causes, evaluation and repair of cracks in concrete structures

Causes, Evaluation and Repair of Reapproved 1998 Cracks in Concrete Structures Grant T.. The causes of cracks in concrete structures are summarized The proce-dures used to evaluate crac

Causes, Evaluation and Repair of

Reapproved 1998

Cracks in Concrete Structures

Grant T H&orsen*t Chairman

Randall W Poston Secretary

Wii Hansen

M Nadim Hassoun Tony C Iiu$

Edward G Nawy Harry M Palmbaum

Keith A Pashina

Andrew Scanlon$

Ernest K Schrader Wimal Suaris Lewis H Tuthill Zenon A Zielinski

* Contributing Author

t Member of Task Group which prepared these revisions

$ Principal Author

0 Chairman of Task Group which prepared these revisions

Note: Associate members Masayatsu Ohtsu, Robert L Yuan, and Consulting Member LeRoy Lutz contribute to the revision of this document.

The causes of cracks in concrete structures are summarized The

proce-dures used to evaluate cracking in concrete and the principal techniques for

the repair of cracks are presented The key methods of crack repair are

discussed and guidance is provided for their proper application

Keywords: autogenous healing; beams (supports); cement-aggregate reactions;

concrete construction; concrete pavements; concrete slabs; concretes;

consol-idation; corrosion; cracking (fracturing); drilling; drying shrinkage; epoxy resins;

evaluation; failure; grouting, heat of hydration; mass concrete; methacrylates; mix

proportioning; plastics, polymers and resins; precast concrete; prestressed

concrete; reinforced concrete; repairs;resurfacing; sealing settlement (structural);

shrinkage; specifications; structural design; tension; thermal expansion; volume

1.2-Cracking of plastic concrete

1.3-Cracking of hardened concrete

ACI Committee Reports, Guides, Standard Practices, and

Commentaries are intended for guidance in designing,

plan-ning, executing, or inspecting construction and in preparing

specifications References to these documents shall not be

made in the Project Documents If items found in these

documents are desired to be a part of the Project

Docu-ments, they should be phrased in mandatory language and

incorporated into the Project Documents.

Chapter 2-Evaluation of cracking, pg 224.1R-9

2.1-Introduction2.2-Determination of location and extent of concretecracking

2.3-Selection of repair procedures

Chapter 3-Methods of crack repair, pg 224.1R-13

3.1-Introduction3.2-Epoxy injection3.3-Routing and sealing3.4-Stitching

3.5-Additional reinforcement3.6-Drilling and plugging3.7-Gravity filling3.8-Grouting3.9-Drypacking3.10-Crack arrest3.11-Polymer impregnation3.12-Overlay and surface treatments3.13-Autogenous healing

AC1 224.1R-93 supersedes ACI 224.1R-90 and became effective September 1, 1993.

Copyright d 1993, American Concrete Institute.

All rights reserved including rights of reproduction and use in any form or by any means, including the making of copies by any photo process, or by any elec- tronic or mechanical devices, printed or written or oral, or recording for sound

or visual reproduction or for use in any knowledge or retrieval system or device, unless permission in writing is obtained from the copyright proprietors.

224.1R-l

Cracks in concrete have many causes They may affect

appearance only, or they may indicate significant

struc-tural distress or a lack of durability Cracks may

repre-sent the total extent of the damage, or they may point to

problems of greater magnitude Their significance

de-pends on the type of structure, as well as the nature of

the cracking For example, cracks that are acceptable for

buildings may not be acceptable in water-retaining

struc-tures

The proper repair of cracks depends on knowing the

causes and selecting the repair procedures that take these

causes into account; otherwise, the repair may only be

temporary Successful long-term repair procedures must

attack the causes of the cracks as well as the cracks

themselves

To aid the practitioner in pinpointing the best solution

to a cracking problem, this report discusses the causes,

evaluation procedures, and methods of repair of cracks

in concrete Chapter 1 presents a summary of the causes

of cracks and is designed to provide background for the

evaluation of cracks Chapter 2 describes evaluation

tech-niques and criteria Chapter 3 describes the methods of

crack repair and includes a discussion of a number of

techniques that are available Many situations will require

a combination of methods to fully correct the problem

Preface to the 1991 Revision

Following the initial publication of ACI 224.1R in

1985, the Committee processed two minor revisions One

revision, published as ACI 224.lR-89 simply updated the

format of recommended references A second minor

revi-sion contained minor technical revirevi-sions and editorial

corrections in the document, and added a new section to

Chapter 3, regarding the use of high-molecular-weight

methacrylates as sealer/healers

During 1990 a Committee 224 Task Croup reviewed

the document and recommended the revisions contained

herein Chapter 1 has been altered in only minor detail

The introduction to Chapter 2 has been revised

exten-sively, and additional minor revisions have been made to

the rest of the Chapter In Chapter 3, the section on

routing and sealing has been rewritten to include flexible

sealing and overbanding of cracks, and it is updated to

reflect current materials and construction practices

Sec-tion 3.2 on epoxy injection has been revised to be

some-what more general and reflect current practice The

for-mer section on high-molecular-weight methacrylates hasbeen moved to Section 3.7 and retitled “Gravity Filling.”This recognizes the point that “high-molecular-weightmethacrylate” is a material, and not a method Refer-ences are presented in Chapter 5; citations throughoutthe text have been revised to employ the author/dateformat Several new references have been added.Additional revision of the report is ongoing Commit-tee 224 invites comment from the readers and users ofthis report on new developments, or alternate viewpoints

on the Causes, Evaluation, and Repair of Cracks in crete Structures.

Con-CHAPTER 1-CAUSES AND CONTROL

OF CRACKING 1.1-Introduction

This chapter presents a brief summary of the causes ofcracks and means for their control Cracks are categor-ized as occurring either in plastic concrete or hardenedconcrete (Kelly 1981; Price 1982) In addition to the in-formation provided here, further details are presented inACI 224R and articles by Carlson et al (1979), Kelly(1981), Price (1982),, and Abdun-Nur (1983) Additionalreferences are cited throughout the chapter

1.2-Cracking of plastic concrete 1.2.1 Plastic shrinkage cracking-"Plastic shrinkage

cracking (Fig 1.1) occurs when subjected to a veryrapid loss of moisture caused by a combination of factorswhich include air and concrete temperatures, relativehumidity, and wind velocity at the surface of the con-crete These factors can combine to cause high rates ofsurface evaporation in either hot or cold weather.”When moisture evaporates from the surface of freshlyplaced concrete faster than it is replaced by bleed water,the surface concrete shrinks Due to the restraint pro-vided by the concrete below the drying surface layer, ten-sile stresses develop in the weak, stiffening plastic con-crete, resulting in shallow cracks of varying depth which

Fig 1.1-Typical plastic shrinkage cracking (Price 1982)

Fig 1.2-Crack formed due to obstructed settlement (Price

Fig 1.3-Settlement cracking as a function of bar size,

slump and cover (Dakhil et al 1975)

may form a random, polygonal pattern, or may appear as

essentially parallel to one another These cracks are often

fairly wide at the surface They range from a few inches

to many feet in length and are spaced from a few inches

to as much as 10 ft (3 m) apart Plastic shrinkage cracks

begin as shallow cracks but can become full-depth cracks

Since plastic shrinkage cracking is due to a differential

volume change in the plastic concrete, successful control

measures require a reduction in the relative volume

change between the surface and other portions of the

concrete

Steps can be taken to prevent a rapid moisture loss

due to hot weather and dry winds (ACI 224R, ACI

302.1R, ACI 305R) These measures include the use of

fog nozzles to saturate the air above the surface and the

use of plastic sheeting to cover the surface between

finishing operations Windbreaks to reduce the wind

velocity and sunshades to reduce the surface temperatureare also helpful, and it is good practice to schedule flatwork after the windbreaks have been erected

1.2.2 Settlement cracking - After initial placement,

vibration, and finishing, concrete has a tendency to tinue to consolidate During this period, the plastic con-crete may be locally restrained by reinforcing steel, aprior concrete placement, or formwork This local re-straint may result in voids and/or cracks adjacent to therestraining element (Fig 1.2) When associated with rein-forcing steel, settlement cracking increases with in-creasing bar size, increasing slump, and decreasing cover(Dakhil et al 1975) This is shown in Fig 1.3 for alimited range of these variables The degree of settlementcracking may be intensified by insufficient vibration or

con-by the use of leaking or highly flexible forms

Form design (ACI 347R) and vibration (and tion), provision of a time interval between the placement

revibra-of concrete in columns or deep beams and the placement

of concrete in slabs and beams (ACI 309.2R), the use ofthe lowest possible slump, and an increase in concretecover will reduce settlement cracking

1.3-Cracking of hardened concrete 1.3.1 Drying shrinkage-A common cause of cracking

in concrete is restrained drying shrinkage Dryingshrinking is caused by the loss of moisture from thecement paste constituent, which can shrink by as much as

1 percent Fortunately, aggregate provides internal straint that reduces the magnitude of this volume change

re-to about 0.06 percent On wetting, concrete tends re-toexpand

These moisture-induced volume changes are a teristic of concrete If the shrinkage of concrete couldtake place without restraint, the concrete would notcrack It is the combination of shrinkage and restraint(usually provided by another part of the structure or bythe subgrade) that causes tensile stresses to develop.When the tensile strength of concrete is exceeded, it willcrack Cracks may propagate at much lower stresses thanare required to cause crack initiation

charac-In massive concrete elements, tensile stresses arecaused by differential shrinkage between the surface andthe interior concrete The larger shrinkage at the surfacecauses cracks to develop that may, with time, penetratedeeper into the concrete

The magnitude of the tensile stresses induced by ume change is influenced by a combination of factors, in-cluding the amount of shrinkage, the degree of restraint,the modulus of elasticity, and the amount of creep Theamount of drying shrinkage is influenced mainly by theamount and type of aggregate and the water content ofthe mix The greater the amount of aggregate, thesmaller the amount of shrinkage (Pickett 1956) Thehigher the stiffness of the aggregate, the more effective

vol-it is in reducing the shrinkage of the concrete (i.e., the

shrinkage of concrete containing sandstone aggregatemay be more than twice that of concrete with granite,

basalt, or limestone (Carlson 1938)) The higher the

water content, the greater the amount of drying

shrink-age (U.S Bureau of Reclamation 1975)

Surface crazing (alligator pattern) on walls and slabs

is an example of drying shrinkage on a small scale

Crazing usually occurs when the surface layer of the

concrete has a higher water content than the interior

concrete The result is a series of shallow, closely spaced,

fine cracks

Drying shrinkage can be reduced by increasing the

amount of aggregate and reducing the water content A

procedure that will help reduce settlement cracking, as

well as drying shrinkage in walls, is reducing the water

content of the concrete as the wall is placed from the

bottom to the top Using this procedure, bleed water

from the lower portions of the wall will tend to equalize

the water content within the wall To be successful, this

procedure needs careful control of the concrete and

pro-per consolidation

Shrinkage cracking can be controlled by using

contrac-tion joints and steel detailing Shrinkage cracking may

also be reduced by using shrinkage-compensating cement

The reduction or elimination of subslab restraint can also

be effective in reducing shrinkage cracking in

slabs-on-grade (Wimsatt et al 1987) In cases where crack control

is particularly important, the minimum requirements of

AC I 318 are not always adequate These points are

dis-cussed in greater detail iu AC I 224R, which describes

additional construction practices designed to help control

the drying shrinkage cracking that does occur, and in

AC I 224.3R, which describes the use and function of

joints in concrete construction

1.3.2 Thermal stresses-Temperature differences within

a concrete structure may be caused by portions of the

structure losing heat of hydration at different rates or by

the weather conditions cooling or heating one portion of

the structure to a different degree or at a different rate

than another portion of the structure These temperature

differences result in differential volume changes When

the tensile stresses due to the differential volume changes

exceed the tensile stress capacity, concrete will crack The

effects of temperature differentials due to different rates

of heat dissipation of the heat of hydration of cement are

normally associated with mass concrete (which can

in-clude large columns, piers, beams, and footings, as well

as dams), while temperature differentials due to changes

in the ambient temperature can affect any structure

Cracking in mass concrete can result from a greater

temperature on the interior than on the exterior The

temperature gradient may be caused by either the center

of the concrete heating up more than the outside due to

the liberation of heat during cement hydration or more

rapid cooling of the exterior relative to the interior Both

cases result in tensile stresses on the exterior and, if the

tensile strength is exceeded, cracking will occur The

ten-sile stresses are proportional to the temperature

differ-ential, the coefficient of thermal expansion, the effective

modulus of elasticity (which is reduced by creep), and the

degree of restraint (Dusinberre 1945; Houghton 1972,1976) The more massive the structure, the greater thepotential for temperature differential and restraint.Procedures to help reduce thermally-induced crackinginclude reducing the maximum internal temperature, de-laying the onset of cooling, controlling the rate at whichthe concrete cools, and increasing the tensile strength ofthe concrete These and other methods used to reducecracking in massive concrete are presented in ACI207.1R, ACI 207.2R, ACI 207.4R, and ACI 224R.Hardened concrete has a coefficient of thermal expan-sion that may range from 4 to 9 x lOA F (7 to 11 x 10”C), with a typical value of 5.5 x lOA F (10 x lOA C).When one portion of a structure is subjected to a tem-perature-induced volume change, the potential for ther-mally-induced cracking exists Designers should give spe-cial consideration to structures in which some portionsare exposed to temperature changes, while other portions

of the structure are either partially or completely tected A drop in temperature may result in cracking inthe exposed element, while increases in temperature maycause cracking in the protected portion of the structure.Temperature gradients cause deflection and rotation instructural members; if restrained, serious stresses canresult (Priestly 1978; Hoffman et al 1983; ACI 343R).Allowing for movement by using properly designed con-traction joints and correct detailing will help alleviatethese problems

pro-1.3.3 Chemical reaction-Deleterious chemical

reac-tions may cause cracking of concrete These reacreac-tionsmay be due to materials used to make the concrete ormaterials that come into contact with the concrete after

it has hardened

Some general concepts for reducing adverse chemicalreactions are presented here, but only pretesting of themixture or extended field experience will determine theeffectiveness of a specific measure

Concrete may crack with time as the result of slowlydeveloping expansive reactions between aggregate con-taining active silica and alkalies derived from cement

hydration, admixtures, or external sources (e.g., curing

water, ground water, alkaline solutions stored or used inthe finished structure.)

The alkali-silica reaction results in the formation of aswelling gel, which tends to draw water from other por-tions of the concrete This causes local expansion andaccompanying tensile stresses, and may eventually result

in the complete deterioration of the structure Controlmeasures include proper selection of aggregates, use oflow alkali cement, and use of pozzolans, which them-selves contain very fine, highly active silicas The firstmeasure may preclude the problem from occurring, whilethe later two measures have the effect of decreasing thealkali to reactive silica ratio, resulting in the formation of

a nonexpanding calcium alkali silicate

Certain carbonate rocks participate in reactions withalkalies which, in some instances, produce detrimental ex-pansion and cracking These detrimental alkali-carbonate

reactions are usually associated with argillaceous

dolomitic limestones which have a very fine grained

(cryptocrystalline) structure (ACI 201.2R) The affected

concrete is characterized by a network pattern of cracks

The reaction is distinguished from the alkali-silica

reaction by the general absence of silica gel surface

deposits at the crack The problem may be minimized by

avoiding reactive aggregates, dilution with nonreactive

aggregates, use of a smaller maximum size aggregate, and

use of low-alkali cement (ACI 201.2R)

Sulfate-bearing waters are a special durability problem

for concrete When sulfate penetrates hydrated cement

paste, it comes in contact with hydrated calcium

alumi-nate Calcium sulfoaluminate is formed, with a

subse-quently large increase in volume, resulting in high local

tensile stresses that lead to cracking which causes

development of closely spaced cracking and

deteriora-tion ASTM C 150 Types II and V portland cement,

which are low in tricalcium aluminate, will reduce the

severity of the problem The blended cements specified

in ASTM C 595 are also useful in this regard In severe

cases, some pozzolans, known to impart additional

resis-tance to sulfate attack, could be used after adequate

testing

Detrimental conditions may also occur from the

appli-cation of deicing salts to the surface of hardened

con-crete Concrete subjected to water soluble salts should be

amply air entrained, have adequate cover of the

rein-forcing steel, and be made of high-quality, low

per-meability concrete

The effects of these and other problems relating to the

durability of concrete are discussed in greater detail in

ACI 201.2R

The calcium hydroxide in hydrated cement paste will

combine with carbon dioxide in the air to form calcium

carbonate Since calcium carbonate has a smaller volume

than the calcium hydroxide, shrinkage will occur

(com-monly known as carbonation shrinkage) This situation

may result in significant surface crazing and may be

especially serious on freshly placed surfaces during the

first 24 hours when improperly vented combustion

heaters are used to keep concrete warm during the

winter months

With the exception of surface carbonation, very little

can be done to protect or repair concrete that has been

subjected to the types of chemical attack described above

(ACI 201.2R)

1.3.4 Weathering-The weathering processes that can

cause cracking include freezing and thawing, wetting and

drying, and heating and cooling Cracking of concrete

due to natural weathering is usually conspicuous, and it

may give the impression that the concrete is on the verge

of disintegration, even though the deterioration may not

have progressed much below the surface

Damage from freezing and thawing is the most

com-mon weather-related physical deterioration Concrete

may be damaged by freezing of water in the paste, in the

aggregate, or in both (Powers 1975)

Damage in hardened cement paste from freezing iscaused by the movement of water to freezing sites and byhydraulic pressure generated by the growth of ice crystals(Powers 1975)

Aggregate particles are surrounded by cement pastewhich prevents the rapid escape of water When the ag-gregate particles are above a critical degree of saturation,the expansion of the absorbed water during freezing maycrack the surrounding cement paste or damage the aggre-gate itself (Callan 1952; Snowdon and Edwards 1962).Concrete is best protected against freezing andthawing through the use of the lowest practical water-cement ratio and total water content, durable aggregate,and adequate air entrainment Adequate curing prior toexposure to freezing conditions is also important Allow-ing the structure to dry after curing will enhance itsfreezing and thawing durability

Other weathering processes that may cause cracking inconcrete are alternate wetting and drying, and heatingand cooling Both processes produce volume changes thatmay cause cracking If the volume changes are excessive,cracks may occur, as discussed in Sections 1.3.1 and 1.3.2

1.3.5 Corrosion of reinforcement-Corrosion of a metal

is an electrochemical process that requires an oxidizingagent, moisture, and electron flow within the metal; aseries of chemical reactions takes place on and adjacent

to the surface of the metal (ACI 201.2R)

The key to protecting metal from corrosion is to stop

or reverse the chemical reactions This may be done bycutting off the supplies of oxygen or moisture or by sup-plying excess electrons at the anodes to prevent theformation of the metal ions (cathodic protection).Reinforcing steel usually does not corrode in concretebecause a tightly adhering protective oxide coating forms

in the highly alkaline environment This is known aspassive protection

Reinforcing steel may corrode, however, if the linity of the concrete is reduced through carbonation or

alka-if the passivity of this steel is destroyed by aggressive ions(usually chlorides) Corrosion of the steel produces ironoxides and hydroxides, which have a volume much great-

er than the volume of the original metallic iron (Verbeck1975) This increase in volume causes high radial burstingstresses around reinforcing bars and results in local radialcracks These splitting cracks can propagate along the

bar, resulting in the formation of longitudinal cracks (i.e.,

parallel to the bar) or spalling of the concrete A broadcrack may also form at a plane of bars parallel to a con-crete surface, resulting in delamination, a well-knownproblem in bridge decks

Cracks provide easy access for oxygen, moisture, andchlorides, and thus, minor splitting cracks can create acondition in which corrosion and cracking are acceler-ated

Cracks transverse to reinforcement usually do notcause continuing corrosion of the reinforcement if theconcrete has low permeability This is due to the fact thatthe exposed portion of a ba.r at a crack acts as an anode

At early ages, the wider the crack, the greater the

cor-rosion, simply because a greater portion of the bar has

lost its passive protection However, for continued

cor-rosion to occur, oxygen and moisture must be supplied to

other portions of the same bar or bars that are

elec-trically connected by direct contract or through hardware

such as chair supports If the combination of density and

cover thickness is adequate to restrict the flow of oxygen

and moisture, then the corrosion process is self sealing

(Verbeck 1975)

Corrosion can continue if a longitudinal crack forms

parallel to the reinforcement, because passivity is lost at

many locations, and oxygen and moisture are readily

available along the full length of the crack

Other causes of longitudinal cracking, such as high

bond stresses, transverse tension (for example, along

stir-rups or along slabs with two-way tension), shrinkage, and

settlement, can initiate corrosion

For general concrete construction, the best protection

against corrosion-induced splitting is the use of concrete

with low permeability and adequate cover Increased

con-crete cover over the reinforcing is effective in delaying

the corrosion process and also in resisting the splitting

and spalling caused by corrosion or transverse tension

(Gergely 1981; Beeby 1983) In the case of large bars and

thick covers, it may be necessary to add small transverse

reinforcement (while maintaining the minimum cover

re-quirements) to limit splitting and to reduce the surface

crack width (ACI 345R)

In very severe exposure conditions, additional

pro-tective measures may be required A number of options

are available, such as coated reinforcement, sealers or

overlays on the concrete, corrosion-inhibiting admixtures,

and cathodic protection (NCHRP Synthesis 57) Any

pro-cedure that effectively prevents access of oxygen and

moisture to the steel surface or reverses the electron flow

at the anode will protect the steel In most cases,

con-crete must be allowed to breathe, that is any concon-crete

surface treatment must allow water to evaporate from the

concrete

1.3.6 Poor construction practices-A wide variety of

poor construction practices can result in cracking in

concrete structures Foremost among these is the

com-mon practice of adding water to concrete to improve

workability Added water has the effect of reducing

strength, increasing settlement, and increasing drying

shrinkage When accompanied by a higher cement

con-tent to help offset the decrease in strength, an increase

in water content will also mean an increase in the

tem-perature differential between the interior and exterior

portions of the structure, resulting in increased thermal

stresses and possible cracking By adding cement, even if

the water-cement ratio remains constant, more shrinkage

will occur since the relative paste volume is increased

Lack of curing will increase the degree of cracking

within a concrete structure The early termination of

curing will allow for increased shrinkage at a time when

the concrete has low strength The lack of hydration of

the cement, due to drying, will result not only in creased long-term strength, but also in the reduced dur-ability of the structure

de-Other construction problems that may cause crackingare inadequate formwork supports, inadequate consolida-tion, and placement of construction joints at points ofhigh stress Lack of support for forms or inadequate con-solidation can result in settlement and cracking of theconcrete before it has developed sufficient strength tosupport its own weight, while the improper location ofconstruction joints can result in the joints opening atthese points of high stress

Methods to prevent cracking due to these and otherpoor construction procedures are well known (see ACI224R, ACI 302.1R, ACI 304R, ACI 305R, ACI 308, ACI309R, ACI 345R, and ACI 347R), but require special at-tention during construction to insure their proper exe-cution

1.3.7 Construction overloads-Loads induced during

construction can often be far more severe than those perienced in service Unfortunately, these conditions mayoccur at early ages when the concrete is most susceptible

ex-to damage and they often result in permanent cracks.Precast members, such as beams and panels, are mostfrequently subject to this abuse, but cast-in-place con-crete can also be affected A common error occurs whenprecast members are not properly supported duringtransport and erection The use of arbitrary or conven-ient lifting points may cause severe damage Lifting eyes,pins, and other attachments should be detailed or ap-proved by the designer When lifting pins are impractical,access to the bottom of a member must be provided sothat a strap may be used The PCI Committee on QualityControl Performance Criteria (1985, 1987) provides addi-tional information on the causes, prevention and repair

of cracking related to fabrication and shipment of precast

or prestressed beams, columns, hollow core slabs anddouble tees

Operators of lifting devices must exercise caution and

be aware that damage may be caused even when the per lifting accessories are used A large beam or panellowered too fast, and stopped suddenly, results in animpact load that may be several times the dead weight ofthe member Another common construction error thatshould be avoided is prying up one corner of a panel tolift it off its bed or “break it loose.”

pro-When considering the support of a member for ment, the designer must be aware of loads that may beinduced during transportation Some examples that occurduring shipment of large precast members via tractor andtrailer are jumping curbs or tight highway corners, torsiondue to differing roadway superelevations between thetrailer and the tractor, and differential acceleration of thetrailer and the tractor

ship-Pretensioned beams can present unique cracking lems at the time of stress release-usually when thebeams are less than one day old Multiple strands must

prob-be detensioned following a specific pattern, so as not to

place unacceptable eccentric loads on the member If all

of the strands on one side of the beam are released while

the strands on the other side are still stressed, cracking

may occur on the side with the unreleased strands These

cracks are undesirable, but should close with the release

of the balance of the strands

In the case of a T-beam with a heavily reinforced

flange and a highly prestressed thin web, cracks may

develop at the web-flange junction

Another practice that can result in cracks near beam

ends is tack welding embedded bearing plates to the

casting bed to hold them in place during concrete

place-ment The tack welds are broken only after enough

pre-stress is induced during pre-stress transfer to break them

Until then, the bottom of the beam is restrained while

the rest of the beam is compressed Cracks will form near

the bearing plates if the welds are too strong

Thermal shock can cause cracking of steam-cured

con-crete if it is treated improperly The maximum rate of

cooling frequently used is 70 F (40 C) per hour (ACI

517.2R; Verbeck 1958; Shideler and Toennies 1963;

Kirk-bride 1971b) When brittle aggregate is used and the

strain capacity is low, the rate of cooling should be

decreased Even following this practice, thermally

in-duced cracking often occurs Temperature restrictions

should apply to the entire beam, not just locations where

temperatures are monitored If the protective tarps used

to contain the heat are pulled back for access to the

beam ends when cutting the strands, and if the ambient

temperatures are low, thermal shock may occur

Temper-ature recorders are seldom located in these critical areas

Similar conditions and cracking potential exist with

precast blocks, curbs, and window panels when a rapid

surface temperature drop occurs

It is believed by many (ACI 517.2R; Mansfield 1948;

Nurse 1949; Higginson 1961; Jastnebski 1961; Butt et al

1969; Kirkbride 1971a; Concrete Institute of Australia

1972; PCI Energy Committee 1981) that rapid cooling

may cause cracking only in the surface layers of very

thick units and that rapid cooling is not detrimental to

the strength or durability of standard precast products

(PCI Energy Committee 1981) One exception is

trans-verse cracking observed in pretensioned beams subjected

to cooling prior to detensioning For this reason,

pre-tensioned members should be depre-tensioned immediately

after the steam-curing has been discontinued (PCI

Energy Committee 1981)

Cast-in-place concrete can be unknowingly subjected

to construction loads in cold climates when heaters are

used to provide an elevated working temperature within

a structure Typically, tarps are used to cover windows

and door openings, and high volume heaters are

oper-ated inside the enclosed area If the heaters are locoper-ated

near exterior concrete members, especially thin walls, an

unacceptably high thermal gradient can result within the

members The interior of the wall will expand in relation

to the exterior Heaters should be kept away from the

exterior walls to minimize this effect Good practice also

requires that this be done to avoid localized dryingshrinkage and carbonation cracking

Storage of materials and the operation of equipmentcan easily result in loading conditions during constructionfar more severe than any load for which the structurewas designed Tight control must be maintained to avoidoverloading conditions Damage from unintentional con-struction overloads can be prevented only if designersprovide information on load limitations for the structureand if construction personnel heed these limitations

1.3.8 Errors in design and detailing-The effects ofimproper design and/or detailing range from poorappearance to lack of serviceability to catastrophicfailure These problems can be minimized only by athorough understanding of structural behavior (meanthere in the broadest sense)

Errors in design and detailing that may result inunacceptable cracking include use of poorly detailedreentrant comers in walls, precast members and slabs,improper selection and/or detailing of reinforcement,restraint of members subjected to volume changes caused

by variations in temperature and moisture, lack of quate contraction joints, and improper design of foun-dations, resulting in differential movement within thestructure Examples of these problems are presented byKaminetzky (1981) and Price (1982)

ade-Reentrant comers provided a location for the centration of stress and, therefore, are prime locationsfor the initiation of cracks Whether the high stressesresult from volume changes, in-plane loads, or bending,the designer must recognize that stresses are always highnear reentrant comers Well-known examples are windowand door openings in concrete walls and dapped endbeams, as shown in Fig 1.4 and 1.5 Additional properlyanchored diagonal reinforcement is required to keep theinevitable cracks narrow and prevent them from pro-pagating

con-“;

Fig 1.4-Typical crack pattern at reentrant corners (Price 1982)

/

Fig 1.5-Typical cracking pattern of dapped end at service

load*

The use of an inadequate amount of reinforcing may

result in excessive cracking A typical mistake is to lightly

reinforce a member because it is a “nonstructural

mem-ber.” However, the member (such as a wall) may be tied

to the rest of the structure in such amanner that it is

required to carry a major portion of the load once the

structure begins to deform The “nonstructural element”

then begins to carry loads in proportion to its stiffness

Since this member is not detailed to act structurally,

unsightly cracking may result even though the safety of

the structure is not in question

The restraint of members subjected to volume changes

results frequently in cracks Stresses that can occur in

concrete due to restrained creep, temperature

differen-tial, and drying shrinkage can be many times the stresses

that occur due to loading A slab, wall, or a beam

re-strained against shortening, even if prestressed, can easily

develop tensile stresses sufficient to cause cracking

Pro-perly designed walls should have contraction joints

spaced from one to three times the wall height Beams

should be allowed to move Cast-in-place post-tensioned

construction that does not permit shortening of the

prestressed member is susceptible to cracking in both the

member and the supporting structure (Libby 1977) The

problem with restraint of structural members is especially

serious in pretensioned and precast members that may be

welded to the supports at both ends When combined

with other problem details (such as reentrant comers),

results may be catastrophic (Kaminetzky 1981; Mast

1981)

Improper foundation design may result in excessive

differential movement within a structure If the

differ-ential movement is relatively small, the cracking lems may be only visual in nature However, if there is amajor differential settlement, the structure may not beable to redistribute the loads rapidly enough, and a fail-ure may occur One of the advantages of reinforced con-crete is that, if the movement takes place over a longenough period of time, creep will allow at least someload redistribution to take place

prob-The importance of proper design and detailing willdepend on the particular structure and loading involved.Special care must be taken in the design and detailing ofstructures in which cracking may cause a major service-ability problem These structures also require continuousinspection during all phases of construction to supple-ment the careful design and detailing

1.3.9 Externally applied loads-It is well known that

load-induced tensile stresses result in cracks in concretemembers This point is readily acknowledged and ac-cepted in concrete design Current design procedures(ACI 318 and AASHTO) Standard Specifications forHighway Bridges) use reinforcing steel, not only to carrythe tensile forces, but to obtain both an adequate dis-triiution of cracks and a reasonable limit on crack width.Current knowledge of flexural members provides thebasis for the following general conclusions about the var-iables that control cracking: Crack width increases withincreasing steel stress, cover thickness and area of con-crete surrounding each reinforcing bar Of these, steelstress is the most important variable The bar diameter

is not a major consideration The width of a bottomcrack increases with an increasing strain gradient betweenthe steel and the tension face of the beam

The equation considered to best predict the mostprobable maximum surface crack width in bending wasdeveloped by Gergely and Lutz (1968) A simplified ver-sion of this equation is:

w = 0.076 #Ifs (d,A)” x 1O-3 (1.1)

in which w = most probable maximum crack width, in.;

B = ratio of distance between neutral axis and tensionface to distance between neutral axis and centroid ofreinforcing steel (taken as approximately 1.20 for typicalbeams in buildings); f, = reinforcing steel stress, ksi; d,

= thickness of cover from tension fiber to center of barclosest thereto, in.; and A = area of concrete symmetricwith reinforcing steel divided by number of bars, in2

A modification of this equation is used in ACI 318,which effectively limits crack widths to 0.016 in (0.41mm) for interior exposure and 0.013 in (0.33 mm) forexterior exposure However, considering the informationpresented in Section 1.3.5 which indicates little cor-relation between surface crack width for cracks transverse

to bars and the corrosion of reinforcing, these limits donot appear to be justified on the basis of corrosioncontrol

*From Alan H Mattock and Timothy C Chan (1979) "Design and Behavior

of Dapped-end Beams,” Joumal, Prestressed Concrete Institute, V 24, NO 6, Nov.-Dec., pp 28-45.

There have been anumber of equations developed for

prestressed concrete members (ACI 224R), but no single

method has achieved general acceptance

The maximum crack width in tension members is

larger than that predicted by the expression for flexural

members (Broms 1965; Broms and Lutz 1965) Absence

of a strain gradient and compression zone in tension

members is the probable reason for the larger crack

widths

On the basis of limited data, the following expression

has been suggested to estimate the maximum crack width

in direct tension (ACI 224R):

w = 0.10 f s (d c A)0.33 x 10 -3 (1.2)Additional information on cracking of concrete in direct

tension is provided in ACI 224.2R

Flexural and tensile crack widths can be expected to

increase with time for members subjected to either

sus-tained or repetitive loading Although a large degree of

scatter is evident in the available data, a doubling of

crack width with time can be expected (Abeles et al

1968; Bennett and Dave 1969; Illston and Stevens 1972;

Holmberg 1973; Rehm and Eligehausen 1977)

Although work remains to be done, the basic

princi-ples of crack control for load-induced cracks are well

understood Well-distriiuted reinforcing offers the best

protection against undesirable cracking Reduced steel

stress, obtained through the use of a larger amount of

steel, will also reduce the amount of cracking While

reduced cover will reduce the surface crack width,

de-signers must keep in mind, as pointed out in Section

per-pendicular to reinforcing steel do not have a major effect

on the corrosion of the steel, while a reduction in cover

will be detrimental to the corrosion protection of the

reinforcing

CHAPTER 2-EVALUATION OF CRACKING

2.1-Introduction

When anticipating repair of cracks in concrete, it is

important to first identify the location and extent of

cracking It should be determined whether the observed

cracks are indicative of current or future structural

prob-lems, taking into consideration the present and

antici-pated future loading conditions The cause of the

crack-ing should be established before repairs are specified

Drawings, specifications, and construction and

main-tenance records should be reviewed If these documents,

along with field observations, do not provide the needed

information, a field investigation and structural analysis

should be completed before proceeding with repairs

The causes of cracks are discussed in Chapter 1 A

detailed evaluation of observed cracking can determine

which of those causes applies in a particular situation

Cracks need to be repaired if they reduce the strength,

stiffness, or durability of the structure to an unacceptable

level, or if the function of the structure is seriouslyimpaired In some cases, such as cracking in water-re-taining structures, the function of the structure willdictate the need for repair, even if strength, stiffness, orappearance are not significantly affected Cracks in pave-ments and slabs-on-grade may require repair to preventedge spalls, migration of water to the subgrade, or totransmit loads In addition, repairs that improve theappearance of the surface of a concrete structure may bedesired

2.2-Determination of location and extent of concrete cracking

Location and extent of cracking, as well as information

on the general condition of concrete in a structure, can

be determined by both direct and indirect observations,nondestructive and destructive testing, and tests of corestaken from the structure Information may also be ob-tained from drawings and construction and maintenancerecords

2.2.1 Direct and indirect observation-The locations and

widths of cracks should be noted on a sketch of thestructure A grid marked on the surface of the structurecan be useful to accurately locate cracks on the sketch.Crack widths can be measured to an accuracy of about0.001 in (0.025 mm) using a crack comparator, which is

a small, hand-held microscope with a scale on the lensclosest to the surface being viewed (Fig 2.1) Crackwidths may also be estimated using a clear comparatorcard having lines of specified width marked on the card.Observations such as spalling, exposed reinforcement,surface deterioration, and rust staining should be noted

on the sketch Internal conditions at specific crack tions can be observed with the use of flexible shaft fiber-scopes or rigid borescopes

loca-Crack movement can be monitored with mechanicalmovement indicators of the types shown in Fig 2.2 Theindicator, or crack monitor, shown in Fig 2.2 (a) gives adirect reading of crack displacement and rotation Theindicator in Fig 2.2 (b) (Stratton et al 1978) amplifiesthe crack movement (in this case, 50 times) and indicatesthe maximum range of movement during the measure-ment period Mechanical indicators have the advantage

Fig 2.1-Comparator for measuring crack widths (courtesy

of Edmound Scientific Co.)

Newly Mounted Monitor

Monitor After Crack Movement

(a)-Crack monitor (courtesy of Avongard)

CRACK ON GIRDER FACE

SEE ISOMETRIC SECTION AT

\

(b)-Crack movement indicator (Stratton et al 1978)

Figure 2.2

Fig 2.3-Pachometer (reinforcing bar locator) (courtesy of

James Instruments)

that they do not require moisture protection If more

detailed time histories are desired, a wide range of

transducers (most notably linear variable differential

transformers or LVDT'S) and data acquisition systems

(ranging from strip chart recorders to computer-based

systems) are available

Sketches can be supplemented by photographs

docu-menting the condition of the structure at the time of

investigation Guidance for making a condition survey of

concrete in service is given in ACI 201.1R, ACI 201.3R,

ACI 207.3R, ACI 345.1R, and ACI 546.1R

2.2.2 Nondestructive testing-Nondestructive tests can

be made to determine the presence of internal cracks

and voids and the depth of penetration of cracks visible

at the surface

Tapping the surface with a hammer or using a chain

drag are simple techniques to identify laminar cracking

near the surface A hollow sound indicates one or more

cracks below and parallel to the surface

The presence of reinforcement can be determined

using a pachometer (Fig 2.3) (Malhotra 1976) A

num-ber of pachometers are available that range in capability

from merely indicating the presence of steel to those that

may be calibrated to allow the experienced user a closer

determination of depth and the size of reinforcing steel

In some cases, however, it may be necessary to remove

the concrete cover (often by drilling or chipping) to

identify the bar sizes or to cahbrate cover measurements,

especially in areas of congested reinforcement

Receiving Transducer

Direct Transmission Increased Path Length Due to Dlscontlnuity Partial Depth Crack Alternate Configuration

a) Pulse transmitted through member

Fig 2.4-Ultrasonic testing through-transmission technique

If corrosion is a suspected cause of cracking, the *

easiest approach to investigate for corrosion entails theremoval of a portion of the concrete to directly observethe steel Corrosion potential can be detected by electri-cal potential measurements using a suitable referencehalf cell The most commonly used is a copper-coppersulfate half cell (ASTM C 876; Clear and Hay 1973); itsuse also requires access to a portion of the reinforcingsteel

With properly trained personnel and careful tion, it is possible to detect cracks using ultrasonicnondestructive test equipment (ASTM C 597) The mostcommon technique is through-transmission testing usingcommercially available equipment (Malhotra and Carino1991; Knab et al 1983) A mechanical pulse is trans-mitted to one face of the concrete member and received

evalua-at the opposite face, as shown Fig 2.4 The time takenfor the pulse to pass through the member is measuredelectronically If the distance between the transmittingand receiving transducers is known, the pulse velocity can

be calculated

When access is not available to opposite faces, ducers may be located on the same face [Fig 2.4(a)].While this technique is possible, the interpretation ofresults is not straightforward