Deterioration, maintenance, and repair of structures

242 77 66MB

English Pages 373 [383] Year 1965

Report DMCA / Copyright


Polecaj historie

Deterioration, maintenance, and repair of structures

Citation preview



B.E., M. Eng., P-£.; Member,

American Society of Civil Engineers; Associate,


Formerly: Adjunct Assistant Professor of Civil Engineering, New York University; Instructor in Civil Engineering, Yale University.






San Francisco




Architecturai Llbrary








Copyright © 1965 by McGraw-Hill, Inc. All Rights Reserved. Printed in the United States of America. This book, or parts thereof, may not be reproduced in any form without permission of the publishers.

Library of Congress Catalog Card Number



SY¥S5% 19-220


The major problems

in design and construction of civil engineer-

ing works are well known and, in general,

are carefully considered

in practice. In addition, design live-load provisions incorporate a margin for overload, design specifications provide a substantial safety factor, and design assumptions are conservative. As a result,

a structural collapse, other than a mishap occurring during con-

struction, is a rarity, and both engineer and owner rest secure, knowing that a structure, if at least reasonably well designed, will adequately support the loads to which it is subjected. However, while they may be perfectly adequate for the applied loads, many structures, after being put into use, develop serious

problems of maintenance. recur

in similar


Further, the same problems consistently under



of exposure,

which suggests that the defects might be the result of an inadvertent but repeated use of unsuitable details and/or practices in design and construction. As will be developed in the text, this is actually the case, and the primary purpose of this book is to consider the cause and repair of defects, not resulting from errors, but which occur in structures proportioned without error and by conventional design procedures and which, therefore, must be considered as having been adequately designed. It also will be developed that the use and repetition of these un-

satisfactory details and practices is due, simply, to the fact that



designers and construction personnel are not aware that they are troublesome and so do not incorporate corrective provisions into subsequent construction. That this should be so is readily under-


Design, construction, and maintenance usually are per-

formed by separate departments

and frequently by separate agencies

or firms. Moreover, the maintenance problem does not show up for several years, by which time the parties responsible for the design

and construction may be engaged in other duties, deceased, or otherwise detached from concern with the matter. As a result, liaison between the designer, the construction personnel, and the maintenance engineer is poor, and designers and construction people, however well-intentioned, have little opportunity to learn from the poor performance of their work. There is also little opportunity to learn from the poor performance of the work of others

due to a natural reluctance to advertise one’s difficulties.

Accordingly, a second purpose of this text is to describe some of the experiences of the author and of others in detecting and repairing deterioration and so to establish some of the necessary liaison between the designer, builder, and maintenance engineer and to acquaint designers and construction personnel, in general, with potential problem areas of which, lacking experience in maintenance, they might be otherwise unaware. For each of the principal classes of construction (steel, concrete, and timber), the data are presented in two parts. The first part discusses types of deterioration, identifies their cause or causes, and describes details and procedures in design and construction which may be employed to prevent or minimize their occurrence. Examples are given of details and practices which are known to or are likely to give unsatisfactory results. The application and importance of the following basic principles are emphasized: (1) Selection of the proper construction materials as related to site conditions, conditions of exposure, and use of the structure.

(2) Attention to the details of design and their importance with

respect to the overall design concept.

(3) Insistence on proper construction practices.

The second part of the text discussion of each class of construction concerns itself with the detection and correction of such deterioration as does occur. Details or references as to suitable methods of repair are presented. Procedures employed in specific

instances are detailed, and the success (or failure) of these pro-

cedures is noted. The necessity for the provision and careful execution of a properly conceived program of inspection and minor preventive maintenance is emphasized as a means for arresting and, to some extent, preventing serious forms of deterioration. A

chapter concerned with the strengthening of existing structures has



been included, as this subject is ancillary to that of rehabilitation. A number of photographs have been included in the text to present the visual characteristics of several of the more common forms of deterioration and to provide the reader with a basis for recognizing the occurrence of defects and their probable causes. Examples of deterioration are very common, but to the unpracticed eye may all look much the same. Further, the untrained observer may not understand the significance of what is seen. It is like a problem in geology. The skilled geologist is able to relate appearance of land forms to varying geologic agents. Similarly, in maintenance engineering, practice is required to recognize the subtle differences which represent the action of different deteriorating agents. The photographs provide touchstones for this purpose and should be carefully studied. In the author’s



of preventing,


ing, identifying, and correcting structural deterioration involve the exercise of a high level of ingenuity and engineering creativity. The organization and execution of a proper maintenance program likewise requires competent attention. If these facts be recognized, and maintenance be treated as a necessary and fully professional specialty of engineering, much will have been accomplished toward correcting the basic deficiency in maintenance, i.e., failure during the stages of design and construction to recognize and to provide against potential sources of deterioration. The author acknowledges the valuable experience gained as an associate in the firm of Praeger-Kavanagh- Waterbury and, particularly, the tutelage of Capt. Emil H. Praeger. This experience has contributed much of the necessary background for this book, and many of the examples and illustrations are taken from projects accomplished by this firm. Thanks are expressed to Mr. Irving Zand for assistance in preparing a number of the photographs, and to Mr. Jerome S. B. Iffland and Mr. Harold L. R. Kistler, Jr., for innumerable and invaluable suggestions.






Prevention. .

Repair Steps. ee Be é ea 3 as BEE Workmanship... .... 0... 0 cece eee ee ee eee eee eee Summary: so ee ee 8 6 6 ci seaiarererensnerecenanmen 60 a 6 oO ww Chapter Notes 1. Sources of Reserves of Strength in Structures....... 2. Validity of Load Tests as Indicative of Structural Strength.

References. ........++eeeeeeee 2. STEEL STRUCTURES...



Types and Causes of Deterioration.

Preventive Measures......... Repair





eee eee

e cece eens


about the Occurrence


Corrosion: ¢ + = i. ¥ ¥ she eee WS WSS BEES E EE EERE References... . 2 6 eo ee ee nieieiennnceninieie ae ee eee




of Deterioration...

Diagnosis of Cause Flow Charts


for Diagnosis... .





References...... 4. CONCRETE


Introduction ...........02 eee eee Methods

of Repair...


20 20 25 35 44


Some Case Histories............ Chapter Notes 1. Some Pertinent Generalizations

13 17 19

........0. 0c e eee eens


55 57



5. CONCRETE STRUCTURES: REPAIRING SPALLING AND DISINTEGRATION ....... 00. eee eect eee eee tee eens Repair Procedures Some Case Histories Chapter Notes 1. Compatibility of Materials and Sections .............% 2. Sample Specification Provisions for Prototype Jackets ... 3. Sample Specification Provisions for Removal and Replacement of Permanent Forms for Concrete Jackets . .



.. eee eee eens



Restoring a Disintegrated Surface





218 229 231 233 236


Correcting a Settlement Condition ............- se eeees Some Case Histories.............. eee e ee ee eens References..... Pee eee eee eee nee eee eet eee eee

DO CRY ose ee oe 6 8 6 8 & wrerienenieneriexeceraenexexesinie Marine Borers . Insects .... Shrinkage Deterioration of Hardware Replacing and Reinforcing Timbers ... Some Case Histories .





...........-. eee ee ee eeeee

Sheet Pile Structures. .........5.0e 000 ee ee teen ee Precast Concrete Piles .... 2... ee eee eee eee eee see Expansion Joints. ..... 2... eee Some Case Histories . eee References. .... tees 9.






General Principles ........ Strengthening the Superstructure . Strengthening the Substructure. Illustrative Examples....... wate References. ........ 2.00 e eens Index




oe ............

215 215



The deterioration and maintenance

of engineering structures is a

common and serious problem, involving considerable cost and inconvenience to industry and to the public. For the engineer, the problem involves two basic aspects, prevention and repair. Except for the common requirement of a knowledge of what defects may occur, the two aspects are separable and

will be so considered in this text. A. PREVENTION

Of the two considerations, prevention and repair, prevention is the more important. In fact, the need for adequate attention to potential maintenance problems during the design and construction

phases of a project

is emphasized throughout the text.

The trouble-

some details and construction procedures described herein, of which many will seem obvious, nevertheless are of frequent occurrence, and while the defects themselves may seem minor, the consequences

can be serious.’

Accordingly, it will be apparent that the design engineer must select materials for his design which are suitable for the exposure 1


Deterioration, Maintenance, and Repair of Structures


at the site, must detail the structure in a manner which

will prevent the occurrence of serious deterioration

(at least for

the assumed service life of the work), and, through the inspection

staff, must insist on proper construction. Clearly, these three points—proper materials, proper details, and proper construction— require a knowledge of what is improper and imply a knowledge

of the various forms of deterioration which occur and an understanding of their causes. Data on these aspects of the problem are set forth in Chapters 3 and 7 and in portions of Chapters 2 and 8. The engineer concerned with the design or inspection of new construction should direct his attention to those portions of this volume. The remaining portions of the text are for use by those concerned with the maintenance of existing constructions.

B. REPAIR STEPS If prevention of deterioration has been unsuccessful, then the structure must be abandoned, replaced, or repaired; usually it is repaired.

The execution of such a repair is an



matter involving five basic steps: (1) finding the deterioration, (2) determining the cause, (3) evaluating the strength of the existing structure, (4) evaluating the need for repair, and (5) selecting and implementing a repair procedure. Step 1.

Find the Deterioration

It is, perhaps, unnecessary

to state, but before a repair can be

effected, there must be a realization

that something is wrong, and

this realization must come before it is too late to make a repair, i.e., before the structure has collapsed. In practice, this apparently obvious and simple statement may present a very subtle problem. For example, as will be detailed later, timbers and timber piling

can be damaged by

insects or marine

organisms, virtually to the

point of collapse, without exhibiting any external evidence which would be apparent except to a trained observer.

Even a common

defect like corrosion

of steel can be difficult

to detect because it occurs, principally, in the most inaccessible parts of the structure. The reason is simple. The accessible parts

are painted, but the inaccessible Also, it is common to find that perfectly sound have been rotted verely. Here, the reason is that

parts often are neglected. heavy timbers which appear to be away internally, often quite sethe rot, which is due to dampness,

starts at and progresses from the abutting faces of adjacent timbers or at the interface between timber and masonry, where the mois-

Introduction and General Considerations


ture is retained. These faces are not visible, the rot is hidden, and it is necessary to sound the timbers with a hammer, or to bore or cut them, in order to discover the condition. Those faces of the timber which are visible and can be readily inspected are exposed to the atmosphere, are dry, do not rot, and look fine. There are many similar examples which could be presented. The point to be made is that the engineer charged with or interested in maintenance must be trained, technically, in where to look, how to look, and what to look for, before he can even be expected to realize that there is trouble. Knowing what, where, and how to look also requires a knowledge of the various types of deterioration which can occur and of the basic causes of deterioration—and experience. This information is contained in Chapters 2, 3, 7, and 8, and before checking a suspect construction, the maintenance engineer should acquaint himself with those sections of the text. Step 2. Determine the Cause This is, by far, the most difficult and important step of all. It is not possible to evaluate the need for repair or to select a repair procedure with assurance of satisfactory results unless the cause is understood. This does not mean that the specific cause must be identified. Indeed, particularly with concrete, often this cannot be done, either because there are insufficient data to pinpoint the trouble or because there are several agents simultaneously at work. What can be done, though, is to eliminate possibilities until only a few remain and then select a repair procedure which will correct the existing condition and prevent further deterioration by any and all of the suspect aggressive agents. For example, Case History

8-1 describes a case of deterioration

of steel sheetpiling which

could have been due to four separate causes. Since it was not possible to determine which of the four agents was responsible for the damage, a repair procedure was selected that would prevent further deterioration by any of them. The results proved entirely satisfactory. However, inability to pinpoint the cause meant that the cost of the repair was somewhat greater than what it might have been if the cause had been precisely determined. This is generally the case, and normally, for economic reasons, it is advisable to make every reasonable effort to identify the probable cause or causes of the difficulty as precisely as possible. In this connection, it should be noted that failure to understand the cause of a defect can lead to the selection of a repair procedure which would be harmful, rather than helpful. There are no rules or set procedures for determining the cause or causes of deterioration. Each case is an individual problem and




and Repair of Structures

must be individually diagnosed. However, with experience, some patterns appear. For example, cracks in walls due to foundation settlements usually run diagonally. The cement paste of concrete subject to sulfate attack has a characteristic whitish, dead appearance. Cracks due to corrosion of reinforcement run in straight, parallel lines at uniform intervals and usually show evidence of rust staining. Also, one soon learns where to look for corrosion in steel structures and where to expect rot in timbers. In general, however, the diagnosis is difficult, and there is no help but to have an intimate knowledge of what can go wrong and to eliminate possibilities until some answer appears. This will require study of the data in Chapters 2, 3, 7, and 8, as previously noted. Just a few tips: Inspect the structure. Study it. Do not be ashamed to just sit and stare at it. Observe it in bad weather as well as in good. Compare it with other constructions inthe area or elsewhere, and try to analyze what there is in this case that is abnormal. Be patient. Repair problems are not solved by assembly-line techniques. Study the problem. Be thorough, and allow enough time to do the job. There are few bona fide emergencies in structural work, so resist being pressured into any snap decision. Also, it is important to investigate the problem sufficiently deeply to discover any hidden or latent defects and not just to repair the surface deterioration, meanwhile covering up some deeper -seated problem.

Step 3. Evaluate the Strength of the Existing Structure Usually, the structure being investigated is in active use, and it

is necessary

to determine, as quickly as

possible, if it is safe to con-

tinue to use it, or if the facility should be restricted to some less severe usage. If the strength has been extremely impaired, it may even be necessary to abandon it, or to install temporary, emergency shoring. Even if the structure is not in active use, its strength and safety still are of primary interest. Evaluating

the strength

of a deteriorated construction

can be a

major engineering problem in itself. Fortunately, however, in the majority of instances this is not the case, it being quite obvious that the strength is still adequate. For example, spalling of the cover over the reinforcement in a concrete slab or tied column is not serious, unless the bars are badly corroded, since the cover generally is not relied upon to contribute strength to the section and because the bond stresses are low. Similarly, deteriorated expansion joints produce leakage and/or local spalling of the concrete without causing any serious danger to the structure as a whole; or the deterioration may be limited to diaphragms, fascia plates, or similar nonstructural components.

Introduction and General Considerations


Another case of an obviously adequate structure, which is often encountered, is where the structure is currently supporting the maximum load of probable occurrence, as in the instance of a fully loaded parking garage or storage or warehouse structure, or in the case of a construction carrying active railroad traffic. Cases where the structures have become so distorted that the strength is obviously inadequate also offer no problem. It is where there is some question as to the adequacy of the strength that there is a problem, and for this problem there are

three methods of approach as follows:

a. Fixed-percentage Method. The first, and simplest, approach is to assume that all members which have lost less than some predetermined percentage of their strength (not section, but strength) are still adequate and that all members which have lost more than this amount of strength are inadequate, i.e., require repairs. This is a very common criterion. The allowable loss of strength is variously set anywhere from 15 per cent on up, depending on the structure, the original design criteria, and the conditions of use. The 15 per cent figure is widely used, and is a good figure for major elements of bridges and buildings which are proportioned on the basis of calculated stress. Higher values are applicable when evaluating the condition of piling, details such as stiffeners and bearing plates, and members where failure will not result in collapse. There are no general rules. The value to be used depends on the judgment of the engineer, on how closely the structure was originally designed, and on whether the design was based on an elastic analysis, was predicated on ultimate strength, or took into account plastic redistribution of the stresses. Was any allowance made in the original design for sacrificial metal? Have the loads to be carried increased or decreased? There are many factors to be considered. This method assumes, of course, that the original design itself was adequate. This is usually, but not always, true and is one of the weaknesses of the approach.

b. Analysis of the Actual Stress Condition.

The second approach

is to make a detailed stress analysis of the structure, as it stands, including allowances for loss of section where it has occurred. This is a difficult and expensive operation, but is worthwhile if it appears that extensive and costly repairs will be necessary. Usual procedure is to make a preliminary evaluation of the deterioration based on the concept of maximum allowable per cent loss of strength, as described above, and, if it appears that major repairs will be required, to reevaluate the strength based on a detailed stress analysis, which should consider all conditions contributing to such



of these




in Chapter

Deterioration, Maintenance, and Repair of Structures


Note 1. This analysis should go beyond that which normally would be prepared for an original design because, while an ultrarefined analysis is not economically justified in new construction (the few pounds of steel or yards of concrete which might be saved are not worth the increased engineering costs), where there is a repair problem, a little reserve of strength can mean the difference between


or no repair,




in lieu of re-

pairs, or may mean the difference between continuing the use of, or abandoning, the structure. Caution: If a repair is to be omitted or limited, or if a structure is to continue to be used on the basis that it has a reserve of

strength in excess of that indicated by the original design computa-

tions, the following conditions must be met: (1) The strength of the structure must be adequate to support, with a suitable factor of safety, not just the existing loading condition, but the maximum probable load, present or future. (2) Any possibility of the existence of significant residual stresses in the structure due to damage, or to other factors in its past history, should be explored. The presence of residual stresses tends to decrease the capacity of members to resist buckling and is an extremely important point, which is frequently overlooked (see

also Chapter 9). It is like being cautious about buying a used car

which has been in a bad accident. It may have been repaired ina manner that appears to be satisfactory, but may carry some latent defect which could cause unexpected and embarrassing consequences. If the structure has been distorted, do not force it back into alignment. Either take it apart, straighten the members under unrestrained conditions, and reassemble it, or assume that the distorted members

do not have

strength to support

any loads additional

to the loads which they are already supporting and strengthen accordingly.

(3) Any future deformations of the structure which will be re-

quired to develop the needed reserve of strength should be con-


These deformations should not cause an unsightly condi-

tion or cause objectionable side effects like closing of expansion joints, leakage due to heavy cracking, or exposure of reinforcement to rust. As examples: (a) movement of the structure will be required to develop passive earth pressure, and (b) deflection must occur before the structure can develop resistance due to catenary

action. (4)


reserve of strength being utilized (or the apparent re-

duction in applied load) must be real and permanent.

Do not count

on the passive pressure of earth, which may be inadvertently or unkowingly excavated; on weight, which may be removed; or on

sheltering by adjacent structures, which may be demolished.


Introduction and General Considerations


necessary, post a load limit, and enforce it. If the problem at hand is one of foundations, compute soil pressures on the assumption

that structures will be erected on all adjacent property.

(5) Further deterioration of critical members must be fully and effectively stopped. Members which have not been seriously damaged need not, necessarily, be protected, however. If the structure is to be entrusted to the care of a competent maintenance staff, or if the failure of the less critical members would not result in serious damage, their protection may be deferred with the resulting financial advantages described in Step 4. If such is not the case, then they too should be effectively preserved. (6) Where the required reserve of strength lies in some structural action not contemplated in the original design, the stress analysis must be predicated on this action. The details must be carefully checked. For example, check any lapped splices in reinforcing bars which are to act as arch ties to be certain that there is no possibility of loss of bond between the spliced sections. Check that there is something to receive the thrust from the arches or the tension from the catenary. Check the end-span conditions.

c. Load Test.

The third method for checking the structural

adequacy of doubtful members is a load test. Load tests may be required by the local building official, but they should only be performed where computation indicates that there is a reasonable margin of safety against collapse, lest the test bring the structure down. Load tests tend to show strengths much greater than computed strengths when performed on three-dimensional, i.e., actual, structures. This is partly due to the interaction of the adjacent framing, to the differences in the effects of long-term loading versus the short-term test loads, and to difficulties in correlating concrete cylinder strengths with in-place concrete strength, plastic redistribution of the load, and many other factors. But it is precisely because they do show greater strengths than are indicated by computation that they are desirable as an aid in evaluating the strength of the construction. In repair work, every little bit of strength is important. Accordingly, the use of load tests is recommended, but only with a full and clear understanding of their limi-

tations and range of applicability as discussed in Chapter Note 2.

Step 4.

Evaluate the Need for Repair

When the cause of the deterioration has been determined and the strength of the existing structure has been checked, a decision

must be made whether (a) to permit deterioration to continue, (b)

to take measures to preserve the structure in its present condition without any attempt to strengthen it, (c) to strengthen the construc-


Deterioration, Maintenance,

and Repair of Structures

tion, or (d) if deterioration is exceptionally severe, to reconstruct or, possibly, to abandon it. This decision is rendered on the basis of considerations of safety, economics, and appearance, subject to the following general principles. A different decision or approach may be appropriate for different elements of the same structure.

Case a. Analysis Shows That the Structure Still Has Adequate

Strength. If the appearance of the existing condition is objectionable or if further deterioration will produce an objectionably unsightly condition, repair the structure, now. If appearance is not a problem: (1) Put the condition under observation to check if it is dormant or progressive. (2) If it is dormant, neither repair nor preservative treatment is required; in fact, it is better left alone. (3) If it is progressive, check the feasibility and relative economics of permitting deterioration to continue and performing a repair at some later date, and of making the repair right away. First, consider the feasibility of deferring a repair. If the strength of the structure is just barely adequate, and deterioration is progressive, of course, repair cannot be deferred. Also, unless there is an organized and competent maintenance staff to whom the future inspection and care of the structure can be safely entrusted, repair cannot be deferred. Similarly, if it takes several years to appropriate maintenance funds and to perform the work, repairs must be made immediately. Assuming, however, that it is feasible to defer the repair, consider the relative economics of doing the work now, or later. If

the work is deferred, deterioration will become worse and the re-

pair will be more costly. Also, if the structure must be given some form of preservative treatment in order to defer the repair, the cost of this treatment will be in addition to the ultimate cost of the work. On the other hand, these increased and additional costs will be partly or wholly offset by the investment value of the expenditure which has been deferred. For example, assume that it would cost 50 per cent more to provide a preservative treatment for certain members of a structure and later to repair them than it would cost to repair them right away. This 50 per cent figure would represent the cost of the preservative



30 per cent),

an allowance

for the fact that

the repairs will be more expensive if deterioration is more severe (another 20 per cent), and an allowance for anticipated increases or decreases in construction costs. Assuming the investment value of the deferred expenditure as 6 per cent per annum, with C representing the cost of immediate repair, the economic picture is as follows:


Introduction and General Considerations Cost of preservative treatment ---- 0.30C Cost of immediate repair 1.00C --- 1.20C Cost of later repair Cost which would be deferred by use of preservative treatment now and repair later------------ 0.70C

The break-even point can be calculated from the usual compound interest formula

A=P(1+i)" where A

is the cost whose


is to be realized


P is the

amount of the deferred expenditure (0.70C), i is the interest rate

(0.06), and n is the number of years that the repair must be deferred to return the value of the increased cost of the work. For the case cited above, nis 13 years. If the preservative treatment will defer repair longer than this period, it is economically desirable to repair later. This principle is of primary importance and is considered in greater detail in the following chapters. Case b. Analysis Shows That the Strength of the Structure Cur-

rently Is, or Shortly Will Be, Inadequate.

Either repair or rebuild

the structure, abandon all or part of the facility, or consider a change of use. The first thing to decide is whether or not the deteriorated structure (or structural element) is really necessary. For example, parapets give a lot of trouble. As a result, whereas parapets were once standard features in building design, many buildings are now built without them. Instead, a fence or railing is provided, or ten-

ants are denied access to the roof.

Another source of headache is

canopies, which are frequently unnecessary. Surprisingly often, an owner will discover that he really does not need some building appurtenance, or even an entire building, when he finds out how much it will cost to repair it. Also, it may be found that the required loading conditions are not as severe as were contemplated in the original design, and the structure may be figured for a lesser live load. However, if the decision not to repair or to limit repair is predicated on an assumption of a reduced live load, be sure to record this decision for future consideration by others and, in the case of a building, to file for a change of occupancy. Assuming, however, that the facility is needed, a rule of thumb in frequent use is to repair it if the repair costs 50 per cent or less of the replacement cost and to replace the structure if the cost is greater than this percentage. Of course, this assumes that the

facility can be put out of service


it is being




Deterioration, Maintenance,

and Repair of Structures

times this is not possible and a repair must be executed, whatever the cost. If the decision is to repair the structure, proceed with Step 5. Step 5. Select and Implement a Repair Procedure Compared with Steps 1 through 4, the selection of a suitable method of repair is easy. It is almost entirely a matter of selecting the least expensive method which will adequately do the job. The following generalizations apply: (1) Economics does not mean first cost only. Consider the total cost, which includes first cost, maintenance costs, and the investment value of deferred costs, as previously noted.

(2) Do the job in time.

Do not wait for an emergency.

A proper

repair job requires consideration and care. This takes time. repairs get more expensive as the condition gets worse. (3)

If the defects



few and isolated,



on an

individual basis is indicated. If the defects are general, consider a basic revision of the design. (4) Be sure that the repair prevents further development of defects. Alternately, consider the possibility of incorporating a margin of excess strength in the repair to provide for continuing or future deterioration. (5) If the structure has been dangerously weakened, the repair must restore the strength, not merely impede further decay. (6) Determine whether appearance is a problem. If so, the number of applicable types of repair is limited, and it may be necessary to cover the repair with some form of architectural facade. Of course, any facade should not cover inner or latent defects or impede access to the underlying construction. (7) Make sure that the repair work, during construction, will not interfere with essential operations of the facility. If there will be such interference, provision must be made for the continuance of such operations, just as provision is made for maintenance of traffic in highway construction. This can be a big item of cost, time, and difficulty, and must not be neglected. (8) Repair frequently involves adding section to a member. This causes the member to be stiffened, and the distribution of live load and live-load moments in the structure will change, penalizing certain members and relieving others. These changes may be substantial and serious. Take care that they do not adversely affect other structures or parts of the same structure. A stress analysis of the revised condition should be made. In addition, do not interfere with drainage, clog weep holes, or create a vapor barrier which will trap moisture within, rather than exclude moisture from, the struc-

Introduction and General Considerations


ture. Be sure that expansion joints remain functional and that problems of access to utility lines are not created. As a general rule, do not change anything, omit anything, block any opening, or abandon anything unless its purpose is known. Even then, it is good practice, before blocking or abandoning any openings or utility lines, to get permission (in writing) from the owner and from any authority having jurisdiction. After selecting a suitable method of repair, and after considering all the ramifications of its application, the last step is to prepare plans and specifications and to proceed with the work. Sometimes, particularly where the engineering and construction are to be performed by the same firm, it is a temptation to proceed with the work without submitting to the expense of preparing plans and specifications. Unless an emergency exists, it is best not to proceed without preparing these documents. Except for the simplest cases, repair jobs involve accommodating a number of special conditions. Detailing such accommodation should be planned and considered beforehand. Field adaptations and changes are expensive and not always satisfactory.

C. WORKMANSHIP The principal cause of deterioration of structures is some relatively small carelessness or oversight in the details of design or

in the construction. Accordingly, if the condition has come to a point where correction is required, it is folly to permit similar carelessness in the repair work. The following chapters describe in some detail methods for repairing structures. Experience has demonstrated that, if good results are to be obtained, the specifications, practices, and steps described must be followed completely, carefully, and exactly in every detail. It is not always possible to give reasons why one detail or method of repair works and another does not. The engineer frequently is subject to pressure to modify some detail or method in order to make the work easier or more rapid, and sometimes finds himself in a position where he is required to defend the detail or method which he has specified without being able to present definitive reasons for having specified it. If he has selected a repair procedure with care and consideration and on the basis of experience

(whether his own or that of others), the engineer should be firm and not hesitate to indicate that some procedure has been specified on the basis of experience, and that another proposed procedure, although theoretically feasible, is undesirable because it is unproved. Except for simple work like plating, replacement of sections, or


Deterioration, Maintenance, and Repair of Structures

painting, it is advisable to call for prototype installations, i.e., samples of the proposed repair to be installed under conditions simulating those in the actual work. The contractor’s experience in performing the specified repair procedure may be limited, and possibly nil. Even if the contracting firm has experience, the workmen on the job probably do not. Also, since most repair procedures involve extensive manual operations, the foremen and workmen must be fully instructed concerning the procedures and the reasons for their use. The equipment, materials, and all details both major and minor should be tested. The prototype installation provides opportunity to satisfy all these requirements. The engineer should provide full-time, competent inspection of the repair work. Repair jobs are not routine, and the very best personnel should be assigned.

The specifications should contain a clause requiring dependable

and capable workmen, and the engineer should not hesitate to insist on conformance with its provisions. The contract should also provide for inspection of the completed work including provisions for removal and replacement of sections of permanent forms or removal and replacement of portions of the finished work. Payment for such removals and replacements should be provided in the contract, preferably on a unit price basis, with no payment where inspection following removal indicates the underlying work to be


One last point: It may be dreary, boring, and even annoying to have to be concerned with little, picayune details like wire brushing the concrete surface, wetting it, rubbing in a neat cement wash, placing concrete in half-inch layers, etc., but it must be done, or there is no sense in doing the job at all. Repair may not be a glamour job, but it is an essential one, one which requires very careful attention.

D. SUMMARY To summarize, the approach of an engineer to a problem of structural deterioration and maintenance should be identical to the approach of a doctor to his patient. It involves detection, diagnosis, and cure, with an overriding requirement for prevention. In order to proceed, the engineer must have a knowledge of the various forms of deterioration which are known to occur (the symptoms), the cause or causes of the various types of decay (the disease), and how to

correct the condition (the cure).

Ancillary to this is the need in new

or existing construction (the healthy structure) for proper design, construction, and maintenance in order to prevent decay.

Introduction and General Considerations


The reader will obtain maximum benefit from this volume by approaching it with the above in mind, acquainting himself first with the symptoms and diseases, and concerning himself with the cure only after the cause has been diagnosed.



1. Sources of Reserves of Strength in Structures It is often found that a badly deteriorated structure, apparently greatly reduced in strength, continues to support the applied loads with no discernible indications of distress. Some of the more important reasons for this are as follows: (1) Structures are commonly proportioned on the basis of stylized procedures of computation called structural design. These procedures are intended to represent, as nearly as possible, actual physical conditions as they occur in the structure. Because of mathematical difficulties, however, certain simplifying assumptions are made. These assumptions are, necessarily, conservative and result ina margin of safety in the construction. (2) Members are proportioned for maximum stress conditions. The section required at points of maximum stress, frequently, is carried for the full length of the member to minimize the costs of fabrication or form work, or for aesthetic reasons. If the deterioration of a member is localized and does not occur at a point of maximum stress, the strength of the section may not be impaired. (3) The design is usually based on an elastic analysis. Reanalysis on the basis of ultimate strength, of plastic redistribution of moments, or of relief of moment based on the concept of the beamline ? frequently will indicate a greater capacity. (4) Design live loads are seldom realized in practice. For example, a common value for design floor loading in a garage for passenger vehicles is 75 pounds per square foot. The actual equivalent uniform load does not exceed 40 pounds per square foot. A similar condition applies with office or residential occupancy. Even in industrial and storage structures, the design load specified in a building code seldom includes a consideration of the necessity for aisle space. The actual and the design loading conditions should be compared.


The design may incorporate excess strength by way of sacri-

ficial metal; by sizing the members to the next heavier section, or to a lighter but stronger section; or by satisfying requirements for minimum thickness of metal or maximum deflection. (6) The strength of the structural materials, particularly con-


Deterioration, Maintenance,

and Repair of Structures

crete, may increase with age, or, if it is a steel structure, there may have been some strain hardening. In fact, whereas it has long been customary in steel design to relate load-carrying capacity to the properties of the material as determined from standardized tests on small specimens, it is now well known that the mechanical properties of steel may be significantly improved by cold working during forming operations. In fact, this increased strength is commonly utilized when proportioning designs of light-gauge metal. The actual strengths of materials can be checked by taking cores or

coupon specimens.

If checking for the effects of cold working, bear in mind that the

effects in a relatively thin, wide section (like a wide flange) will be

greater at the corners of the flanges than in the web. In a compact section, the increased strength may occur more uniformly.

(7) The structural action may differ from that conventionally

assumed for purposes of design. Ordinary beams and slabs are a common case in point. These are proportioned on the basis of flexural behavior. However, except for large ratios of span to depth, pure flexural action is not achieved, and the member resists the load, at least partly, by arching or catenary action. Witness, for example, conventional design coefficients for draped mesh slabs.° The amount of reinforcement required is substantially less than that which would be computed on the basis of flexural theory. These relatively short-span slabs act, essentially, as catenaries. Similar action can occur in any concrete beam or slab where there are bent up bars or inclined stirrups (see Figure 1-1a). Other common examples are subway construction using shortspan wall or roof arches and short-span, heavy types of construction such as piers, wharves, and the like, where the span/depth ratio is small. For such construction, the spans can act as tied arches with the bottom reinforcement acting as the tie. All that is necessary is

that the bottom reinforcement be anchored in or be continuous over

a point of restraint such as a fixed-pile bent or pier, which is the

usual case (see Figure 1-1b).

In steel designs, also, plates exhibit a reserve of strength due to catenary action. In fact, it is common to take advantage of this behavior in designing the plates for the hulls of ships. (8) In some cases the design load represents a temporary or construction condition, and the service loads are of lesser magnitude. For example, consider a retaining wall. If the wall is well drained, maximum lateral pressure will occur during and shortly after backfilling, the active pressure decreasing with time. Another example is that of a hydraulic fill. The lateral pressure decreases as the fill drains. Borings will help in evaluating actual, in-place soil properties at the time that repair is required.

Introduction and General Considerations


Inclined $ tirrups.

Negative Reinforcement

Positive Reinforcement | apport

1) Construction

ines of Tensile Stress





lof P stress follow pattern of





2) Structural Action Resultiny trom Development of Defects Figure




action due to cracking of reinforced









Support 1) Construction

ines of Compressive




Stress (Arch




‘Spalled Surface Lines of Tensile Stress (Arch Tie)

Exposed Reinforcin, Bare 2) Structural Action Resulting from Development of Defects



Arch action due to spalling of reinforced concrete member.


Deterioration, Maintenance, and Repair of Structures

(9) Conventional design procedures do not consider all actions of the structure, such as arching of the fill behind an anchored bulkhead wall




in the frictional


of a

cofferdam fill due to application of a lateral load;* composite action of the floor finish with the slab and framing, of the walls with the framing, or of embedded or fireproofed steel framing;® and many other conditions. (10)





of stiffness

incorporate excess material.



than strength and, accordingly,

on the basis

Tierod Reaction

Ky= Tan? as)

1) Construction


Sheet Pile Bulkhead


Active Pressures predicated on

2) Design Forces

Figure 1-2a. Changes in forces flection of the sheetpiling.

Redistribution of active ressures due . |to deflection of ulkhead and ching action in the fill. Note that moments in sheet piling are reduced, but thag tieri fengion ie ine creased

3) Actual Forces

acting on a sheet-pile bulkhead due to de-

Steel Sheet Piling MPUS Section

Notes: 1) Design strength predicated on each sheet pile acting individually. Scetion Modulus = 5.4 in. per foot of wall. 2) interlocks rust tight and can develop shear, neutral axis shifts to N,A. 2 and Section Modulus is increased to 7.1 in > per foot of wall, an increase of 32%. Figure


Increase of strength of the steel sheet-pile wall due to aging.

Introduction and General Considerations


2. Validity of Load Tests as Indicative of Structural Strength® It is a matter of common experience that load tests consistently show a structure to have substantially greater strength than is indicated by computation. Assuming that the load test is of the usual type, i.e., the criteria for satisfactory performance are that the gross and net deflections shall not exceed some fixed amount or some function of the span, the principal reasons for this apparent discrepancy are: a. Nonelasticity of the Structure. The load-deflection curve of the structure—especially of timber and reinforced concrete members, but also of steel members—is not elastic except at relatively low loads. This is due to nonlinearity of the stress-strain curve, to

slip in the connections, and to other factors.

b. Volume Changes. Time-dependent deformations occur in concrete and timber structures due to influences other than externally applied loads. Such influences include creep, shrinkage, temperature change, and possible chemical effects. These deformations do not occur during the relatively short duration of a load test, but do occur in actual use.

c. Frame


Structures are usually analyzed as two-dimen

sional frames when actually they are three-dimensional frames. The variations in the distribution of moments and torsions through out the building may be substantial and exert strong influences on the amount of the deflection of the members. d. Floor Diaphragm Action. The floors act simultaneously with the three-dimensional frame in carrying the load. This action is by no means as simple as the procedure of delegating certain portions of the floor to act with the frame members by the usual Tbeam theories. Even for an ideally isotropic, elastic mechanism this problem has been solved for only rather simple cases. The contribution of the floor system to the deflection of a member is presently indeterminate, but it is known to be substantial. e. Contribution of Walls. The presence of walls and partitions stiffens the structure in a manner similar to the floor diaphragms. This contribution, normally, is neglected in design. f. Moment Redistribution. The development of plastic hinges and the redistribution of the moments in a continuous frame begin to take place well before the ultimate load is reached. Some tests indicate that this redistribution starts at approximately one-half of the ultimate load capacity. This naturally varies with the structure and the details involved; however, this phenomenon may take place at the relatively small deformations that occur under the usual load test. Load-carrying capacity and deflections would not have any simple relationship in this circumstance.


Deterioration, Maintenance,

and Repair of Structures

g. Modulus of Elasticity. Knowledge of this constant is essential for the calculation of the theoretical deflection. For steel it is reasonably well known and constant. For concrete or timber it is not. For example, the modulus of elasticity of concrete depends on whether the concrete is in a wet or dry condition. The variation

may be as high as 30 per cent in favor of the dry concrete.


concrete in a building being load tested is usually in a dry state, and this factor would have an appreciable influence upon the deflection. This same argument applies to varying degrees of seasoning in timber structures. Also, again for concrete, the modulus of elasticity is a function of the square root of the compressive strength. The strength increases with time. Accordingly, the structure gets stiffer as it gets older. h. Actual Strength of Steel. In order to meet minimum ASTM requirements, steel is produced with yield strengths in excess of that required. Values 20 or 30 per cent above the minimum yield

values are not uncommon.

This influences the deflection as well

as the strength (of both steel and concrete members)

the stiffness. i, Location, Type, and Amount of Reinforcement.

by affecting

In a reinforced

concrete member, the stiffness is a function of the amount and location of the reinforcement. This is especially true for variations in the center half of the span. Also, the negative bars at the ends act the same as haunches and reduce somewhat midspan deflections. Moreover, arch and catenary effects result from the presence of

the bars.

In addition, most concrete structures are designed to be underreinforced. Tension failures rather than compression failures (although both ultimately result in a compression failure of the concrete) are desirable. Consequently, a load test is essentially a test of the reinforcement, not of the concrete strength, and the dependence of the measured deflection on the amount and position

of the reinforcement is accentuated. Conclusion: The primary function of a structure is to carry load.

Logically, the true measure of its load-carrying ability is a load test. There are, however, two complications. First, the load test does not reflect the effects of time-dependent variations in the properties of the structure. Second, since it cannot be carried to fail-

ure and since the results of load tests on actual structures almost

invariably cannot be related to the results of computation, the load test does not indicate the safety factor present in the structure, and, as noted in the text, a minimum safety factor should be maintained. Accordingly, it is suggested that a load test be accepted as a

Introduction and General Considerations


valid indicator of structural strength only if (1) the applied test load equals or exceeds the anticipated load, in use, times a suitable safety factor, and


the test load remains

in place


is observed

long enough so that some idea of the probable effects of creep, etc., can be evaluated. Note: The above discussion is applicable to load tests as a measure of the requirement for repair and is not intended as a criterion for acceptance

of new



Elmer E. Gunnette, ‘‘Lessons Learned from Construction Failures,’’ Proceedings of the National Engineering Conference, American Institute of Steel Construction, 1961. 2. Basil Sourochnikoff, ‘‘Wind Stresses in Semi-rigid Connections of Steel Framework,’’ Transactions of the American Society of Civil Engineers, vol. 115, p. 382, 1950. 3. Section C26-620. 0 (d) of Building Laws of The City of New York, 4.

5. 6.


‘‘Field Study of a Cellular Bulkhead,’’ Journal of the Soil Mechanics and Foundations Division, Proceedings of the American Society of Civil Engineers, February, 1962, p. 67, discussion by S. M. Johnson. R. H. Wood, ‘‘Studies in Composite Construction,’’ Part II, National Building Studies, Research Paper No. 22, Her Majesty’s Stationery Office, London, 1955. J. S. B. Iffland, ‘‘Commentary on Load Tests’’ (Unpublished commentary prepared in connection with preparation of Building

Code for City of New York.)




In this and in subsequent chapters it will be assumed that the design has been executed with sufficient care that defects due to overstress are not a problem. With this premise, there are five basic types of deterioration to be considered when dealing with structures of steel. These are corrosion, abrasion, loosening of connections, fatigue, and impact. The symptoms and basic causes of these deteriorating agents are as follows: 1, Corrosion Corrosion may be defined as the conversion of metals, by natural agencies, into compound forms and is, by far, the major maintenance problem in steel structures. Deterioration due to this agent may be readily distinguished. The symptoms are a pitted, oxidized surface, usually showing loose flakes or scales of oxide and a typically reddish-brown rust-colored appearance. Figures 2-1 through 2-4 show some typical conditions. The nature and mechanism of corrosion are set forth in detail in Reference 1. In brief, there is a chemical or electrochemical reaction which converts the metallic iron into an oxide or other compound. In the case of steel, this compound generally is poorly 20

Steel Structures


Figure 2-1. Typical example of corrosion. The photograph shows corrosion of the anchor bolts and bearing plates for a bollard support.





of corrosion.




a condi-

tion of corrosion found at the intersection of the bottom chord and the post of a through truss bridge. This is usually a vulnerable location, since there

is a tendency

to form

a trap

for dirt and debris,

as shown.

bonded to the parent metal and flakes off easily, and the area of the section is reduced. The decreased section and the accompanying stress concentrations correspondingly reduce the load-bearing capacity of the member. Resistance to fatigue loadings (repetitive stress conditions) also is impaired. In extreme cases, the member may snap or buckle.





Repair of Structures

Figure 2-3. Corrosion of cover-plated stringer and diaphragms of highway bridge due to a leaky relieving joint. A relieving joint in the deck slab was located right over the line of diaphragms. The joint leaked, causing the condition shown. Note that the splice plate on the underside of the flange has been buckled by the corrosion (A) and that the heads on the rivets closest to the joint have been greatly reduced in size (B). The top flanges of the diaphragms also have been severely attacked, but the portions of the stringers remote from the location of the joint are in good condition. Joints are a weak spot in any construction and, in inspection for deterioration, should be checked first.

Figure 2-4. Corrosion of channel cap on a steel sheet-pile bulkhead. The use of steel channel caps is an ill-advised detail. Concrete caps are much more satisfactory.

Steel Structures


It is not necessary that the designer or the engineer concerned with the maintenance or repair of structures have a detailed knowledge of the mechanics of corrosion. Those few generalizations which are pertinent to structural work are summarized in Chapter Note



it is necessary

to realize

that corrosion

occur, and to know where it is most likely to occur, it, and how to repair the condition when it develops. are


in Sections

B and C

of this



how to prevent These matters

2. Abrasion (Erosion) Deterioration of steel sections due to abrasion also may be readily identified—and may be easily distinguished from deterioration due to corrosion—by the worn, smooth appearance of the abraded surface. Where the abrading agent is no longer active and has been supplanted by a corroding agent, the evidence of abrasion is less obvious, but usually can be detected by the general depression of the abraded area as compared with the surrounding section. Abrasion of steel structures is associated with the working of moving parts

in contact;




to wave



with portions of members immersed in a moving fluid. The case most frequently encountered is that of groins or piled structures in the surf zone. In fact, on a sandy beach, the continual wash of water having a substantial content of suspended sediment can cut through a steel section a half inch or more thick in a few years. Flue linings, particularly where the flue gases have a high ash content, are another frequent source of difficulty. Flumes also are often damaged in this manner. In desert locations, exposed steel may show a polished surface after a storm, due to abrasion by wind-borne dust and debris. Abrasion of steel, whatever the direct cause, can be prevented or corrected by armoring and streamlining as described in paragraphs 4 and 8 of Section B. 3.

Loosening of Connections

Rivets and ordinary bolts in connections for steel structures subject to shock or impact loading tend to work loose with time. The magnitude of the problem may be appreciated by reference to Ruble,’ who quotes a figure of 800,000 field rivets redriven each year by maintenance crews during the repair of railroad bridges

in the United States.

Loosening of the connections induces slip in the joints, causes distortion of the structure, creates areas of extreme stress con-

centration, and increases the vulnerability of the structure to


Deterioration, Maintenance,

and Repair of Structures

fatigue failure. Accordingly, every effort should be made to prevent such occurrence, and if connections do come loose, they should be promptly repaired. Crane supports, railway structures, and supports for reciprocating machinery are particular sources of difficulty in this regard. Little or no trouble is encountered in buildings and other structures subject to static loads or where the dead load/live load ratio is large. Accordingly, the connections of steel structures and elements thereof which are subject to impact loading should be checked periodically using standard inspection procedures used for checking new work. Loose rivets should be cut out and replaced, either with new rivets or with high-strength bolts, and loose bolts should be retorqued or replaced. There is some indication that structures assembled with highstrength bolts, if the bolts are properly installed, are relatively proof against loosening of connections, even under severe service conditions. Experience is still somewhat limited, however, the first extensive installations of this type of fastener having been

made circa 1950. 4,


Fatigue may be defined as the fracture of a structural member due to repetitive, fluctuating load occurring at stresses at or below usual allowable design values. Fatigue occurs in the same types of structures and under the same conditions described in paragraph 3, Loosening of Connections. The symptoms are small fractures, oriented perpendicular to the line of stress, and are serious sources of danger, largely because the resulting fractures may be extremely difficult to detect. If not detected, however, fatigue cracks may result in collapse without warning. Accordingly, members subject to repetitive loading must be diligently inspected. Repair of members showing fatigue cracks requires restoration of the lost strength. Such strengthening usually is effected by plating. 5. Impact Exposed steel sections are sensitive to damage from the impact of moving objects, much more so than are sections of concrete or heavy timber. This sensitivity is due to the use, in steel design, of sections having relatively thin flanges and other projecting elements. Impact damage is characterized by local distortion of the affected members, usually in the form of a crimp or a bow of short wavelength. However, buckling of compression members and flanges due


Steel Structures

to overstress also occurs as a bow, sometimes also of short wavelength. It is essential to differentiate the two, inasmuch as impact damage is more or less superficial and may be repaired readily, while buckling due to overstress is evidence of more deep-seated problems, which may require modification of the design. Sometimes, the differentiation is not easy to make. Of course, if the distortion occurs

as a sharp


i.e., a peaked

or pointed


or occurs

in a tension member or tension flange, then the trouble is probably impact. On the other hand, if an S curve has been formed with the arcs of the curve displaced on either side of the original axis of the member, then the cause is probably buckling. Between these extremes, positive identification of the cause of distortion is difficult, unless it can be established by questioning those familiar with the history of the structure that an impact has, indeed, occurred. If there is doubt, a stress analysis should be made of the affected members. Where such analysis indicates that the distorted mem-

bers or portions of the members

are stressed close to the buckling

range, take the pessimistic view point: Assume that a condition of overstress exists, and make the repair accordingly. Cases of impact damage are common. They may be repaired by strengthening the member, either plating or encasing it. If the impact could possibly recur, and if the weight and bulk permit, encase-

ment is the better solution.

B. PREVENTIVE MEASURES 1. Keeping the Structure


Corrosion will be much accelerated if dirt or debris is allowed to accumulate in contact with the member. The reason is that the dirt or debris retains rain or wash water (or may even soak up moisture from the atmosphere) and maintains this moisture in contact with the steel surface. Further, in industrial areas the accumulated dirt often consists of soot containing a heavy concentration of corrosive sulfur compounds. Also the accumulation of dirt tends to hide underlying defects and make proper inspection difficult. No matter how well the steel may be painted or otherwise protected, it is essential to sweep, blow, or hose off the dirt; frequently from accessible parts, periodically from parts which are less accessible. The importance of regular inspection and cleaning in preventing corrosion is well established. In fact, regular cleaning and inspection are more important than painting. Do not make the mistake of thinking that the accumulated dirt forms a protective coating.




and Repair of Structures

2. Painting Painting is the general, all-purpose method for protecting steel against corrosion and is used for all applications other than those involving special problems of accessibility, thin section, large surface presentment, or particularly severe conditions of exposure. For these latter problems, additional protection—in the form of encasement, use of corrosion-resistant alloys or sacrificial metal, or cathodic protection—is required. Paint technology and application is a science in itself, and a detailed discussion is beyond the scope of this book. Accordingly, the following discussion is limited to a few fundamentals which are required for background. When faced with a specific problem, the reader should consult Reference 3 (‘‘Steel Structures Painting Manual’’). This manual contains specification provisions covering the various coating systems, surface preparation, application, and other considerations involved in painting, and is so organized that, knowing the environment and the type of structure, the user may select a painting system suitable for the application. a. Surface Preparation. Prior to painting, the surface should be clean and free of loose contaminants. This involves solvent cleaning to remove grease, oil, and dirt, followed by removal of scale and rust (by hand scraping or by wire brushing if loose, or by use of power tools if more adherent). Very tight scale and rust must be removed by pickling, blast cleaning (sand or shot), or flame cleaning. Criteria for selection of the proper method and related specification provisions are contained in Reference 3. The life of the coating is largely dependent on the degree of surface preparation,



and thorough

cleaning producing


system having substantially increased length of protection. b. Types of Paint. A thorough discussion is contained in Reference 3. It is necessary that the various coats of paint and the several components of each coat (primers, thinners, pigments, vehicles, etc.) be compatible with each other. One way to insure this is to require that they be like products of the same manufacturer. c. Application. After cleaning, the prime coat should be applied

before the surface can become recontaminated. In the case of flame cleaning, this means before the surface has cooled to or below the air temperature the metal)

but not while

(so that there will be no condensation on

the surface

is so hot as

to damage


primer or coating material. Except for certain emulsions, the surface should be as dry as possible, and should be neither very cold nor very hot. Usual limits on application designate that painting shall not be carried on when the surface temperature is below 40°

Steel Structures


or above 140°F. Each coat of paint must be thoroughly dry before application of the next coat. Paint may be applied by brush or spray with equally satisfactory results. Dipping, mopping, roller coating, or flow coating are used only when authorized or for certain types of bituminous coatings. d. Inspection. This is an important requirement that is occasionally overlooked. The cleaning operation should be carefully observed to assure its thoroughness and completion. Once they start spreading paint, painters do not like to be interrupted, and sometimes paint gets sprayed over dirt and other items not thoroughly bonded to the steel surface.

The painted surface should be checked after each coat for the

occurrence of pinholes, holidays (the omission of paint from local areas), and blisters, and these should be touched up. Each coat

should have

a detectable variation in color shade to facilitate in-


The thickness of the several coats should be checked. Paintfilm gauges (both wet and dry) are available for this purpose. Electrical flaw detectors are available for the detection of small defects. The final surface should be checked for damage before acceptance and final payment. e. Repainting. In repainting, be sure that the new paint is compatible with the existing paint and that the old coating is ‘‘touched up’’ or ‘‘spot painted’’ before applying the new. The proper interval for repainting may be determined by the fixed-percentage method, as described in Chapter 1, when, say, 15 per cent of the painted area shows distress. Do not apply several successive repaintings without removing the old coatings, because a heavy paint skin will tend

to check.

3. Other Coatings a. Bituminous Paints. The formulation, applicability, and application of bituminous paints are discussed in Reference 3. The basic considerations are similar to those described in paragraph 2, Painting.


Zinc Coating.

This coating is used where longer life protection

is desired than can be provided by usual methods not a permanent

the galvanized




of painting.

and in moist,


(It is


coating itself is usually protected with a good-quality


is also useful

for subaqueous

it gives fairly good protection. For structural work it is customary



to specify zinc coating by the


Deterioration, Maintenance,

and Repair of Structures

hot-dip process (galvanizing, ASTM A-153), because the resulting coating is thicker than that applied by other processes such as sherardizing, electroplating, or spraying. This is desirable, since the protective


of zinc coatings

is approximately



their thickness. All fabrication, especially bending, burning, and welding, should be performed before galvanizing, because these operations will crack or burn the protective zinc.

4, Encasement The previous paragraphs described the prevention of corrosion by the application of relatively thin coatings. The trouble with thin coatings, however, is that they must be renewed at more or less frequent intervals. Permanent or semipermanent protection may be provided by encasing the entire member with concrete or plastics, or by sheathing it with nonferrous metals or other noncorroding materials. a. Concrete. Concrete is often used for the encasement of steel sections. If the weight and bulk are not objectionable and if the encasement is properly constructed, economical and very satisfactory


can be provided.

Concrete encasement is most frequently used for protection of waterfront structures in the tide zone and below the waterline, for buried pipe structures, for lining of pipes, for filling pipe and tubular columns, for protection of structural elements which will be inaccessible in the final work, and for structures subject to particularly corrosive atmosphere such as those exposed to locomotive blast or those in chemical plants. An example is shown in Figure


In addition to corrosion protection, an encasement of concrete is used for the prevention of damage due to abrasion and, of course, for fire protection. Placement of the concrete coating may utilize forms, or the concrete may be sprayed on. If forms are used, the provisions of Chapter 5, Section B, paragraphs 3, Filling the Forms, and 4, Finishing the Jacket, are applicable. For subaqueous installations, the provisions of Chapter 5, Section B, paragraphs 2, Forms, and 5, Remarks, apply. For sprayed-on applications, the provisions of Chapter 5, Section C, should be followed. The thickness of the concrete cover over the steel should conform to the minimum requirements established in Chapter 3, and a good bond between concrete and steel should be assured. To this end, the steel surface should be cleaned of all loose rust, scale, or dirt, and the concrete should be a dense, sound, rich mix. If the quality of the concrete is poor, the resulting protection will be poor.

Steel Structures


Some form of beam wrapping (usually a wire mesh of thin rods at two- to four-inch spacing) must be provided. The AISC specification* states that steel sections to be encased in concrete shall not be painted. However, in the author’s experience, for exterior exposures, it is advisable to paint or coat the first 6 inches of embedment to provide against localized moisture penetration along shrinkage cracks. Cautions: (1) Concrete encasement is not a concrete fill. Use at least 3,500 pound per square inch controlled-quality concrete, and insist on first-class workmanship throughout. (2) Even a well-constructed encasement of concrete will not protect the steel from corrosion due to electrolysis, and where the steel may be required to discharge stray currents to the ground, some form of grounding system must be provided. b. Reinforced Bituminous Coatings (Wrappings). These provide excellent protection against corrosion and are widely used for the encasement of buried members in highly corrosive soils suchas river bottoms, marshlands, cinder fills, fills containing organic debris suchas garbage, and in tidal regions. Wrappings are particularly used to protect pipe, the tie rods and fittings that anchor re-

taining structures, and elsewhere where high-quality, long-lasting

protection is desired. Installation consists of coating the metal surface to be protected with hot coal-tar primer and enamel and covering it with one or more spiral-wound wrappings of felted or other suitable fabric, saturated with a waterproofing, bituminous mixture. The fabric serves to reinforce and prevent injury to the coating and to increase the coating thickness. Detailed and excellent specifications covering the installation and materials for reinforced bituminous coatings are contained in NAVDOCKS Specification 34Y, Department of the Navy, Bureau of Yards and Docks. The coating should be protected during backfilling by covering the protected member with a half round section of tile, sheet iron, or similar device or, for vertical or inclined surfaces, by a protection course of brick or block. The protected members should not be in contact with massive clayey or silty soils, which tend to adhere and contract upon drying and may produce rupture of the coating. Also, the member should not rest on rocks, boards, or similar debris, because the coating may be punctured by working back and forth over the hard spot as the structure breathes. It is well to specify a backfill of sandy material around the encasement. Where the member to be coated is long and flexible, provide intermediate supports at, say, 30-foot intervals to prevent sagging of the member. c. Other Materials. Concrete and asphalt wrappings are, by far,




and Repair of Structures

the most common materials used for encasement of steel. However, certain other materials have found occasional use for specialized applications. For example: A packing of urethane foam covered with a polysulfide liquid was used for corrosion protection of the cable clamps during operations of double-decking the George Washington Bridge.* The foam was used as a caulking and filler and the polysulfide liquid as a waterproofer. An 18-gauge sheathing of nickel-copper alloy has been used to cover and protect steel sections of several offshore drilling rigs for installations in the Gulf of Mexico. Monel metal is regularly used to sheath piling and framing of offshore mooring platforms in the tide range and in the splash zone. Sprayed-on vinyl-plastic sheeting was originally used for ‘‘mothballing’? Armed Forces material and equipment in the period immediately following World War II. In more recent years, the technique also has come into use for fireproofing and protection of steel. There are a number of compounds of proprietary manufacture which are satisfactory for this use, and reference should be made to the manufacturer for details and specifications pertinent to the specific application. There are also MIL Specifications (Bureau of Ordnance) which describe the basic materials. However, several of the coatings of this type are not permanent and require rejuvenating applications at periods, usually of several years, and this must be considered when selecting a formulation for use.

5. Corrosion-resistant Alloys The rate of corrosion of steel is a function not only of environment but of composition as well. For a given environment, the rate of corrosion of a steel structure may be decreased by the use of corrosion-resistant alloys in lieu of ordinary carbon steel (ASTMAT or A36). a. Atmospheric Exposures. Applicable corrosion-resistant alloys include copper-bearing steel (ASTM-A7, A36, or A373 with a minimum copper content of 0.2 per cent) and certain (not all) of the highstrength, low-alloy steels similar to ASTM-A242. Copper -bearing steel may be specified, using the ASTM designation and modification, as indicated above. Most of the low-alloy steels applicable for purposes of corrosion resistance are proprietary items, however, and the engineer should inquire of the individual manufacturer in order to select a product suitable for his

purpose. It is not sufficient to specify ASTM-A242. steels show a resistance to atmospheric ior to that of copper-bearing steel.

Some low-alloy

corrosion which is super-


Steel Structures

Except under extremely severe exposure conditions, usual practice is to use corrosion-resistant alloys, which cost a penny or two a pound more than A36 steel, in critical members only, rather than in a general application for an entire structure. For example, their use is indicated (1) for moving parts and assemblies such as expan-

sion plates, for end dams, and for rocker and roller bearings; (2) for assemblies of relatively small members such as gratings, which expose a large surface area and would therefore, be expensive to maintain; (3) for members which would be relatively inaccessible in the final construction; and (4) for scuppers, trash racks, and similar members, which are subject to particularly severe exposure conditions due to frequent wetting aggravated by frequent packing with wet or damp debris. In addition, corrosion-resistant steel is sometimes specified for portions of marine structures, or for structures in aggressive industrial atmospheres such as occur near pickling tanks and the like. b. Buried and Subaqueous Installations. Although only limited data are available, it appears that the use of copper-bearing or high-strength, low-alloy steels has little advantage over the use of ordinary carbon steel for buried or subaqueous installations or wherever the steel is continously wet. However, alloys high in silicon, chromium, or nickel do have markedly superior corrosionresistant properties for such applications and theoretically may be used where their cost can be justified. Caution: Except for certain special architectural applications, corrosion-resistant steels should be painted, galvanized, or otherwise protected, as are ordinary carbon steels, the use of corrosionresistant metal being, principally, an added precaution provided to reduce the maintenance problem associated with members in severe exposures and with members which are difficult of access.

6. Sacrificial Metal This term denotes thicknesses of metal which are provided over and above the requirements of stress. The purpose is to provide an allowance for corrosion loss in excess of the structural requirements for the section. The areas of application are the same as described above for corrosion-resistant steels; indeed, insofar as purely structural applications are concerned, the two techniques are similar. The minimum thickness requirements found in many codes and design specifications for steel structures are ramifications of this principal. For example, Standard Specification for Highway Bridges (AASHO) calls for minimum 0.400-inch metal in webs of steel piles and




in pile splices.


other design


Deterioration, Maintenance, and Repair of Structures

specification have similar provisions elements.

in connection with foundation

Usual procedure when considering the use of sacrificial metal is

to calculate the member size, based on stress; estimate the rate of loss of metal due to corrosion (published data on this subject are plentiful); and increase the thickness of metal by an amount equal to the rate of loss times the desired service life of the construction. Be sure to allow for loss of metal from both faces of the member, and consider that the rate.of loss on each face may be different, as, for example, where there is fill against one side of a section of sheetpiling. The caution cited under paragraph 5 (Corrosion-resistant Alloys) with regard to painting or galvanizing of the metal is applicable.

7. Cathodic Protection Corrosion in aqueous environments or in damp soil is primarily electrochemical in nature and is due to a current passing from anodic areas of the metal into solution and returning to the metal at cathodic



type of corrosion

can be prevented

by impres-

sing a countercurrent on the metal in a sufficient amount to neutralize the aggressive electric currents. Cathodic protection, which consists of the electrical connection of a sacrificial anode to the structure to be protected, serves this function by neutralizing the corroding current and forming layers of insoluble reaction products on the new cathodic areas. In structural applications, cathodic protection is used, principally, for protection of buried steel (such as pipe or piling), for protection of the submerged portions of marine structures such as piling and bracing, for protecting lock and dam gates, for the interiors of water tanks, and for the exteriors of buried tanks. Cathodic protection, however, will not prevent corrosion of structures unless the metal to be protected is surrounded by an electrolyte such as water or damp soil and is ineffective in protecting structural elements above the low-water line or in very dry soil. The principles of cathodic protection are discussed in some detail in Reference 8 with a ‘‘do it yourself’’ approach. However, although the principles are simple, in order to be effective and economical, their application



and experience.



of a reputable corrosion engineer are indicated for major installations. Caution: In cathodic protection, the effects of the induced currents on adjacent structures must be considered. These adjacent structures may be damaged unless they are adequately bonded to the new system or other means of protection are provided.

Steel Structures


8. Armoring Armoring is used as a means for protecting the structure against damage due to abrasion. Concrete encasement, as described in paragraph 4, is an excellent form of armoring. Other devices include wrought iron protection plates, timber facings, the provision of sacrificial metal, claddings of hard metals, and the use of abrasion-resistant paint coatings such as vinyls, neoprene, and baked phenolic finishes. (Provisions relating to these abrasion-resistant paint coatings are contained in Reference 3.) Where there is a real problem with deterioration due to abrasion, encase

the members

in concrete.


protection plates

also provide excellent service, but are expensive. The other devices are, at best, suitable only for light to moderate exposures. When armoring is to be provided, streamline the encasement. The importance of streamlining is noted by Schaufele (Reference 6), who describes experiences with H piling in the breaker zone, where steel piles of structural shape were cut off by abrasion due to sand in suspension, but rounded shapes survived. 9. Influence of Design Details Much difficulty with deterioration of steel structures can be avoided by attention to the design details, particularly: (1) Either parts of the structure should be accessible for maintenance, or, if not accessible, the members should be encased or provided with some form of permanent protection—unless located in a heated and well-protected interior and in an area remote from plumbing and from potential sources of leakage through roofs or sidewalls. Inaccessible dead spaces under roofs, under the ground floor, or in closed abutments, should not be permitted. Provide access doors. If the space is too small for a man to enter and work, consider such a space as inaccessible and protect the members, as

described above.

Inaccessibility is one of the most troublesome characteristics that can be built into a structure, especially one of steel. (2) Select structural shapes which will have a minimum of exposed surface. For example, from the standpoint of minimizing corrosion, a T section is better than double angles, and a box beam is better than an H section. Where stress conditions permit, compact sections are preferable to expanded members. (3) Avoid shapes or details which will catch dirt or debris. If economically feasible, use rounded sections. (4) Eliminate pockets, low spots, and crevices which will trap





Repair of Structures

water. Channel sections should have the legs turned down, should be provided with drain holes which are large enough not to be clogged with leaves or debris, or should be filled with concrete. If there are web stiffeners on the section, drain the pockets between stiffeners (Figure 2-5). Wide flange and I sections which are used with the web horizontal should be pitched to drain or should be provided with drain holes. (5) Column bases should be protected with concrete encasement or pedestals projecting above the ground line or floor level. Pitch the adjacent concrete surface to drain away from the steelwork. Concrete construction supporting or overhanging steel framing should be provided with drips, scuppers, and drainage grooves, as described in Section B of Chapter 3. (6) Slots and holes in horizontal surfaces of the steelwork must be either plugged so that they do not clog or opened up enough to provide positive drainage. If concrete is used to fill a hole, be sure to use an expanding mortar lest shrinkage of the concrete permit moisture penetration throughthe shrinkage crack which tends to form between the fill and the edge of the hole. Figures 2-6 and 2-7 show a detail of this type which failed, and the corrected detail. Drain Hole

Drain Hole (between stiff ener plates)

Use of Drain Holes (Boxed Channel Sections)

Useof Drain Holes

(Channels Used Back to Back)

Use of Concrete Fill Figure 2-5. Draining structural sections. shapes is similar.

Treatment of wide flange or I

Steel Structures


(7) Avoid details which include narrow crevices which cannot be sealed or painted. A common case is the use of double angles or channels placed back to back but spaced apart by the gusset. In exposed structures, this detail invariably gives trouble unless the joint between



is sealed




(8) For riveted or bolted joints, or for sections placed back to back, inhibit water penetration between adjacent plates or shapes by assuring that all adjacent metal surfaces are drawn up tight. For bearing-type connections, prime the faying surfaces before assembly. (9) Full-penetration butt welds are preferable to lapped joints, partly because of water penetration between the plates and partly because it is impossible to remove all the flux from the faying surface. The flux adds to the corrosive attack of any intruding moisture. The same thing applies to partial-penetration butt welds. (10)


or tubular



be concrete-filled,

or sealed,

airtight. (11) To avoid galvanic corrosion be sure to isolate dissimilar metals, particularly the aluminum and copper alloys commonly used for mullions, muntins, flashing, and trim. Zinc chromate paint or washers and gaskets impregnated with zinc chromate are excellent for this purpose.



the two



be of

stainless steel or be provided with some neutral bushing. (12) The spaces around steel members should be either ventilated

or sealed.

Observance of the above details will involve little or no increase in cost of the structure and will result in substantially decreased



C. REPAIR PROCEDURES The previous section has dealt with precautions in design and


in maintenance


will prevent

or inhibit the occur-

rence of deterioration in steel structures. Assuming, however, that significant deterioration has occurred and that decision has been made to repair the deteriorated elements, the following methods are applicable. Note: Repairs to steel structures almost invariably involve some degree of strengthening. Accordingly, the provisions of Chapter 9 (Strengthening an Existing Structure) apply and must be considered when applying the techniques described below.


Deterioration, Maintenance,

and Repair of Structures

(a) Figure 2-6. Railing posts heaved by frost penetration. (a) An overall view of the condition. (b) A detail of one of the posts. Note that the post has been


about 4 inches.

into the socket

in which

The condition was caused by the penetration of water the post was





is shown

in Figure 2-7. A similar condition is caused by insufficient travel in the expansion joints in the railing. The compressive forces created in the rails develop

a vertical




the posts.

In this case,

the fact

that the end post is heaved is the clue that this has not been the case. These photographs show the subtlety of damage that can result from moisture penetration

and the need

to seal all potential


of such


1. Plating (Doublers) Where abrasion and corrosion are localized in a few members or portions of members of a structure, a convenient and economical way to compensate for the loss of section is to splice new metal across the areas of deterioration. This can be accomplished by taking plates or rolled sections, running them alongside and past the deteriorated


or member,




into the struc-

ture in areas where the structure is still sound. Figures 2-9 to 2-11 show some typical details. This procedure is known as plating, or, where the replacement metal consists of sections similar to those of the original structure, the added sections are sometimes referred to as doublers. Plating is also convenient for repairing members which have cracked, buckled, or suffered local crushing. The technique is not applicable where the appearance of the additional plates would be aesthetically objectionable. If such is the case, replacement is a better solution.

Fabrication and installation of plating follow standard procedures

for new



that care


be taken to provide

close fit between the new and existing members.






Steel Structures


fe-Railing Post

Moisture penetrated along interface and mortar fill. In steel between freezing weather, this moisture se photographs, froze. For results,

Railing Post (Cut and Reset)

New Base Plate

New Anchor Bolts. (set in

Metal Sleeve

Non-shrink Grout)

Mortar Fill Original Detail

Improved Detail Figure 2-7.


Deterioration, Maintenance, and Repair of Structures

Gap as required to g


fit gueset tt




ER. G0

we Detatt

‘Laring Bare

|Continuous rin y




4 fa me



Betler Detail

Bede Dotan

Figure 2-8. Details for double-angle members used in exposed structures. Details for double-channel members are similar.

be ground and fitted to the fillets of the existing sections.

ing is most desirable. Before

Seal weld-

installation of the plates, the contact surfaces between the

new and existing members

should be cleaned and the whole assem-

blage cleaned and painted after plating, as will be described under

Miscellaneous Considerations.

Particular attention should be given

to those surfaces of the existing member which will be in contact with the new metal in order not to trap any rust, scale, or dirt between the two sections. The contact surfaces should be primed and

painted with special care before plating, using additional coats of paint as judgment dictates. A question which frequently arises is what to do with the existing members which are no longer needed for stress. Usual practice is to leave the member in place. This both eliminates the cost of re-

moval and takes advantage, in the repaired structure, of whatever strength the existing member may still have. Caution:

Plating is not necessarily

a permanent



encased or otherwise permanently protected, corrosion, both of the original section and of the new plates, will continue. Basically, unless protected against further corrosion, this method is a form of sacrificial metal, and excess section must be provided to allow for future corrosion loss.

2. Replacement Where the entire member is severely corroded, if there is no

room to add plates or doublers replace the member.

or if appearance

is a consideration,

Steel Structures

39 occurs between

tio and replace existing rivets Jas required to install new angles. High strength bolts may be substituted for rivets.

Corrosion of web members usually occurs betwe: angles, only. Flanges are steeply pitched, shed dirt, and do not corrode.

a) Elevation



‘Existing Angles

to make up current Jand anticipated loss of strength. ..

Replacement Sections (See Note on Section A-A ) f

existing angles.

b) Section A-A

¢) Section B-B

(Section A! - a! similar) Figure 2-9. Plating details for strengthening truss composed of doubleangle members. Note that the introduction of the replacement sections may create an increased eccentricity in the connection. This must be considered in proportioning the member.

3. Encasement with Concrete This


is described

in the previous section as a means

preventing deterioration due to corrosion and abrasion.


It is also a

good, effective way to reinforce a deteriorated section, provided


Deterioration, Maintenance,

and Repair of Structures

Two Plates




Two Angles

Lepilte a) Strengthening a Flange



Typ.) Channel Sections Plug Welds Fills ae required

b) Strengthening the Web Figure 2-10.

plate girder.

° Strengthening the Entire Member

Plating details for strengthening wide flange section or welded

For I sections, provide bevel fills as required to match the

shape of flanges. Where reinforcement is placed on one side of the member only, check for torsion due to unsymmetrical bending.

that there is clearance in the structure for the enlargement of the section, there is strength to carry the added weight, and the encasement will not be unsightly. 4, Miscellaneous Considerations a.




have deteriorated, they may

be repaired by plating or replacement, as described above. Where only the fasteners are deficient or deteriorated, they may be replaced, by reaming out the holes and inserting stronger fasteners, or welds or high-strength bolts may be added. Where the connections are merely loose, if bolted, the bolts may be retorqued except for high-strength bolts, where

retorquing is not

Steel Structures









a) Strengthening

the Flange

b) Strengthening the Web


Flange e each eid New Stiffener Sections

Remove existing stiffeners and replace ae shown.

Limits of Corrosion of Flange

Length as required to transfer streas from existing to new flange angles

c) Strengthening the Flange where it is not Accessible for Adding Cover Plates Figure 2-11.

Plating details for strengthening riveted plate girder.

desirable. High-strength bolts which work loose should be replaced. Loose rivets also should be replaced.

Beware of corrosion of the main structural material under the

rivet heads or under the heads, nuts, or washers of the bolts. Where this occurs, ream or drill the corroded material away from around the hole, and install oversized bolts or rivets. b. Angles or Channels, Back to Back. Corrosion between members placed back to back is a relatively common problem. If the loss of metal is not so severe that strengthening is required, further difficulty can be prevented by blasting out the rust and scale

(i.e., with a sand blast), and seal welding to prevent intrusion of

moisture (see Figure 2-12). ce. Bearings. Figures 2-13 and 2-14 show a condition found when inspecting a steel-stringer bridge. This condition is not common, but it is not rare either. In fact, articulated components of structures (like bearings and expansion joints) frequently are subject to unanticipated movement. Repair

may be effected as shown in Figure 2-15.


Deterioration, Maintenance,

and Repair of Structures Blast out rust,

dirt and debris, and seal weld.





Corrosion will attack in these locations.

Blast out rust, dirt and debris, and seal weld. Original Condition


Figure 2-12. Detail for repair of localized corrosion of angles or channels used back to back. For other conditions see Figure 2-9.

Figure 2-13. Excessive movement in a stringer bearing. The sole plate (A) has been displaced to the extent that it is about to fall off of the bearing plate (B). This condition was caused by a backward tilt of the abutment, plus insufficient travel in the expansion joint. However, the stringer projected sufficiently back of the end of the bearing plate that it was possible to effect a repair by installing a safety plate between the stringer and the top of a new


shelf formed

on the abutment





Steel Structures

Figure 2-14.


General view of condition shown

in Figure


The dis-

placed bearings are indicated by the arrows. The backward tilt of the abutment is shown by the relative inclination with respect to the bearing stiffener on the end stringer (A). Note that a section of the top of the wall

(location rested

B) has been torn loose by the drag of the roadway slab which

on the wall.

the end of the bridge wall.


detailing eliminates

this occurrence

slab on the end diaphragms,

by resting

rather than on the back-

Another thing to watch is the condition of the bearing surfaces and of the guides, if it is an expansion bearing. The contact between the sole and bearing plate is intended to be a finished surface. After several years of exposure, these surfaces are likely to be pitted by corrosion. If this pitting is severe, the plates may be removed, ground or milled smooth, and reinstalled (using shim plates to make up the loss of thickness due to grinding). If the guides and guide grooves become worn, there will be a tendency for the bearing to slew from side to side unless there is a heavy dead-load reaction. This can be repaired by depositing weld metal, as is done for hard facing. If the expansion plates are locked, check the pintles on the fixed plates. The longitudinal forces which are intended to be dissipated

by movement

of the expansion



to shear

the pintle,

or, more often, the sole plate will ride up on the pintle, bending it. d. Painting. Whenever a steel structure is repaired, clean all existing surfaces to bare metal, then make the repairs, and, finally, give all new and existing surfaces a suitable protective coating of


Deterioration, Maintenance, and Repair of Structures

New Stiffener

Original End


of Beam



Displaced Center Line of Bearing

L Gap : Heer Bearing


Rolled Section

(f Requited to velop Shear)

}— Anchor Bare New Bearing


Drill and Grout|

Anchor Bars Figure 2-15a. plate.


for correcting


of stringer bearings—safety

paint or other material. The philosophy is that if there has already been trouble with the structure to the extent that repair is required, more trouble is likely to occur, and it is worthwhile giving both the new

and existing metal

D. SOME 2-1.

a good




A Deteriorated Pier

a. Type and Cause of Deterioration. Figure 2-16.shows a cross section of a pier having a steel framed substructure (H piles with single-and double-angle bracing in and above the tide range). Figures 2-17 to 2-19 show typical views of the observed deterio-


Damage to this structure was the result of corrosion. The deterioration followed a clear and consistent pattern. A band of heavy corrosive attack occurred from about 1 foot below mean high water to a level about 4 to 5 feet above this mean, i.e., in the upper tide range and in the splash zone. In this range, many of the steel sec-

Steel Structures



New Stiffener Relocated Sole| Plate Displaced Position of jole Plate



Displaced Center Line Bearing

|Original Center Line lof Bearing Figure 2-15b. Method for correcting cate sole plate. Connection -

movement of stringer bearings—relo-

Single Angle Diagonal bracing

Concrete Deck


ansverse Horizontal Bracing

ngitudinal Bracing

Steel Piles



Typical cross

section of pier described

in Case

History 2-1.

tions were corroded entirely through, many others were reduced to mere slivers of sheet metal, and all had suffered more or less severe loss of section. Below the mid-tide range and down to the mud line, corrosion was moderate, consisting principally of heavy, but localized, pitting with a total loss of section of about 15 to 25 per cent, which is substantial, but by no means critical, considering the


Deterioration, Maintenance, and Repair of Structures

(a) Figure 2-17. of a typical



pile before



of steel piles. (a) The condition

of the rust scale.

of the web and flanges after its removal. about 75 per cent, without failure.

(b) and

(c) The


The section had been reduced

low working stresses used for piling. Below the mud line, the pile sections were virtually free of corrosion. Above the band of heavy deterioration, corrosion was relatively light, except for certain areas. These exceptions also followed a significant pattern. At the head end of the pier, where wave slap against the bulkhead caused a heavy spray condition, the upper framing was heavily corroded. A similar but less severe condition occurred at the outboard end of the pier and along the side exposed to the prevailing wind, again due to splash resulting from wave action. b. Repairs. (1) The bearing piles were jacketed with concrete

(see Figure 2-20).

The jackets extended about 2 feet above and

below the limits of major corrosive attack in order to effect a transfer of load from the steel to the concrete section and back again. The sections could have been plated, but the extent and the amount of underwater work, as well as considerations of future maintenance, made this uneconomical and undesirable. Note the handwork required to place the jackets around the intersecting bracing.

Steel Structures


Figure 2-17b.




Deterioration, Maintenance, and Repair of Structures

Figure 2-18. condition



History 2-1—condition of transverse braces.


moval of the rust scale.



of the rust scale.

(b) The


(a) The

after the re-

By taking advantage of the stiffening effects of the concrete

jackets, it was possible to eliminate the diagonal bracing (except for a few bents at the outboard end and in the middle of the pier, where the bracing was restored by the use of doublers). (3) Isolated sections of the upper framing, where deterioration was severe, were plated to restore their strength.

Steel Structures


(4) All steel framing above mean high water which was not en-

cased was given two coats of an asphaltic paint. c. Discussion. This case presents several notable points. First is the range of occurrence of corrosion and the variation in degree of its severity. Severe corrosion occurred in the upper tide range and in the splash zone. Where the splash zone reached a higher level, because

of increased


or wave


the severe


rosion also reached a higher level. This is the usual case and provides a measure of the limits within which steel framing for waterfront structures should be provided with special protection, i.e., in the tide and splash zones. Due consideration should be given to the exposure conditions, including fetch and probable wave height, when estimating the extent of the splash zone, and conditions producing increased wave slap (such as proximity to the bulkhead, firewalls, or similar



not be neglected.

As will be noted

in the following chapter, proximity to the splash zone also has a deleterious effect on concrete construction, and, where severe exposure conditions occur, the deck framing (or the deck level, if feasible)


be raised

as high as possible.

Above the splash zone, deterioration was light. Where the framing was continually submerged, corrosion was moderate, and below the mud line, the member had not beendeteriorated. These, too, are the usual conditions, although occasionally there will be a band of severe corrosion at the mud line and another just below the lowwater level. The effects of the presence of marine growths on the framing are discussed in the Chapter Notes. A second point to be noted is that the loss of section of the piling

below water was about 15 to 25 per cent.

Although this was sub-

stantial, it was not serious, because piling is proportioned at relatively low working stresses. Many designers are impatient with these low working stresses, which apply not only to piling and to marine construction, but to certain types of industrial constructions and to exposed framing in general. They see substantial economies to be effected by applying the design stresses and practices common in building work. This impatience is not justified. Deterioration is a serious and very real problem. It occurs, and until such time as the art and practice of maintenance reaches a much higher level than at present, the designer must assume that if trouble can occur, it will occur. He must provide enough margin so that the structure will not collapse before someone has the opportunity to realize that something is wrong and has the time to do something about it. The third thing to be considered is the elimination of most diagonal bracing. Within practical limits, whatever protective measures

might have been taken, this framing always would have been a


Deterioration, Maintenance, and Repair of Structures

(a) Figure 2-19. Case History 2-1—corrosion of steel beam. (a) The appearance of the beam before removal of the rust scale. As will be noted, except for some pitting and tuberculations, it looks relatively sound. (b) The same view after the rust scale had been partly removed by hammering. Note that the entire appearance has been changed, revealing a very serious condition of corrosion. The flange has been reduced to sheet metal, with holes indicated by the arrows. These photographs illustrate the fact that the severity of corrosion in a steel member should not be judged until after the rust scale has been removed.

maintenance problem. Obviously, the best thing to do was to eliminate it. Fortunately, in this case, it was possible to do so. In many cases it is not, but the principle remains. The best solution to a problem of deterioration is to eliminate the source of the problem. The


of repair


in his work,


the original


sign, correct any deficiencies in that design, and, insofar as possible, improve the design details. 2-2.

A Deteriorated


a. Types and Causes of Deterioration. Figure 2-21a shows an elevation of an old-time, rolling lift bridge of medium span. At the time of investigation this structure was about 45 years old, and had

Steel Structures




been damaged by corrosion, impact, and a general warping of the structure as a whole. Heavy corrosive attack occurred in several locations. The stiffeners and bottom flanges of the track girders of the fixed span

(location A in Figure 2-21a) were heavily attacked, the outstanding

legs of the bearing stiffeners being entirely corroded through. This corrosion was due to an accumulation of dirt, leaves, and debris on the shelf formed by the bottom flanges of the girders. The accumulation acted like a sponge to retain water, thereby hastening deterioration. A similar condition prevailed with the bottom flanges of the floor beams and with some of the gusset plates for the lower lateral bracing. The bracing itself, which consisted of tension rods, was in good condition, the round rods shedding the water and debris as contrasted to the flat surfaces of the gussets and the main framing, which tended to accumulate the dirt. The stringers were less severely corroded, a number having been previously replaced. Other important trouble spots in this structure were the sidewalk brackets and the diagonal braces for the upper chords of the trusses









angles, 3/8 inch back to back, and connected to gussets of that dimension. The angles were severely corroded.


Deterioration, Maintenance,

and Repair of Structures

sting Concrete Deck


Existing Braces


\Cancrete or


umped Concrete Fill

#8 Reinforcing Bar Section A-A

Hand- packed ‘Concrete

Existing Braces to Remain

[Pumped Concrete Fin

fexisting Steel Pile

v Elevation Figure 2-20.

Detail of pile repair described

in Case History 2-1.

Warping of the structure consisted of about a 1 1/2-inch lateral

displacement of the center line of the bridge at the open end of the span and the elevation of the end of one truss at the open end of the

cantilever about 3 inches above the other truss.

Steel Structures





Floor Grating brore

> Trotk Girder re.

Figure 2-21a.

Longitudinal section of bridge described in Case History 2-2.

Figure 2-21b.


section of bridge described in Case History 2-2.

b. Repairs. The loss of strength of the corroded members and of the buckled flanges of the members damaged by impact was restored by use of plates and doublers, using details similar to those shown in Figures 2-9 to 2-11. The steel work was then painted. Since the abutment at the open end of the bridge had to be rebuilt for reasons other than deterioration of the steel framework, the warping was corrected by matching the bearing plates on the new abutment

to the warped


of the span,


than by trying to

force the steel framing back into alignment. c. Discussion. The significant points to be noted in this case


as follows: (1) Deterioration due to corrosion could have been substantially reduced


by improved


including simple cleaning and


Deterioration, Maintenance, and Repair of Structures (2)



attack occurred

in the very



would be suspect. For example, the projecting flanges of the beams, the gussets, and other flat horizontal surfaces, which retained water and debris, were severely corroded. The rounded surfaces, vertical surfaces, and overhanging surfaces, which shed water and debris, were in good condition. The double-angle members were heavily attacked. (The particular susceptibility of such members to corrosive attack has been noted previously and their use is undesirable for exposed



The warping of this structure is also significant.

riveted framed



subject to an impact load. with cantilevers.


not perfectly


the cantilever,



This is not usually important, there

warp if the load is not precisely uniform variable


Bolted or

In this particular

is a tendency,



either to

or to creep if the load is case,

one truss

of the

movable span was loaded more heavily than the other because of the presence of the sidewalk. This truss ended up 3 inches lower than the other,


the center

heavily loaded side.



shifted to the more

Some Building Problems

Except where near

line of the bridge


there may



be a leaky roof or an exterior wall, or area,


of a steel building

frame is seldom encountered, simply because the construction is sheltered from the weather. However, where the framing is exposed,

(for example, a roof tank or the framing near a large door), it suf-

fers as in the cases previously cited. In addition, an important problem, which is more often encoun tered in the maintenance of building structures than with exposed constructions, is that small, subtle bits of deterioration can have disproportionately important consequences. Figure 2-22 illustrates such a problem. The figure shows a large closed shed in which exposed, inclined braces are provided to resist the lateral loads. The shed, including the exposed braces, when inspected, was in excellent condition


for the anchor

bolts at A.


being close

to the

ground and on a concrete base which was not positively pitched away from the base of the framing, were practically corroded through. The shed and the equipment it contained were worth millions. The anchor bolts were worth only a few dollars. Yet the deterioration

of this small detail endangered the entire building.

Another point of interest. which is also peculiar to enclosed structures, is that deterioration is likely to be localized and, worse, concealed. For example, a small roof leak may corrode right through the chord of a roof truss without being detected until it is

Steel Structures


Gusset ‘Horizontal Plate

Bolts virtually

corroded through inthiearea



Stiffener Plates wow

See Detail A

Detail A Figure 2-22.

Details of structure described in Case History 2-3.

too late, particularly if the space under the roof is not used. Similar corrosion can occur in embedded wall framing and other encased members. Accordingly, detection is the real problem in mainte-

nance of buildings.

Reference 7 (‘‘The Durability of Lightweight Types of Steel Construction’’) describes a survey made in 1940 of lightweight steel framing members in buildings in varying areas and climatic exposures within the United States. The survey showed that, with some exceptions, the steel was in good condition. The exceptions were in steel members over a chemical storeroom, in the flanges of exterior exposed members, in members exposed over a shower room, in a beam beneath a leak in a shower room, and in members installed in a damp pipe trench. These locations, of course, are precisely where one would expect trouble and illustrate the need for maintenance


in critical


E. CHAPTER NOTES 1. Some Pertinent Generalizations about the Occurrence of Corrosion (1) The rate of corrosion is influenced by the chemical composition of the steel and may be reduced by the use of certain corrosionresistant alloys, as described in Section B, paragraph 5.

(2) Corrosion can take place in the presence of dissolved oxygen




and Repair of Structures

as well as in the free atmosphere; i.e., corrosion can occur in buried and submerged exposures. Corrosion in submerged locations is



to the dissolved


content of the water.

For a proposed installation, where there are limited or no data on corrosion rates, some idea of the corrosion problem can be obtained by comparing the dissolved oxygen content of the water at the site under consideration with that at other locations where the performance of submerged steel is known. The influence of the other factors mentioned



also must

The presence

be considered,

of moisture

of course.

is essential for the occurrence


corrosion. For a given exposure, corrosion will be more severe the more humid the atmosphere. Exposed and unprotected steel struc-

tures are quite durable in arid regions but are rapidly attacked in humid climates. It is of interest to note that corrosion on the sunny side of the structure, i.e., the south side or the side which is not shaded. is less severe than on the north or shaded side. This is because the moisture on the sunny surfaces is more rapidly evaporated. The quantity of moisture required for corrosion to occur is not great. Corrosion reactions will proceed, albeit slowly, on surfaces where

the film of moisture is present in such minute quantity as to be inyisible. (4) The rate of corrosion increases with increasing temperature.

Particular attention must be given to tropical and high-temperature installations. (5)


in velocity

or turbidity of the corroding


causes an increase in the rate of corrosion, principally because it causes more oxygen to be brought into contact with the metal. (6) For a discussion of submerged corrosion, see Case History 2-1, (7) In marine installations, the presence of a limited growth of barnacles or similar organisms tends to increase the amount of pitting, partly because of chemical reagents originating from the organisms and partly because of an uneven or erratic growth which promotes the occurrence of local galvanic cells. However, heavy growths


to suppress


by holding corrosion


in place and protecting the surface from contact with the oxidizing agents. Marine growths should not be indiscriminately removed. (8) In tropical atmospheres, fungi do considerable damage to metal structures, and application of a fungicide is indicated. (9)



in submerged

be caused by the presence (10)

In subaqueous

and buried

of accidental,




stray electric currents.


for marine


it is frequently observed that the precipitated rust forms a dense and adherent coating, which creates an effective barrier against further corrosion. This coating is usually a greenish-black color as

Steel Structures


contrasted to the usual reddish brown of ‘‘rust.’’ It is usually this phenomenon that accounts for the seemingly improbable lack of corrosion of exposed reinforcing rods often observed in deteriorated concrete piling (see Figure 3-33). This coating should not be removed when repairing the piles (see Chapter





Dissimilar metals in contact in an electrically conducting create a potential

source of local corrosion and should be

(12) Note that different environmental conditions can prevail in different parts of the same structure. The design (or the repair) should be tailored to reflect these differences.

(13) Corrosion of buried structures is much accelerated by the presence of corrosive soils. Soils known to be corrosive include organic deposits such as peat and organic


silt, and river and harbor

Cinders also are highly corrosive because of the presence of

residual combustion products, which,

produce acids.

in the presence of moisture,

Garbage fills are also very common and are partic-

ularly troublesome.


Frank N. Speller, ‘‘Corrosion: McGraw-Hill

Book Company,

Causes and Prevention,’’ 3d. ed.,

New York,


2. E. J. Ruble, ‘‘Symposium on High-strength Bolts: Part IT,’ Engineering Conference, American Institute for Steel Construc3. 4.

tion, 1950. ‘‘Steel Structures

Painting Manual,’’ vols. I and II, Steel Struc-

tures Painting Council, Pittsburgh, Pa.

“Specification for the Design, Fabrication and Erection of Structural Steel for Buildings,’’ American Institute for Steel Construction, adopted Apr. 17, 1963.

5. ‘‘Urethane and Polysulfides Help Protect Bridge Cables,’’ Engi6.

neering News-Record, May 10, 1962, pp. 54, 55. J. F. Schaufele, ‘‘Erosion and Corrosion in Marine Structures, Elwood, California,’ Proceedings of the First Conference on Coastal Engineering, Council on Wave Research, University of California, p. 326.

7. “The Durability of Lightweight Types of Steel Construction,”’ American Iron and Steel Institute, 1958. 8. Lindsay M. Applegate, ‘‘Cathodic Protection,’’ McGraw-Hill Book Company,

New York,



A. INTRODUCTION There are three basic visual symptoms of distress in a concrete structure: cracking, spalling, and disintegration (which may be defined as a general decay of the surface involving loss of the cement paste and loosening of the particles of coarse aggregate). Each of these basic symptoms in itself is fairly obvious and may be readily detected and differentiated from the others. However, each occurs in several forms, each form having a different significance. Moreover, in a given structure, the three basic indicators of distress may occur not only in combination, but with several forms of each symptom being manifest simultaneously. As a result, diagnosis of the cause of deterioration of concrete is a very subtle problem and a much different problem from that of diagnosing the cause of distress in a steel or timber structure, where the relation between symptoms and cause is usually quite clear. As described in Chapter 1, this problem is solved by assessing all possible causes of the observed condition and eliminating possibilities. To apply this procedure requires, first, a list of the 58

Concrete Structures:


of Deterioration and Preventive Measures


agents and processes which cause deterioration of concrete and some understanding of how they act and affect the concrete matrix. These data are set forth in Section B. The next step is to diagnose the probable cause using the aforementioned process of elimination (Section C), and the last step is to select and implement a repair procedure. The reader should approach the following text in the steps thus described, not proceeding to the diagnosis until the mechanics of deterioration

are understood


not attempting

until the diagnosis has been completed. B. CAUSES

Table 3-1.




incident to construction

causes of deterioration in

Causes operations


Internal settlement of the concrete Setting shrinkage Premature removal of shores


Temperature stresses a. Variations in atmospheric



Variations in internal temperature


of moisture

by the concrete

5. Corrosion of the reinforcement a. Corrosion due to chemical agents b. Corrosion due to electrolytic attack


of Deterioration

Localized settlements of the subgrade Movements of the formwork

Drying shrinkage








Weathering Shock waves Erosion

a repair


For a tabulation of the more concrete, see Table 3-1.


to select


Poor design details a. Reentrant corners b. Abrupt changes in section ec. Rigid joints between precast slab units



og) 1D



and Repair of Structures


Leakage through joints Poorly detailed drips and scuppers Inadequate drainage

Insufficient travel in expansion joints Unanticipated



in piers, columns,

or abutments


Incompatibility of materials or sections . Neglect of plastic flow



in design

1. Occurrences

Incident to Construction Operations

Improper procedures or carelessness during any phase of the construction operation may result in concrete of inferior quality. Such concrete will be more susceptible to deterioration than that produced by strict adherence to ‘‘good practice’ and, of course, should be avoided. While it may open the gate to other aggressive agents, however, poor construction practice is seldom in itself a direct cause of deterioration, except for the following instances:

a. Localized Settlements of the Subgrade.

If there are local soft

pockets in the subgrade on which the concrete is placed, or if there are any air pockets or hollows under the building paper, there will be a localized settlement of the concrete due to the weight of the plastic mass. If this settlement occurs after finishing of the con-

crete surface, cracks will ensue (see Figure 3-1).

The occurrence

should be prevented by giving proper attention to compacting and draining the subgrade. Keep the workmen from walking on the building paper. Eliminate trapped air pockets. Also, cracks of this type will be closed when finishing the concrete surface, unless the final finishing pass is made directly after placing the concrete.



of Concrete Pour


Depression of Subgrade due to Weight of Wet

Subgrade of Concrete Pour


Figure 3-1.

Cracking due to settlement of subgrade during construction.




of Deterioration

and Preventive Measures


Therefore, insofar as is practical, delay the final finishing for as long as the concrete surface remains workable. The cracks, having been closed, do not reappear. b. Movements of the Formwork. Any movements of the formwork which occur between the time that the concrete begins to lose its fluidity and the time that it has fully set will cause cracks to appear in the structure. These cracks may be internal and invisible by surface inspection. As such, they are potentially dangerous in that they form a water pocket in the concrete mass, which, upon freezing, will spall the concrete surface. Equally dangerous, corrosion of the reinforcement can result from such water pockets. Prevention of such Displacement of Form

Surface of Concrete


Concrete Form


Crushing of Wale

Form ‘orm Ti Tie

Displacement of Form.

(b) Figure 3-2. concrete.

Cracking due to movements

of the forms during setting of the


Deterioration, Maintenance, and Repair of Structures

cracks requires concrete

that surfaces of the form

be coated

to prevent



in contact with the

of moisture

and consequent

swelling and that the forms be properly designed (see Reference 1), particularly with respect to the details and deflections. The permis-

sible rate of rise of the concrete level in the forms should be noted

on, or accompany, the formwork drawings, which should be prepared. Nails, preferably, should be in shear to prevent loosening. Check all details and workmanship periodically during the pour. c. Vibrations. Cracking of concrete due to vibrations which take place during the set is a common occurrence. The source of vibration may be passing traffic, pile driving, blasting, delayed vibratory compaction, or accidental vibration of the forms caused by impact of equipment or by careless workmen. To prevent this occurrence,

do not let the workmen use the forms as a workbench while the concrete is setting. Do not let the concrete truck back against the forms. In granular soils (loose or medium compact), keep the pile driver at least 50 feet away from the concrete being placed. In massive clay or silt formations, do not drive piles in the area until the concrete has fully set. Do not walk heavy equipment past the pour while it is hardening. When working adjacent to a heavily traveled highway, consider detouring traffic. d. Internal Settlement of the Concrete Suspension. Fluid concrete, before attaining initial set, is subject to settlement of the heavier particles through the fluid matrix. Since the surface concrete hardens first, if such settlement is prevented locally, cracking will occur. Such settlement will be prevented by the presence of reinforcement, which is supported in place. This causes surface cracks as shown in Figure 3-3a. Worse, where the reinforcement consists of a heavy mat of closely spaced bars, instead of surface cracks, a plane of general separation may be formed under the mat (see Figure 3-3b). The presence of such a plane of separation invites damage due to freezing of water or due to corrosion of the steel. Prevention. Surface cracks, such as are shown in Figure 3-3a,

can be closed by delayed finishing.


the curing opera-

tion as soon as possible after placement of the concrete is also beneficial in that the set of the surface concrete will be delayed, thereby reducing the amount of differential settlement between the surface and interior volume of the suspension. This is one of the reasons

that prompt





is important on hot, windy days or in arid climates. dense,



is recommended.


of the concrete

The use of a



is a must. Delayed finishing and prompt curing will not prevent or correct the formation of an internal plane of separation, however. Accordingly, for thick masses and heavy reinforcement, the engineer should

Concrete Structures:


of Deterioration and Preventive Measures


Surface of Concrete

ele AY bee Lp


Reinforcing Bar

of the Concrete due to Settlement (a) Movement of the


Concrete due to Settlement ees

of Concrete


Separation in Concrete

b) Figure 3-3. sion.

Cracking of concrete due to settlement of the concrete suspen-

consider the formation of such a plane of weakness

inthe structure

as a probability and provide for its occurrence by the use of revibration or by decreasing allowable bond stress. In this connection, attention is called to the provisions of the ACI Building Code (Reference 2) for reduced allowable bond stresses on top bars in slabs more than 12 inches in depth and for top bars in footings. These provisions reflect the probable occurrence of such

a plane of separation. e.

Setting Shrinkage.


changes during the initial setting of

the concrete tend to cause the formation of shallow surface cracks.


Deterioration, Maintenance,

and Repair of Structures

These cracks have the characteristic appearance of alligator scales. Preventive measures are the same as those described in paragraph d for preventing surface cracks. f. Premature Removal of Shores. With the modern pressure for speed and economy in construction, there is a tendency to remove the shores or the forms before the concrete has attained sufficient strength. When this occurs, the concrete often cracks, sometimes severely. The solution is simple. Leave the shores and forms in place


until the concrete

is strong


Drying Shrinkage

The chemical reactions incident to the hardening of concrete occur over an extended period of time (perhaps several years) and involve a decrease in volume, known as drying shrinkage or simply shrinkage. If the structure is restrained against the free occurrence of these volume changes, stresses are created which may cause cracking of the concrete mass. Prevention. The amount of drying shrinkage can be reduced by using stiffer mixes and leaner mixes, Type I or II cement in lieu of Type III (high-early-strength), and standard-weight in lieu of lightweight aggregates. With a given mix and type and brand of cement, however, little can be done to reduce the amount of drying shrinkage other than (1) to eliminate the restraints on the structure by the frequent use of construction and contraction joints, (2) to provide an adequate amount of reinforcement to distribute and reduce the size of those cracks

which do occur, and (3) to chill the aggregate and mixing water.

With the latter technique, the trick is to lower the mean temperature of the concrete during placement to a level below that of the atmosphere, effectively producing a temperature rise to offset the equivalent temperature drop conventionally assumed to represent the

shrinkage (see Reference 3).




a. Variations in Atmospheric Temperature. Variations in the temperature of a hardened concrete mass will result in changes in shape and volume of that mass. If the free occurrence of such changes in shape and volume is prevented by restraint of the structure, stresses are created, and if such stresses produce tension in

the concrete section, cracking will result (see Figure 3-4). Consider, for example, a concrete slab on ground, that has been constructed during the late summer or early fall. During the winter, the mean atmospheric temperature and the temperature




Figure 3-4.



of Deterioration and Preventive Measures


in concrete deck slab due to shrinkage and tempera-

ture stresses. This is an exposed concrete structure and was constructed 1,000 feet long without an expansion joint. The results were two ruptures,

as shown, located at the third points. Note that failure occurred precisely at the change of section where the bottom bars are bent up. Note also the satisfactory condition of the adjacent concrete, with form marks still plainly visible.

the concrete in the slab length of the slab would This decrease in length the supporting ground. movement


could drop 70°F or more. Each 100-foot tend to decrease in length by about 1/2 inch. is resisted by friction between the slab and If the friction is sufficient to prevent the

is usually

the case),

the concrete


subject to a tensile stress of about 1,300 pounds per square which is far in excess of its strength.





Deterioration, Maintenance, and Repair of Structures

Another case of common occurrence is the foundation wall for a building (see Figure 3-5). Here, in winter, the wall tends to shorten. The footing, being well below the ground surface in a moist environment where the temperature is relatively stable, tends to remain unchanged in length, and the floor slab which is in a heated environment

(the interior of the building) also tends to remain unchanged

in length (or even to lengthen). The result is a pattern of cracks, such as that shown in the figure. A similar situation occurs where the exterior walls are intersected by the upper floors or roof.


that, in this case, the distress is caused by a nonuniform change in temperature, i.e., by a thermal gradient. Note also that a thermal

gradient may be caused by a lag in the reaction of the interior of the concrete mass to a rise in surface temperature, and that in lower

latitudes such gradients may amount to as much as 10 degrees per


Cracks in Floor Slab (Note maxi mum opening at

wall with minimum opening in heated interior of building)

Grade Line Extension of Crack in the Sidewalls


Wall and Footing

Crack in Wall. (Dissipates in wall below grade) Grade Line Figure 3-5.



in a building.

A third case of importance is that involving dissimilar materials

such as a new concrete

surfacing on an old concrete



coefficients of thermal expansion of the two materials probably will not be the same, and a change in the temperature of such a composite member, even if uniform, will create a warping of the section and may cause a severe stress condition. Prevention. The problem is very similar to that of preventing

drying shrinkage (see Section B, paragraph 2, Drying Shrinkage), and consists of providing joints to relieve the restraints

in the

Concrete Structures:


of Deterioration and Preventive Measures

structure and providing reinforcement to distribute the In the case of the problem of pavement, as described usual to provide contraction joints at intervals of about and temperature reinforcement of about 0.2 to 0.25 per are





the provision


stresses. above, it is 30 to 35 feet cent. There

of numerous

joints is impractical. One example is a building. Here, the solution is to provide whatever expansion joints are feasible, to locate the joints judiciously in critical areas, and to take care of the rest of the problem with reinforcement. A major problem is to determine the required amounts of temperature reinforcement. Stress calculations are impractical. In general, use the minimums of 0.2 and 0.25 per cent specified in the ACI Building Code.” However, the structure should be visualized (as per the above example of the foundation wall) to detect probable areas of increased temperature stress, and greater amounts of temperature steel should be used in these critical areas. Locations known to be troublesome include (1) canopies; (2) parapets; (3) cornices; (4) places where the framing is such that the floor slabs span the short direction of the building, as in schools and dormitories; (5) openings (the usual 2-#4 or 2-#5 bars frequently are inadequate); and (6) the quarter or third points of the foundation wall and first-floor slab. One other thing which can be done to reduce temperature effects on concrete


is to use






span roofs, the provision of insulation offers an attractive alter-

native to the provision of expansion the enclosing construction.



the roof and

b. Variations in Internal Temperature. Volume changes in concrete also can be caused by variations in the internal temperature. For example, increases in internal temperature of mass pours such as dams, mat foundations, and turbine-generator footings, resulting from heat generated by the concrete during hardening, are well known. Less well known, but of importance, is the use of aggregates having coefficients of thermal expansion markedly different from the average value for concrete of 0.0000065 inch per inch per degree Fahrenheit. In the former case, the concrete is subjected to a change in volume due to the development of the heat. In the latter no heat is generated, but any volume changes induced, whether by variations in internal

or external



not uniform.

In either


if the volume changes or changes of shape are restrained, stresses and cracks occur. Prevention. Preventive measures are the same as described for the prevention of drying shrinkage, with the added requirement that the possible occurrence of aggregates having incompatible coeffi-


Deterioration, Maintenance,

and Repair of Structures

cients of thermal expansion must be considered. This is difficult to anticipate, and the best assurance is simply to use cement and aggregates from known and proved sources. When working with untried or unproved sources, careful laboratory testing of the materials to determine their thermal properties, as well as the usual tests for soundness, organics, etc., is advisable. 4.

Absorption of Moisture by the Concrete

In varying degrees, all concrete is porous. In fact, frequently it may be observed that in different parts of the same structure, constructed of the same materials, by the same contractor, to the same specifications, one part will be severely deteriorated and others will be sound. The usual reason is the difference in amount of water absorbed by the concrete, due to differences in exposure and differences in porosity resulting from variations in workmanship. As the moisture content of the concrete increases, it swells. Schaufele,


on the behavior

of concrete

in marine


ments along the California coast,° cites instances where concrete cylinders originally 13 feet in diameter have grown 6 inches and where cylinders of 1/4-inch steel plate, 6 feet in diameter and filled with concrete, have ruptured. Champion* similarly reports expansions ranging from 0.01 per cent for good concrete to 0.5 per cent for poor concrete, the amount depending on the age, porosity, type of aggregate, and initial moisture content. If the swelling is prevented, cracks and spalling occur. Prevention. In general, it is impractical to prevent swelling due to increased moisture content. The solution is either to allow for expansion of the concrete in structures subject to alternate wetting and drying, or to keep the concrete continually wet by encasing it in something like a heavy timber jacket, which will prevent drying of the underlying mass. This latter technique has many other advantages, particularly in marine environments, and is described in more detail in Chapter 5.

5. Corrosion of the Reinforcement a.

Corrosion Due to Chemical





Except for an occasional





forcement. The reinforcement is deliberately and almost invariably placed within a few inches (often a fraction of an inch) of the surface. If the reinforcement is exposed to circulating air and water, it will corrode. The volume of the oxide produced by corrosion is about eight times that of the parent metal, and the result is that the concrete cover is cracked and spalled (see Figures 3-6 to 3-10).

Concrete Structures:


of Deterioration

and Preventive Measures


Figure 3-6. Spalling of concrete surface due to corrosion of reinforcement. This photograph shows the visual evidence of this type of defect in the various stages of its development. In the foreground, the concrete surface is still sound, except for a few cracks running parallel to the bars. In the middle of the panel, the cover is beginning to spall. The area in the background shows the full development of the condition. The lower mat of reinforcement is exposed and the bars are corroded. Note how the cracks and spalls run parallel to and spaced with the bottom bars. The dark stains are damp areas in the concrete. Note also that the worst deterioration occurs in the damp areas and that the dry sections of the soffit are still in fair condition.

Prevention. Prevention of corrosion of the reinforcement requires that the steel be kept from contact with circulating water in the presence of oxygen or containing dissolved oxygen. This can best be accomplished by encasing the bars in a dense concrete mass and providing adequate cover in accordance with Table 3-2. Avoid details in the design which would promote ponding of water. Do not use trough sections without large and frequent weep holes. Rail slots and ballast troughs must be drained by positive means. Slope horizontal surfaces at least 1/16 inch per foot, and preferably 1/8 inch per foot. Particular attention should be paid to sloping the tops of parapets. Slope the adjacent grade away from the structure. In addition, keep weep holes open and see that roofs, balconies, ledges, and similar features drip clear of the lower structure.


Deterioration, Maintenance, and Repair of Structures

(b) Figure 3-7. Spalling of concrete ceiling construction insidea building. This structure was located in a humid, semitropical environment. As may be noted, buildings, as well as exposed structures, suffer damage due to corrosion of reinforcement.

Concrete Structures:


of Deterioration

and Preventive Measures


Table 3-2 Exposure conditions

Follow ACI Building Code (Reference

Interior Exterior



Building walls,

roofs, and

(2) (3)

grade with good drainage Below grade Submerged

similar locations above

Marine environments (1) For members in the tidal and splash zones (2) For members which are

submerged Above

splash zone

Figure 3-8. Spalling forcement. The dark edges of the cap, the in these pile caps is





Minimum cover over main bars

Follow ACI

Building Code (Reference 2)

Follow ACI Building Code (Reference 2) 1 1/2 in. for slabs, 2 in. for beams and girders 2 in. for slabs,

2 1/2 in. for beams


girders, 3 in. for piles and columns

3 in. from a point

10 ft. below mean

low-water level to mean low-water

level; elsewhere use 2 1/2 in. 11/2 in. for slabs, 2 in. for beams girders


of the edges of a pile cap due to corrosion of the reinspots are rust stains. Except for the spalling of the concrete is in excellent condition. The reinforcement just the two corner bars plus bars in the slab.


Deterioration, Maintenance,

and Repair of Structures

Volume of reinforcing bar after corrosion

Surface of concrete



resulting in spalling).

Figure 3-9. Cracking and spalling of concrete due to corrosion of the reinforcement.

Figure 3-10.


Detail of a ‘‘pop-out’’ due to local corrosion of a reinforcing

Note the corroded condition of the bar and that the greatest depth of

the pit is adjacent thereto. These are typical manifestations of the condition. Note, also, that the pop-out starts at the mid-diameter of the bar, with sound concrete beyond. This, too, is typical and raises the problem

of how far to cut back the concrete when making a repair. (Chapter

5) for a discussion

of this


See the text

Concrete Structures:


of Deterioration and Preventive Measures


b. Corrosion Due to Electrolytic Attack. In the presence of moisture, particularly if the moisture contains salts, concrete is electrically conductive; and the presence of stray electric currents can create electrolytic action on the steel, causing violent corrosion. The sources of the stray currents are frequently accidental, such as earth returns or leakages; or they may be deliberate, such as cathodic protection. As a guide to prevention, it should be considered that sulfates, chlorides, and carbonates will act as aids to corrosion. Seawater is a common source, and using it as mixing water for concrete is to be discouraged, because of difficulty in ensuring that the salts are completely digested during hydration of the cement. Calcium and magnesium chlorides, used to accelerate the set of the concrete or as antifreeze agents, also will promote corrosion if present in excess amounts. In addition, if the reinforcement is not of uniform composition, corrosion will attack the intersections of stirrups with

main bars




points of contact between

the chairs or tie wires and the bars. Accordingly, reinforcement in a given structural section should be all the same grade and, preferably, from the same mill. The cautions described in paragraph a, above, also are applicable.

6. Chemical Reactions The reactions occurring in and the behavior of concrete during and after hardening have engaged the attention of physicists and chemists for at least a century and still are not entirely explained. Accordingly, it is not feasible to go into the problem in detail in this text, but merely to set forth some of the more common aggressive agents, a few of the more common reactions, and some devices for preventing or inhibiting chemical attack.

a. Unsound Materials.

The use of unsound or fouled materials in

concrete results in volume changes, pattern cracking, and all manner of unpredictable defects. Accordingly, the following discussion presumes that the materials used for the construction conform to ASTM C150 for the cement and to ASTM C33 for the aggregates; that the mixing water is clean and potable; that the admixtures (if any) are types and brands of common and proved use; and that the materials were not fouled before or during placing of the concrete. b. Some of the More Important Aggressive Agents. Commercial cements are alkaline and are attacked by acids, by organic compounds which can be hydrolized to acids, and by some alcohols. Ground water having an acid content due to the presence of decayed vegetation is sometimes a cause of difficulty in this regard. Dairy floors are a problem due to the presence of lactic acid in any milk


Deterioration, Maintenance, and Repair of Structures

which spills and sours. The floors in breweries, wood-pulp mills, and food-processing plants are similar sources of difficulty. Salts containing ammonium and magnesium ions attack concrete by reacting with the calcium. In particular, the magnesium exchange takes place when concrete is exposed to seawater and is one of the mechanisms whereby seawater attacks concrete. Sulfate solutions react with tricalcium aluminate hydrate, which is a normal constituent of concrete, forming calcium sulfoaluminate

hydrate (which is the mineral ettringite).

This reaction is accom-

panied by a substantial expansion and causes cracking and disruption of the concrete mass. Although most commonly encountered in concrete exposed to seawater, sulfate attack can occur in the presence of combustion products (if moisture is present), in water draining from mines and industrial sites, or, in fact, wherever sulfate solutions come in contact with a hardened portland cement matrix. Difficulty should be anticipated whenever the environment of the concrete


sulfate solutions

in concentrations



SO.) in excess of 0.10 per cent. The tricalcium aluminate in the cement also reacts with chloride ions, which is another reason for not using salt water for mixing water (see also Section B, paragraph 5, Corrosion of the Reinforcement). Soft water tends to leach out the lime in the cement paste, leaving a porous Silica skeleton. However, this reaction is slow and is seldom a cause of trouble unless the water is forced through the concrete




High temperatures (over 300°C) will drive off the water of hydration in the cement gel and cause loss of the cementing action. Bacterial action is often cited as a cause of deterioration of concrete.


it is an indirect cause.



action of marine

on wood.

in sewers


similar locations, the concrete is attacked because the bacteria generate hydrogen sulfide, which is weakly acidic, and sulfuric acid, which is a very strong acid. However, biological attack in the form of borings by marine organisms does occur. Chapter 7 describes the destructive

Mollusca also bore into concrete. on concrete





in tropical waters




instances include attack and on concrete


jackets in Los Angeles harbor. Although the potential damage may be severe, the occurrence is so infrequent that special preventive measures


not warranted

indicates a need.




Carbon dioxide is another material which reacts

with concrete if the concrete is fresh.

in the area


This is the reason for the com-

mon prohibition against letting the exhaust gases from salamanders or other heaters come into contact with newly placed concrete.




of Deterioration and Preventive Measures


Deterioration of concrete also may occur through chemical reaction between cements of high alkali content and the mineral constituents in certain aggregates. This reaction results in localized ‘‘pop-outs’’ of the concrete surface, map cracking, and an overall expansion of the concrete mass. In connection

with this


it should be noted

that most


sive chemical agents, to produce significant attack on concrete, must be in solution form and above some minimum concentration. Concrete is seldom attacked by solid dry chemicals. Also, for maximum effect, the chemical solution needs to be regenerated, i.e., circulated in contact with the concrete mass. This is the reason that concrete in submerged environments is more susceptible to attack and that concrete subject to aggressive solutions under differential pressure is vulnerable. The pressure gradients tend to force the aggressive solutions into the matrix. If the nonpressure face is exposed to free evaporation, a concentration of salts tends to accumulate


at that face,


in increased




Figure 3-11. Concentration of salts due to evaporation. The cracks in this wall are clearly marked by deposits of salts formed by evaporation of the solution penetrating from the pressure face. The construction pictured is the sidewall of a drydock, well below the water level. The concentration of solution tends to promote aggressive chemical attack on the concrete at the face of the wall, and removal of the encrustations showed some of the cracks to be slightly chased by solution.




and Repair of Structures

The general symptoms of chemical attack on concrete are disintegration and spalling of the concrete surfaces and the opening of cracks and joints. There is also a general disruption of the concrete mass and swelling of the structure. The aggregate particles protrude from the matrix, and there is a loss of cementation in the cement paste (see Figures 3-12 to 3-15).

Figure 3-12. Chemical deterioration of concrete in bridge pier. This bridge pier is located in a highly polluted tidal river. The condition shown was caused by chemical attack. Note the protruding aggregate, usually associated with this type of deterioration. The aggregate, being more inert than the cement-sand matrix, is not attacked and stands proud from the surrounding surface. Note also that the upper limit of the deterioration is near the high-water mark. This pier was repaired by jacketing, as shown in Figure 5-3.

Where the reaction produces an internal swelling of the concrete mass (alkali-aggregate reaction, for example), the symptoms consist of the formation of a pattern of cracks which develop by opening and deepening until sections of concrete are spalled away. Where the expansion is not restrained (as in pylons, railings, wingwalls, etc.), the crack pattern is random. Where the expansion is restrained along one or more axes (as in a bridge pier or an abutment where the load resists vertical motion), the cracks occur as a



View of concrete pile deteriorated by seawater-sulfate attack.

Note protrusion of aggregate particles from limit of deterioration near high-water mark


cement-sand matrix; abrupt (black, horizontal band on the

and spalling of corners due to corrosion of corner bars.

series of parallel openings with lateral expansion of the concrete perpendicular to the axis of restraint (see Figure 3-16). Prevention. The most important requirement is to use good, sound,




to the provisions

of References

6 and 7. Concrete of good quality prevents the intrusion of aggressive chemical solutions and is clearly and consistently more resistant to chemical attack than is poor concrete. Even good concrete can be attacked to some degree by most of the aforementioned agents, however, and the following precautions in design and construction (and in repair) should be observed: (1) Concrete for use in a sulfate environment should be made with a sulfate-resistant cement. In particular, concrete for marine exposures should be made with a portland cement having a content of tricalcium aluminate of not more than 8 per cent. Cements conforming to ASTM Standard Specification C150, Types II, IV, and V, meet




Figure 3-14.

Section of deteriorated concrete






illustrate the visual

water-sulfate attack.


be used



removed with

Repair of Structures



of concrete

pile shown in




3-15 by sea-

this requirement. Ordinarily, Types I and III cements do not qualify. Concrete for use in environments having very high sulfate contents (over 0.20 per cent, expressed as SOx, or over 1,000 parts per million) should be made with cement having a still lower content of the tricalcium aluminate, say 5 per cent, for which cements conforming to ASTM Standard Specification C150, Type V qualify. (2) In acid environments, the use of limestone aggregate will somewhat inhibit attack by neutralizing a portion of the acid attacking the binder. A more positive answer, however, is some form of protective coating. Rubberized coatings are commercially available. Bituminous coatings work nicely for exposure to sulfate or weakly acid solutions. Impregnation of the concrete with a bituminous solution is a form of bituminous coating in occasional use for the protection of structures exposed to seawater-sulfate attack. (See Reference 4, Chapter 8.) A facing of ceramic tile set in acid-proof mortar is an excellent precaution. Even a little thing like frequent rinsing down of the concrete surface with water is a big help. (3) The cracks which do occur in concrete exposed to an aggressive



be kept


by applications

nous sealers to prevent the penetration of chemical

of bitumi-


Concrete Structures:


of Deterioration

and Preventive Measures


Figure 3-15. (See also Figures 3-13 and 3-14.) Disruption (by hand) of concrete specimen shown in Figure 3-14. Note pulverization and lack of cementation in sand-cement matrix and lack of bond with aggregate. These characteristics are a clue to the occurrence of chemical disintegration (or weathering) in concrete. (4)

To prevent

the alkali-aggregate



that the

maximum alkali content of the cement (calculated as the percentage of Na.O plus 0.658 times the percentage of K20) shall not exceed 0.60 per


it having



that concrete



the usual stone aggregates and with cement having a lesser alkali content normally is not affected by this reaction to an objectionable degree. (If high-alkali aggregates are used, further reduction in the alkali


of the cement

is required,

of course.)

is not to be used indiscriminately, however.




or less



Most portland

as a natural




the occurrence of these constituents in the clays and shales used their manufacture, and an arbitrary restriction on the amount of alkali may substantially increase the cost of the cement for the



Deterioration, Maintenance, and Repair of Structures

Spalled section

[Bridge Pyion f— Cracking

4) Movement Unrestrained IS

ee Fine | Cracks |

Open Cracks —


Figure 3-16. Physical of concrete mass.




Movement Restrained on Vertical Axis

appearance of deterioration due to internal swelling

project. Actually, the best means for assuring freedom from deleterious activity of the aggregates is to take them from proved sources or to prequalify the aggregate by some means, as, for example, by use of the testing procedure described in ASTM Standard Specification C33. The use of air entrainment also is beneficial in reducing the amount of expansion resulting from the reaction.

7. Weathering As has been stated, to some extent all concrete is porous and will absorb moisture. Having absorbed this moisture, if exposed to sub-

Concrete Structures:


of Deterioration and Preventive Measures

freezing temperatures, the moisture will freeze and expand, resulting hydraulic pressure will tend to cause the concrete to crack. Upon thawing, the cracked surface will spall. This ess, repeated for many cycles, causes the concrete surface

integrate (see Figures 3-17, 3-18, 3-32, and 3-33).


and the surface procto dis-

Prevention. Since the basic difficulty is expansion of the water absorbed by the concrete mass during freezing, minimizing the porosity by use of a dense, sound concrete is the best preventive. In particular, the water/cement ratio should not exceed 6 gallons per bag of cement. The use of air entrainment is a must. The effectiveness of this agent in improving resistance to weathering is well established, having, in fact, been proved by literally thousands of tests (see Figure 3-19). Insulating the concrete against freezing is another useful technique. The most common instance of its application is in a facing of granite to prevent deterioration, particularly on concrete piers in the tide zone or near the waterline. A facing of this type is a fine, all-purpose expedient, protecting against erosion and chemical attack, as well as against weathering. A timber jacket, as described in Section B of Chapter 5, is another excellent solution. - The use of an absorptive form lining to increase the density of the concrete surface is recommended for particularly severe exposures, such as areas which are partly buried or are subject to fluctuating or shallow immersion or to a spray condition. A surface sealant, in the form of an epoxy, or impregnation of one of the types described in Section I of Chapter 5 also is recommended. The type of aggregate seems to have some influence on the susceptibility of the concrete to attack by weathering. Angular aggregates (such as crushed stone or crushed gravel) appear to behave better than aggregates of rounded gravels. On several occasions, the author has noted the superior performance of concrete in marine environments when made with crushed stone aggregate rather than with rounded aggregate particles. Champion‘ notes similar conclusions. Possibly this effect is due to the fact that rounded particles have the least specific surface and therefore the poorest bond. Also, rounded particles are more easily dislodged from the concrete matrix. The aggregate materials themselves, of course, should be impermeable and of durable quality, from proved


In addition to the above provisions, and as discussed under Poor Design Details, Section B, paragraph 10, the structure should be designed to minimize exposure to moisture and to facilitate drain-



Deterioration, Maintenance, and Repair of Structures

Figure 3-17. Detail of concrete thawing. (Scale in Figure 3-17a

surfaces spalled by cycles of freezing and indicated by lens cap.) Note powdery tex-

ture of matrix and patches of unspalled concrete interspersed in spalled area (A in Figure 3-17a). The presence of unspalled patches is due to areas of better concrete and is found frequently on weathered surfaces.







of Deterioration

of a concrete

and Preventive

wall due


to weathering




sion of the reinforcement. Figure 3-18a shows a general view of the deterioration, and Figure 3-18b shows a detail of the condition. Note the dull, whitish,



of the


an appearance

often noted

in cases

of chemical deterioration and weathering.

8. Shock Waves Concrete is a to spalling when ent transmission materials, such


heterogeneous material and, as such, is susceptible subject to shock waves. This is due to the differrates at which the waves pass through the several as the aggregate, the matrix, and the reinforcement,

its mass.


of difficulty in this regard.




a particular

They are usually heavy sections



Deterioration, Maintenance, and Repair of Structures


2000 1800 1600 400












600 |} 400 200


of Freezing












i 1

\Z 1





/ -



3.0 64



Air Content — Percent

Figure 3-19. Effect of air entrainment on resistance of concrete to freezing and thawing. (Courtesy Alpha Portland Cement Co.)

consequently are subject to hard driving in the process of being


(See Chapter 8.)

Piers and wharves are another potential source of trouble. If not adequately fendered, the impact of berthing vessels can cause damage, particularly at points of stress concentration such as corners,

offsets, or doglegs.

Cases of shock-wave

Foundations for machinery behave similarly. failure due to supersonic



have been

Prevention. Experiences in earthquake areas and with bomb damage during wartime have indicated that the use of heavily reinforced sections will provide a concrete

to shock waves.

structure with excellent resistance

Of course in marine structures this invites corro-

sion of the steel, so that adequate fendering is a better choice.

Shock resistance also can be improved by using concrete made with angular coarse aggregate rather than rounded aggregate; i.e., crushed stone is superior to gravel. In the case of precast piling

(subject to the limitation that the pile must be properly seated), it

is well to use more

jetting and less hammering

in the installation.

9. Erosion (Abrasion) The following cases represent the more frequent occurrences of erosion of concrete. In all cases, the essence of prevention is to

Concrete Structures:

Causes of Deterioration and Preventive Measures


provide a good quality concrete with a smooth, dense surface.

a. Floor Slabs.

While rubber-tired vehicles have largely sup-

planted those having iron-rimmed wheels, vehicle loads have increased greatly, with 16-to 20-ton axle loads being common. Also, handling materials by use of forked-lift trucks having heavy wheel loads is now almost standard practice, and labor costs have so increased that there is a tendency to economize on ‘‘good housekeeping’’ practices, with the result that dirt and debris are not promptly removed from floors, are ground under the wheels of passing traffic, and act as an abrasive which grinds away the floor surface. As a result, abrasion of floor slabs is a common problem.


The recommendations in Chapter 6 with regard to

the use of high-strength concrete; smooth, dense surfaces; air entrainment; curing; and delayed finishing are applicable. In addition, a mineral or metallic dust surface hardener or an abrasionresistant topping is useful, if economically feasible. However, such finish must be used with care and only in accordance with the recommendations of the manufacturer or as described in Reference 8. Alternately, a liquid hardener (say, magnesium or zinc fluosilicate

or sodium



be considered.




dusting and aid in resisting attack by chemical spills. However, they are more suitable for improving a poor surface than for use on well-constructed floors. Abrasion of floor slabs also can be minimized by making the radii available for turning vehicles as large as possible and by careful housekeeping. The occasional use of industrial vacuum cleaning is of value. Where feasible, particularly in piers and similar structures, provide means for flushing down the deck. b. Hydraulic Structures. Concrete flumes are another potential source of difficulty. Solid materials in suspension abrade the floor and walls. Larger particles are tumbled along the bottom of the channel. In addition, at high velocities, cavitation, i.e., erosion not requiring

the presence

of suspended


can occur.

Prevention. Streamline the sections as much as possible and insist on tight, smooth forms and a smooth, steel-trowel finish. Even small holes and imperfections are important. In the case of flumes, it is well to have occasional traps along the bottom to catch heavy particles. Joints, both longitudinal and transverse, should be tight and precisely flush. Expansion joints are a major problem. If the filler material is soft enough to permit expansion, it is probably too soft to show good resistance to abrasion. Such joints simply require maintenance and must be listed in the maintenance schedule. c. Flues. Abrasion may be a serious problem, particularly if the flue conveys gases having a heavy ash content. However, unless the basic combustion process can be improved, there is little that can




and Repair of Structures

be done to avoid this problem aside from the use of linings. If desired, baffles can be used to deflect the ash and gases from the flue walls, but the baffles are subject to even greater come a problem themselves.

d. Waterfront Structures in the Surf Zone.

abrasion and be-

In or near the surf

zone, the breaking waves churn up the bottom and scour any structures located therein with particles of sand and silt like a giant sandblast. The condition continues, practically daily, year after year, for millions of cycles, and will wear away the hardest concrete and the toughest steel. The remedy is to provide steamlined (rounded) sections to minimize the attack, and to allow for such abrasion

as does


by increasing

ing an expendable coating or jacket. tance,


the occurrence

the concrete


of construction


or provid-

avoid honeycomb,



in the surf area,

these will be relentlessly attacked, opened, and enlarged by the




e. Bridge Piers. Abrasion of bridge piers by floating ice also is a frequent occurrence, and the use of wrought-iron protection plates or stone facings


to inhibit such


is a common


Poor Design Details The previous paragraphs

tion resulting from

(1 through 9) have considered deteriora-


in the concrete

not anticipated

in the design.



or inept construction, and attack by one or several types of aggressive environment. Of equal concern is the occurrence of poor design details which, while fully in conformance with the requirements of the design specification, do not work well in practice. If a large number of different structures are examined in different localities, it will be found that deterioration occurs repeatedly in connection with certain details or that certain effects have taken

place which were

A number

of such

details are described in the following paragraphs. Their use should be avoided, and the effects described should be considered in the design. a. Reentrant Corners. This detail (Figure 3-20) creates a condition of stress concentration under the reinforcing bars and can be improved (as shown in Figures 3-20b and 3-20c) by distributing the lateral restraining force required to produce the change in the direction of the bar tension. b. Abrupt Changes in Section. Any abrupt change in section (Figure 3-21) causes stress concentrations which may result in cracking. Such a condition is particularly common in connection with encased steel framing (Figure 3-21a) and can be corrected by dropping the


Concrete Structures:

Causes of Deterioration and Preventive Measures


Reinforcing Bar Balancing Force Required To Change Direction of Bar Tension Crushing and Cracking of Concrete

Poor Detail

Reinforcing Bar


Forces Provided


|Anchorage| for


Bar Tension —/ Betser Detail Figure 3-20. a reentrant


Cause and prevention of crushing and cracking of concrete at corner.

steel to provide the full slab section over the beam. Also, run the slab steel through and provide mesh in the beam encasement. Expansion

or contraction

joints in the structural



be carried through the fill and finish (Figure 3-21b). Openings should be provided with extra reinforcement at corners, and special attention should be paid to pipe openings and ducts, which sometimes are too closely grouped in the slab. c. Rigid Joints between Precast Slab Units. (See Figure 3-22.) Unless special provision is made to extend and splice the top bars, precast slab units are commonly designed for conditions of simple support. However, the rotation of the ends of the slab units will




and Repair of Structures

Crack Note Reduced Section

Encased Steel Framing





Expansion (or construction) Joint

Concrete Floor or Roof Construction

Expansion Joint Not Projected Through Floor Fill and Finish

(b) Figure




due to abrupt


in section

(see also


cause cracking at the juncture between the precast slabs and the cast-in-place fill. Unless floors are covered with asphalt tile or similar flexible materials, this may produce objectionable cracks in the floor finish. For roofs, leakage will occur unless the roof is covered with membrane, asphalt block, or similar material. Flexible jointing and/or a flexible slab finish or topping is required. d. Deflections. (See Figure 3-23.) Usual criteria for allowable live-load deflection vary from 1/240 to 1/360 of the span. For a 20-foot span this amounts to 2/3 inch and 1 inch, respectively. The structure will readily accommodate this movement, but it is a lot of

Concrete Structures: Causes of Deterioration and Preventive Measures


pri Il and/or finish



Crock. Reinforcement









concrete or



* Figure 3-22.


Cracking due to deflections of precast slab unit.

Cracking due to deflection of

floor beam

bearing over lintel

loor System (Behind)

‘Cracking due to

deflection of lintel



Figure 3-23. Cracking due to deflection. (Note the similarity of the crack pattern to that produced by settlement of a foundation.)

deformation to be absorbed in a wall or partition.

As a result, the

partition is called upon to carry an unanticipated load, and if it can-

‘not carry the load, it will crack (as shown in the figure).

The available solutions to this condition are (1) to make the structure stiffer, and so to reduce the deflection and extent of par-

ticipation of the wall;

(2) to make the partition strong enough to


Deterioration, Maintenance,

take the load;

and Repair of Structures

(3) to break the wall loose from the surrounding con-

struction by providing a space or slip joint at the top of the wall and some means to break the bond between the wall and the floor; (4) to use some combination of the above; or (5) to philosophically accept the cracks. The appropriate solution depends on the type of wall construction,

the use,

and the expense

e. Leakage through Joints.


(See Figures 3-24 and 3-25.)


joints are necessary, they must be detailed so that the water will have to follow a long and tortuous path to penetrate. They must be sealed, and water stops should be provided if feasible. f. Poorly Detailed Drips and Scuppers. Where feasible, scuppers, whether in bridge decks (Figure 3-26), through parapets, or through curbs, should be provided with downspouts located so that the discharge cannot be blown against the lower construction. Also, scuppers must be large enough (say, minimum 2-inch dimension) to preclude the possibility of clogging. Overhanging constructions (ledges, copings, etc.) should be provided with drip grooves, drip edges, or drip flashing to prevent the water running down the face of the sheltered wall.

Joint See

Sealer Curb


‘Wearing =}+— Bridge



Course Deck











carefully installed, water percolation should be expected


occur along the interface between concrete and steel, causing corrosion of fascia plate at A and of any anchor: ring fascia plate to


Figure 3-24. penetration

A common of water.


detail which is susceptible to damage

due to the

Concrete Structures:

Causes of Deterioration and Preventive Measures


joint Sealer voint Filler

It must be onticipated that leakage will



through this

ES et

Old Detail

lig’ Compressed

asbestos abutment


with graphit:


‘Abutment Backwall

Figure 3-25. Modification of detail at bridge abutment to inhibit penetration of water through the expansion joint.

g. Inadequate Drainage. This is one of the most common errors in detailing of concrete structures. The construction must not pond water. Horizontal surfaces, even the tops of walls, should be pitched to drain, and large and frequent weep holes must be provided in trough or depressed sections or any other low spots. Weep holes should not discharge over the exposed face of the concrete. The surface water should drain away from the structure, not toward it, and drainage from higher ground should not be permitted to flow over the top or face of a wall. h. Insufficient Travel in Expansion Joints. (See Figure 3-27.) This

defect causes spalling in localized areas adjacent to the joint. See also Chapter


i, Unanticipated Shear Stresses in Piers, Columns, or Abutments.

Expansion-bearing assemblies

tend to become

immobilized with time


Deterioration, Maintenance, and Repair of Structures

Orift of scopoer discharge Ave med awe


corrosion foining oFofprembearings ond

Pier— D —_——




2) Poor Practice Figure 3-26.

Figure 3-27.


jpeing of scupper och o slash Gh sinuchure ee discharge lor provide splashfo sewer blocks:

land sex// on ground | \Cor TF rivers.


6) Better Practice Proper treatment of scuppers.

The results of insufficient travel in an expansion joint.

Concrete Structures:


of Deterioration and Preventive Measures


from the effects of corrosion, accumulations of dirt and debris, etc. As a result, the longitudinal thrust, instead of being resisted by the fixed bearings only, becomes partly resisted by the supports for the expansion bearings, and a shear is imposed on the supporting concrete. The resulting diagonal tension causes cracking of the type indicated in Figure 3-28. The proper means for preventing this condition is the use of stirrups, as shown in the figure. Although, theoretically,



not required,

as a practical


pensive and well worthwhile. j. Incompatibility of Materials or Sections.

(1) Compatible Sections.




Massive concrete sections abutted by and

tied to thin sections just do not work together. Unless adequately reinforced, cracking such as that indicated in Figures 3-29 and 3-30 will develop. In the case shown in Figure 3-30, the condition is actually aggravated by the heavy reinforcement in the flanges, since the presence of the reinforcement tends to resist the shrinkage deformations, whereas in the lightly reinforced web these deformations are not so resisted. In cases of this type, it is advisable to provide longitudinal reinforcement of, say, 0.5 per cent or more in the web, even though, theoretically, it is not needed. Force induced by friction in bearing | (accentuated by corrosion or fouling of bearing surfaces), |

Expansion Joint| Filler


A >

Reaction to

longitudinal] force



ae see L


4 : “+


Sliding Expansion




Cracking resulting from drag of bearings which have been designed to be free to expand.

Prevention - Provide stirrups and vertical bare, as shown.

Figure 3-28a.

Cracking of beam



Deterioration, Maintenance, and Repair of Structures

Slab rests on backwall without provision for relieving reaction. Longitudinal shear is induced by bond or frictional resistance, resulting in cracking, as shown. Prevention - Relieve slab reaction as shown

in 3-28a,


Sliding Expansion Bearing Reaction to longitudinal force

Bearing Pad


Figure 3-28b.

Cracking of slab seat.

New Concrete Deck Slab

Existing or New

Mass Structure

Figure 3-29. structure.

Cracking due to an abrupt change in section of a concrete

Concrete Structures:

Causes of Deterioration and Preventive Measures


raaas Reinforcement

dyacking due to shrinkage and/oy temperature. Correct by prdviding longitudinal web reinfofcemeat.

Flange Reinforcement (Acts as restraint against temperature and shrinkage effects.) Figure 3-30.


cracking in flanged concrete girder.

Another common incidence of incompatible sections is a separate, rigid wearing course used on a bridge deck. Experience indicates that this detail does not work well. The sun shining on the upper deck surface expands the wearing course more than the structural slab below, causing a separation between the two courses and local heaving of the wearing surface. The wearing surface, being thin and lightly reinforced, cannot take the load and breaks up under the pounding of traffic.

(2) Materials.

The importance of the use of compatible materials

is discussed in Chapter Note 1 of Chapter 5. There is one condition which occurs peculiarly in concrete, however, and which must be considered. This is lamination of the mass as a result of excessive vibration or wet mixes. Concrete, as left to harden in the forms, frequently is not uniform. Water and fine materials tend to rise and collect at the top of the pour and to flow ahead of the heavier materials filling in the distant corners of the form, particularly if the form is excessively vibrated, if the mix is excessively wet, or if the vibrator is used to move the concrete about. Similarly, excessive spading or vibration of the forms results in a rich layer of cement paste adjacent to the forms. The net result is that the concrete is laminated and, as discussed herein, tends to delaminate as a result of thermal changes or changes in moisture content. These practices

should not be countenanced.

Cold joints (Figure 3-36) produce similar effects.


Deterioration, Maintenance,

and Repair of Structures

(3) Differential Thermal Effects on Tank Structures.


the situation shown in Figure 3-31. The sun shining on one side of the tank tends to change the circular shape to an elliptical one. This deformation, as shown in Section A-A, is restrained at any points of


| lL


Walls —-}

Deformation of wall




Buried Floor Slab

exposed to the sun

i Ipone


Vertical Section

Section B-B

(The undistorted shape)

Note inward displacement of sides. i INo deformation

Section A-A

Note Bulge

Jat point of

ithermal stability

Sections A. Figure 3-31.

Effects of differential thermal expansion on a tank.

Concrete Structures:

thermal and


of Deterioration and Preventive Measures


stability (in the buried floor slab, in the shade of the eave,

on the shaded


which cause cracking.

of the tank).

The stresses





are reversible, depending on

whether the differential temperature change is a rise or a fall. Prevention consists of providing adequate dowels between the floor and walls and vertical reinforcement in both faces of the walls and in the floor at the junction with the walls. The portion of the walls of the tank near the liquid level are sub-

ject to a similar effect due to the thermal stability produced by heat

absorption in the liquid.

Also, the concrete in the walls below the

liquid level is saturated by the fluid, whereas that portion above the

liquid level tends to dry. As described in paragraph 4, Absorption of Moisture by the Concrete, this leads to a differential swelling of the tank walls, and horizontal and vertical cracking concentrated

near any relatively permanent liquid level may be anticipated.


this case also, prevention consists of providing adequate reinforcement horizontal and vertical, in the walls. The same thing happens with rectangular tanks and with the walls

of a building unless they are properly tied into the abutting compo-

nents of the structure or are broken up into panels which are short enough so that they do not develop an appreciable bow. k. Neglect of Plastic Flow. Flow, or the slow deformation of concrete under stress, is a well-known phenomenon which is sometimes neglected. The result is that the structure may suffer unexpected deformations. These deformations cause opening of joints and cracking of supported or abutting constructions and must be considered in the design.

11. Errors in Design It is not the purpose of this text to discuss basic errors in design. Design errors do occur, however (see References 9 and 10), and the symptoms of their occurrence are cracking and spalling—the same symptoms that the other deteriorating agents described herein prod-

uce. So, of course, it is necessary to consider them in the list of potential sources of deterioration when diagnosing the probable cause of some



and they are so considered

in Section C.


As described in Chapter 1, diagnosis of the cause of deterioration in a concrete structure is largely a matter of eliminating possibilities until some conclusion appears, and the following portion of this




and Repair of Structures

Figure 3-32. Deterioration of precast concrete piling in a marine environment. This photograph illustrates several important points. First, note that the bars still show the deformations and that the corners of the piles are not spalled in the splash zone (as compared with the edges of the pile caps). Therefore, this is not a problem of corrosion. Second, damage to the piles does not occur below low water and ends just below high water. This rules out attack by chemicals in the water (including sulfate attack). The light patches on the piles are areas recently spalled, indicating that deterioration is progressive and active. The condition shown was attributed to weathering plus abrasion by floating ice.

section consists of a suggested procedure for such diagnosis. Unfortunately, the procedure does not always lead to a unique conclusion. Partly, this is due to limitations in the present state of our knowledge. More often, it is due simply to a lack of data, particularly on the history of the structure. For example, more often than not, such basic data as type of cement; mix; sources of aggregate; climatic conditions during construction; and, for piles, driving resistance and lengths will be unavailable. However, despite this handicap, following the guideposts contained herein, drawing on the individual’s experience, and using judgment and ingenuity, much can be done. Further, one improves with practice, so that clues which may be missed the first time a condition is encountered will be readily detected the next time.




of Deterioration

and Preventive



Figure 3-33. Culmination of the condition shown in Figure 3-32. Observe the condition of the piles in the near row of verticals. Nothing remains but the bars. Can there be any question of the need for a substantial safety factor in exposed structures? Note also that the bars, remarkably, are still in relatively good condition (the structure was about 17 years old when this photograph







by jacketing.

Before proceeding to the diagnosis, however, one word of caution: Unless the cause becomes very obvious, do not stop part way

through the procedure. The trouble may be due to several deleterious agents acting simultaneously, and it is no use to identify one and not be aware of the effects of the others. Step 1.

Check for an Error in the Basic Design

The first thing to check is whether the condition is caused by overstress resulting from some pertinent deficiency in the basic design. It is a tribute to the profession that this is so seldom the case, but sometimes it does occur. To check, proceed as follows: (1) First, consider what types of stress could have caused the observed symptoms of deterioration. For example, tension causes cracking, usually without spalling, and usually just one or a few cracks are enough to relieve the restraints causing the cracking. On


Deterioration, Maintenance, and Repair of Structures



Figure 3-34. Diagonal tension crack in a concrete beam. (a) An overall view of a condition exposed during demolition of a closed abutment for a highway



is shown




the position

in larger detail in (b).

of a diagonal

tension crack

(The lines of the crack have been

accentuated for clarity.) Note in (a) that the stirrup spacing is irregular and that the crack occurs precisely where the spacing is greater than aver-




one or two stirrups were left out, or they were

carelessly spaced. An apparently small deficiency, but the consequences could have been most serious. These photographs illustrate the need for

care in construction, the other hand,

including every detail.



is almost



by spalling and shredding before the restraints

are relieved.


or both

in torsion or shear






However, none of these basic conditions of overstress in themselves cause disintegration of the concrete surface. So if the basic symptom of distress in the structure is disintegration, the cause is not likely to be underdesign. Similarly, if the basic Symptom is cracking without spalling, rule out excessive compressive stress and, tentatively,




If the structure




spalling, rule out excessive tensile stress (see Figure 3-38). Also, try to relate the locations of the defects in the structure to the probable types of overstress. Remember, the webs take the shear, the flanges take the bending, and torsional stresses are a maximum in the extreme fibers of the cross section. Consequently,

the appearance of shear cracks in the flanges would be inconsistent with the basic stress pattern, as would be indications of overstress

Concrete Structures: in the web

Causes of Deterioration and Preventive Measures

due to tensile or compressive





is in

the case where the member (usually the web) is subject to a combined state of stress. This has to be checked by analysis. (2) Next, knowing what types of stress may be at fault, compare them with the areas and elements of the structure in which the trouble occurs. For example, if the difficulty is tentatively diagnosed as tension cracking, and the cracks occur in the compression flange, an inconsistency occurs and overstress is probably not the problem. Similarly, spalling is inconsistent with the stress conditions in tension areas, and crushing is inconsistent with the stress conditions in tension or shear areas. Also consider at what locations the defects occur in the members. In every structure some Sections are highly stressed by the design conditions,






if the locations

of the areas


deterioration correspond to the regions of high stress. In this connection, do not forget to check the locations of the cracks or spalls against the locations of the cutoffs or bends in the reinforcing bars, and do not forget about the combined state of stress which occurs in the web near points of bearing. (3) If no inconsistency has yet occurred, consider the orientation of the defects. Tension cracks should run roughly perpendicular to the line of stress. Shear usually causes failure by diagonal tension, and the cracks run diagonally in the web. Foundation settlements also usually manifest themselves by diagonal cracking, although the pattern



not be oblique

(see Reference

in the immediate

alize the basic


are consistent

with the pattern





of any wall

11), or near midspan (Figure


in the structure

similar parasitic





of temper-

and see if they

of defects.

(4) If an inconsistency is encountered at some point in the above procedure, the probability of the occurrence of some basic design deficiency is remote and may be tentatively eliminated. If everything has checked out smoothly, however, then the trouble may very well be something in the design, and a careful stress analysis should be made. Note that the elimination of overstress as a cause and as accomplished in this step is a tentative elimination only. If, after following the procedure to its end, the observed conditions still do not conform

to the factors

as yet uneliminated,

and check for some unusual circumstances

Step 2.

of stress.

Relate the Potential Causes to the Three



that underdesign

of the observed


to this step

Basic Symptoms

has been eliminated as a probable


the next


is performed

by relating the list of causes of deterioration to the three basic


Deterioration, Maintenance, and Repair of Structures

Figure 3-35. Condition of cap beam of structure shown in Figure 3-32. Cracks of this type were found in almost all of the cap beams and were the result of relaxation of the supports due to deterioration of the piles. Repair consisted of restoring the bearing capacity of the piles by jacketing and sealing the cracks by opening them and refilling with pneumatically



symptoms. Reference is made to Table 3-3. Observe, for example, that if the symptom is disintegration of the concrete surface, all but three possible causes of deterioration have been eliminated. If one of the symptoms is spalling, there are only seven basic possibilities. If the trouble is cracking, there are still eight potential causes to be considered. Step 3.

Eliminate the Possibilities Which Are Readily Identified

These possibilities include the following: a. Corrosion of the Reinforcement. This agent may be readily identified. The cover spalls off the bars, and the bars rust. In the early stages, the deterioration occurs as a series of parallel cracks running along the reinforcement. A little later, a cleavage plane forms along the level of the reinforcing mat, and rust staining occurs along the cracks. In the last stages, the cover spalls off the bars. To diagnose, check if the locations of the cracks correspond to the loca tions of the reinforcing bars. Also, remove some of the cracked or loosened concrete cover and see if the bars are rusted. Chase down

Concrete Structures:

Causes of Deterioration and Preventive Measures 103

some of the bars to see if the limits of corrosion correspond to the limits of cracking and spalling of the concrete. Check to see if the concrete beyond the plane of the reinforcing mat is sound. If the answers are all yes, it is a clear case of corrosion of the reinforcement. Now, find out if it is a chemical or electrolytic reaction. Ex-



of the bars.

If the corrosion


over short,


Table 3-3 Principal symptoms produced

Basic cause

Cracks | 1. Occurrences incident to construc-

Spalling |


Probable status of deteriorating agent 8 ag










tion operations


Drying shrinkage

3. Temperature



Variations in


b. Variations in internal tem-

i Active or



4. Absorption of moisture by the concrete


5. Corrosion of the reinforcement a. Chemical b. Electrolytic

x x

x x

6. Chemical actions








8. Shock waves 10. 11.

Active Active


7. Weathering






Inactive x

Poor design details



Errors in design





Deterioration, Maintenance,

and Repair of Structures

segments of the bars, occurs as pitting, or occurs principally at the intersections of other bars, then electrolytic attack is probably at fault. If corrosion is general, the trouble is probably chemical. Next, find out why this has occurred. If the trouble is electrolytic attack, where are the stray currents coming from? They must be eliminated or the trouble will recur. On the other hand, if the trouble is chemical, the ‘‘why’’ of it does not make much difference. The repair procedure is prescribed, i.e., seal the bars in some material which will prevent further corrosion. As a check on the diagnosis, the density of the concrete may be studied by making tests of moisture absorption and comparing the results with specimens recovered from sections of the structure which are in good condition. If the difference is not significant, the trouble might have been initiated by internal separations occurring in the concrete mass during construction, and a check should be made of this possibility. b. Shock Waves. Unless the damage is old so that the evidence has been destroyed,



to shock waves


is charac-

teristic. Sections of concrete will be spalled, usually leaving the reinforcing cage exposed. The broken surfaces show a fresh, un-



and the bars


not corroded.

Also, the

spalling is likely to penetrate the section deeply, rather than just detaching the surface layers of concrete. Moreover, the structure will probably be a type which is regularly subject to impact, such as a dock or a railway bridge, so that such damage will be immediately suspect. In cases where the structure is not normally subject to shock waves, if one has occurred which was severe enough to cause damage, the occurrence is sure to have been an occasion of great moment which will be recalled by those occupying or using the construction, and will be readily brought out upon questioning. Summary: Summarizing the progress so far, assuming that the results of Steps 1, 2, and 3 have been negative, we have eliminated three possibilities (underdesign, corrosion of the reinforcement, and shock waves), and our original list of eleven basic agents is reduced to eight. Further, if one of the basic symptoms is disintegration, there are now three possibilities to consider; four if the structure shows spalling; and six if the only symptoms are cracks.

Step 4. Make a Detailed Investigation (1) Investigate the history of the structure.

By whom? In what used? What kind of source? What were pour? How was the

When was it built?

season of the year? What type of cement was aggregates were used, and what was their the methods of construction? Was it a tremie concrete cured? Talk to those associated with




of Deterioration

and Preventive Measures


the design and construction, and see if they have any ideas as to the cause. Their ideas often will be very valuable. (2) Check to see if there have been any movements of the structure. Survey the horizontal and vertical alignment. Does it differ

from the original design? If so, how?

Try to check for any localized elevation or depression of presumedly plane surfaces. This is very difficult, but may show something indicative of bulging of the forms during construction or local settlements

of the plastic



(3) Make a survey of the deterioration. Where does it occur? Where is it worst? Plot the cracks. Do they show any pattern? If so, what pattern? Any pattern which can be detected is a valuable clue.

Step 5. Analyze the Available Clues This is the hard part and the most challenging. In general, proceed along the following lines: a. Where the Basic Symptom Is Disintegration of the Surface. The

first thing to do is to check for unsound materials.

Recover samples

of the deteriorated mass, have a laboratory check the materials against the project specifications, and then check the specifications against good practice. If the materials are of types known to be unsound, a conclusion has been reached. If the materials are not of types known to be unsound, but still do not conform to accepted standards of quality, a tentative conclusion has been reached. If the materials are sound, no conclusion has been reached except that one of three agents—chemical attack, weathering, or abrasion—must be at work. To discover which, first check the environmental conditions. If the deteriorated area of the structure is not subject to cycles of freezing and thawing, because it is in either a heated environment or a tropical or subtropical climate, then weathering is ruled out and the trouble must be chemical attack or abrasion. Also, unless the deterioration occurs in a location which is frequently saturated, it cannot be weathering. Next, try to eliminate abrasion. See if the deteriorated area is subject to any such action. If the structure is such that abrasion could be a problem, see if there are any signs of wear or polish of the particles of coarse aggregate. If the action of the abrading agent has not been masked by the pounding of wheels or subsequent weathering of the surface, the aggregate should show some polished surfaces or some striations. See if the particles are crushed or broken, indicating pounding by traffic. Now, check how deep the decay of the concrete penetrates the mass. If the deterioration of the cement paste extends deep into the con-


Deterioration, Maintenance,

and Repair of Structures

crete, then the trouble is probably a chemical attack. For example, in one case in the author’s experience, where the difficulty was sulfate attack on a concrete seawall of gravity design, the concrete could be removed with a pickax to a depth of 2 feet. Frost could hardly have penetrated that deeply, especially without spalling away sections of the mass. If, on the other hand, the depth of decay is relatively shallow, say an inch or two, then the trouble could be either chemical attack or weathering, more likely weathering. Next, look for the usual symptoms manifested by concrete subject to chemical attack, i.e., aggregate particles protruding from the matrix and a loss of cementation in the cement paste (see Figures 3-13 to 3-15). As a further check, remove several samples from the deteriorated and sound portions of the concrete of the structure, and have them subjected to chemical and petrographic analysis. Be sure the samples are from the same pour. Compare the results. If the constituents are present in like proportions, rule out chemical

attack. If the proportions have changed or if a new constituent appears, or if one or more of the original constituents are absent or have been markedly depleted, chemical attack should be suspect, and the constituent involved in the change provides a clue to the reaction taking place.

b. Where There Is Evidence of Swelling (Growth) of the Concrete.

This will have been determined in Step 4. If so, there are three possible causes—chemical reaction, absorption of moisture by the concrete, or a rise in the temperature of the concrete mass. Since all three agents may act on similar types of structures, it is difficult to tell which reaction is occurring. Try reading the internal temperature of the concrete, in the inspection galleries, if there are any, or in a bore hole. The bore hole may also be used to recover cores for chemical and petrographic analysis and comparison with sound samples of concrete taken from the structure, as described in paragraph a, above. Check the environment. Is water available to saturate the concrete mass? If not, growth due to absorption of moisture obviously cannot


It is also important to plot the rate of growth in the different parts of the structure and to try to correlate this information with the possible causes and the specific environmental conditions of the structure. If the volume changes are accompanied by pattern cracking, the use of unsound materials or some similar incident leading to an unanticipated chemical reaction should be suspected. c. Where the Structure Is Spalling. Having eliminated corrosion of the reinforcement and underdesign (Steps 1 and 3) as causes of the condition, if the defects are localized, the greatest probability

Concrete Structures:

Causes of Deterioration and Preventive Measures 107

is that the trouble is some defective design detail such as those described in Section B, paragraph 10. If such is the case, the defect in the concrete and the defective detail will probably be neighbors, so check out the details of the structure near the trouble spot. Another possibility is the occurrence of a chemical attack. Check for the occurrence of ‘‘pop-outs,’’ which are an indication of the alkaliaggregate or similar reaction. If the spalling is general, in the great majority of cases it is due to corrosion of the reinforcement, so go back and check this once more. If it definitely is not due to corrosion, however, it is most likely due to internal temperature changes, weathering, or chemical attack. These possibilities may be checked out and differentiated by the procedures described in paragraphs a and b, above. d. Where the Defects Consist of Cracking. If the trouble were caused by swelling of the concrete due to the absorption of moisture, this would have shown up during the previous steps, so this effect may be eliminated. Accordingly, at this stage of the elimination, five possibilities remain:





shrinkage stresses,


incident to the construction


chemical reactions, and (5) poor design details. First consider temperature

structure and the element the defects occur at some


and shrinkage stresses.


Visualize the

(beam, slab, etc.) which is cracked. Do point of stress concentration, such as near

openings or other abrupt changes in section, or near a cluster of pipe sleeves or duct openings? If so, temperature or shrinkage effects are probable causes. Next, consider any relation between the locations of the defects and the points of restraint in the structure. What stress pattern would be set up by a temperature rise? What pattern would result from a decrease in temperature? Are temperature gradients likely to exist? If temperature gradients could exist, is there any evidence of curling in the structure or of a tendency to curl? Sketch out the stress-flow patterns. Do the cracks run perpendicular to the lines of stress? Does the observed curling of the structural elements conform to the probable patterns of temperature gradients or to the stress patterns which would result from differences in shrinkage due to differences in shape or timing of adjacent pours? If so, the trouble most probably is temperature or shrinkage, and the alternatives are to do nothing or to cut loose the restraints in the structure. The next most likely cause of the difficulty is the design details. As stated above, the details and the defects are usually neighbors and can be so related. If the cracks do not form a pattern in conformance with the occurrence of temperature or shrinkage stresses, consider if their pattern or location is consistent with one or more of the defects occurring


Deterioration, Maintenance, and Repair of Structures

during construction, as described in Section B. If there is or concentric pattern, settlement should be suspect, either overall settlement of the structure or to settlement of the during construction. If the survey made in connection with shows

local bulging of presumably




a radial due to subgrade Step 4

the possibil-

ity of a movement of the forms during hardening of the concrete should be considered. Also, if the cracks run parallel to and near a face of the concrete section, movement of the forms during hardening of the concrete is probable. If the cracks occur as a horizontal cleavage plane under the top reinforcing mat, settlement of the concrete suspension is a likely cause. If the cracks are isolated, the probable cause is some positive overstress, perhaps due to the action of external loads or to the effects

of shrinkage

shrinkage, or some


and temperature.



design deficiency should be suspect.

in particular,

follow the lines of principal

textbooks and are relatively easy to differentiate. Other clues to consider are: (1)


the crack penetrate



the location of the crack correspond


Overload shown


the tension area of the concrete

only, or the compression area as well? (2) Does the crack penetrate the full section? (3) Is the crack deep or shallow, and does it occur internally, externally, or both? For example, if the crack is internal only, movement of the forms should be suspect as should be any other factor tending to cause a volume change in the concrete mass, such as temperature, moisture change, or chemical reaction. Similarly, shallow surface checking is related to shrinkage. with



in pattern, amount of reinforcement, or section? (5) Does the crack line relate to pour sequence, to a section of the structure added at a later date, or to some change of use or occupancy? (6) Does the defect appear to be new or old?

(7) Where and why does the defect stop?

Every attempt should be made to relate these clues to some pattern or to relate a cause and effect.

Step 6. Find Out Why Deterioration Has Occurred Steps 1 to 5 tell, basically, what has occurred. The last step is to find out why it has occurred. What was improperly done in the design or construction which permitted the deteriorating agent to act? To find out, go back to the discussions of Section B relating to means for preventing the action of the agent which has been found to have caused the problem in question, see what provision of the rules for

Concrete Structures: prevention was

Causes of Deterioration and Preventive Measures


in the repair work.

and be sure

For example,

that the violation

in Case History


is not repeated

5-3, the basic

trouble was the use of Type III cement in a marine environment. This resulted in sulfate attack on the concrete. Suppose that it had been decided to repair the disintegrated concrete by jacketing. The rule

of compatibility (Chapter 5) would have required the use of cement in the jacket, with obviously unsatisfactory results. case the rule of compatibility could not have been observed, one of the reasons that jacketing was not adopted for repair particular case.

Type III In this which is in this

Conclusion The preceding discussion has described a highly idealized procedure for diagnosing the cause of deterioration in concrete. It can be expected to work fairly well if the design has been well executed, if the construction was properly performed, and if only one deteriorating agent is at work on the concrete. Sometimes these conditions prevail. More commonly, however, the design is fair, the construction is fair, and several forms of deterioration are simultaneously in action. This is only natural. As soon as the standards of design

Figure 3-36a. Some results of cold joints in a concrete structure. The result of a cold joint in a retaining wall which was constructed without proper attention to bonding the two pours. Water penetration and freezing has spalled the corner of the wall.


Deterioration, Maintenance, and Repair of Structures

Figure 3-36b.

pour joint shows how provided. have been

The result of a cold joint in a bridge deck.

runs through the slab (A) and diagonally down the concrete was run out between pours, with In this case nothing serious has happened, but serious if the joint had occurred in an area of

Note how the

the beam (B). This no end bulkhead the results could high shear stress.

or construction are lowered, the doors are open to deterioration,

and all kinds of aggressive agents rush in with all manner of unex-

pected and unpleasant results. And, of course, for every usual case there are probably one or more unusual cases that do not fit the typical patterns described. The result is that the diagnosis is complicated and, in a sense,

becomes more a matter of possibilities than of probabilities. as indicated in the discussion,

In fact,

even for relatively quite simple cases,

no clear conclusion may appear at all, in which case the repair procedure must protect the structure against the action of a multiplicity of potential aggressive agents. Most important, it should be clearly understood that the discus-

sion in this section of this chapter is not intended nor can it hope to be anything more than a first approximation to the truth, and is not a substitute for the careful collection, consideration, and evaluation of all facts pertinent to the problem

at hand.

In fact, it is presented

only on the basis that a first approximation is better than nothing at all, and the reader is specifically cautioned that careless

use of the

data, out of context or by oversimplification, can be misleading

Concrete Structures:

Causes of Deterioration and Preventive Measures


‘Step 1: Consider Possible

Design Errors Basic Symptom

Disintegration] Spailing| Cracking

Subroutine T (Stress Analysie) t {Conclusion {Conclusion|}}— [Fen


Step 2:

Causes Relate to

Basic Symptoms


Pp 3:





|_| Subroutine 2

(Disintegration) {


[* [iswetting)

{ ‘Conclusion }




[ [spatting)





f Conclusion }

|_| Subroutine 5 (Cracking) }

J Conclusion}

[ No Conclusion] Figure 3-37. Concrete structures. Basic flow chart.

‘Step 6: Why has

Deterioration Occurred? Diagnosis of causes of deterioration.


Deterioration, Maintenance, and Repair of Structures

Construction Shrinkage Temperature Moisture Absorption

Corrosion Chemical Weathering Shock Waves




Eleven Basic Causes

9. Abrasion 10 Design Details N.Design Errors


Step1: Consider cause No. ll (Design Errore) Check



Disintegration] Spalling [Cracking What types of etre could cause the observed symptoms? i Tension Compression Shear Torsion

Are these types of stress consiatent with the locations of the defects ?


I |' ' !

[Are these types of stress [consistent with the orientlation of the defects?


' | 1 T '

' '

[Stress Analysis

[Shows No Apparent Overstres

[Shows lOverstre:

Ten Basic Causes


Construction Shrinkage Temperature Moisture Absorption




Detailed flow chart.

5. 6 7. 8


Corrosion Chemical Weathering Shock Waves

9 Abrasion 10, Design Details



of causes

of deterioration.

Concrete Structure.

: Causes

of Deterioration

and Preventive



Step 2: Relate the Defect to the basic aymptoms (See Table 3-2}

Cracking Remaining Po

Remaining Possibilities:

Remaining Possibilities:

6. Chemical

3b. Internal Temperature

1. Construction





2. Shrinkage





3. Temperature



4. Moisture Absorption

Shock Waves

5. Corrosion

10. Design Details

6. Chemical 8.


Shock Waves

10 Design Detaile

‘Step 3: Check Corrosion and Shock Waves (Causes which are readily iqentified) tive}>-—

—fReeulte Positive}


———ee ‘Summary Symptom - Disintegration Symptom - Spalling

Symptom - Cracking

6. Chemical 7. Weathering

3b. Internal Temperature 6. Chemical

1 Construction 2) Shrinkage

9. Abrasion




Design Details


Temperature 4. Moisture Absorption 5. Chemical 10. Design Details

Investigation [Step 5: Analyze the


fit the Concrete hae Disintegrated

[Possible Causes

6. Weathering 7. Abrasion 9. Chemical



[Check for Unsound] Materials



| | {



Deterioration, Maintenance, and Repair of Structures

' |

[Check Weathe rin,




I | |



| | i 1 1

| 1 1 | | | I |

To Structure Subject to Abrasion and do

Particles of Aggregate show abrasion?|

[Results Negative}=fofResulte Positive] (Check Chemical Attack

| |

Depth of decay, Chemical analysis

Other Pertinent

| | |

No Conclusion,


eee [itthere is Evidence of i Swelling



| |



1 !

Ts Structure Subject to Cycles of Freezing Temperature and to Moisture?


[Check Abrasion


' ! ! | ! I |

Repair to Resist Weathering, Abrasion, and Chemical Attack

rating 2



1 |e. cremicnt

| 4. Moisture Absorption



| | 3b. Internal Temperature t

! '

Check Internal poo Temperatures


| '\ !

Check Moletarel


ther Pertinent

! 1

Evidence a ee


Figure 3-38.








No Conclusion. Repair to reeiet Chemical Attack, or to Accommodate Volume Changes ES


Availability of







' |

I i i

Tnepection Galleries and Bore Holes

ite Negat


1 |

|| [Possible Causes i

' \ \

! I | | i | \


| 1

| |


If the Structure ie Spalling

| |


Possible Causes: 3b. 6. 7. 10,




Internal Temperature Chemical Weathering Design Details


| |





Design Details

| |

[Results Positive] [Check Internal Temperature


Inspection Galleries] and Bore Holes


Te Structure Subject


to Cycles of Freesing Temperature and to Moteture? jeck





Depth of Decay: ‘Chemical Analysis] "Pop Outs"


| |

[Other Pertinent]

Evidence _ No Conclusion Repair to Resist Attack by all Four Agents ‘Subroutine Evidence



if the Defects Coneist|

[Of Cracking

| Jj


| |

I Possible Caus:

| |












|. Design Details

| |


Defects to Stress

| !


| 1 | Figure 3-38.





and Repair of Structures



| ! !

| i

Depth of Cracks and Regions Penetrated





‘No Conclusion. Repair to Correct and Prevent all Possible Causes of Deterioration

[Step 6; Why has Deterioration Occurred? Figure 3-38.


rather than helpful. The reader also is cautioned that, unless a clear and positive conclusion appears, the diagnosis procedure should be carried through to the end lest the effects of some pertinent secondary or contributory aggressive agent be overlooked. In addition, reference

is made

to Chapter

1, which describes



evaluating the strength of an existing deteriorated construction. Such evaluation is essential whatever the type or cause of the observed defects. —

Flow Charts for Diagnosis The procedures described in the text of this section have been set up, for purposes of visual reference and clarity, in Figures 3-37 and 3-38, in the form of a flow chart. The basic flow chart without all of the details and explanations is presented in Figure 3-37, giving an overall picture of the suggested procedure. The basic flow chart also will be useful for purposes of orientation when a new problem is


The detailed flow chart is presented in Figure 3-38. As will be noted, it is long and somewhat complicated. However, not all problems are simple nor, as in this case, do all questions have a clear or easy answer. It is necessary to do the best one can. The detailed flow chart is for reference and development of the diagnosis after the basic chart has been used for orientation. Also, it will be useful as a check list against the possibility of omitting some necessary or pertinent step or detail.

Concrete Structures:

Causes of Deterioration and Preventive Measures



‘‘Pressure of Concrete on Formwork,’’ Bulletin ST3 of Portland Cement Association, Structural & Railways Bureau.


‘‘Building Code

3. 4. 5.

6. 7. 8. 9. 10. 11. 12.

13. 14. 15.



for Reinforced

Journal of the American Concrete

vol. 52, May, 1956.



Institute, Proceedings,

‘Shrinkage Control Cuts Deck Cost,’’ Engineering News-Record, May 10, 1962, p. 35. S. Champion, ‘‘Failure and Repair of Concrete Structures,’’ John Wiley & Sons, Inc., New York, 1961. J. F. Schaufele, ‘‘Erosion and Corrosion in Marine Structures, Elwood, California,’’ Proceedings of the First Conference on Coastal Engineering, Council on Wave Research, University of California. ‘‘Recommended Practice for Selecting Proportions for Concrete,’’ American Concrete Institute (ACI 613). ‘Recommended Practice for Selecting Proportions for Structural Lightweight Concrete,’’ American Concrete Institute (ACI 6134). ‘‘Durability of Concrete in Service,’’ Report of Committee 201, Journal of the American Concrete Institute, vol. 59, no. 12, pp. 1771 et seq., December, 1962. H. Lossier, ‘‘The Pathology and Therapeutics of Reinforced Concrete,’’ National Research Council of Canada, Technical

Translation 1008.

J. Feld, ‘‘Failures of Concrete Structures,’’ Journal of the American Concrete Institute, Proceedings, vol. 54, p. 449, December, 1957; and discussions, vol. 29, no. 12, June, 1958. S. Rosenhaupt, and G. Mueller, ‘‘Openings in Masonry Walls on Settling Supports,’’ Journal of the Structural Division, ASCE, vol. 89, no. ST3, June, 1963. Calvin C. Oleson, ‘‘Abnormal Cracking in Highway Structures in Georgia and Alabama,’’ Journal of the American Concrete Institute, vol. 60, no. 3, March, 1963. ‘Specialist on Sick Structures Speaks Out,’’ Engineering NewsRecord, November 9, 1961, pp. 50-53. G. E. Monfore, and G. J. Verbeck, ‘‘Corrosion of Prestressed Wire in Concrete,’’ American Concrete Institute Journal, Proceedings, vol. 57, no. 5, pp. 491-515, November, 1960. I. L. Tyler, ‘‘Long-time Study of Cement Performance in Concrete, Chapter 12, Concrete Exposed to Sea Water and Fresh Water,’’ American Concrete Institute Journal, Proceedings, vol. 56, no. 9, pp. 825-836, March, 1960.





Deterioration, Maintenance,

P. D. Miesenhelder,

and Repair of Structures

‘‘Effect of Design and Details on Concrete

Deterioration,’’ American Concrete Institute Journal, Proceedings, vol. 56, no. 7, pp. 581-590, January, 1960. B. Tremper, J. L. Beaton, and R. F. Stratfull, ‘‘Corrosion of

Reinforcing Steel and Repair of Concrete in a Marine Environ-

ment, Part II,’’ Bulletin pp. 18-41.

182, Highway Research



Bryant Mather, ‘‘Factors Affecting Durability of Concrete in Coastal Structures,’’ Tech. Memorandum no. 96, Beach Erosion Board, June,


19. S. Halstead, and L. A. Woodworth, forced Concrete

‘‘The Deterioration of Rein-

Structures under Coastal Conditions,’’


actions, South African Institution of Civil Engineers, vol. 5, 20.

no. 4, pp. 115-134, April, 1955. L. B. Mercer, ‘‘General Characteristics of Concrete Cracking,’* Crushed Stone Journal, vol. 28, no. 4, pp. 6-12, December, 1953.


E. M. Rensaa,

22. 23.

‘‘The Cracking Problem in Reinforced Concrete

Buildings,’’ The Engineering Journal, vol. 36, no. 11, pp. 14291434, November, 1953. ‘*Recommended Practice for the Application of Mortar by Pneumatic Pressure (ACI 805-51),’’ American Concrete Institute Journal, Proceedings, vol. 47, no. 9, pp. 709-719, May, 1951. F. M. Lea and N. Davey, ‘‘The Deterioration of Concrete in

Structures,’’ Journal of the Institution of Civil Engineers, vol.

32, p. 248 et seq., 1949. ‘‘The Corrosion of Reinforcing Steel in Cracked Concrete,’’ American Concrete Institute Journal, Proceedings, vol. 43, no. 10, pp. 1137-1151, June, 1947. W. H. Price, ‘‘Erosion of Concrete by Cavitation and Solids in Flowing Water,’’ Journal of the American Concrete Institute,

24, N. Tremper, 25.

vol. 43, 1947,

26. K. A. Adamchik,

‘‘Causes and Prevention of Deterioration of Concrete in Marine Structures in the Zone of Fluctuating Water Level,’’ Corrosion of Concrete and its Prevention, Academy of

Sciences of the USSR, published by the Israel Program for Scientific Translations.


A. INTRODUCTION Cracking is one of the most misunderstood problems of concrete. It is almost always a cause of anxiety and generally is regarded as indicative of defective design or materials. Actually, while the occurrence of large, open cracks often is evidence of poor detailing and/or construction, in the usual case neither the design nor the construction is particularly to blame. It is simply that, except in very elementary structures or for individual elements, cracking is almost omnipresent. This fact is of the greatest practical importance. Concrete structures often are used where watertight integrity is required. However, unless specially detailed and/or constructed, such structures

crack and leak.

This does not mean that a

flood of water is re-

leased. Usually, there is nothing more than a slow, steady drip. Nevertheless, even such minor flow will cause an unsightly efflorescense on the concrete surface, and if the lime-charged solution happens to fall on furnishings or equipment, it quickly destroys the paint or other finish or causes some other form of water damage. The design engineer must consider this factor in his work. Where conditions of exposure and use are such that any cracks which may 119


Deterioration, Maintenance,

and Repair of Structures

form are not subject to a hydrostatic head and, accordingly, cannot leak, or where the damages resulting from some leakage would not be important, it isnecessary merely to provide a sufficient number of relieving or expansion joints so that large, opencracks do not occur (or to provide ample reinforcement to the same effect) and to let the little cracks form as they will. After all, if they do not leak or if a little leakage is not important, the presence of some minor cracking does no harm, and the expense of special details or construction procedures required to minimize cracking is not warranted.

On the other hand, if leakage would be objectionable, prevention

is the necessary




the use

of expansion


precooling of the mix, frequent pour joints, or other techniques as described


in Chapter

If cracking


is as follows:


In addition, waterproofing



if repair


is required,

be pro-

the procedure

Step 1. Determine Whether the Cracks Are Active or Dormant Checking

the activity

of a crack

is done by means

servations utilizing telltales, as illustrated in 4-1a shows how movement can be detected by end of the crack. Subsequent extension of the indicates probable continuance of the activity the defect. The deficiency of this technique is any tendency

for the crack

to close or provide

of periodic


Figure 4-1. Figure placing a mark at the crack beyond the mark originally producing that it will not show any quantitative


on the movement. Figure 4-1b shows a similar device. A pin or toothpick is lightly wedged into the crack and falls out if there is any extension of the defect. The deficiencies, as before, are that there is no indication of closing movement or any quantitative measure of the changes which occur. The strip of notched tape shown in Figure 4-1c works similarly. Movement is indicated by tearing of the tape. An advantage isthat some indication of closure can be realized by observing any wrinkling of the tape. The device is not reliable, however. The tape (whether paper, cellophane, or cloth) is not dimensionally stable under changing conditions of humidity, so that one can never be sure

whether the movements are real or are due to shrinkage or swelling

of the marker. Spark plug feeler gauges may be used to measure the crack opening. However, the apparent size of the opening changes if the gauge point gets dirty or if the feeler is not inserted to a constant depth. The device shown in Figure 4-1d is the most satisfactory of all. Both





indicated and



1/100 inch can be measured using a verniered caliper of the type

Concrete Structures:


Repairing Cracks

Pin (toothpick, for example) lightly wedged into crack. {-] Extension of crack causes pin to loosen and be displaced. 8) Mark End of Crack

b) Pin

Typical Calipers used to measure Distances Between Gauge Points

Zero distance determined

fape. Exteneion of crack causes tape to

Gauge points

¢) Tape

) Gauge Pointe Figure 4-1.


If more



readings are desired,


can be

used. The reference points must be rigidly constructed and carefully glued to the surface of the concrete, using a carborundum stone to prepare the bonding surface before attaching the reference marks. Where extreme accuracy is required, resistance-strain gauges can be glued across the crack. They are, however, expensive, sen-

sitive to changes in humidity, and easily damaged.

In connection with this step, it should be considered that, in a sense, every crack is an active one. For example, if there is any change in the load supported by a member, whether an increase or a decrease, the member must deform to accommodate the change, and the deformation is bound to occur and be concentrated at points

of weakness or articulation, i.e., at the cracks.

The same thing is

true for stresses induced by temperature change, to which every structure is subject in some degree. Thus, the proper differentiation


Deterioration, Maintenance,

and Repair of Structures

between active and dormant cracks is one of magnitude of movement, and the telltales are a measure of the difference. If the magnitude of the movement, measured over a reasonable period of time (say six months or a year), is sufficient to displace or show significantly on the telltales, treat the crack as an active one. If the movements are smaller, the crack may be considered dormant. Step 2. Determine the Cause The procedures described in Chapter 3, Section C, are applicable. Step 3. Select a Method of Repair Selection of an applicable method of repair involves consideration of the following:

(1) Are the cracks active or dormant?

(2) What is the primary purpose of the repair? Is it just to reduce excessive leakage, or must the cracks be made fully waterproof? Is strengthening required? (3) How do the cracks occur? Are they pattern cracks, i. e., large numbers of relatively narrow crevices, or are they large, isolated defects? (4) What is the magnitude and direction of the anticipated future movements? The process of selection proceeds in accordance with the flow diagrams contained in Figures 4-2 and 4-3 and in accordance with the following discussion. ACTIVE


[Pattern Cracks

Required] This is an improbable| occurrence

Isolated Cracks

Not Required] ens il | overlay VE VJ

Required] 1

Mode of occurrence

[Not Required]

Strengthening requirement


|Applicable methods f repair



/ ler


. Stitching? 3. Ext. Stress. 7 2.

Figure 4-2, Note that the following abbreviations have been used: Ext. stress.—external stressing; ?—applicable but not recommended except for very mild or temporary conditions.









OO set cere Strengthening requirement Water condition

T, Rout and Seal]

2, Jud /Neglect 3. . Heal. 4. Overlay

Rout and Seal] Overlay Jud. Neglect? Auto. Heal. 7


Applicable methods of repair

Mode of occurrence

ye strengthening requirement


7 Epoxy

. Stitching ]




J [fi-Ext.Streea] Ext. streee] [1 Ephy

1. Epoxy

+ Epoxy

wis | |+ Epoxy

Rout and Seal] 2, |2, Grout 3 Auto. Heat. ||4, Jud. Neglect ||5.5 Blanketing 6



|. Auto. Healf. Auto. Heal.


4) [

Rout and


Water condition


Applicable methods of repair

Blanketing Jud. Neglect 7 Auto. Heal. ?|

Figure 4-3. Note that the following abbreviations have been used: Ext. stress.—external stressing; auto. heal.—autogenous healing; jud. neglect—judicious neglect; ?—applicable but not recommended except for very mild or temporary conditions.

a. If the Cracks Are Active.

First determine whether the cracks

occur as a few isolated defects or as a general or random pattern, and then decide whether or not it is necessary to restore the tensile strength of the concrete across the cracked section.

(1) Pattern Cracks.

As will be noted from Figure 4-2, and as

discussed in Section C of Chapter 3, pattern cracks are seldom associated with a condition of positive overstress, and strengthening normally is not required. The most commonly applicable repair for a pattern

of active


is an extensible


of one of the

types described in Section B, paragraph 6, of this chapter (Overlays). (2) Isolated Cracks. If the crack is isolated and active, attempting to restore the tensile strength of the concrete across the cracked section will simply cause the trouble to migrate to some other area of the structure. Thus, if the crack cannot be tolerated from structural considerations, it is best to proceed on the basis that an exPansion joint should have been provided in the structure in the original design at or near the location of the crack and that if the designer neglected to provide one, the repair should provide it. If necessary, redesign the framing as required to suit the new stress pattern introduced by the presence of the joint. Where the provision of an ex-


Deterioration, Maintenance, and Repair of Structures

pansion joint is not possible, some good can be done by stitching or posttensioning. In this connection, it should be noted that the crack itself tends to act as a joint and that the structure, at the time it is being investigated, is probably functioning with an articulation which was not provided for in the original design. Accordingly, the stress pattern in the structure has been modified by the presence of the defect, and this must be considered when analyzing the residual stress condition and the need for jointing. If the crack


not create

an unacceptable

it should be tolerated and simply


of overstress,

sealed, preferably by blanketing.

Next, consider what kind of movement must be accommodated in the joint. Is it limited to extension and compression (which is usu-

ally the case with pattern cracks), or is there a component of movement


or longitudinal

of the joint (which

often occurs


isolated defects)? The details of the repair will depend on the types movements which are anticipated. b. If the Cracks Are Dormant. As before, first determine if the



as isolated

or pattern defects,


then decide



or not it is necessary to restore the tensile strength of the concrete across the cracked section. Again, pattern cracks normally are not associated with a condition of positive overstress, and strengthening usually is not required. The applicable methods of repair for each condition are indicated in Figure 4-3. Note in the figure that the use of external stressing alone is not recommended for strengthening where


is a simultaneous


is a good choice,



repair should be accompanied by some since

it helps

and that a posttensioned

form of joint sealing.

to restore

the tensile



across the crack while functioning as a sealer. Autogenous healing may be used as indicated. However, as will be noted in subsequent discussions, the technique should not be used without recognizing the limited degree to which it is capable of restoring the tensile strength of concrete. If the cracks are dormant, naturally there is no need to determine what type of movement is occurring. c. Magnitude of Anticipated Future Movement. Some data on this factor will be provided by the telltale observations described in connection with Step 1. In general, however, the available period for making


is too limited to permit


any quantitative

conclusions, and reliance must be placed on calculations which should consider temperature effects, continuance of drying shrinkage, changes of loading, deflections of the structure, possible settlements of the supports, and any variations in the moisture content of the concrete mass. The temperature effects considered must include those distortions (curling, etc.) due to thermal gradients.

Concrete Structures:



Repairing Cracks



1. Bonding with Epoxies Cracks in concrete may be bonded by the injection of epoxy bonding compounds under pressure. Usual practice is to drill into the

crack from

the face of the concrete

at several


inject water

or a solvent to flush out the defect; allow the surface to dry (using a hot-air jet, if needed); surface-seal the cracks between the injection points; and inject the epoxy until it flows out of the adjacent sections of the crack or begins sure grouting.

to bulge out the surface


just as in pres-

Reference 1 describes the technique and suggests routing the crack (say 1/4 inch wide by 1/2 inch deep) to stiffen the surface seal. The epoxy is injected through holes about 3/4 inch in diameter and 3/4 inch deep

at 6- to 12-inch

centers (the smaller


are used for

finer cracks). Injection is made through tire valve stems fastened in the drilled holes with an epoxy bonding compound, and the injection process



point to point,

being capped except the one being injected. should be maintained for several the finer parts of the crack.


all valves


in the circuit

injection pressure

to force the epoxy



The requirements for materials, mixing, and application are similar to those described in Section H of Chapter 5. The injection of epoxies into cracks in concrete as a means for bonding the broken surfaces represents an application for which there is no real

substitute procedure.

mant (or the cause of cracking is dormant), it will probably recur, structure. Also, the technique is actively leaking to the extent that the cracks are numerous.



the crack

is dor-

removed, thereby making the crack possibly somewhere else in the not applicable if the defects are they cannot be dried out, or where

2. Routing and Sealing This method involves enlarging the crack along its exposed face and filling and sealing it with a suitable material (see Figure 4-4). The routing operation may be omitted, but at some sacrifice in permanence

of the repair.



is the objection

that the surface

of the sealer will be higher than that of the adjacent concrete. This is the simplest and most common technique for sealing cracks and is applicable for sealing both fine pattern cracks and larger isolated defects. The cracks should be dormant, unless they are opened up enough to put in a substantial patch, in which case the repair may

be more properly termed blanketing and is discussed in the following


Deterioration, Maintenance, and Repair of Structures

1/4" Minimum

a) Original Crack

Groove cut with Saw or chipping Tools.

») Routing Figure 4-4.

Joint Sealer,

c) Sealing

Routing and sealing.

paragraphs. The technique is not applicable for sealing cracks subject to a pronounced hydrostatic pressure except where used on the pressure face, in which case some reduction in the flow can be


The sealant may be any of several materials, depending on how tight or permanent a seal is desired. Epoxy compounds may be used, subject to the limitations described in Chapter 5. On roadway pavements it is common to see cracks which have been sealed by pouring hot tar over them. This is a simple and inexpensive expedient where thorough watertightness of the joint is not required and where


is not important.

The various hot-poured joint sealers work very well for this purpose. There are many commercial types, and the manufacturer should be consulted as to the type and grade most applicable for the specific purpose and conditions of exposure. Be certain that the material does

not flow at the temperatures which will be encountered

in the given

exposure, that it will bond to the adjacent concrete surfaces, and, if used in pavement, that it will bear traffic. The routing operation consists of following along the crack with a concrete saw or with hand or pneumatic tools, opening the crack sufficiently to receive the sealant. A minimum surface width of 1/4 inch is desirable. Smaller openings are difficult to work on. The surfaces of the routed joint should be rinsed clean (in any areas where the routing has opened only one side of the crack, be certain to remove any grease, oil, or contaminants clinging to the side which is not chipped) and permitted to dry before placing the sealant. The method used for placing the sealant depends on the material to be used and follows standard techniques. Routing and sealing of leaking cracks preferably should be done on the pressure face so that the water or other aggressive agent cannot penetrate the interior of the concrete and cause side effects such as swelling, chemical attack, or corrosion of the bars.

Concrete Structures:

Repairing Cracks


3. Stitching The tensile strength of a cracked concrete section can be restored by stitching in a manner analogous to sewing cloth. Figure 4-5 illustrates the method. The following points should be observed: (1) Any desired degree of strengthening can be accomplished, but it must be considered that the strengthening also tends to stiffen the structure locally. This may accentuate the restraints causing the cracking and reactivate the condition. (2) Stitching the crack will tend to cause the problem to migrate elsewhere

in the structure.

If it is decided

to stitch, investigate


if necessary, strengthen adjacent areas of the construction to take the additional stress. In particular, the stitching dogs should be of variable length and/or orientation and so located that the tension transmitted across the crack does not devolve on a single plane of the section, but is spread out over an area. Strengthening of the adjacent sections of concrete may consist of external reinforcement embedded in a suitable overlay material (such as pneumatically applied mortar). Note variable length, location, an¢ orientation of doge so that tension acrose crack is distributed in the concrete rather than concentrated on a single plane.

Holes drilled in concrete to receive dogs. Fill holes with non- shrink grout.

Figure 4-5.

Repair of crack by stitching.


Deterioration, Maintenance, and Repair of Structures

(3) Where there is a water problem, the as well as stitched so that the stitches are the stitching itself will not seal the crack. pleted before stitching is commenced both corrosion

and because

the presence

crack should be sealed not corroded, and because Sealing should be comto avoid the aforementioned

of the dogs


to make

it dif-

ficult to apply the sealant. An exception should be made in the case of active cracks where the structure must be stabilized before sealing lest the movements of the concrete break the seal. Sealing preferably should be done from the pressure face so that corrosion of the old reinforcement will stop. (4) Stress concentrations occur at the ends of cracks, and the spacing of the stitching dogs should be closed up at such locations. In addition, consideration should be given to drilling out a hole at each end of the crack to blunt it and so to relieve the concentration of stress.

(5) Where possible, stitch both sides of the concrete section so

that further movements of the structure will not exert prying or bending action on the dogs. In bending members it is possible to stitch one side of the crack only, but this should be the tension side of the section, where movement is originating. If the member is in a state of axial tension, then a symmetrical placement of the dogs is a must, even if excavation or demolition to opposing sides of the section.

is required

to gain


(6) If the stitching is to supplement the strength of the existing section, the deformations must be compatible. The dogs must be grouted with a nonshrink or expanding mortar, so that they have a tight fit, and movement of the crack will cause the simultaneous stressing of both old and new sections. If this is not possible, proportion the stitching to take the entire load without participation by

the existing materials.

The holes for the legs of the dogs should be

completely filled with grout. (7) The dogs are relatively thin and long and so cannot take much in the way of compressive force. Accordingly, if there is a tendency for the crack to close as well as to open, the dogs must be stiffened and strengthened by encasement in an overlay or by some similar


4. External Stressing Development of cracking in concrete is due to tensile stress and can be arrested by removing these stresses. Further, the cracks

can be closed by inducing a compression force sufficient to overcome

the tension and to provide a residual compression. The compressive force is applied by use of the usual prestressing wires or rods. The principle is very similar to stitching, except that



Repairing Cracks


the stitches are tensioned, and the points noted under the discussion of stitching, particularly with regard to migration of the problem area, must be considered. An additional problem is that of providing an anchorage for the prestressing wires or rods. Some form of abutment is needed for this purpose, such as a strongback bolted to the face of the concrete; or the tendons can be passed through and anchored in connecting framing (see Figure 4-6). The effects of the tensioning force (and of any eccentricities introduced in anchoring the wires) on the stress conditions in the structure should be

analyzed. The compressive force also may be applied by wedging, i.e., opening the crack and filling it with an expanding mortar; by jacking and grouting; or by actually driving wedges. Figure 4-7 shows the principle. If an expanding mortar is to be used, the discussion of Section A of Chapter

__ f=

5 must

be considered.

“7 Toes.

| Ske

a =]


Tension Tie



a) To Correct Cracking of Slab

b) To Correct Cracking of Beam Figure 4-6.


rod tensioned

by torquing nuts (or using turnbuckles).

of external stressing.


Deterioration, Maintenance, and Repair of Structures Ende (Ties

of Tension Ties embedded in new



\ L\\

° .3 *


Existi wall

New Pilasters


To Correct Cracking of Wall Figure 4-6.


5. Blanketing Blanketing is similar to routing and sealing, but is used on a larger

scale and is applicable for sealing active as well as dormant cracks and joints. The following are the principal types of blanket joints:

a. Type I—Where an Elastic Sealant Is Used.

The type of construc-

tion to be considered is shown in Figure 4-8. The sealant material is one which returns to its original shape when not under an externally induced stress, i.e., acts elastically. The recessed configuration (Figure 4-8a) is used where the joint is subject to traffic or a pressure head. The type shown in Figure 4-8b is applicable where there are no traffic or pressure problems and is somewhat less costly. The deformation of the sealant due to movement of the joint is

illustrated in Figures 4-8c and 4-8d.

As may be noted, a bond

breaker should be used at the bottom of the chase (or adjacent to

the crack opening, as in Figure 4-8b) so that the sealant is free to

deform. Otherwise, the stress in the top fibers would be increased some 100 per cent compared with what it would be with the bond





Repairing Cracks


Load (Symmetrical, but partially applied)

HILTNLA Wedges, Expansive Mortar, Jor Jacke-used in key cut in extrados.

Wedges, Expansivi Mortas,or Jackeused in key cut in intrados

B_ expansive Force

a) Diagram of Structure

Deflection due to| expansive forces

riginal Centerline and idealized net deflection curve resi from deformations Jdue to load and expansive forces. estes to load


b) Defection Diagram Figure 4-7. An example of the use of wedging, for external stressing.

jacking, or expansive keys

breaker, and the bottom fibers will tend to tear. include polyethylene,



(1) Dimensions of the Chase

and foil.

Good bond breakers

(a) Width. The required width of the chase is related to the depth of the sealant, its physical properties, and the amount of movement which

is anticipated.


it is required that the sealant not tear.

This means that the tensile stress due to movement in the crack should not exceed the cohesive capacity of the material. The tensile stress is a function of the amount of movement in the joint and of the ratio of the depth of the sealant to its unbonded width in the chase. Figure 4-9 shows how the strain in the fibers of the sealant may be determined

from the anticipated

from Reference figure

is the per





is taken

The per cent linear expansion to be used in the

cent expansion

of the unbonded

section of the seal-

ant, and not the per cent increase in width of the crack or joint.) The per cent strain in the sealant determined from Figure 4-9 may be converted into stress by multiplying by the elastic modulus of the sealant material, bearing in mind that this modulus may vary with the rate of movement and with temperature and, therefore, should be obtained from the manufacturer for the proposed conditions of use.


Deterioration, Maintenance, and Repair of Structures

Strip Sealant

0) Sealed Chose,

b)Sirip_Seolont Movement


| Pet voint -—

[of voint




. / Tensile


Bond Breaker-


Tearing of jealant

Crack ¢)Detormation of Sealont (with bond breaker)

S)Deformotion of Sealant (without bond breaker)

Figure 4-8. Joint with an elastic sealant. (Values range from about 20 to over 50 pounds per square inch.) The stress, so calculated, should not exceed the tensile or bond

capacity of the sealant material divided by a suitable safety factor

(say 4 or 5). The tensile stress in the fibers of the sealant, calculated as described in the above paragraph, must be resisted by the concrete on the sides of the chase. This, also, should be calculated and should not exceed, say, 0.02f¢ to 0.025f¢, that is, the modulus of rupture divided by a suitable safety factor. A safety factor of 4 or 5 is recommended where the joint to be sealed is a chase, sawed or chipped in sound concrete. If the joint is a formed discontinuity between adjacent pours, an even greater safety factor should be used, because the formed face tends to accumulate a rich but weak mortar or a laitance. If the necessary strength cannot be achieved readily in the concrete, the sides of the chase may be armored with edge angles, expansion-dam assemblies, or some similar device. In any event, the width of the chase should be at least four, and preferably six or more, times the anticipated amount of movement

in the joint, not only for purposes of avoiding excessive tensile and

bond stresses, but, in the event of closing movement of the crack, to eliminate compression set (i. e., compression beyond the recoverable capabilities of the material) or excessive extrusion of the mate-

rial from the joint.

(See also Figure 4-10.)

Concrete Structures:

Repairing Cracks



0: x


0: Pia










(oint width

when filled)




Lr é.

Strain along



© 180





" Linear expansion, %


Figure 4-9. Relation between unit strain in joint sealant and per cent movement in joint. (Courtesy of Raymond J. Schutz.)

(b) Depth. The required depth of the chase is related to the width, as described, in order to prevent excessive stress in the sealant and concrete. It also will depend on the pressure head acting on the seal and the width of the crack or joint which

has to be spanned.

To this

extent, the layer of sealant must function as a bridge, and its ability to act in this manner is related to its strength and elastic (or creep)






be checked with the

manufacturer, and if there is an active pressure, the ability of the sealant to bridge the opening must be checked by computation.) The usual minimum value for the depth of the sealant is 3/4 to 1 inch, except in cases of the type shown in Figure 4-8b. If the depth required for purposes of bridging the existing or anticipated opening is excessive with respect to limiting tensile stresses in the sealant material,

the width of the unbonded

section of the chase

increased to effect the required reduction. internal pressure,

this means



When working against an

that the effective span of the



Deterioration, Maintenance, and Repair of Structures

Figure 4-10. Shape factor in an expansion joint. (a) An overall joint. (b) A detail of the deterioration. Note that the joint failed crete rather than in the sealer. The condition may be corrected the joint to change the shape factor or by armoring the edges of with edge angles.

view of the in the conby widening the joint

Concrete Structures:

Repairing Cracks


so more

is increased,


depth is required,

and so on, in circles.

In some cases the effects are divergent, and an effective seal cannot be achieved without resorting to some special measures to reduce the acting pressure or to armor the joint. In extreme cases it may be necessary to use some other type of detail. Note that it is not necessary to center the chase on the crack. (2) Preparing the Chase. The chase usually is cut square, as shown

in Figure





be chipped

as smooth,


level, and as clean as possible to facilitate breaking the bond between sealant and concrete. The sides of the chase, on the other hand, should be prepared to provide a good bond with the sealant material. To this end, they should be dry. Application of a blow torch witha fan flame may be required for this purpose. All loose,


or otherwise



adjacent to the crack should be removed, even though the resulting shape is irregular and the volume to be removed is large. It is no use

to try to bond the new


to poor




is to do the work

at the right time

rather than to plug the leak. If the pressure head on the plug is large, it may be relieved by using weep pipes (with or without headers) as shown in Figure 4-11d. If the concrete adjacent to the chase is porous, dampness will spread to the surface of the slot from the crack by capillary action. In such case, a bond coat of waterproof mortar laid on the sides (and bottom,

if necessary)

of the chase

is required.



son is that if the trouble is leakage, the water is apt to bypass the repair by flowing through the poor matrix. If the resulting chase is smaller than the minimum dimensions specified above, enlarge the opening. If the removal of poor concrete has resulted in an opening larger than that desired, rebuild to the desired dimensions with drypack or replacement concrete, following the rebuilding procedures described in Chapter 5. Provide sufficient time for the patch to cure completely before filling the joint with the sealant. If there is difficulty in drying the chase against the inflow of water from the crack, a temporary mortar plug made with a flash-set additive can be placed to stop the flow (Figure 4-11c). The plug will last a few weeks or months until there is enough movement to break it loose. If the crack is too fine to plug, it can be cut open for this purpose, but it must be remembered that since the sealant ultimately must bridge the width of the plug, the plug should be kept as narrow as possible. In some cases, the water flow is seasonal or intermittent, and the proper

' | {

bond coat

must be properly dried and cured before proceeding to place the sealant. (3) Sealant Materials. A number of compounds are available com-

' ! ‘ '


Deterioration, Maintenance,

and Repair of Structures

mercially which are expressly designed for sealing crack openings of this type. The problem is to select the proper one. The first consideration

is the amount

of movement


and the extremes

of temperature at which such movements will occur. The sealant material should be capable of deforming the required amount under the applicable conditions of temperature. In this connection, it should be considered that most sealants are organic compounds, degrade with age, and will not perform after exposure for several years as they do fresh from the laboratory. There is also a fatigue factor, but, in general, this is taken care of in the safety factor of 4 or 5 previously recommended. The material should be able to take traffic, where this is a consideration; be resistant to chemical spillage, if this may occur; or be capable of being pigmented, if desired. For some sealant materials, the rate of deformation is important. In structures like bridge decks, the joints are subject to a very rapid deformation due to impact. Also, it is very common to find that the bearings lock and suddenly release when the longitudinal force builds up enough to overcome the friction. The sealant materials must be able to take such sudden surges of movement. On the other hand, in tunnels, water tanks, and the interiors of buildings, movements are slow, and the sealant need not have the reserve of strength and elasticity to absorb the shock. This factor must be checked with the manufacturer of the proposed sealant material before deciding on an appropriate


(4) Placing the Sealant. After cutting the chase, attach the bond breaker to the bottom of the slot with some sort of adhesive so that

it will not be displaced when filling the joint. Next, if bonding of the selected sealant material is required, prime the sides of the slot,

and then place the sealant, following the manufacturer’s directions. Cautions. (1) Joints of this type (Type I) should not be covered

with a mortar



to a metal

tread plate,

or otherwise

bonded to a material on the face of the slot, since the necking down of the joint (shown

in Figure

will accommodate


(2) This movements the surface movement



be able to occur


type of joint has limited application where subject to parallel to the direction of the joint or perpendicular to of the chase, or to bending. It is primarily for use where consists of axial extension or compression, although it bending

or transverse

larly if the detail shown in Figure 4-8b is used.

b. Type II—A Mastic-filled Joint.



The detail is similar to that

shown in Figure 4-8a, except that the bond breaker is omitted and the sealant is bonded to the bottom as well as to the sides of the chase. The sealant is a mastic rather than a compound having elastic properties.

Concrete Structures: Repairing Cracks


This type of joint is for use where the anticipated movements are small and where trafficability or appearance are not considerations. The advantage is that the mastic is less costly than the elastic type of sealant material.

c. Type III—A Mortar-plugged Joint. The detail is shown in Figure 4-11. Figure 4-11a shows application where the joint must

resist a pressure

head acting on the face of the plug.



shows the detail where the anticipated pressure head comes from behind. Figures 4-11c and 4-11d show details for sealing the crack

against leakage using a temporary mortar plug and for relieving the pressure on the plug by draining the water through weeps connected to a header embedded in the joint.

(1) Dimensions of the Chase (a) Depth. The required depth is the sum of the required thicknesses of mortar plug and sealant. The mortar provides the strength for the joint, functioning as indicated in Figures 4-1la and 4-11b. Note that the plug resists the pressure

Pressure Heed Head Pressure TLL Ly Ni

= ae



on the joint by arching the

/ Relieving Groove

Arch compression Hicicse’ tr ereasire ead on merlar play


thine stress iN AE Shear| Sf mortar Plug Mastic Sealant

Crack Pressure

a) Where

induced bi


is from b)

Where Interior




.. 7

Pressure of

undercut Chose)

uick-Set aejorlar Plug


Quick-Set geese




0 Leaking Crock Figure 4-11.

is from


‘Meader Pipe


Draining o Leoking Crock

Mortar-plugged joints.


Deterioration, Maintenance,

and Repair of Structures

load to the sides of the chase. Where the pressure acts on the face of the joint, static balance requires the development of tensile and shear stresses between the plug and the sides of the slot. Compres sion and


are developed


the pressure


is internal.

Development of tensile and shear resistance requires that there be a good bond between the plug and the original matrix or that the plug be anchored to the sides of the chase, as described below. Assuming a maximum 45° distribution of the load, the thickness of the plug should be at least half the width of the chase. However, the tensile and shear stresses should be checked, and, if excessive, the plug should be thickened to reduce the distribution of the load to some less acute angle. Determining the thickness of the sealant is primarily a matter of experience. It can be any reasonable amount. One-quarter to onehalf inch is a common figure. Greater thicknesses should be used where



are expected,

with as much

as 2 or 3 inches

used where the cracks or the anticipated movements are of comparable size. Champion (Reference 3) recommends a minimum overall

depth of 1 1/2 inches for the chase.

(b) Width. The required width of the chase is whatever is required to work in the slot. Two or three inches is normal, depending on the depth and accessibility. (2) Shape of the Chase. Theoretically, the edges of the chase should be undercut so that the mortar plug, which dries quicker and more thoroughly at the surface, does not shrink differentially and so pull away from the sides and bottom of the chase. In practice, however, the refinement is seldom necessary, and it is satisfactory practice

to cut the edges


to the existing concrete


face with the bottom of the chase chipped as flat as possible (see Figure 4-12). The following cases are exceptions: (1) When working on the nonpressure face of a crack which subsequently will be subject to a pressure head, it is advisable to undercut the sides of the chase so as to lock the repair into place. (See Figure 4-12b.) (2) When the repair and the adjacent concrete will be finished with a reinforced overlay or heavy unreinforced overlay, it is advisable to avoid an abrupt change of section, which would promote cracking of the overlay. A beveled chase with sides sloped at, say, 1:1 or 2:1 is preferable. (See Figure 4-12c.) Note: Unless there is to be an overlay, as described in the preceding paragraph, do not permit the formation of a dished or wedgeshaped chase (see Figure 4-12d). Such a section will tend to curl because of temperature change or drying shrinkage and will separate from the original concrete section. The thinner the edge of the patch, the greater the tendency to curl. The use of a beveled chase is per-

Concrete Structures:

Repairing Cracks


Start Edge cut with Concrete Saw

Plug of Mortar plus Sealant Plug of Mortar plu Sealant

Start edge cut for chase with pass of concret saw where feasible Crack— Note that chase need not be centered on crack.

Chipped Flat, but Roughened




b) Undercut Chase (For use when sealing ag internal Pressure head)


Configers figuration)

[Note that rigidity of ithe overlay prevents the curling shown in Figure d.

Note curling due to differential shrinkage of the mortar plug resulting from irregular shape and| volume.

ont Lp lto2 Crack

¢) Beveled Chase (For use where repair ie to be covered with an overlay) Figure 4-12.

d) Dished Chase (Not recommended)

Type III repair—configuration of the chase.

missible, however, where there is a heavy overlay, since the rigidity of the overlay tends to prevent the curling as described above. It is not necessary that the chase be centered on the crack.

(3) Preparing the Chase.

cleaned of to fill the of Type I, replacing

All surfaces of the chase should be

deposits which would impair the bond of the material used joint, and they should be completely dry. The provisions paragraph (2), relating to removal of unsound materials, concrete, and plugging, drying, and sealing the joint, apply.



(4) Filling the Chase.


and Repair of Structures

The sealant may be bitumen or any one of

several mastic joint sealers commercially available. The compound should be one which will flow viscously under strain without snapping. The mortar plug must be thoroughly bonded to the sides of the chase and should be grooved with the groove centered over the crack. If the crack follows an irregular plan, the groove must follow. If the crack breathes, the groove should be full depth of the plug and wide enough to take the closing movement without shutting tight. If the movement of the crack is limited to extension, a thin groove of partial depth, as shown in the figure, is preferable.

(5) Anchoring the Mortar Cover.

Theoretically, the mortar cover

could arch the pressure load in the crack to the sides of the chase by bond alone. In practice, unless the head of water is limited to a few feet, it is difficult to get enough bond strength to get the necessary arch action, and it is best to anchor the mortar plug, in place. In many cases, this happens automatically, because the depth of the chase has to be 2 or 3 inches and, therefore, penetrates behind the plane of the reinforcement in the concrete section being repaired. A simple header bar, perhaps 1/4 inch in diameter, will anchor the cover and span it between the main rods (see Figure 4-13). If this is not feasible, consider doweling the cover into the parent mass, either by drilling into the sides of the chase or, perhaps, by anchoring back into the open crack.


(1) This type of repair (Type II) is not applicable where

the anticipated


of the crack are too rapid to permit

required slow, viscous flow in the sealant.


(2) Caution 2 stated in connection with the Type I repair is ap-


Mortar Plug Bond Coat



gap in bar so that



@s required



Reinforeitty Bart * ENGR



or close,

New header bar

(say 4 in. diam). Tack weld to

existing bars.

Sealant. Nofe that chase

13 cul deep enough $0 thot

Seolanr is entirely behind bors. Figure 4-13.

Anchoring the mortar plug.

Concrete Structures:


Repairing Cracks

d. Type IV—A Crimped Water Bar. The applicable details are shown in Figure 4-14. Where the joint has to bear traffic, a mortar cover is required, as shown in Figure 4-14b. The depth, width, in-

stallation, and anchoring of the mortar cover follow the discussion given under Type III repair. For repairs not made against a pressure head, or where the pressure is nominal and the joint does not have to bear traffic, the water stop can be laid directly on the concrete surface without any chasing except as required to install the crimped section and to form the


It is desirable to bond the wings of the water bar to the con-

crete on which they rest.

An epoxy bonding agent is useful for this


A crimped water-stop joint sealant is not applicable for use where the crack is subject to a heavy pressure head from inside the joint or where movement occurs as a shear along the length of the crack. In the one case, the pressure would tend to bulge out the crimp, and

in the other, the longitudinal movement would tend to tear the stop. Accordingly, this type of joint is primarily for use where the anticipated movements are limited to a simple extension or contraction and where the pressure head is either small or acts from the face of the joint. However, while not specifically intended for such application, a rubber-type or similar crimped joint material will take a Mastic



Bond with epoxy or ofher adhesive a) Where

Joint is not Subject to Traffic Mortar



both faces or






Where Joint is Subject_to Troffic or Where Appearance Qc Theft Require Conceaiment of Woter Bar Figure 4-14,

Crimped water bar.




and Repair of Structures

little longitudinal movement, and the crimped stop will function under a limited amount of relative transverse displacement. When

a sheet



is used, the



canized, or otherwise made continuous; or else each splice. If the water bar is nailed in place, of materials compatible with the water bar are nail holes do not leak. Of course, all nail holes a nail.




not be permitted.

be soldered,


the joint will leak at flathead roofing nails required so that the should be filled with

Where a copper water bar is used, the material is sufficiently valuable to tempt thieves, and a joint of the type shown in Figure 4-14a is inappropriate. e. Dealing with the Reinforcement. Whatever the type of detail used, when cutting the chase it is probable that some of the reinforce ment will be exposed. If this is the case, cut deep enough so that the sealant will be behind the reinforcing, clean the bars, and paint them with bitumen as a protection against moisture penetration and corrosion. Be sure to paint all the exposed surfaces. If the crack is an active one with substantial movements, cut the bars (Figure 4-13) so that they do not impede the movement of the joint. In any event, the reinforcement should not be left in a position where it is partly embedded in two different matrices, i. e., part in the mortar plug and part in the sealant, or part in the joint and part in the original concrete. f. Remarks. Blanketing is expensive and is applicable only to the repair of major cracks. Accordingly, the job must be well done, or it is not worthwhile. In particular (except where a crimped water bar is used)

the bond between


and concrete


be first-class

and should be checked. Grab an exposed end of the sealant material with a pliers and pull on it. It should strip away a skin of mortar from the concrete or tear through the sealant, not separate cleanly at the interface. At isolated locations, test the bond of the mortar plug, if one is used. Strike it smartly with a hammer or a maul. If it breaks, the break should not occur as a cleavage plane along the sides of the chase. If the bond appears to be poor, strip the questionable area down and repeat the application, checking the dryness of the surfaces of the chase. 6. Overlays The use of overlays for repairing spalled and disintegrated concrete is described in Chapter 5, and for resurfacing pavements in Chapter 6. Overlays may also be used to seal cracks and are very useful and desirable where there are large numbers of cracks and treatment

of each

individual defect would be too expensive.

Concrete Structures:

Repairing Cracks


Form key with precast concrete or mortar plugs set in bitumen. The bitumen is to break the bond between plugs and hole so that plugs will not be cracked by subsequent movement of the opening. If a particularly good seal is required, drill a second hole and plug with bitumen, alone, using the first hole as a key and the second as a seal.

Hole drilled in stem of wall, centered on and following down crack, Size of hole depends on width of crack, Use 2" to 2-1/2" minimum diameter. Crack


Figure 4-15.

Sealing a crack by drilling and plugging.

a. Active Cracks. Sealing of active cracks by use of an overlay requires that the overlay be extensible. Note the designation— extensible, not flexible. The occurrence or prolongation of a crack automatically means that there has been an elongation of the surface fibers of the concrete. Accordingly, an overlay which is flexible but not extensible, i. e., can be bent but cannot be stretched, will not seal a crack that is active. A two- or three-ply membrane of roofing felt laid in a mop coat of tar, with tar between the plies, the whole covered with a protective course of gravel, concrete, or brick, functions very well for this purpose. The type of protective course depends on the use to which it will be subject. Gravel is typically used for applications such as roofs, and concrete or brick is applicable where fill is to be placed against the overlay. An asphalt block pavement also works well and


Deterioration, Maintenance, and Repair of Structures

may be used where the area is subject to heavy traffic, such as a roof used for parking cars. Caution: If the cracks are subject to longitudinal movements parallel to their axis, the overlay will wrinkle or tear. Be very careful in repairing such joints. Blanketing may be a better solution. b. Dormant Cracks. If the cracks are dormant, almost any type of overlay may be used, provided that it will take the traffic to which it is subject and that it is either adequately bonded or thick enough so that curling due to differential deformations is not a problem (see the discussion

of overlays

in Chapter




ing into increasingly frequent use for this purpose.

are com-

1. Grouting Grouting of cracks can be performed in the same manner as the injection of an epoxy, and the technique has the same areas of application and the same



the use

of an epoxy

is the

better solution except where considerations of fire resistance or cold weather prevent such use, in which case grouting is the compar-

able alternate.

The procedure is described in Reference 4. It is similar to other grouting methods, and consists of cleaning the concrete along the crack;

installing built-up seats at intervals astride

the crack


provide a pressure-tight contact with the injection gun); sealing the crack between the seats with a cement paint or grout; flushing the crack to clean it and test the seal; and then grouting the whole. The grout itself is high-early-strength portland cement, used neat. An alternative

and better



it can be performed,


to drill down the length of the crack and grout it so as to form a key (see Figure 4-15). However, the technique is only applicable where the cracks run in a reasonably straight line and are accessible at one



is not often the case,

but sometimes

it does occur


walls or similar constructions where the cracks are due, basically, to shrinkage or temperature. The grout key functions to prevent relative transverse movements of the sections

of concrete


to the crack

or joint (which


be important in some applications). It will also reduce any heavy leakage through the crack or loss of ground from behind a leaking wall. The seal will not be perfect, however,andif such sealing is required, the drilled hole should be filled with soft bitumen in lieu of grout, or, if the keying effect is essential, the bitumen can be placed in a second hole, the first being grouted. The drilled hole should preferably be 2 or 3 inches in diameter and flushed to clean out the crack and permit better penetration of the sealant or grout.

Concrete Structures:

Repairing Cracks


8. Autogenous Healing The inherent ability of concrete to heal cracks within itself is termed ‘‘autogenous healing’’ and is a phenomenon which has been known for some time. It has practical application for sealing dormant cracks, such as in the repair of precast units cracked in handling, reknitting of cracks developed during the driving of precast piling, sealing cracks in water tanks, and sealing of cracks which are

the result of temporary conditions or inadvertencies similar to those described under Judicious Neglect. The effect also provides some increase in strength of concrete damaged by vibration during set and of concrete disrupted by the effects of freezing and thawing. The mechanism whereby healing occurs is the carbonation of the calcium oxide and the calcium hydroxide in the cement paste by carbon dioxide in the surrounding air and water. The resulting calcium carbonate and calcium hydroxide crystals precipitate, accumulate, and grow out from the cracks. The crystals interlace and twine pro-

ducing a mechanical bonding effect, which is supplemented by a chemical bonding between adjacent crystals and between the crystals

and the surfaces of the paste and aggregate. As a result, some of the tensile strength of the concrete is restored across the cracked section, and the crack may become sealed. Saturation of the crack and the adjacent concrete with water during the healing process is essential for the development of any substan-

tial strength.

Submergence of the cracked section is desirable.


ternately, water may be ponded on the concrete surface so that the crack is saturated. Tests by Lauer and Slate (Reference 5) indicated

restoration of up to 25 per cent of the normal 90-day tensile strength

by healing under conditions of submergence

ponding period.

in water for a corres-

Similar healing under conditions of indoor atmos-

pheric exposure without submergence of strength.


little or no restoration

The saturation must be continuous for the entire period of healing.

A single cycle of drying and reimmersion will produce a drastic reduction in the amount of healing strength. Healing should be commenced as soon after the crack appears as possible. Delayed healing results in less restoration of strength than does immediate correction. The section may be permitted to dry after healing is complete, provided it is not subject to any pronounced thermal changes or to decidedly arid atmospheric conditions. Drying will produce some loss of strength, presumably due to shrinkage stresses induced thereby, with the amount of loss of strength being less for betterhealed specimens than for those which are poorly bonded. As a

corollary to this, and since the structure is doubtless under some


Deterioration, Maintenance, and Repair of Structures


of restraint,

if it is possible

to localize the saturation

of the

concrete right around the crack, do so, as there will be a reduction in the subsequent drying shrinkage and resulting shrinkage stresses. Healing will not occur if the crack is active and is subject to movement during the healing period. This means the structure must be under a constant, static load to repair itself. Healing also will not occur if there is a positive flow of water through the crack, which


and washes


the lime deposit,

is so slow that complete evaporation occurs causing redeposition of the dissolved salts. Autogenous


is primarily

tures in moist (preferable




the flow of water

to the

repair of struc-

at the exposed face


free from ag-

gressive constituents which would dissolve the crystal deposits. 9. Judicious Neglect

Dormant cracks, such as those due to shrinkage or to some inadvertency in construction such as premature removal of the forms or settlement of the sills supporting the shores, frequently are selfsealing. This does not imply an autogenous healing and gain of strength as described above, but merely that the cracks clog with dirt, grease, or oil, or perhaps a little recrystallization occurs, and so on. The result is that the cracks are plugged (or sealed, if you will), and problems which may have been encountered with leakage, particularly if leakage is due to some intermittent cause rather to a continuing pressure head, will disappear without doing any repair.


a chance

this is not engineering, but it works,

that it will occur,

if aesthetic


and where there is



if there is time to give it a try, it is by far the best answer.


10. What Not To Do When




the following



a. Do Not Fill the Crack with New Concrete or Mortar. Theoret-

ically, using the techniques of surface preparation and the mixes and methods of placement described under Replacement of Concrete in Chapter 5, it would be possible to open up the crack and put in new concrete or mortar. In practice this seldom works well. While a good bond can be obtained, the bond is neither of the magnitude of the tensile strength of uncracked concrete nor equal to that attainable using epoxies, sealants, or similar materials. Further, the new and old materials are unlikely to be entirely compatible, and the dimensions and volume of the repairs are likely to be too small to render the incompatibility unimportant. As a result, a crack which

Concrete Structures:

Repairing Cracks


is cut out and filled solid with new concrete almost invariably reopens in time, even where there has been no perceptible movement in the structure. b. Do Not Use a Brittle Overlay To Seal an Active Crack. It has been stated that overlays used to seal active cracks must be extensible, must not be brittle, and must not be inextensibly bonded to the adjacent concrete surface. The admonition is worth repeating. Epoxies,








coatings may be used to repair dormant cracks but do not work for sealing cracks which are active. If there is any movement, they crack sympathetically with and reflect the underlying defects. Even an asphaltic



be expected

to give

full protection


less there is traffic to knead it so that any reflected cracks will be closed. c. Do Not Fail To Relieve the Restraints Causing the Crack. Do not forget this important point. In most cases, the repair is not intended or likely to be stronger than the original concrete section. Therefore, either the repair must be capable of accommodating future


of the structure,

or these




eliminated. d. Do Not Surface-seal Cracks over Corroded Reinforcement without Encasing the Bars. This does little good. As has been noted, even good concrete is not absolutely watertight. If the bar has already corroded, there is a cleavage plane in the concrete mass next to it. This ponds water. Corrosion will continue—perhaps slower than otherwise, but continue



for temporary



the bar, remove the rust, and encase it. e. Do Not Bury or Hide a Joint So That It Is Inaccessible. Joints, including repaired joints, should be accessible for inspection and maintenance. Joints are a tricky business, particularly if, as is often the case, the movements which occur are not limited to simple extension or compression. An occasional poor performance should be expected, and it is well not to bury the trouble.

REFERENCES 1. ‘‘Guide for Use of Epoxy Compounds the American Concrete Institute, vol. September, 1962. 2. Raymond J. Schutz, ‘‘Shape Factor in neering, October, 1962. 3. S. Champion, ‘‘Failure and Repair of Wiley & Sons, Inc., New York, 1961. 4.

C. C. Carlson,


with Concrete,’’ Journal of 59, no. 9, pp. 1132-1133, Joint Design,’’ Civil EngiConcrete

in Repairing

Structures,’’ John

and Obscuring







and Repair of Structures

in Concrete Structures,’’ Journal of the American Concrete Institute, vol 52, no. 2, pp. 6-8, February, 1944. K. R. Lauer and F. O. Slate, ‘‘Autogenous Healing of Cement Paste,’’ Journal of the American Concrete Institute, vol. 27, no. 10, pp. 1083-1097, June, 1956.





and Therapeutics

of Reinforced

Concrete,’’ translated by D. A. Sinclair, Technical Translation 1008. National Research Council of Canada, Ottawa, Canada,


l. J. A. Roberts and H. E. Vivian, ‘‘Improvements and Repairs to Concrete,’’ Constructional Review, vol. 33, pp. 31-33, October, 1960. 8. B. W. Steele, ‘‘Cracks in Concrete,’’ Journal of the American Concrete Institute, vol. 18, pp. 629-633, February, 1947.


The previous chapter described







for repairing cracks

in themselves

are seldom



tive of structural danger; accordingly, repair usually does not involve strengthening; and so the repair procedures which were described


the cracks


largely superficial,


being intended

to seal

against an objectionable flow of water or to improve


of the construction.

Repair of a structure showing spalling and disintegration is an entirely different matter. Here, it is usual to find that there have been substantial


the reinforcement.

of section




Both are matters of concern from a



viewpoint, and repair generally involves some urgency and some requirement for restoration of lost strength. This is an important

difference and must be borne in mind when considering application of the techniques of repair which are described in this chapter. The principal methods used for repair of spalling and disintegration are jacketing, use of pneumatically applied mortar, prepacked concrete, drypack, replacement of the concrete, and the application of overlays of several types. The procedures are as follows: 149



Deterioration, Maintenance,


and Repair of Structures


1. Preparing the Surface of the Concrete Which Is To Be Repaired a.



all unsound,

otherwise undesirable concrete.

concrete concrete






it is not obvious that sound

has been reached, a rule of thumb in common which is difficult to remove is good concrete.

use is that Generally,

this is a satisfactory rule to follow. There are exceptions, however. For example, concrete cover over reinforcement will spall easily when the chipping hammer is applied. In fact, it is easy to chip away perfectly good concrete by using a pointed tool or a heavy hammer, and by following along the natural plane of separation formed by the reinforcing mat. If care is not exercised, the extent of the work may be substantially and needlessly increased because of this factor. On the other hand, it is sometimes difficult to remove relatively porous concrete from overhead surfaces because of a lack of purchase in working the hammer overhead. In such circumstances, scaffolding and work platforms should be provided for the workmen, and, where possible, consideration should be given to suspending the chipping hammer from the overhead. The use of pneumatic equipment for removing unsatisfactory concrete is advisable. Gad points are preferable to chisel points, because the former tend to leave cleavage planes which are free of crushed pieces of aggregate, whose presence would tend to impair the bond condition between the new and existing concrete. Another


to be considered

in the problem

of how



crete to remove is that, in many cases, particularly where the deterioration consists of spalling of the concrete cover due to corrosion of the reinforcement,

it will be found that the plane

of division

between loose and sound concrete occurs just on the plane of the reinforcement. In such a case, a decision will be required whether or not to cut back the concrete

and expose

the bars.



should be anticipated and the decision rendered in the specifications for the work, since the cost of removing the additional concrete is substantial. The answer may be derived from consideration of the following: Technically, it is not good practice to have a joint between new and old materials right at the plane of the reinforcement. The two materials will not be the same, and a potential cleavage plane is created in which moisture could be ponded. It would be folly to deliberately create such a condition right at the plane of the reinforcement, thereby inviting a repetition of the spalling which is being repaired. Therefore, the bars should be completely encased except under conditions where penetration of moisture through the

Concrete Structures:

Repairing Spalling and Disintegration

new or existing concrete

is unlikely to occur, or where

quantity of penetrating water is very small.

that, in practice, such conditions are rarely terioration of concrete usually is associated water.) Basically, then, this means that the unless all faces of the concrete section are


the probable

(It should be noted

encountered, since dewith the presence of bars should be exposed either effectively water-

proofed or normally would not be in contact with water (such as for interior surfaces; in a case where the concrete surface is raised far above the water level; or for vertical surfaces adequately

shadowed by overhanging canopies or similar construction).


there is any doubt that these conditions exist, if there currently is any corrosion of the face of the bars remote from the repaired sur-

face, or if there is any possibility that there can be water penetra-

tion from behind the plane of the reinforcement,

expose the bars

completely (at least 3/4 inch all around), and encase them fully in a new matrix.


(See also Remarks,

under Pneumatically Applied

If it is not deemed necessary to encase the bars fully, it is still

advisable to key the new work to the existing concrete by exposing more than half the circumference of thé bars (say 3/4 of the cir-

cumference). Remember

Corner bars always should be fully exposed.

that, to the man



to remove

it, inferior


crete can begin to look very sound. Do not be impatient, and provide adequately for the cost involved. Shortcut performances should not be accepted. If there is a doubt, remove concrete until no fur-

ther doubt exists.

b, Shaping the Hole.

(See Figure 5-1.) Where repair is confined

to relatively narrow or small,

deep areas—along



filling holes left by form ties, or, in general, where a relatively

small cavity limits of the into place. For large cally placed

is to be filled by hand using drypack or mortar—the cavity can and should be undercut to lock the repair

areas and where repair is to be by use of pneumatimortar, the edges of the area to be repaired should

not be undercut but should be cut back sharp, perpendicular to the face of the existing concrete surface, and at least 1 inch deep (or back of the reinforcement, whichever requires the greater depth). On horizontal surfaces, and for extended areas,

it is desirable

to make the first pass of the cut at the edges of the repair area with a concrete saw, because this gives better definition than hand or pneumatic tools.

The saw cut should not be so deep as to inter-

cept the reinforcing bars. For vertical surfaces, where the hole is deep and large enough to

warrant repair by filling with new concrete, the edges at the sides and bottom of the repair area should be cut normal to the concrete


Deterioration, Maintenance,

and Repair of Structures

Limits of removal of concrete

Undercut to lock drypack in place. ‘This can be acc:






rocking the bit.


ished by usiny

tooth bit an



jon two opposite

hand hammer


drill bit

sides (or all around) to

lock drypackin place.

Cavity before chipping



As required to bond to reinforcement but Gof less than 34" (Except as noted






Figure 5-1. Preparing the surface of the concrete which is to be repaired (shaping the hole).

surface with the top edge sloped about 3:1 to permit placement of the concrete without forming a void or air pocket at the top of the repaired area. Do not leave feather edges. Where the hole goes completely through the concrete section, remove the concrete from both sides to prevent the spalling which would occur if the tool were punched through from one side only. Corners of the repaired areas should be rounded to get best results. c. Cleaning the Surface. That surface of the existing concrete which is to be bonded to the new work should be cleaned and moistened just prior

to placement

of the new


or mortar.

Following a preliminary rinsing, cleaning by use of sandblast is desirable and is preferable to wire brushing. Where wire brushing is specified, power-operated equipment will do a better job than manual equipment.





Repairing Spalling and Disintegration


be rinsed


after sandblasting


or wire

brushing. An etch with muriatic acid, followed by brushing and rinsing, may be considered in lieu of the sandblast. It is important

to perform

the cleaning

operation just prior


placing the concrete or mortar in order to avoid fouling of the surfaces by rising tide, splash, or windblown dust or debris. The settlement of soot on horizontal surfaces in industrial areas is particularly troublesome. A lapse of more than 2 or 3 hours is usually too much. Bond between the new and old concrete cannot be obtained if there is a film of oil or loose material on the bonding surface. d.


the Surface.

After the cleaning,

the surface

to be

repaired should be saturated and then allowed to approach dryness just before placing the new concrete. Saturation can be accomplished by spraying or by packing the holes to be repaired with wet burlap which is kept damp by occasional sprinklings. Either way, the surface should be kept moist for several hours, preferably overnight, to assure saturation. The spray should be stopped and/or the burlap removed an hour or two before

placing the new



the surface



color will change. Except as noted below, just before the surface is dry (i.e., just before the color is as light as that of completely dry surfaces which are near by), it should be given a slush coat of mortar having the same proportions as the matrix of the replacement concrete. This slush coat can be wire-brushed into the surface, rubbed in by hand, or placed pneumatically, but must be well worked into the existing surface and should be soft enough to do so. The slush coat should be a thin film, in no event more than 1/8 inch thick. Coarse particles not firmly embedded in the mortar should be removed. The replacement concrete should be placed directly after placing the slush coat. In hot, dry weather, it may be necessary to fog the surface just ahead of the slush coat. Where repair is effected by use of pneumatically applied mortar, the slush coat is not necessary, and where repair is to be effected by use of drypack, the slush coat definitely should be omitted, because it makes the drypack material too wet and increases the shrinkage, thereby preventing the development of good bond. With drypacking, moist surfaces are necessary, i.e., they should

show a small amount of free water.

The surfaces should then be

dusted lightly and slowly with cement, using a small dry brush until they have been covered and darkened by absorption of the free water into the cement. There should be no dry cement present on the surfaces, and such cement, if present, should be removed.





Repair of Structures

2. Existing Reinforcement Existing reinforcement which is to be incorporated into the new work should be cleaned of all corrosion, oil, dirt, and similar foreign matter. Much of this will be accomplished by the sandblasting used to clean the concrete surface. In some cases, however, the sandblast will not clean that surface of the bars which faces away from

the nozzle,

and this surface

should be inspected and, where ing.

If new


ting reinforcement,

and the underlying


is to be added,


if it has not broken,


cleaned by wire brushthat the exis-

is still carrying load.


possible, leave it in place. Also, if the load on the structure is not relieved (see Chapter 9), new reinforcement added to the structure

will carry a lesser stress than the existing bars, and the concrete stresses

will have to increase

in order

for the new


to parti-

cipate in the action of the structure. The resulting stress conditions can be important and should be investigated. All unnecessary


placing the new concrete.


3. Inspection before Placing New

etc., should

be removed



The specifications for the new work should provide for the in-

spection and approval of the condition of the existing concrete faces before new concrete may be placed.


Compatibility of Materials


and Sections

Do not attempt to repair a structure with materials which are dissimilar to those in the existing structure unless provision is made either to tie the new and existing matrices together or to permit their free separation. This important principle is discussed in Chapter Note 1. 5. Requirement

for Good-quality Concrete and Mortar

This factor has been and will be emphasized



is no sense

in repairing


repeatedly in this


with more


the same. The usual standards of practice are set forth in References 1 and 2 of this chapter and References 2 and 6 of Chapter 3

(Standards of the American Concrete Institute),

particular attention to the following:

In addition, pay

(1) Do not use more cement than is needed for strength.

is a tendency to equate more


for too much


cement with better concrete and to




or other deficiencies



Repairing Spalling and Disintegration


by adding still more. Cement is rather like medicine, in that the proper amount is helpful whereas too little or too much may be ineffective or even harmful. In the first place, rich mixes shrink more


lean mixes.


if the environment

or aggregates


such that chemical attack is possible, excess cement just adds fuel for the reaction. (2) Select clean, stable aggregates, and use the largest aggregate that can be placed in the new concrete section. This will tend to reduce the water/cement ratio and the amount of fines. (3) Be absolutely insistent on proper cleaning and placement of reinforcement and on removal of all existing concrete which is unsound. (4) Control the temperature of the materials and immediate environment during placement and curing. This is in order to reduce shrinkage and prevent freezing. 6.


The deterioration to be repaired probably will be due in some degree to insufficient cover. Be sure that the cover provided in the new



to the requirements


in Chapter


7. Appearance One of the principal reasons for making repairs to concrete structures is to improve the appearance. The following cautions apply. a. Patches. Where the extent of the repair is limited, unless it is carefully performed, the repair can do the appearance of the structure more harm than good. Do not let the workmen rush in and patch the concrete right after stripping the forms and before the engineer has had a chance to look at it. If there is an area of honeycomb to be repaired or the holes for the tie rods are to be filled, require that the patch be made with a mix that will resemble the concrete


in the structure,

not with

a rich mortar,


certainly not with a different brand (or type) of cement or with a

different kind of sand. Use a lean, stiff mortar rendered lightcolored by using a portion of white cement, if necessary. If wood forms are used, finish the patch with a piece of wood, not a steel trowel. Require the construction of a sample wall of each of the major types of exposed concrete to be used in the work, and require that the contractor develop a patch mix that will resemble the main concrete mass. These provisions should be included in the speci-


Where there are fins, bevels,

or rough spots requiring the use of


Deterioration, Maintenance, and Repair of Structures

a stone, apply the stone to the fin, bevel, or rough spot without touching the rest of the concrete. Where required, ‘‘print’’ form marks on the repaired surface by use of a sharply grained form board, placed on the surface and struck with a hammer. Best of all, if the condition is not severe, leave it alone.

b. Repairs to Extensive Areas. Where exposed concrete (or mortar) is to be used for repairs to extensive areas of the existing structure, much can be done to give the work a pleasing appearance. The surface of the work may be broken up into relatively small


each unit at a different

level or relief.

In general,


various planes should be offset by at least 1 inch, preferably more. Panels may be fluted or textured. Simple scoring should be avoided as it does not provide sufficient relief to give necessary light and shadow to the surface. Uniformity in the work is a second important consideration. Use the same types of sand and cement throughout the work on any one panel or series of like panels. Keep the mix the same. Small variations in the sand/cement ratio, in the amount of mixing water,or in the proportions of fine and coarse aggregate can produce significant changes in the color of the final work. Unless it is desired to deliberately produce these color variations as an architectural effect, measure the quantities of ingredients by use of gauge boxes or by weight. Do not proportion the mix by shovel measurements or proportion the water by the slump or by the kickback of the mixing drum. Curing methods must be constant. When using pneumatically applied mortar, keep the same crews together, since the appearance of the final work will be dependent upon the technique not only of the nozzleman, but of the backup laborers and mixers as well. Do not permit construction joints in any but approved locations. Approved locations should be shown on the plans and should be limited to points of changes in level or relief. Be specially cautious of cold joints, and be sure that the spacing of construction joints is small enough that the work can be completed in a single shift. Do not commence operations if it is probable that the weather will turn bad before the work is completed. A heavy rain can ruin the appearance of fresh concrete. Do not work when there is a heavy wind unless curing is to be by use of a spray-on compound or other material which will not be blown about. Insist on the employment of experienced and qualified finishers. The construction of sample, prototype panels is very desirable. Even with all these cares and attentions, be cautioned that it will be difficult to get really good architectural appearance in a repair job, and an architectural facing of brick or stone masonry should be considered.

Concrete Structures:

Repairing Spalling and Disintegration


8. Admixtures The use of admixtures is an important factor in the repair of concrete structures, much more so than in conventional concrete work.


is because

the repair

is usually a thin section,


in an aggressive environment. Under such circumstances, minor deficiencies in workmanship or concrete technology can produce major defects in the finished product. The proper use of the correct admixtures can significantly increase the margin for error in both respects. The principal admixtures required for use in repair work are: a. Air-entrainment agents. Whether by additive or by use of an air-entraining


air entrainment

of the concrete

or mortar

used in the work is advisable wherever the new material will be exposed to weathering (freezing and thawing). The beneficial effects of air entrainment in this regard are well established (see Chapter 3) and apply to mortar as well as to concrete. In this connection it should

be noted that the benefits

of surface


as des-

cribed in Section I, apply to air-entrained concrete as well as to sections without air entrainment. b. Retarders and Densifying Agents. The use of a retarder and densifying agent is advisable. In warm weather, retarding the set reduces cracking due to settlement of the concrete suspension

(Chapter 3, Section B, paragraph 1d) and allows for delayed finish-

ing, which is important in connection with closing surface cracks. This type of admixture also increases the workability of a given mix, thereby reducing the water requirement for a given slump. These effects increase the density and strength of the concrete, and make the concrete less permeable. Some of the products which are commercially available will effectively reduce shrinkage of the concrete. Others, despite the reduced water/cement ratio, produce a net increase in shrinkage. c. Accelerators. Accelerators speed the rate of setting and hardening of concrete and are useful where working to seal against the flow of water, when working against the tide, or when working on surfaces

exhibiting running water.




cluding Type MlIl—high-early-strength cement) increase the heat of hydration of the mix, increase the shrinkage, and should not be used unless absolutely essential. d. Expanding Mortars. There are two basic types of expanding mortars in common use. Both are represented by commercial compounds, which are readily available. The first relies on the addition of aluminum powder to the matrix, producing hydrogen bubbles, which cause expansion of the mortar. The gas pressure developed is relatively low and easily confined by the forms, and


Deterioration, Maintenance, and Repair of Structures

the expansion takes place along the line of least resistance, i.e., toward the open top of the form. In extreme cases, a foamy laitance may be formed at the top of the concrete. This reaction is of little effect after the concrete

is hard.


while the setting

shrinkage is readily overcome, and a net expansion may be produced, the drying shrinkage may not be substantially reduced, and this type of agent may not be fully effective in wedging a repair in place (unless the repaired area is undercut). The second type of expanding mortar contains iron filings or powder, plus a catalyst. The idea is that the iron is to oxidize in place after the concrete has set, expanding, increasing the volume, and offsetting not only the setting shrinkage but the drying shrinkage as well.


as to the actual realization

of this effect

vary. The presence of moisture is required for oxidation to take place. It is difficult to see how moisture can penetrate thick sections, especially in sheltered or interior exposures. However, in thin sections, open to the weather, there is a good possibility that the desired oxidation and expansion will occur, and this type of expanding agent is commonly used for grouting anchor-bolt holes, setting expansion anchors, and for making patches. A difficulty is that of rust staining of adjacent surfaces, and this must be thoroughly investigated before using this type of mortar for patching a surface where appearance is of great importance. A third type of expanding mortar currently is being developed for applications in self-stressing concrete. The mixture produces concrete that expands slightly, in predetermined amounts, as it cures. In one mixture, the basic expanding agent is reported (Reference 3) to be calcium sulfoaluminate, which is the same constituant that produces expansive reactions in concrete exposed to seawater. This technique is too new to comment on its longterm performance. For applications where the desired postsetting expansion can be attained with assurance, the use of an expanding mortar can bea very valuable tool. Patches and sections of replaced concrete can be wedged tightly in place, and shrinkage cracks can be avoided. External stressing can be achieved, as described in Chapter 4. However, varying degrees of expansion can be obtained by varying the amount of additive, and care should be observed not to add so much that the repaired surface will be out of plane.

e. Waterproofing Admixtures.

Waterproofing admixtures are of

value in a coating designed to inhibit corrosion of the reinforcement by preventing contact with an aggressive environment. However,


proofing agent,



the same

if used, is merely


and the water-

an additional safety factor.


Concrete Structures:

Repairing Spalling and Disintegration


waterproofing admixture is not the equal of a membrane or similar construction,

since any attempt to waterproof the concrete


nally is doomed to failure unless the concrete can be kept completely free of cracks;

an ideal which,

Chapter 4, is most difficult to attain.

as has been discussed


Waterproofing admixtures are of greater value in thin sections than in heavy sections. They are never a substitute for good, dense concrete.

f. Other Admixtures

and Special Cements.

The use of flyash or

natural cement will retard the set of the concrete. These ingredients are usually substituted for a part of the portland cement which would be required. Do not just add them to the regular mix. The mix should be specially designed when their use is contemplated. Use of a vacuum process or the application of shock to the fresh concrete will remove a part of the water and improve the strength, durability, and density, provided that the shock is not applied too

late in the setting process and the concrete is not damaged rather

than improved. Certain slag cements ical attack.

can be used to improve resistance

to chem-

The use of aged cement will decrease shrinkage of the concrete.

This is a simple and practical technique. Concrete mixed with new cement shrinks more than concrete of the same proportions mixed with old cement. If proper storage facilities are available, if the cement can be screened just before use to remove any lumps, and if time considerations permit, storage of several weeks will be


Another technique

for reducing shrinkage of concrete

is to mix

it and let it stand before using it. The mix should not stand so long


it will set before

it can

be finished



but if the wea-

ther is not particularly warm or dry, an hour’s aging is not excessive in summer, and two hours or more is not excessive in winter.

B. JACKETING The use of jacketing is principally applicable to the repair of deteriorated columns, piers, and piling. It is especially useful where

all or a portion of the section to be repaired is under water.



Deterioration, Maintenance, and Repair of Structures

Description Jacketing consists of restoring or increasing the section of an

existing member (principally a compression member) by encasement

in new concrete.

The original member

need not be of con-

crete. Steel and timber sections may be jacketed as discussed in Chapters 2 and 7. Figure 2-20 and Case History 2-1 show application of the technique to the protection of steel H piles. Figure 7-7


its use to protect and fireproof timber



of concrete piles is described in Case History 5-1 and details of a typical repair are shown in Figures 5-12 to 5-14. Figures 5-2 and

5-3 show jacketing of a concrete pier. The method is applicable for protecting a section against further deterioration as well as for strengthening.

2. Forms The form for the jacket should be provided with spacers to assure clearance between it and the existing concrete surface. For jacketing straight members, the form may be a single length sec-

— Bridge Truss


“aa Pumped Grout Fin

Limit of Deteriorated |Concrete





lLashings eee

|2"Creosoted | T &G Perma-

|nent Form

" As reqd,to





displacement of Toe of Form Sheeting


Figure 5-2. Example of jacketing for repair of pier of a swing bridge. (Courtesy of Praeger-Kavanagh- Waterbury.)



Repairing Spalling and Disintegration

Figure 5-3.

Repair of condition shown


in Figure 3-12.

tion. Where the member has a bend or curve (as with a timber pile), it should be oversized or may be segmented. The form may be temporary or permanent and may consist of timber, wrought iron, gauge metal, or precast and exposure, as follows:

a. Timber



depending on the purpose

For marine environments

or elsewhere where

it is desired to protect concrete (whether new or repaired work)

from chemical reaction with its environment or from weathering, the use of permanent timber forms is recommended, provided the appearance

of the form

a fire hazard.

is not objectionable

and does

not constitute

The form serves several useful purposes.

ates the concrete




and thawing.

It insul-



moisture by the timber tends to keep the underlying matrix continually damp and thus prevent volume changes due to alternate wetting and drying. The form inhibits chemical reaction between the concrete and its environment by preventing circulation of the environment in contact with the concrete, and the resilient jacket tends to reduce abrasion of the concrete surface due to windblown debris,



or ice action.

The form lumber should be given a preservative treatment and


be tongue

and groove.




be substantial,


and spaced close enough together to prevent bulging of the lumber due to the hydrostatic pressure of the fluid concrete. A mortartight fit is required, or sand streaks will result from leakage of the cement through the joints. A rubber or other suitable seal should be provided

at the bottom

of the form,



the work

is be-

low water, a diver should be employed to check the lower sections




and Repair

of Structures

of the forms for leakage during placement of concrete and to tighten or seal them, as required. The installation of permanent, timber forms also is recommended for new work. Here, the forms can be clamped in place on the finished concrete, and it is not necessary to make them mortartight. The lumber will swell upon wetting and give a secure fit. The value of this detail can be appreciated by comparing Figures 3-33 and 5-4. The two sets of piling are in the same waterway, have similar exposures, and are within 2 miles of each other. Both were constructed in the mid-1930s. After about 25 years, the piling installed with protective jackets was found still to be in excellent condition. Those installed without jackets were almost completely

destroyed in less than 20 years.

For permanent forms the bands should be of corrosion-resistant metal or of adequate section to resist corrosion. Do not use wire rope, unless the banding is to function on a temporary basis only,

Figure 5-4. Timber jackets placed on new precast concrete piles. The piles shown in this photograph are about 25 years old and are located in the same waterway about 2 miles from those shown in Figure 3-33. Compare the condition. These piles are in excellent shape. This is the result of the use of timber jackets which were installed at the time that the trestle was built. Note that some of the bands have corroded through and the jackets have fallen off, but that the concrete, including that in the tidal range, is still in

excellent condition.

Concrete Structures:

Repairing Spalling and Disintegration

and some other means


is provided to secure the form sheets


the wire rope rusts out (see Figure 5-2). Galvanizing of the bands is recommended.

b. Wrought Iron. Wrought iron makes a very satisfactory, permanent form but is expensive and so is limited to installations where the additional cost is justified by the increased life of the repair or where it is desired to protect the concrete from severe abrasion,


as that due to heavy


of moving


c. Precast Concrete. Precast concrete forms (Figure 5-5) are not as serviceable as timber forms with regard to protecting the underlying structural concrete. They are used, however, where timber forms would constitute an unacceptable fire hazard or where the use of timber jackets is impractical because of the presence of borers. Concrete pipe jackets also find frequent use for fireproofing tim-

ber piles (see Figure 7-7). d. Gauge Metal.

Gauge metal and other temporary forms

ure 5-6) are used under the following circumstances:


(1) Where the repair will always be in the dry and anticipated


attack is mild

(2) Where it is desired to strip the forms for the sake of appear-


(3) Where the proposed life of the structure or repair is limited (4) Where exposure conditions are so mild that the degree of

protection provided by a timber or concrete

(5) Where timber forms will not last


is not needed

For example, in some tropical marine environments, attack by borers is so severe that timber jackets, albeit creosoted, will not last more than a few years. Similarly, metal jackets corrode rapidly. For such a case, there is no choice but to leave exposed concrete, and the use of less costly, temporary forms is justified. Alternatively, precast concrete forms may be used. In either case, however, the concrete section should be increased beyond the value required for stress to allow for some future deterioration.

3. Filling the Forms Filling of the forms can be accomplished by pumping grout, by using prepacked concrete, by using a tremie, or, for subaqueous work, by dewatering the form and placing the concrete in the dry.

a. Preparing the Concrete Surface.


The provisions of Section A

In addition, in marine work, after the forms

are placed,


interior surface of the form and the surface of the underlying con-

crete, timber, or steel should be flushed with a solution of potassium permanganate to kill, soften, and remove any organic growths


Deterioration, Maintenance,

and Repair of Structures

Original Pile Size Grout or Concrete


+3" Galv. W. |. Bond 5+0" mox. ¢.c., but no

14x 4'x4" Concrete end blocks to be cost

of jocket.

Ito precest

less thon 2 per sect.|









44-44 Welded Wire Mesh


ond glued

Paint outelde of form with Epoxy Cement before serting into


lower unit.

fo be fabricoted: sections of any convenient length, not tess thon 6-0"

"x18" Gow wW.l. Band 4°31" (2


per face)




Vertical Section Figure 5-5.

Jacketing—precast concrete form.




Concrete Structures: Repairing Spalling and Disintegration

ewe dy

Sheet Metall Stee! Shell

be-1" Typ

Grout or Concrete Fill


Rubber Seal ‘Bent





Pile, Size

Concrete Cotlor

44-44 Welded lwire Mesh




Jacket Forms to be fobricated in sections of ony convenient length, not fess thon 6-0".

ae ols 7] & are wl

Connect sections of form by use of inter sleeve. For smooth wall forms bond


pf o

|—__+ | °

sieeve to upper ond

lower form sections


by use of Epoxy.

For corrugated forms,

use Epony or screwed connection.

Plate Spacers

Noo °





’ Nr










Figure 5-6. Jacketing—sheet metal form.





in the




cleaning the concrete,

areas a


steel and actual filling of the forms. even weeks,

and in some



and Repair of Structures


This period may be days or growth

of weed



shellfish will accumulate in that length of time. The potassium permanganate solution remains in the forms for an hour or more depending on the growth which has developed and is flushed out of the form with clean water, by overflowing, just prior to filling.


the purple color dissipates from the overflow water. b. Filling the Form by Pumping Grout. Except where the volume to be filled is large or where it is particularly necessary to minimize shrinkage (in which cases prepacked concrete may be a desirable alternate), this method offers an advantage compared to




reliable results

dence upon the skill ienced workmen and obtained with any of The use of a grout to one

away the dead growths

part cement,

and is continued

can be obtained with


less depen-

of the workmen. However, with skilled, expercompetent supervision, good results can be the methods described below. consisting of between two and three parts sand

by volume,

is recommended.




is preferred for thinner sections and the leaner mixture for heavier sections. Grout leaner than 3:1 causes excessive sand streaks unless great care is taken with the forms and placing. Grout of the proportions indicated will have adequate strength for all practical


The grout should be placed as soon as possible after the rinsing operation and under an air pressure sufficient to assure a smooth and continuous flow. It is not necessary to pump the grout in from the bottom of the form. Satisfactory results can be obtained by inserting the grout pipe or hose into the form from the top. However, the discharge end of the pipe or hose should be inserted until the bottom of the form is reached before any grout is pumped into place. As the level of the grout rises in the form, the pipe or hose should be slowly withdrawn, care being taken that the end of the pipe or hose be continuously immersed in the newly placed matrix. Kinking of the grout hose and other sources of sudden surges in grout flow should be avoided, insofar as is possible, since they cause sand pockets. In general, grout should be placed in a smooth, continuous operation. Forms should be filled to overflowing, the grout allowed to settle for about 20 minutes, and the forms refilled to overflowing. The outside of the forms should be vibrated during placing of the grout. Placing of grout through a single hose is adequate if the grout does not have to travel more than 3 to 4 feet laterally or funnel through any opening less than 3 inches. If it is necessary that the grout travel greater lateral distances or funnel through small

| \

Concrete Structures:

Repairing Spalling and Disintegration


(b) Figure



Jacketing deteriorated concrete piles using metal

of photographs


of The





(a) Con-

dition to be repaired. (b) Cleaning pile. Note use of grinder. This same grinder was used by a diver to clean pile sections below the water line.

openings, use twin or multiple hoses or pipes spaced about 5 to 6 feet on centers. Call for a manifold between the grout mixer and the several discharge lines.




and Repair of Structures

(a) Figure 5-7. (Continued) (c) Setting the form. Form consisted of half-round sections of corrugated metal about 2 feet long. Joints in the form were sealed with mastic. Note bottom form (lower section of photograph), Form was assembled on the pile in sections and lowered into place. (d) Filling the form.



filled by pumping,

using intrusion mortar.



Repairing Spalling and Disintegration


(e) Figure 5-7. (Continued) (e) Finishing the jacket. The hand-placed collar of mortar is shown. The exposed surface of the pile above the jacket was coated with a bituminous mastic.


Use of Prepacked Concrete.


of prepacked




Filling of concrete jackets by

is similar

to other


of this

method as described in Section D. d. Use of a Tremie. Unless the void to be filled is uniformly large, the tremie material should be grout, not concrete. The grout proportions and the required precautions during placement to those for grout


difficulties should be anticipated: The


pipe must

be immersed

by pumping,

but the following

in the fluid grout

at all times

and the pipe and hopper kept filled with grout so that the seal is not broken. Practically speaking, this is hardly possible. If the grout is sufficiently fluid to run into the crevices to be filled, it is too fluid to stay in the tremie pipe or hopper, and runs out like water down a drain. A continuous flow of mixed grout into the hopper is therefore


The penetration into crevices in the work is less positive than with a pressure system such as pumped grout or prepacked con-


If more than one grout pipe is required, it is impractical to keep the several hoppers filled. With a pumped grout, the manifold takes care of this problem. Unless the grout hose is straight or less than about 4 inches in


Deterioration, Maintenance, and Repair of Structures

diameter, when it is filled with grout it is rigid and cannot be withdrawn without breaking the seal. Because it does not require specialized equipment, there will be pressure to substitute tremie placement for placement by pumping. Do not yield to such pressure. Theoretically, a satisfactory job can be accomplished. Practically, it is improbable unless the volume of the work is small enough to be placed in a single charging of the hopper.

e. Placing the Concrete Fill ‘‘in the Dry.’’

This discussion is

pertinent to repairs to portions of structures, principally piling, below water. Repairing marine piles and foundations in the dry requires some

form of cofferdam or limpet which can be dewatered.

(See Refer-

ence 4.) If the work can be accomplished at a cost comparable with the cost of doing the work by the pumping or prepacked methods,

it is preferable to do it in the dry. Better preparation of the concrete surface can be accomplished. Workmen can see what they are doing. Visual inspection is possible. The working surface is not fouled by fluctuating tide levels, and the reinforcement, if any,

can be placed more


However, if the cost of doing the work in the dry is substantially greater than the cost of doing the work without dewatering the forms, the author does not consider the extra expenditure to be warranted, having obtained uniformly excellent results on several projects doing the work in the wet.

In this connection, even if it is decided to do the work in the dry, do not forget to include permanent forms and to consider their cost. Permanent forms are not provided just to hold the concrete during placement, but serve several other functions, as described in para-

graph 2a. f. Handpacked Jackets. A new method offering interesting possibilities for jacketing piles below water level consists of using an additive in the grout which gives it quick-setting properties, enabling the grout to be applied to steel, wood, or concrete piles at or below water level, either pneumatically or by hand. The grout can

be made to stay in place without forms of any kind, even in the face of heavy wave action. The method is outlined in Reference 5, but experience data are limited.

4. Finishing the Jacket The top of the jacket should be finished with a collar of pneu-

matically-projected or hand-placed


The collar should be



Repairing Spalling and Disintegration


sloped to drain and should be faired into the existing concrete in such a manner that a smooth transition between repaired and exis-

ting work will result (see Figure 5-13).



a. Options. It is advisable to describe several acceptable methods for filling jackets in the specifications for the work and to permit the contractor to select that method which he feels is most economical.




Filling the



are sev-

eral methods which can be relied upon to give excellent results. The most economical one for a particular job depends on the con-


his equipment,

his experience,



and the site

conditions. b. Prototype Jackets. Chapter 1, Section C, refers to the need for prototype installations. Nowhere are they more needed than with filled jackets. Sample specification provisions relating to prototype jackets are contained in Chapter Note 2. c. Removal and Replacement of Forms. Inspection of repair work

is essential.

If permanent







spection would be impossible unless provision were made to remove and replace some sections of form. Sample specification provisions relating to removal and replacement of forms are contained in Chapter Note 3. This should be a separate, unit price item of the contract.

d. Reinforcement.

Provide wire mesh

(or equivalent) reinforce-

ment for the jackets. The original trouble was probably due to exposure of the concrete surface to an aggressive environment. The mesh will tend to prevent the formation of major cracks in the new matrix. Since the material in the jacket is a rich mix and usually not compatible with the material which it is supposed to protect, such cracking can be a major problem. e. Admixture. Addition of an admixture designed to reduce setting shrinkage is recommended. f. Shoring. Sometimes repair work is not undertaken until the structure is dangerously weakened. The chipping, sandblasting, and other operations incident to preparation of the surface to be repaired will temporarily weaken the structure further. Do not start preparing the surface of a whole line or group of piles, piers, or columns indiscriminately, lest a collapse be triggered. Provide shores. Where this is impractical, repair a few members at a time,

at scattered


until sufficient strength




stored to work the job on a production basis. g. Heat of Hydration. The rich mix used to fill the jacket and the insulating quality of the heavy form will cause the concrete fill to


Deterioration, Maintenance,





and Repair of Structures

heat of hydration.

In fact,

if the

forms are stripped shortly after being filled, the surface of the concrete fill may be uncomfortably hot. Do not be alarmed. No serious damage has been done. However, when the concrete has cooled, check to see if the bands on the forms need tightening.



Pneumatically applied mortar is used for the restoration of concrete surfaces where the deterioration is relatively shallow. It can be used on vertical and overhead as well as on horizontal surfaces and is particularly useful for restoring surfaces spalled due to corrosion of the reinforcement. Although this method of repair has had widespread use for perhaps half a century, its efficacy is still the subject of some debate. Criticism


is directed

(1) The material is porous.

at the following points:

(2) It shrinks more than conventional concrete. (3) The coating may inadvertently include pockets of rebound or



and hide hollow


(4) The quality of the work is very sensitive to the skill of the


(5) The work is expensive.

As for expense, all repair with new construction. In its pneumatically applied mortar repair. The other criticisms by using the method where it

be done.

(See Reference 6.)

work is expensive when compared limited field of application, the use of is competitive with other methods of are valid, but with proper care, and is applicable, a first-class job can

1. Description

Pneumatically applied mortar (also called shotcrete and gunite)

is a mixture of portland cement, sand, and water, shot into place by compressed air. In structural applications, the sand and cement are mixed dry in a mixing chamber, and the dry mixture is then transferred by air pressure along a pipe or hose to a nozzle, where it is forcibly projected onto the surface to be coated. Water is added to the mixture by passing it through a spray injected at the nozzle. The flow of water at the nozzle can be controlled to give a mix of desired stiffness, which will adhere to the surface against which it is projected.

Concrete Structures: Repairing Spalling and Disintegration


2. Applications In general, pneumatically applied mortar is used for relatively thin coatings, say up to 3 or 4 inches, occasionally more, and for the repair or coating of relatively large areas. Small and/or scattered areas usually are more economically repaired by mortar, placed by hand. Use of the material generally is limited to applications involving structural repair, where its ‘‘wooly’’ appearance is not objectionable. Normally, it is not used for redecorating or restoring the appearance of architectural surfaces, because uniformity of appearance is relatively poor, and a high degree of surface planeness cannot be achieved. It also is not applicable for coating surfaces which must be smooth for hydraulic purposes, for streamlining,

or for trafficability (such as flooring).

It is best not to use pneumatically applied mortar for restoring surfaces having offsets which would require forming by use of timber battens, because the work is necessarily carried up to the batten and left overnight, the batten removed, and the work then continued. This leaves a cold joint at a critical location—an open invitation to the penetration of water. Despite these limitations, however, the applications of this material are numerous and include protective coatings for steel, masonry, and rock; linings; encasement of exposed reinforcement (particularly where the reinforcement is exposed because of spalling of the concrete cover); filling out spalled areas; and coating riprap



Preparing the Concrete Surface

The provisions of Section A apply. The final surface must be rough to afford a keying effect. Anchor bolts tying the new work to the old concrete are essential. As discussed in Chapter Note 1, incompatibility of pneumatically applied mortar with the old concrete is a major problem, and the keying effects of the rough surface and the doweling effects of the anchor bolts are necessary to assure interaction between the two materials (see Figure 5-17). 4.

Materials and Proportions

Sand for pneumatically applied mortar should be uniformly graded, as for conventional concrete. Hard particles are desirable, since there is a tendency to grind and crumble the grains as they

pass through the discharge hose. per cent moisture

The sand should contain 3 to 5

for efficient operation of the equipment.




and Repair of Structures

Because of impact, a certain amount of the material being projected against the surface to be coated will bounce off. This mater-

ial is known sand

as ‘‘rebound.’’


and has

It consists primarily of the coarse

a much

as projected from the nozzle. of some


20 to 30 per cent. Because



Practically speaking,

is unavoidable.

of the rebound,


In fact,

the cement

it frequently

than the mix

the occurrence amounts

content of the mortar,

to in

place, will be substantially greater than that of the materials as fed to the mixer. Thus, if a 3:1 mix is desired in place, a 4:1 mix fed into the mixer might be adequately rich. It is difficult to specify mix proportions for pneumatically

jected mortar of rebound,

because of uncertainties


of difficulties

and variations

in making


in the amount



cylinders, and because there is little control over the water/cement ratio of the material, in place. In practice it is usual to use as much water as possible without causing fallouts or sloughing, as this minimizes the rebound. As a rough estimate, this might mean a water/cement

ratio of 0.5 to 0.6.

Starting with this assumption,

and assuming a 20 to 30 per cent rebound, the proportions required for the mix, as fed to the mixer, can be roughly approximated for any desired strength. In practice, the author usually calls for a 3:1 or 3 1/2:1 mix, by volume, to be fed into the mixer. An airentraining agent or cement is used. 5.

Mixing and Placing The following sample specification illustrates the important


to be considered.

Mixing: The sand and cement shall be thoroughly mixed in the dry state. Only approved mixing and placing equipment shall be used. All oil or other rust inhibitors shall be removed from parts of the equipment which will come in contact with the mortar before the equipment is used. The time of mixing shall not be less than 1 1/2 minutes.

Any mixed


not used within

1 hour of mixing


be wasted. Mixing drums and blades shall be inspected at frequent intervals, and all materials caked on the drum or blades shall be removed. Worn blades shall be replaced. Air Pressure and Water Pressure: The compressed air flowing into the propelling device shall be maintained at such a pressure that it is adequate to carry the dry mixture through the material hose at such a velocity that the mixture is projected from the nozzle ina forceful stream. The hydrating water fed into the mixture at the nozzle body shall be maintained at a pressure in excess of the air pressure and such



Repairing Spalling and Disintegration


that there is an adequate positive flow into the nozzle. On vertical or overhead surfaces particular care shall be taken to keep the plastic

mix dry enough to avoid sloughing or falling, without causing excessive rebound. Placing: Except where cramped working space or other local conditions interfere, the nozzle shall be held between 2 feet and 4 feet from

the surface

to be coated,

and held in such a position

that the

flow of material will strike it at as near to a right angle as possible. In shooting vertical or sloped surfaces, the placing shall be started at the bottom and carried up. Also, on such surfaces the pneumatically projected mortar shall be placed in layers of such thickness

that the weight

of the plastic


does not cause

it to sag.

When more than one layer is to be used to complete the final thickness of the work, the delay between application of the successive layers shall be ample to prevent sagging or fallout of the mass, but not so long that the underlying

oped a glaze coating.


a proper interval.) Before any pneumatically


has completely

set and devel-

Thirty minutes to one hour is usually projected


is applied, care


be taken to remove any sand or rebound clinging to the surfaces. No pneumatically projected mortar shall be applied to a surface on which there is running or free water. When shooting around reinforcing rods or anchor bolts, the nozzle shall be moved from side to side and angled to place pneumatically projected mortar back of the rod. At the end of the day’s


or at similar

stopping periods,


pneumatically projected mortar shall be tapered to a thin edge. Before shooting the adjacent section, this tapered portion shall be thoroughly cleaned and wetted. No square joints will be allowed except at the edges of repaired areas. All operators of equipment for mixing and placing of pneumatically projected mortar shall be experienced men. Minimum crew shall include a rebound man, who shall be in addition to the nozzleman

and machine operator. Operations shall be suspended when wind velocity is such that it blows away the spray from the nozzle and prevents proper control of the consistency. At the edges of repaired areas, the pneumatically projected mortar shall be faired, flush-face, into the existing concrete. The placing of the mortar shall be performed in such a manner that existing weep




and other necessary


ses in the existing surface are not blocked or filled. Wood

blocks, concrete

nails, or other approved


or recesor markers

shall be attached to surfaces to be repaired, so that the required thickness of mortar and the level of the final surface will be clearly indicated to the workmen placing the mortar and to the inspection forces.

Such blocks,

nails, or markers

shall be removed

after the


Deterioration, Maintenance, operations are completed. an approved

and Repair of Structures

The resulting cavities shall be plugged



Finish: Pneumatically projected mortar shall be shot on about 1/2 inch from the final surface and all irregularities with a sharp edged trowel. Pneumatically projected mortar be screeded. A finishing or flush coat shall then be applied the pneumatically projected concrete to a final surface. All matically projected concrete shall be applied with a minus of 1/8 inch. The pneumatically projected concrete shall be a natural finish. (Note: Screeding produces tends to impair the bond.)

a smoother

to a line cut away shall not to bring pneutolerance left with



Curing: As soon as dry patches begin to appear on the surface of the newly placed mortar, curing by use of a water spray or application of two coats of an approved sealing compound shall be commenced. Minimum curing period shall be 7 days.

6. Remarks a. Encasing the Reinforcement. The provisions of Section A apply. However, due to the porosity of pneumatically projected mortar, full encasement of the bars is particularly advisable. (See Reference 7, p. 135, and discussion thereof.) b. Mesh and Anchor Bolts. Provide them. They are needed to distribute the cracks in the new work and to keep the new work in intimate contact with the surface being repaired. They also serve to keep the cleavage plane at the interface closed. c. Rebound. Do not underestimate the importance of removing rebound. Unless care is used, numerous and surprisingly large pockets of rebound will occur, creating traps and reservoirs for moisture. It may not be pleasant, but the inspector must follow where the nozzleman is working to check and be sure that the rebound man is doing his job. d. Fallout and Sagging. On any extensive job, some local fallout or sagging of the mortar will occur. The entire surface of the new work

should be sounded




a hammer

and all hollow


cut out

and repaired. e. Defects in the Surface Being Repaired. All defective concrete can and should be removed from surfaces to be repaired. However, the cracks in that surface cannot be eliminated. If these cracks are active, corresponding cracks will appear in the covering coat of pneumatically applied mortar. Experience has demonstrated that the thin mortar skin will not stop the cracking of the larger mass, nor will it serve to waterproof the existing surface against a head of water. It will help to seal against a spray condi-



Repairing Spalling and Disintegration


f. Sealing Shrinkage Cracks. The development of fine shrinkage cracks in pneumatically applied coatings should be expected. These cracks can be sealed easily and relatively inexpensively by periodic coatings

of linseed oil, silicones,

cribed in Section I.


and similar


as des-


The use of prepacked concrete is well adapted for certain types of repair work, particularly under water and elsewhere where accessibility is a problem. 1. Description Prepacked concrete is made by filling forms with coarse aggregate and then filling the voids of the aggregate by pumping ina sand-cement grout. A neat cement grout is sometimes used for special work. As the grout is pumped into the forms, it will fill the voids, displacing any water in them, and form a concrete mass.

2. Applications Prepacked concrete is used for refacing of structures, jacketing as described in Section B, filling of cavities in and under structures, and underpinning and enlarging piers, abutments, retaining walls, and footings. It is also used for stabilization of rubble and mound structures such as breakwaters. 3. Composition Coarse aggregate should conform to the requirements for ordinary concrete, must be carefully selected, and must be clean. The void content of the aggregate mass should be minimized to reduce the required volume of the more costly grout. In normal practice, the voids ratio ranges from 35 to 45 per cent. The




fine sand,




pozzolanic material of low mixing-water requirement, an ‘‘agent’’ designed to increase the fluidity and to inhibit early stiffening of the grout, and mixing water.


Deterioration, Maintenance,

and Repair of Structures

4. Placing the Grout The forms must constitute a closed, watertight system, vented at the top only. If not watertight, the travel of the grout cannot be controlled. If not effectively vented, a back-pressure will be created, with the result that the concrete fill may contain voids. Preparation of the surface of the existing concrete should conform to the requirements of Section A. Pumping of mortar should commence at the lowest point and proceed upward. The grout pipes should not be more than 5 feet on centers, and the grout level in the mass should be brought up uniformly, as determined by observations of grout levels in the grout pipes. A positive head of at least 3 to 5 feet should be maintained in the grout pipes above the level of the outlets. Placing of grout should be a smooth, uninterrupted operation, and a positive head should be maintained in the grout pipes after the forms have been filled and until the grout has set.

5. Remarks The use of prepacked concrete can produce excellent results, but it is a specialized operation which should be attempted only under the careful supervision of competent, experienced personnel. If not performed carefully and rigorously, defective work will result and probably go undetected. With any repair project, it is desirable to employ sample or prototype installations, and such constructions should be completed before beginning the work. This is particularly important when considering any repair by the prepacked method. Provisions for the recovery of cores and/or the removal and replacement of forms to permit inspection of the concrete are highly desirable. Sample provisions of this type are given in Chapter Notes 2 and 3.



This method consists of replacing the defective concrete with new concrete of conventional proportions, placed in a conventional man-



1. Applications Replacement of defective concrete with new concrete of conventional proportions is a satisfactory and economical solution where

Concrete Structures:

Repairing Spalling and Disintegration


the volume of material to be replaced is relatively large, where the repair occurs in depth, i.e., several inches deep (and at least beyond the reinforcement), and where the area to be repaired is accessible.


the work

is inaccessible,


as below

corresponding solution is the use of prepacked concrete if the volume requirement is not excessive). Repair

by replacement

is particularly

indicated where



(or grout, a water-

tight construction is required and where the deterioration extends completely through the original concrete section. Piers, walls, hydraulic structures, and similar heavy structures above grade and water level are the usual areas of application. If economically justified, replacement of the concrete is probably the best of the several methods of repair described herein, at least insofar as durability and soundness of the repaired work are concerned. The new section can be reinforced and made self-sustaining. With proper construction and contraction joints, using waterstops where required, the repair can be rendered substantially watertight. Thin sections, which generally give trouble due to temperature changes, are avoided. In fact, the greater thickness reduces the magnitude and frequency of temperature variations in

the underlying parent mass and, most important, at the interface between the new and repaired sections. Furthermore, the propor-

tions of the new concrete can be made section

being repaired,


to resemble those of the


the degree

of incompati-

bility of the two materials and reducing the damaging effects of temperature changes, changes in moisture content, and differential elastic or inelastic strains.


Preparing the Existing Concrete Surface

The provisions of Section A apply. The shape of the excavated area should be such as to permit vibration of the concrete during placement and to ensure filling the cavity without entrapping air pockets. 3.

Proportioning the New


There are two cases to be considered: (1) The first case is where the new concrete occurs in sections which are large in length and width as compared with the thickness,


the material

is used to resurface

an existing construc-

tion, or where the compatibility of the two materials is important in preventing separation along the interface. In such a case, if possible, match the mix used in the existing work to be repaired,




and Repair of Structures

particularly with regard to size, type, and quantity of coarse ag-

gregate and with regard to the water/cement ratio, but be sure that the slump is the minimum consistent with placing the new concrete.


paired was

this is impossible because the deterioration being recaused by the use of a poorly

More often, however, mix,






the cause of the difficulty is a satisfactory


too little cover,

too much



quate compactive effort resulting in honeycomb or in excessive settlement of the plastic mix—or other poor construction practices or poor design details as described in Chapter 3. (2) The second case is where the new material is used in massive sections; where it is used in sections of limited length and width as compared with thickness; where the new work is tightly confined by existing concrete; or where the strains of temperachange,

or shrinkage,

ratio may

be similar

or the elastic strains

due to

applied load do not produce any pronounced tendency toward a shear or separation along the interface between the two materials. In Such a case, the compatibility of the two materials is of lesser importance, and the mix should utilize as large a coarse aggregate and as low a slump as are consistent with proper placement. The


Air entrainment is important,

to that for new

(3 to 6 per cent) is advisable.

the brand,



type, or blend of cement


appearance be varied


match the existing work. Care must be observed to provide uniform batching of materials. Batches

will be small

and numerous,



in batches


cause variations in appearance of the finished work. For this reason, where the required yardage permits, batch enough concrete at one time to fill the entire hole.



(1) For open top forms (for example, for repairs to horizontal

surfaces and tops of walls) no special features are required other than that forms should be mortar-tight and closely fitted and/or gasketed where they adjoin existing concrete surfaces to prevent leakage which would result in honeycomb and sand streaking. (2) For closed forms (for example, replacing or patching a section of a vertical wall remote from the top of the wall) it is feasible to give the concrete plug a ‘‘forced fit’’ by applying pressure to the forms by use of a pressure cap in the chimney used to fill the form

(see Figure 5-8). Tight forms are essential. Joints between form

boards, between panels, at juncture with existing concrete, and around holes for tie rods should be caulked. Forms must be substantial. Tie rods should have large sturdy washers and should be

Concrete Structures:

Repairing Spalling and Disintegration


Pressure on new concrete exerted by tightening bolt

Existing Concrete

Movement of Pressure Block

Slope to Vent

[— Pressure Block

Tie Bolts

j~— Wale

i; wates

New Concrete Form

Existing Concrete





of pressurized


fitted with blocks under the wales to minimize deflections of the wales and give better bearing on the form sheeting. Pressure is applied to the newly placed concrete by tightening the tie rods through the pressure cap. Pressure should be applied immediately after the void has been filled. If form vibrators are used, pressure should be reapplied after intervals of vibration.

5. Placing and Curing When concrete is being placed in a vertical, closed form more than 18 inches high, placement should be in lifts not exceeding 12 inches,

and the forms

should be constructed

in horizontal


to permit such placement. Provide sufficient openings of adequate size in the forms so that




and Repair of Structures

it is not necessary to run the concrete horizontally in order to fill the form. For walls and similar structures, the vents or chimneys should run the full width of the cavity. Thorough vibration, both internal and of the forms, is essential. However, do not vibrate closed forms unless accompanied by the application of pressure, lest a void or film of water be left at the top of the repair. When filling the top of a closed form, do not let the depth of concrete in the chimney exceed 2 to 3 inches, since a greater depth tends to dissipate the pressure. Conventional curing procedures are applicable.

F. DRYPACK Drypacking is the hand placement of a very dry mortar and the subsequent tamping or ramming of the mortar into place, producing an intimate contact between the new and existing work. Because of the low water/cement

ratio of the material,


is little shrink-

age, and the patch remains tight and is of good quality with respect to durability, strength, and watertightness. No special equipment is required. The necessary hand tools are available to all concrete finishers.

1. Applications Drypack is used for filling small, relatively deep holes, such as those resulting from the removal of form ties, and narrow slots cut for repair of cracks. The use of drypack is not advisable for filling or patching large, shallow areas; for extensive areas where it is necessary to fill behind reinforcing bars; or for repairs requiring large volumes of material.

2. Preparing the Existing Concrete Surface The provisions of Section A apply.

3/4 inch deep and should

Bonding surfaces

be undercut

should be rough.

Holes should not be less than as shown

in Figure


3. Mix The usual mix is one part cement to 2 1/2 to 3 parts of fine sand

(passing a No. 16 screen).

Only enough water is added to produce

a mortar which can be molded into a ball by a slight pressure of the hands and will not exude water but will leave the hands dry.

(See Reference 8.)

Concrete Structures:

Repairing Spalling and Disintegration


4. Placing For best results, skilled workmen are required. Drypack should be placed in compacted layers not more than 3/8 inch thick. The surface of each layer should be scratched to bond with the succeeding layer, and each layer should be compacted by use of a hardwood stick and a hammer. The stick is placed against the mortar and struck with the hammer. For small areas the end of the stick is placed against the mortar and the tamping begun at the middle of the area and progressed toward the edges to produce a wedging effect. For larger areas, a T-shaped rammer may be used, laying the flat head of the T against the mortar and hammering on the stem. Successive layers of drypack are placed without interval, unless the material becomes spongy, in which case there should be a short wait until the surface stiffens. Do not attempt to dry out the material by ramming in a dry layer. Alternate wet and dry layers must not be used. Holes should be filled flush, and finished by striking a flat-sided

board (or the flat of the hardwood stick) against the surface. rammers trowel.


should not be


used. Do not strike off the work with a


The use of overlays to seal cracks was described in Chapter 4 and to repair pavement in Chapter 6. An overlay also may be used to restore a spalled or disintegrated surface or to protect the existing concrete from further attack by aggressive agents in its environment. Overlays used for these purposes include concrete (or mortar), bituminous compounds, and epoxies, and should be bonded to the existing concrete surface. The use of unbonded overlays is limited to restoring pavement and is not considered herein. The use of epoxies is described in Section H. The use of concrete and bituminous compounds is considered in this section.

1. Applications An overlay can do much to inhibit attack from aggressive agents in the environment around the concrete, but is useless against (1) internal attack due to chemical reactions, (2) deterioration due to thermal strains, or (3) moisture penetration coming from a face remote from that being treated. Also, the work is of limited value where the overlay can be breached by the movement of active cracks in the parent mass. Except for heavy plastic coatings (such as bitumens), an overlay


Deterioration, Maintenance, and Repair of Structures

should not be used where watertightness is required (i.e., as a substitute for membrane or similar waterproofing). Unless the coating is thick and soft (as in a mortar-plugged blanket joint, described in Chapter 4), if the overlay bonds to the concrete, as it should, and must, to protect the concrete against the further development of spalling and disintegration, it cannot be sufficiently flexible to bridge thin, active cracks in the underlying mass. Also, it is virtually certain that the overlay materials will not be elastically and thermally compatible with the underlying concrete, and will, therefore, tend to crack.

Accordingly (except for pavements), when used to repair spalling

or disintegration, an overlay is usually considered, not as a structural repair, but as a means for prolonging the life of the underlying concrete section. For example, in any repair job involving resurfacing of extensive areas where the concrete cover has spalled (for whatever cause), there will be amplé areas which are still relatively sound with only occasional cracks or spalls. Some of these areas may contain excellent concrete. Most will be just a

little better than the neighboring areas which have spalled and will



in a few years.



to repair these


now would require chipping relatively hard concrete and would cost a great deal. From an economic standpoint it may be better to let the area deteriorate, provided that it will not collapse, and repair it later. First, the expenditure is deferred, thereby returning the value of investment of the deferred expenditure. Second, the final repair probably will not cost much more than if the repair is attempted



will be more

steel to replace,

but less


to chip. Thus, the overlay acts as preventive maintenance and so may be economically attractive. (See Chapter 1, Section B.) For example, suppose that the cost of a later repair is substantially the same as that of doing the work at the present time, and assume that the cost of the overlay is about 15 to 25 per cent of the cost of chipping, replacing, and encasing the deteriorated steel (which is a usual value). Taking the investment value of the deferred expenditure as 5 per cent per year, the overlay has only to increase the

life of the structure 4 or 5 years to pay for itself. Any additional

increase in life of the structure represents a profit. Another use for an overlay is to give areas which have not been repaired an appearance similar to those which have been repaired or, simply, to improve the appearance of a structure which has deteriorated to a minor degree.

2. Preparing the Concrete Surface The provisions of Section A apply.

Concrete Structures:

Repairing Spalling and Disintegration


3. Materials a. Pneumatically Applied Mortar.

Pneumatically applied mortar

is frequently used. The thickness should be at least 1 to 1 1/2 inches, depending on the severity of the exposure conditions, the thickness which can be placed in a single layer, and the cost. Pneumatically applied mortar is pervious and will not endure as well as hand-troweled material, but is less expensive, particularly if the overlay is placed in conjunction with the repair, also using pneumatically projected mortar, of adjacent spalled areas.

b. Sand-Cement Slurry.

This consists of two or more brush

coats in the proportions of 1 or 1 1/2 parts of fine sand to 1 part of cement, mixed with water. A waterproofing agent may be added, if desired. A finish coat of neat cement (using white cement, or cement plus pigment, where necessary to match existing work) may be used to give the work a smooth or colored appearance. The total thickness of the coating is generally about 1/8 inch. c. Sand-Cement Plaster. This consists of two or more ‘‘cast on,’’


or floated


is about 1/4 to 3/8 inch thick.

of sand-cement




The first layer, or tack coat, might

be a 1:1 mix with subsequent coatings being of increasingly leaner mix. The final coat is usually about 3:1 proportion. For additional information and details on overlays using sand-cement plasters or slurries, reference is made to the excellent coverage contained in

S. Champion’s book. °® d.

Epoxy Resins.

See Section H.

e. Bituminous Coating of Reinforcement.

As described in Chap-

ter 1, it is sometimes possible to effect a repair in a particularly economical manner by maintaining the structure in a status quo, i.e., simply by preventing further deterioration. A common case of this type occurs where the damage consists of spalling of the concrete cover due to corrosion of the reinforcement. Frequently, it will be found that the bars have been entirely exposed over extensive areas of the structure, apparently without any discernible reduction in the load-carrying capacity. The reasons for this are set forth in the Chapter Note following Chapter 1. If the conditions set forth there can be met, it is possible to prevent further deterioration by chipping back the concrete to expose all corroded reinforcement (and all faces of exposed reinforcement) and coating the bars with a heavy application of bitumen. The bars should be cleaned of corrosion, preferably by use of a sandblast, before applying the coating. Splices should be mechanically secured. Caution: Be sure that the conditions of Chapter 1, relating to extension

of deterioration,


of redistribution

of stresses,






Repair of Structures

adequacy of details of design can be met, lest a dangerous condition be overlooked or preserved. Also, beware of creep in the steel if the bond of the bars is broken for any great length of bar. 4.


a. Pneumatically Applied Mortar. apply.

The provisions

b. Sand-cement Slurries and Plasters:

of Section C

A slurry is brushed on to

the surface to be coated. The first coat should be well worked into the existing surface to improve the bond. Subsequent layers are applied by brush as soon as the previous coat has sufficient strength to support the added weight. The layers of a sand-cement plaster coating may be placed by troweling or floating. On vertical surfaces, bond may be improved by ‘‘casting’’ the material forcibly against the surface to be repaired, similar to the action when applying pneumatically applied mortar, except that the operation is performed by hand. The final coat may be finished as with ordinary plaster, leveling and planing as required. On horizontal surfaces, the techniques of Chapter 6, Repairing



Epoxy Resins.




See Section H.


be used.

5. Remarks As has been noted, an overlay is not intended to be a structural reinforcement unless bonded to the underlying mass and adequate in thickness. Its basic function is to reduce attack on the concrete by aggressive agents in its environment and to restore appearance. If the surface being repaired has areas of defective concrete which are not removed or if the surface is intersected by active cracks, the defects or cracks will appear in the overlay. If the intrusion of water from the reverse face occurs, the overlay will be stained, spalled, or otherwise damaged. However, properly applied, the technique can give excellent long-life protection, albeit not to the degree attainable using the more substantial and basic methods of repair described in the previous sections. A service life of 10 to 20 years is readily attainable, with longer periods possible—depending on the site conditions, details, and workmanship.

H. EPOXY RESINS Epoxy resins are organic compounds, which, when activated with suitable hardening agents, form strong, chemically resistant structures having excellent adhesive properties. They may be used as

Concrete Structures:

Repairing Spalling and Disintegration


adhesives or binders to bond new concrete patches to existing surfaces or to bond together cracked or severed portions of an existing concrete section. Once hardened, for all practical purposes, the compound will not melt, flow, or bleed. The use of epoxy compounds in repairing cracks has been described in Chapter 4.

1. Applications a. Bonding Cracked or Severed Portions of a Concrete Section.

Theoretically, if a section of concrete has been severed from the parent mass, the displaced portion could be reconnected by coating the contact, in-place, and severed surfaces with the adhesive compound and joining them. In practice, this is seldom economically justified, and common practice is to patch the spall with new con-

crete (possibly using an epoxy patching compound).

b. Patching Spalled Surfaces. The epoxy resin may be used as an adhesive to bond a patch consisting of portland cement concrete or may be substituted for the portland cement and used as the binder in the new


For small


of material,

for thin sections,

or where conditions require that the patch be subject to traffic before there is time to permit proper curing of a conventional concrete patch, the epoxy compound should be substituted for the portland cement

and used as the binder.


it is more


ical to use conventional concrete bonded to the original work with a coating of epoxy adhesive.

2. Preparing the Existing Concrete Surface The provisions of Section A apply. It is most important that the surface be strong. The epoxy itself is very strong, and any weakness in the existing concrete will invalidate the strength of the new patch. A method for testing the strength of the existing surface is described in Reference 10. It also is most important that the bonding surface be scrupulously clean and dry. For some compounds, complete dryness is required. For others some dampness can be tolerated. In any event, it is well to have the surface as dry as possible.* Where a capillary or hydrostatic condition is suspected, a rubber mat or polyethylene sheet laid over the patch area, weighted, and left overnight will *although most epoxies will not properly cure in the wet, for particular problems, specially formulated compounds can be obtained" which will cure in the presence of water or even when completely submerged.


Deterioration, Maintenance, and Repair of Structures

serve to test the amount of moisture percolating through the surface. The cover is raised in the morning. If moisture has collec-

ted under the cover, it may be necessary to dry the surface, arti-

ficially and in depth, before applying the patch or to use a patching compound which will bond to a moist surface.

3. Materials Commercial formulations are used.

all manner

of modifiers

Such formulations contain

as required to adapt the compound


varying specific applications. Accordingly, a specific epoxy compound should be used only for the purpose for which it is intended

and only in accordance with the manufacturer’s directions. Moreover, many of the formulations are quite new, so that, when selecting one, it is advisable to insist on field performance data to sup-

port the laboratory results.

Where the patch consists of concrete prepared with an epoxy binder, in general the aggregates must be surface dry, otherwise the mixture will not harden. Also, there should be enough binder

in the mix so that there is an excess to wet the bonding surface. Alternatively,

the bonding surface may be primed with the adhesive.

4. Mixing and Placing Thorough

mixing of the components

of the adhesive

is required.

The California Division of Highways uses an interesting technique

to ensure proper mixing

(see Reference


It is specified that

the components be tinted in contrasting colors and that the com-

bined adhesive be mixed to a uniform color without visible color streaking. The amount of material mixed in a given batch should not exceed

the volume


by the manufacturer,


larger batches are more rapidly cured because of the heat gener-

ated by the larger volume. Where an epoxy is used as the binder in a concrete patch, mixing may be accomplished by hand or machine methods. Care should be observed not to contaminate one component by mixing with tools

used in another component. The sand or aggregate should not be added until after the binder has been mixed.

Where the epoxy is used as a primer or bonding adhesive, it may be applied with squeegees, rollers, or brushes. Care should be ob-

served to rub it in well enough to avoid entrapped air pockets. Hardening of the adhesive on the applicators will occur,

should be inexpensive and disposable. cleaned in a solvent wash.

and they

If they are not, they may be

Application in the amount of 30 to 50

Concrete Structures:

Repairing Spalling and Disintegration


square feet per gallon is usual, with considerably heavier application on rough surfaces. Spray application may be considered if suitable precautions are taken to apply the patch after the spray solvent has been completely evaporated but before the adhesive has set. This can be very difficult to accomplish, however, and improper timing will cause unsatisfactory results. Patches must be placed before the adhesive or binder has started to harden. The patch is pressed or tamped into place and finished by troweling or screeding. Curing follows standard procedures. When working on vertical or inclined surfaces, forms may

be used.

The time


for hardening of the epoxy compound depends

on the thickness of the patch and on the air temperature.


temperatures and larger masses cause hardening to occur more rapidly. Accordingly, hardening can be accelerated by the applica-

tion of heat to the compound or patch. Tenting and heating may be used, the patch may be covered with a black polyethylene film to

absorb solar radiation, infrared heaters may be used, or the patch may simply be covered with a metal plate, which is then moderately heated with a blowtorch. Under normal circumstances, at the

time of application, the temperature of the epoxy compound should

be between 70 and 90°F and of the concrete surface, above 60°F. 5. Remarks

a. Thermal Compatibility. While only limited data are available, it is advisable to exercise caution and to provide frequent joints when patching relatively large areas, because there is indication of substantial thermal incompatibility between concretes mixed with portland cement and those using epoxy resin as the binder. b. Cold Weather. Epoxy materials exposed to cold weather tend

to become brittle.

This matter should be thoroughly investigated

before using this material in applications where there will be such exposure. The installation may be impractical or may require special compounds or formulations.

c. Strength of the Bond.

The strength and bond of a properly

formulated epoxy adhesive is greater than that of the concrete. This has been demonstrated many times in test specimens subjected to bending, tension, and torsion. Failure occurs in the concrete

rather than in the adhesive. d. Dangers. Workmen should be protected from direct contact

with the epoxy material.



and disposable gloves

are desirable. Coveralls should be worn. Any skin areas accidentally contaminated with the resins should be cleaned immediately.



can be harmful,

and if work

is being done indoors,


Deterioration, Maintenance, and Repair of Structures

Figure 5-9. Epoxy patching of concrete surface. (a) The patched area prepared to receive the epoxy patching mix. The edges of the area to be repaired have been neatly saw cut and the area cleaned of loose concrete, dirt, and debris to sound matrix. This is workmanship at its best. (b) The completed patch.

Concrete Structures: Repairing Spalling and Disintegration


proper ventilation must be provided. Similarly, where equipment is to be cleaned, or cured epoxy removed by heating the compound, care should be observed to avoid inhaling the fumes.

e. Fire Resistance.

Epoxies are organic compounds and, as

such, are deteriorated by heat. They have a poor fire resistance, and repairs made with the usual formulations should not be used where the fire-resistance rating of the construction is of critical importance.

f. Wet Weather During Application.

As noted, the bonding sur-

face should be dry to achieve good adherence of the epoxy. On jobs exposed to the weather, the work should cease if it rains and not resume until the surface has adequately dried. However, rain does not particularly harm a semihardened epoxy surface. g. Saturation of the Patch. It is usual to keep the binder content of the patch as low as possible

to reduce

the cost of the mixture.

For exposed work, if the binder content is not adequate to saturate the patch, it will be porous and, if subject to freezing and thawing, will spall. There are two ways to correct this. One is to design a mix which will be saturated. The second, and in most cases the better way, is to go over the patches with the epoxy after they are

made and to effect a seal by filling them with as much binder as

they will absorb through surface applications. h. General. The use of epoxies as adhesives or binders for patching spalled surfaces is an alternative to the procedures of surface preparation described in Section A. If properly performed, the procedures described in Section A will give satisfactory results. However, considering the increasing difficulty in getting quality craftsmanship in construction, the use of an epoxy, if climatic and other conditions permit, is a good substitute, since the application does not require the careful preparation and workmanship necessary to get good bond in a conventional patch.




The durability of concrete can be substantially improved by preventive maintenance in the form of a weatherproofing surface treatment. These treatments are used to seal the concrete surface and to inhibit the intrusion of moisture or chemicals. Well-designed structures made with top-quality concrete do not require such treatments, but considering how easy it is to get something less than top quality in concrete and that surface treatments are relatively inexpensive, they are generally regarded as good insurance and are in common use. However, care must be taken not to use surface sealants where such use would seal moisture into the


Deterioration, Maintenance, and Repair of Structures

concrete by preventing the evaporation of water penetrating from unsealed surfaces. In such a case, the sealant would do more harm than good, and this condition must be scrupulously guarded against. Types of surface treatment in common use include:

1, Oils a. Linseed Oil. An antispalling compound consisting of a mixture

of 50 per cent boiled linseed oil and 50 per cent mineral or petroleum spirits, by volume, is commonly used. The coatings must be applied to clean, dry surfaces, free of dust or loose particles, pre-

ferably during warm weather (at or above 70°F).

Application is by spray, using motorized equipment

for large

jobs or a weed spray unit (either hand or power) for small areas.

The rate of application is a matter of experience. However, at least two coats are required, with a third coat used where inspection after the first coat is dry shows the concrete to be particularly porous. Successive coats should not be applied until the previous coat has thoroughly dried.

b. Petroleum Oils.

Petroleum oils are also used for weather-

proofing but weather out rather rapidly and are sometimes unsightly. They are applicable, however, where not exposed to the weather or traffic, such as for sealing the surface of a structural bridge slab which is covered by a separate wearing course. The Department of Public Works of the State of New York has used a modified asphalt cutback solution for this purpose, although their more recent practice is to use a silicone solution.

2. Silicones Silicones are in common use to seal concrete masonry against absorption of moisture. Treatment consists of a single spray application of a dilute silicone solution to the surfaces to be protected. Mechanical spray equipment is used on extensive horizontal surfaces. Hand-operated equipment, including brushing, may be used on vertical surfaces or small areas. The provisions of Section A relating to surface treatment are applicable. One and only one application of the solution is made, and no attempt is made to respray treated areas, since the material, once dry, will repel itself. An exception is where the spray bar will not cover the entire surface to be treated, in which case successive passes should overlap to assure complete coverage.


The silicone solution is likely to be strongly caustic,

and the eyes, skin, and clothing should be suitably protected. If contact does occur, the affected area should be flushed with water.


Figure 5-10. Protective surface treatment. The surface treatment in this case was a silicone water repellent. The photograph was taken after a heavy rain. The area at the right, which was treated with the repellent, is still dry. The dark color of the untreated area at the left shows the relative degree of water absorbed. (Courtesy of the General Electric Co.)

Care must be exercised against excessive inhalation of the solvent vapor during spray application. Care also must be taken to avoid fire or explosion, since the flash point of the solvent solution is

likely to be quite low, say 100°F.

Silicones are very widely used for the purpose described. For example, the 1962 edition of the Standard Specifications of the Highway Department of the State of New York calls for application to all exposed concrete surfaces except the underside of superstructure slabs and the tops of superstructure slabs which are to receive a separate wearing course. The material also is applicable for sealing exposed masonry in building construction. 3.


Epoxies are used for sealing areas of concrete subject to particularly severe exposures, such as bridge seats, exposed sections of piers and walls near the waterline or spray level, or precast concrete piles in the tide range. The methods of application are described in Section H.



J. SOME 5-1.



Repair of Structures


A Deteriorated Wharf

a. Description. The construction of this wharf is shown in Figure 5-11. As will be noted, it consists of a cast-in-place, reinforced-concrete


pile platform


on precast

piles. Typical views showing the deterioration are shown ures 3-6, 3-8, 3-32, and 3-33.


in Fig-

b. Types and Causes of Deterioration (1)



Deterioration of the piling consisted of spalling of the

and corrosion

El. +10, 0.

of the reinforcement,


both effects



Concrete Deck



L ero!





1at x 14 Piles

L Precast Concrete

mt tiff

11 | Grouna tt Gro





i! xh

iv, it it 1

20" Precast Concrete Sheet Piling

2-16 Rods (Typ. )

Precast Concrete Piles. Typical Reinforcement 4-1"9 corner bare plus 3/8" stirrups. Section A-A Figure






Concrete Structures:

Repairing Spalling and Disintegration


within a relatively narrow range extending from a foot or so above mean high water to a level a few inches to 2 or 3 feet below mean low water. Within this range the concrete of many piles was entirely missing, leaving only the four vertical corner rods in place. Some of these rods were severely corroded, but the majority, surprisingly enough, were in relatively good condition, showing that dark greenish-black color of adherent rust, as described in Chapter 2, Chapter




rods were


and twisted be-

cause of displacement of the deck, but, again surprisingly, most were straight and uniformly spaced, indicating that there had been no substantial settlement of the deck despite the obvious loss of support. This case is an excellent example of the ‘‘reserve of strength’’ to be found in structures, as discussed in Chapter 1, Chapter Note 1. Above the range of severe deterioration, the piles had spalled, but principally at the corners of the section. This spalling was due to corrosion of the vertical bars. At depths of immersion more than 2 or 3 feet below mean low water, the concrete was in good condition. It was concluded that the observed deterioration was due, principally, to weathering, with some damage probably contributed by abrasion due to drift and floating ice. Corrosion of the reinforcement also contributed to the spalling of the concrete cover layers, but was inferred not to have been a primary causative agent, since the deterioration of the concrete extended 6 or more inches beyond the plane of the bars. Chemical reaction also was ruled out as a primary causative agent, since the concrete below low-water level was

in good


Inspection indicated some crushing of the concrete due to overstress resulting from the reduction in pile cross section, but this condition was not widespread. (2) Cap Beams. Deterioration consisted of spalling of the sides and corners of the concrete section due to corrosion of the reinforcement. This was the result of absorption of water by the concrete, the soffits of the cap beams being less than 2 feet above mean high water and therefore subject to frequent wetting by splash or heavy wave action. The majority of the cap beams also had developed heavy vertical cracks (up to about 1/8 inch wide) due to loss of support occurring as the piles were deteriorated. (3) Deck Slab. The soffit of the deck slab was spalled because of corrosion of the reinforcement. The degree of spalling varied widely. Some bays were spalled over their entire surfaces. Others were partly spalled and partly in good condition. In all cases, the spalling did not extend much beyond the plane of the reinforcing


Deterioration, Maintenance, and Repair of Structures

bars and was limited to the area between the faces of the cap, fascia, and backwall beams.

(4) General Conditions.

The concrete was made with a hetero-

geneous gravel aggregate. Some micaceous, friable, and otherwise unsound particles were noted. In many locations, particularly on the slab soffit, there was clear evidence that the requirements for

minimum cover over the reinforcing bars had not been observed.

In some cases, the reinforcing mat apparently had been depressed flat against the bottom form. Some of the form lumber was still in


In a number of locations wood blocks and debris had been

left in the forms and were found embedded in the concrete. The concrete was generally rather porous, and the deck was built rather low to the water, the clearance of the top of the deck above mean high water being only about 5 feet. c. Repairs

(1) Piles.

soted timber,


The piles were jacketed in a permanent form of creofilled with grout.

The details are shown

in Figure

The surface of the existing piles was prepared for jacketing by chipping away all loose and disintegrated concrete. It was then cleaned by sandblast. Even the portions below water level were sandblasted, the relatively small back pressure exerted by the 2or 3-foot submergence being overcome by holding the nozzle close against the concrete. The top of the jacket was finished with a collar of pneumatically applied mortar, this material being selected because it was being used on the job for repairs to the cap beams and was, therefore, available. The collar was simply built up to the underside of the cap and the load transferred from the deck to the new jacket by bearing. The lower end of the jacket form was set to lap the undeteriorated portion of the pile about 2 feet, in order to transfer the load from the jacket back into the pile by bond. The timber form was made of 2-inch thick creosoted timber, using tongue and groove boards in order to get a tighter fit. The corners of the form were sealed with a rubber gasket about 1/8 inch thick. The bottom was sealed with sponge rubber stripping supplemented by some of the rubber gasket material. Sealing the bottom of the forms was difficult and required the services of a diver, who checked each form as it was being filled, packing the seal and tightening the clamps as required. It was necessary that the lower clamp be relatively close to the bottom of the form to aid in making the seal. Nevertheless, in general, a perfect seal was not obtained, and it was necessary to add a portion of gravel in the lower part of the form to act as a filter and impede the leakage of the grout. The gravel ended up as a partly consolidated fill

Concrete Structures:

Repairing Spalling and Disintegration


about 4 to 6 inches deep, and the form, therefore, was lengthened correspondingly to make up for loss in bond strength and load transfer from jacket to pile. The

existing reinforcement



by use of the sandblast,

and where the existing bars were bent, broken, or severely corroded, they were spliced with new bars of equal area. The pile was then wrapped with galvanized wire mesh, the forms set, and the interior of the forms flushed with a solution of potassium permanganate, rinsed, and then filled.

Figure 5-12. Case History 5-1—concrete jacket exposed by removal of the form. It may be noted that the concrete jacket is well made except at the lower end, where a leak in the bottom seal caused the grout to bleed out of the form leaving the mesh and the deteriorated, lower section of the pile exposed. The condition was corrected by making a new jacket of larger diameter around the unsatisfactory one. The importance of sealing the lower end of the form will be apparent. This usually requires the services of a diver. Note the jacket forms in place on the piles in the back-



Deterioration, Maintenance, and Repair of Structures of





mentor éehncen surface o;

form ord

op of

Limit of pile



es back



of le

fiers Gk



Fill Hose


@" mor di.

New reing:

ee ee



csti exis


Teeinf Bor



Form fe remain

in place



wfie}st-Creosoted wood form




= t my

thot pile is ares

welded wire

(Style 44/88



fog evel et