure and axial tension 134floor slabs 59

hollow core 59floor unit support and continuity 2floor unit support and continuity 5-18floor units 5

creep 7, 14, 17deflection and vibration control 17diaphragm requirements 15ductility 12fire resistance 16fire resistance rating 16flange-supported double tees 6flexural continuity 6horizontal shear 6

minimum bearing lengths 5moment fixity 5seating details 11seating lengths 7service load deflection 17shear strength 12, 14shrinkage 7, 14, 17simple support 5support details 5continuous support 6temperature changes 7vibration 17volume changes 7

foundation 51frame 70frame connections 2, 21-38frame connections, North America and Japan 35friction loads 93

Ggrade 300e reinforcement 84, 85grade 500e reinforcement 84, 85gravity loading 34grout 73-84

bond 73cement-based grouts 73, 74cement-based mortar 77correct usage 84durability 73epoxies and polyesters 73epoxy-based mortar 77high-wetting grouts 75most appropriate grout 73placement conditions 73rate of strength gain 73required final strength 73selection of grout 73shrinkage 73viscosity 73

grouted and welded bars 3grouted and welded connections 73grouted connection 73

bond 73cement-based grouts 73, 74cement-based mortar 77choice of duct size 78clearances 75construction aspects 83decanting 75duct diameters 79durability 73embedment/lap length 76epoxies and polyesters 73epoxy-based mortar 77high-wetting grouts 75horizontal starter bars 75horizontal voids or spaces 76horizontally grouted bars 76most appropriate grout 73non-destructive test method 84on-site pull-out tests 84outlet ports 80placement conditions 73post-grouting 75pre-grouting 75rate of strength gain 73required final strength 73

Index • 141

selection of grout 73shrinkage 73specific connections 78strength capacity of grouted bars 84tremie tube 75vertical starter bars 74, 76vertical voids 75viscosity 73workmanship 83

grouting situations 74grouting specific connections 78

Hhollow-core floor 15

shear keys 16shear ties 16tension forces 16

Iimpact of vehicles 60imposed deformations 91induced actions, thermal, creep and shrinkage 125industrial buildings 35inertia effects 91interaction of plastic hinge zones and diaphragms 67interface tolerances see tolerancesinternal restraint actions in diaphragms 62

volume changes 62inter-storey drift 63inter-story deflection 111

Jjoint core 22

potential plastic hinge region 23, 28shear friction 25shear keys 25vertical shear 25vertical sliding 23

joint types 85-87non-symmetric 85symmetric 85

jointed wall system 3, 46

Llap or splice regions 46lateral load resistance 51light-weight concrete 95limited ductility 111

Mmaximum possible crack width 127maximum tolerances see tolerancesmechanical connectors 88

cadweld 88swaged sleeves 88threaded bars 88

minimum tensile edge distance 95moment resisting frame structures 31, 61moment-resisting connections 117monolithic construction 35monolithic wall system 3, 41mortar seating pads 6

Pperformance of grouted connections for system 2 123

bond strength 123

cementitious grouts 123pinned joints 33plastic hinge 111

compression zone 111plastic hinge rotation 111reversing plastic hinge 111shear force 111

plastic hinge formation 111plastic hinge rotation 33plastic hinge zones 67, 96

flexural cracks 97post-drilled anchors 97

plastic hinging 119post-drilled fixings 96potential plastic hinges 60precast beam 21, 117

anchorage 23corrugated steel ducting 26development length 23

precast concrete construction 101precast concrete structural wall 56

creep 56horizontal joints 56jointed systems 56lap bar reinforcement 56monolithic wall 56precast panels 56seismic movements 56settlement 56shrinkage 56thermal movement 56

precast concrete structural wall construction, types 41precast concrete support details 5precast concrete unit connections, grouted and

welded bars 73-88precast diaphragm 65, 70

beams 65bond forces 65bond resistance 65bond stresses 65continuity bars 65creep 70delamination 65delamination failure 65elongation 65extruded dry concrete 65flexibility 70internal restraint actions 70internal restraint actions 70prestressed 65robustness 70shrinkage 70structural redundancy 70strut and tie 70thermal effects 70topping concrete 65transfer diaphragms 70

precast floor unit seating 7appropriate seating 9construction methodology 7seismic effects 7tolerances 7transverse load distribution 7volume changes and thermal effects 7

precast floors 61, 67precast walls 46, 60

cantilevered shear walls 51


capacity design 46ductility 51horizontal joints 46horizontal reinforcement 49horizontal reinforcing 46jointed construction 46jointed precast concrete wall systems 46lap length 46monolithic construction 46monolithic joint 51mortar joint 51plastic hinge zone 46shear reinforcement 46steel hair pins 51steel sleeve splices 46vertical joints 46vertical reinforcement 46

product tolerances see tolerancesprofile tolerances see tolerancesprogressive collapse 60proprietary inserts 96punching shear failures 63

Qquality assurance 83

milestone 83

Rrecommendations

connections between precast concrete units by grouted and welded bars 87diaphragms 70embedded steel connectors 98frame connections 35structural wall elements 56tolerances 104

reinforced concrete columnsbearing failure 28cast-in-place joint 26diagonal reinforcement 28gravity loading 26grouted column bars 28grout-filled ducts 28hooked lap 28potential plastic hinge region 28precast t units 30splice length 28straight bar lap 28straight bar splices 28

reinforced concrete frames 21strong column-weak beam 21

requirements of diaphragms 59robustness 60serviceability limit state 59ultimate limit state 59

ribbed units, support types see support types for ribbedunits

rigid diaphragm 63

Sseating details for flooring units 11

type 1 support 11type 2 support 12type 3 support 14

secondary beams 33secondary structural elements 91

seismic effects 8, 60elongation 9membrane action 9seating lengths 9steel corbels 8

seismic grade reinforcing bars 84, 85selection and types of grout 73selection of fittings, criteria 129shallow concrete inserts 96shear deformation 67shear failure 67shear load 95shear transfer 37shrinkage 117, 125, 127sliding bearings 6

linkage bars 6spalling, allowances for effects 115specified tolerances 79stainless steel types 131steel connectors, embedded 91steel embedments 91-98

anchor spacings 93axial deformations 93axial tension 91bolted connectors 94brittle fracture 97capacity design 94, 95cast-in-steel sections 91concrete inserts 91concrete strength 93cyclic loads 94deformation capacity 91deformation capacity 93ductility 93edge distances 93elongation 93embedment length 93, 94, 95flexibility 93headed steel inserts 96loads 92localised movements 91long-term performance 94mechanical interlock 91member elongation 91overstrength 94post drilled fixings 91pull-out failure 94, 97shear resistance 98shrinkage and creep 93siesmic loading 91spalling of concrete cover 97temperature effects 93temperature-induced structural deformations in fires 91tensile or yielding strength 94weld plates 91, 95

steel, cold worked 131galvanising 131

strain aged embrittlement 132strong beam-weak column 30

post-elastic deformations 30structural actions, on steel embedments 92structural integrity 60structural wall 41

capacity design principles 41dimensional limitation rules 44ductility 42

Index • 143

flexural strength 44inelastic deformations 41jointed 41monolithic precast structural wall 41, 43, 46reinforcement 44splices 45strength 42tilt-up wall construction 41wall elements 41

structural wall elements 3, 41-56structural wall elements, examples 49-55strut and tie 63, 65, 66, 70support types for precast concrete floor units 5support types for ribbed units 14system 1 type connectors 117

creep and shrinkage 117dead load 117moment-resisting connections 117precast beams 117

system 1, precast beam units between columns 21system 2 type connections 119

diagonal tension cracking 119performance of grouted connections 123plastic hinging 119“tension shift” effect 119

system 2, precast beam units through columns 26system 3, precast t or cruciform shaped units 30

Ttemperature profiles 125tensile stresses 65tensile yielding 111“tension shift” effect 119thermal gradients 125

thermal movements 6thermal strain 127tolerances 3, 101-109

beam-column junction 104clearances 104connection details 104erection tolerances 101, 103, 104for in-situ construction 101for precast components 101interface tolerance 104maximum tolerances 104permissable tolerances 104product tolerances 101, 104profile tolerances 101, 103

UU-beams 31

bond 31interface shear strength 31

Wwall 60, 70weathering details, structural wall elements 56weld specifications 88welded connection 73, 84-88

preheating 88tolerances 88weld specifications 88welding near bends in reinforcement 88welding of reinforcement 88

welded reinforcing bars, connections between precastelements 84

welded splices 84welding of reinforcing steel at connections 73


in Buildings

The First Edition of Guidelines for the Use of Structural Precast Concretein Buildings was published by the Centre for Advanced Engineering(CAE) in 1991 and was reprinted in 1992. The material that led to thiswidely-used book was produced by a Study Group of the New ZealandConcrete Society and the New Zealand Society for EarthquakeEngineering.Following release of the revised New Zealand Concrete StructuresDesign Standard NZS 3101 in 1995, it was decided to produce a newSecond Edition that would not only reflect the requirements of the newcode but would also incorporate appropriate developments of accepted"good practice" within the design and construction industry, and includesome of the relevant research that has been produced since the firstedition was published.The principal aspects of the uses of structural precast concrete coveredare:

• Precast beams (both shell and solid), precast columns and their jointing;

• Beam-column joints, especially if cast-in-place between precastelements;

• Support and continuity of floor slabs;• Jointing techniques and connectors, constructability and tolerances;• Diaphragm actions; and• Behaviour of certain precast concrete wall systems.

Generally, the emphasis is on buildings rather than civil engineeringstructures. Although there is a particular focus on the seismic aspectsof precast concrete construction, many sections refer to gravity loadsas well as volume changes such as creep, shrinkage and thermalactions.

This new edition is intended to assist in providing consistently safeand economical applications of structural precast concrete, and at thesame time allow innovation in design and construction to continue.

Guidelines for the Use of Structural Precast Concrete in ...

Documents

Transcript of Guidelines for the Use of Structural Precast Concrete in ...