Go to Home Page           Conservationtech.com, Building Conservation Technology




NPS-cover.jpg (38563 bytes)







by Randolph Langenbach


PART 2: Conclusion










The problems encountered in developing and gaining approval of an economical upgrade scheme for the Woodrow Hotel provides graphic illustration of how strict conformance to the letter of the code has come to differ so profoundly from rational design which meets the original intent of the code. The source of the problem is the fact that the masonry infill and cladding provides the main lateral force resisting system for this type of building. In order to meet the letter of the code, under the standard code static equivalent design procedures, the new braced frames or shearwalls have to be designed to take most of the lateral earthquake generated forces off of the masonry. This is almost impossible to do. For reasons of simple mechanics of materials (the masonry is not only brittle, but very stiff), it is difficult, if not impossible because of the methods developed for evaluating the strength of masonry in code designed work in Oakland.


Masonry has great compressive strength, but in tension it exhibits very little strength. In frame and infill wall buildings, the masonry is the most rigid element in the system, and as such, the lateral forces on such a building are immediately and almost exclusively felt by the masonry. As with the re-bar in reinforced concrete, the frame in an infill wall building only begins to be subject to significant stresses when the masonry begins to crack. When the masonry does crack, the surrounding steel helps to confine it. In turn, it is the enframed masonry which continues to prevent the pin-connected steel (or the non-ductile concrete) frame from collapsing. The onset of cracking within the masonry panels is thus only the first step in the building's response to the earthquake's enormous forces, and should not be interpreted as "failure" of either the wall or of the building structure.


Professor Richard Klingner, of the University of Texas at Austin has determined from his research on the Woodrow Hotel under the National Science Foundation grant that many of the cracks in the masonry were probably already in existence before the Loma Prieta Earthquake. The stresses in the masonry sufficient to produce tension cracking (as opposed to crushing) are so low that many of the cracks now found in the building were probably caused by other lesser earthquake and wind forces over the building's 80 year history. This only serves to illustrate that it is not the existence or absence of cracks which determines a building's performance, but the ability of the system to remain stable well after cracking begins.


In an earthquake, when the first panel cracks the load is immediately shifted to another panel which similarly exceeds its elastic limit and cracks. As the forces continue, more masonry becomes cracked before the original cracked panel assumes more load. In this way, the energy of an earthquake spreads throughout a wide area of the wall. Before excessive displacements can occur, the steel frame becomes engaged with the panels of brick. When the panels do assume more load, the distortion of the frame causes the masonry to become loaded in compression, forming what is termed an "equivalent diagonal strut". The compressive strength of the masonry loaded in this way is usually far greater than the shear force required to crack the panel at the onset of the earthquake.


Another problem arises from the use of the Uniform Building Code for retrofit design of an infill frame building. The U. B. C. simply does not recognize unreinforced masonry as a cladding or infill material, because it is no longer an approved material for new construction in Seismic Zones 3 & 4. If one only uses this code, this would mean that no loads can be computed as being shared by the masonry. This would be the same as if all of the masonry were removed from the building , loaded onto pallets, and placed back into the building as dead load only. The steel frame of the Woodrow Hotel is pin connected on all floors except the 2nd floor which has some small corner braces. As was then the common practice, the lateral bracing of the structure was originally intended to be provided by the infill masonry. It can easily be understood that if the masonry were loaded on the frame simply as dead load, the bare frame would never have had the strength to remain standing under any significant lateral loading. This model certainly does not represent the actual behavior of the building in the Loma Prieta Earthquake.


To overcome this extreme case, engineers and the city of Oakland have agreed that some minimal values are appropriate for estimating the capacity of unreinforced masonry. Such values can be found in the city's guidelines for interpreting the 10% loss of lateral capacity. [xiii] This document includes a table of "tentatively recommended capacity values and stiffness characteristics for several materials " including masonry. The working stress figure of 5 to 10 psi, working stress, for unreinforced masonry has been adapted from the Uniform Code for Building Conservation, which for bearing wall buildings is an excellent alternative to the UBC. The problem is that the UCBC has not yet been developed to deal with frame and infill wall buildings. In the city's 10% loss guidelines, the figure "tentatively recommended" for the working stress capacity in shear for masonry infilling a frame is 15 psi, with an allowable story displacement of . 02". While the figure for bearing wall masonry represents a discount to less than 10% of the tested ultimate shear values for the masonry, the figure for masonry infilling the frame represents only an arbitrary minor increase from the other number. The artificially low number for bearing wall masonry is based on its brittle behavior. To retain a similarly low number for infill masonry fails to recognize that it is the strength and ductility of the system not that of the material which determines the seismic stability of the building.


The use of some figure for the capacity of masonry in shear is helpful, but the arbitrary use of such a low one means that the computations for the building show the existence of an excessive stress on the masonry walls in the engineering calculations. Also, by setting the story drift limit to . 02",the newly added structure must be extraordinarily rigid in order to meet the guidelines. This rigidity only serves to increase the forces on both the new and existing structure, making it necessary to add enormously strong and rigid elements to existing buildings. Often it becomes a situation of "chasing one's tail", because the weight of the added structure itself then becomes a load which must be resisted.


The question remains as to what number to use for the working stress of infill masonry. [xiv] This will have to be decided as part of the development of a new static equivalent code procedure. The difficulty is that since the the initial cracking of the infill masonry is expected to occur in an earthquake even after a practical upgrade, the working stress number must be high enough to reflect ultimate yield values which are above the elastic limit of the material, or the working stress discount will continue to force over design of the upgrade system. At this time there is no practical static equivalent code design methodology which addresses the unique problems of infill frame buildings.


The irony of all of this is that the underlying intent of the Uniform Building Code is to meet certain objectives of life safety. The "intent of the Code" is described in the SEAOC Blue Book as the goals of good seismic design, namely:


". . . structures designed in conformance with the provisions and principles set forth therein should, in general, be able to: 1) Resist minor earthquakes without damage;2) Resist moderate earthquakes without structural damage, but with some non-structural damage;3) Resist major earthquakes, of the intensity of severity of the strongest experienced in California, without collapse, but with some structural as well as non-structural damage. "


The literal application of the Uniform Building Code fails to address the actual problems which have been manifested in previous earthquakes. While there has never (to my knowledge) been a collapse in and earthquake of a frame and infill wall structure in the United States, there have been many examples of falling debris. The installation of new extremely stiff elements into these buildings may possibly be tantamount to "adding shakers" to the building by increasing earthquake loads on certain existing elements. [xv] While a UCB conforming retrofit may reduce the in-plane stresses on the masonry walls, the out-of-plane shock on those same walls will be increased. By reducing the inelastic response and shortening the period of the structure, the forces on the infill walls, architectural details, parapets and etc. are increased. This will only serve to make it more likely that pieces will fall off of the building into the street.




In a recent article in Earthquake Spectra, Strengthening Buildings to a Life Safety Criterion, Englekirk and Sabol[xvi] make the case for an alternative approach to the retrofit of reinforced concrete buildings from the 1950's and 60's (which I shall call "bare concrete frame buildings" to distinguish them from infill frame structures) by designing to prevent collapse rather than to reduce damage. By adding ductile elements which come into effect after the initial inelastic yielding of the existing structural system, a more economical life safety alternative can be achieved. In describing this system, Englekirk and Sabol emphasize the fact that, while collapse can be prevented by this approach, "a structure so strengthened may be damaged beyond economic repair during a major earthquake. "


For infill wall buildings, this same approach is even more appropriate and necessary, but there is an important difference: the likelihood of irreparable damage is less than for the bare concrete frame buildings. While a concrete frame is difficult or impossible to repair following inelastic yielding (as for example with the damaged freeways), cracked masonry can be successfully repaired. While the masonry contributes to the lateral force resisting system, cracks in it do not compromise the building's primary vertical load carrying system. With infill frame buildings, it is the masonry which takes the first hit, rather than the reinforced concrete frame itself. It is thus the masonry which, by behaving inelastically and providing damping, can serve to protect the concrete frame.


When adapting this approach to frame and infill wall buildings from that proposed for brittle concrete structures, there are some differences which are worth noting as an introduction to the retrofit alternatives described below. For the bare concrete frame buildings, the initial stiffness of the frame would usually be less than the initial stiffness of the masonry infill. Following the onset of cracking, there is a rapid drop off in the resistance of bare concrete structures lacking ductile design, but the infill structures are likely to exhibit a "second wind" as the masonry develops the "equivalent diagonal compression strut" within the frame. While masonry in shear may only be capable of resisting 50 to 150 psi, masonry in compression is capable of stresses in the range of 1,000 to 3,000 psi. Where there are windows, this compression strut can brace the frame at mid-story height causing the columns to begin to bend, bringing additional elastic (and in extreme shaking, ductile) deflection into the system. [xvii]


A dramatic example of the difference between the seismic performance of infill wall buildings and of bare concrete frame buildings can be found in Oakland by comparing the damage to the ca 1950's Bermuda Building to that sustained by the infill wall concrete frame buildings. The Bermuda Building, with its ribbon windows extending around the building, lacked any resistance except that provided by the structural frame itself. While the building did not collapse, the frame was so severely damaged that the building will most likely have to be demolished. None of the much earlier infill wall buildings suffered major or irreparable damage to their structural frames.


Following Englekirk and Sabol's argument, the retrofit of an infill frame building should be designed to provide added ductility in order to bring the entire structure into conformance with the objectives of the code in preventing collapse, rather than by attempting to change the force level required for the onset of damage.


"The design emphasis should be directed toward the attainment of maximum ductility. . . A specific yield level criterion is not required and its inclusion in the design process will only tend to return us to the original dilemma, an economically infeasible solution. "[xviii]


Englekirk and Sabol identify the lack of provision for this approach in the code as one of its failings in retrofit work:


"The code procedure does not lend itself to the separation of the damage level criterion from collapse level criterion and, as a consequence, does not allow the flexibility usually required in rehabilitation work. This is one reason why the statement "bringing it up to Code" is essentially meaningless when one refers to seismic rehabilitation. [xix]


Naturally it is hard to generalize about the complexity of this interaction between the behavior of the masonry and that of the overall building system. Each building will have its differences, but what is apparent is that attempts to avoid the cracking of the masonry by transferring the loads to a new retrofitted structure are doomed to failure.

 The object of the effort should be redirected towards providing a backup system that will:

  1. prevent any possibility of collapse should any of the masonry fall out of the frame, or an existing "soft story" be overwhelmed,

  2. add to the total ductility of the structure, and

  3. prevent the larger scale displacement and damage (ie: crushing, rather than hairline cracking) of the masonry which could lead to falling debris hazards and spoil the architectural features of the building.[xx]

Ductility and damping can be added to a structure in a number of different ways, such as through the installation of eccentric braced frames, or even plywood clad diaphragms and crosswalls. One of the advantages of the dampers over systems which rely on providing ductility to the structure by causing stress beyond the elastic limit in specific structural elements (such as eccentric braced frames) is that the damping is available from the very onset of earthquake shaking. Strong members do not have to be forced to yield first. While this attribute is helpful, it is not essential. Equally valid earthquake hazard mitigation can use any technique which results in added damping in major earthquake shaking (see graph above).

While the damper scheme, by virtue of the existence of the open storefronts, can be used to reduce the impact of the earthquake forces on the upper facade of the Woodrow, it does not eliminate it. On returning to address this facade, and to address the rear facades of the building where the dampers cannot be installed, the issue of what to do about the masonry is still there. Analysis of this masonry based on the elastic limits of the material continue to show overstress. As stated before, the intent of the code allows for structural damage to occur in major earthquakes. In infill wall buildings, this structural damage must involve cracking of the masonry. If one accepts the design methodology proposed by Englekirk and Sabol, then guidelines must be developed for dealing with the cracking of masonry elements on the facade and interior walls of these buildings.


The allowance of cracking then raises the inevitable possibility of falling debris hazards. On the interior of a building, this can usually be addressed by jacketing those hollow clay tile walls which line escape routes such as corridors and stairways. This can be achieved by something as simple as a layer of expanded lath and plaster. If walls are tall and vulnerable to out of plane collapse, some further support may be indicated. What is to be avoided is any proposal that they be removed and not replaced as they were in the Oakland Hotel.

On the exterior the issue is more problematic. Jacketing of masonry would destroy the architectural finishes of the building and would also be extremely costly. What is critical here is the reliance on a retrofit analysis and design which limits the story drift to an amount which would prevent any major degradation of the masonry, and the resulting danger of it being shed into the street below. This reinforces the importance of using dynamic analysis in the design of the upgrade, and also of establishing a back up system which restrains the drift to acceptable limits if that analysis at full earthquake loads indicates that drifts would be excessive.

Finally, the possibility that some smaller pieces could fall to the street remains. This is impossible to avoid altogether, even if the building is retrofitted to conform to the UBC. In the last earthquake, small pieces of masonry fell in San Francisco and Oakland, yet no one was killed or injured by these pieces. One example in San Francisco demonstrates the problem: the Southern Pacific Building on Market Street at Embarcadero lost the corners of many window sills, as well as column capitals at the entrance. This shows the extraordinary stresses which developed from the flexure of the piers between the windows, which pinched off the corners of the terra cotta window sills. No affordable strengthening of the building can avoid this.

Do we therefore tear the building down? Do we strip every building of any projecting architectural ornament? While very few people were hit by debris in the recent earthquake, the possibility remains that they could, even though few people remain standing immediately under walls during an earthquake. Despite the fact that the earthquake struck at about 5 PM, a time when the maximum number of people would be on the sidewalks, it is noteworthy that there were few injuries and no deaths from falling debris off of frame buildings, despite the fact that there was a considerable amount of broken glass, terra cotta and masonry which fell to the sidewalks in both San Francisco and Oakland. Since the maximum danger is thus to those who are running out of a building, then this could be resolved by installing a sturdy canopy over the entrance if a specific danger is determined to exist. Further measures can include Hilti anchors, wall plates, window sill reinforcement, or other building specific designs to reduce the falling debris hazard. This approach is certainly cheaper and even potentially more effective than the restructuring of the whole building. The overall point is that only when we focus attention on the actual realistic problems which have been demonstrated by experience to exist with this or any building type, can we develop specific creative solutions.


Each time that earthquake problems with existing buildings are addressed, the issue of liability emerges. "Liability" has become the controlling factor in much of what we do. Public safety is of course a noble objective, but the issue of liability in recent years has taken on a life of its own. No longer is the question "how can we avoid a hazard?" It has been replaced by "how can we avoid a potential lawsuit?" There is a difference. Engineering in response to the former question can lead to creative solutions, whereas in response only to the second, creativity is stifled.

An engineer is not legally obligated to be original, or to find the least cost solution, or to protect historic fabric. His or her obligation is to protect the public. If the building is heavily damaged in an earthquake, the engineer is vulnerable regardless if he saved money, or preserved important historic fabric. As one engineer stated in a letter to another:

We do not have the same immunity from liability presently enjoyed by cities and counties...It would be disastrous to our profession if strengthened URM buildings were to perform badly during future earthquakes. Remember, 60,000 buildings state wide represents about five to ten billion dollars of direct construction costs. Many lawsuits will follow [the next earthquake], based upon 100% hindsight![xxi]

In the event of damage, there is little to protect the engineer from lawsuits. While adherence to the code helps in an engineer's defence, not even that fully protects the engineer from liability. If he or she departs from the letter of the code in pursuit of a creative low cost solution, the vulnerability to lawsuits is potentially greater. With the upgrade of existing buildings, the liability situation is magnified considerably over that for new buildings because the engineer's work becomes intermingled with the work of the original designer. If problems should emerge in an earthquake, it is the retrofit engineer who must face them. The original designer is "judgment proof." For buildings with masonry infill, where cracking even at low levels of shaking, is inevitable, the problem is magnified still further. Even while the code assumes that structural damage could occur in a major earthquake even in new buildings, many engineers have advocated completely new lateral force resisting systems for infill frame buildings because of the liability which they assume when they touch a building where the existing lateral force resisting system is masonry.

The problems faced by engineers as well as the city officials in both analyzing and developing original solutions for infill frame buildings is best illustrated by the Oakland City Hall. Since the earthquake, more than eight different engineering firms or university faculty members have been employed, with at least two peer reviews, at the cost of over $1 million in fees, in developing the design for the retrofit of this building. The cost of the proposed retrofit exceeds $50 million.


On one level, design professionals must realize that their work always involves some risk, and that professionalism must also include a responsibility to balance the needs of the budget, of historic preservation objectives, with liability issues. In addition, this issue must be addressed on the public policy level. Where the codes don't work for existing buildings, it is the codes which must be changed, not the buildings. (This has been done for bearing wall buildings with the creation of the UCBC, Appendix, chapter 1, but this cannot be used for frame buildings.) Until a new suitable code is developed, a crisis exists as long as either engineering professionals, or public officials, are unwilling to accept alternatives to the strict letter of the prevailing code for new buildings.

If the City of Oakland had not established a policy of requiring damaged buildings to be upgraded as well as repaired, the debate over what is necessary would have been postponed until the implementation of a hazard mitigation ordinance, such as that now being drafted under the State URM Law SB 547. The City has based its policy on the seemingly reasonable claim that structures damaged in the Loma Prieta Earthquake must be the most vulnerable to life threatening damage in the "big one" expected at any time soon. If by only repairing their structures to their original, pre-earthquake strength, the city has sound evidence that a public life-safety hazard exists, then such a mitigation ordinance would be justified.

There are, however, three questions which should be raised in evaluating this position:

(1) Is it true that the so called "big one" will be so dramatically more severe to these same buildings than was the Loma Prieta?

(2) Is it likely that the same buildings which showed damage in the Loma Prieta Earthquake would be the most vulnerable to collapse, or other life threatening damage in the "big one?"

(3) Are there other more cost effective ways of addressing the structural deficiencies of the damaged buildings besides requiring them to meet the provisions of the Uniform Building Code, a code which was never intended to deal with the specific conditions encountered in the retrofit of existing buildings with archaic structural systems?

In the two years following the Loma Prieta Earthquake, it has often been repeated that this was not the "Big One." Engineers have been quoted as saying "if the shaking had continued for another 15 seconds, then certain infill frame buildings] may have collapsed." The entire Repairs Ordinance program in the City of Oakland is predicated on the belief that it is exactly those buildings which suffered damage in the Loma Prieta earthquake which must be most vulnerable to collapse or threat to life safety in the next earthquake. Sadly, and perhaps tragically, this is most probably far from the truth. Even though these buildings were damaged in the Loma Prieta Earthquake, this does not make them the prime life-safety risks in a future major earthquake on the Hayward Fault.

This is true for two reasons: (1) A more local earthquake is not as likely to resonate with the buildings of the same height and flexibility as the distant Loma Prieta event, and (2) within the period of about 1.5 seconds, the Loma Prieta earthquake actually exceeds the 1988 UBC design level (see Loma Prieta Spectra below). In other words, the Loma Prieta earthquake meets the definition of being the "big one" in respect to its impact on these buildings to a degree which is not likely to be either matched or exceeded in even a major Hayward Fault earthquake, because of the likely difference in the period of the shaking.

Cus10134.jpg (54956 bytes)

While one might have hoped that the earthquake would have been beneficial in focusing attention on the broad problem of earthquake safety, most of the attention has been focused solely on the damaged buildings and others of the same type. As a result, owners of masonry buildings have been squeezed between the tremendous loss of market value and the demands to spend money on strengthening, with very little offered in public funds. For most of the damaged buildings, there is just simply not the equity left on which an owner can obtain a loan to do the work. And yet, these damaged buildings, constructed to an engineering design and technology before the current code which proved themselves in earthquake forces which could, by definition, have caused inelastic yielding in a code designed structure. In addition, they are constructed of a system where the inelastic yielding was forced to occur in the masonry infill, which is relatively easy to repair to pre-earthquake strength, rather than in the underlying frame (as it did in the much more recently built Bermuda Building) where it could permanently jeopardize the structural integrity of the building.

As long as unreinforced masonry, whether used as infill or for bearing walls, is treated as having minimal strength before it is cracked, and having basically no strength in shear after it is cracked, design solutions will not be economical, and unless reasonably economical solutions are found, many valuable buildings will be lost in cities such as Oakland. The use of alternatives such as the dampers, or of other "soft" approaches in retrofit design promise not only a chance of affordability, but also of better conservation of historic building fabric.

The use of alternative technologies and designs which rely in part on the lateral resistance of the masonry require a major shift in the City of Oakland's conception of what the masonry in the infill frame buildings is capable of resisting. Sigmund Freeman, engineering consultant to the City of Oakland, once stated: "A masonry building is like a tea cup - when it's cracked, it's broken." While a cracked teacup may leak, a building is not called upon to hold water. The notion that cracks in masonry constitute "failure" is a common misperception which has emerged in the engineering profession mainly from the analysis of reinforced concrete, where cracks may signify vulnerability to collapse. Until this error is corrected, many important and substantial historic buildings are at risk.


  1. Mitigation ordinances should address the problems actually seen to occur in earthquakes related to the building types in question, not every conceivable abstract risk. For example, it is less costly and more direct to address the danger of falling debris directly, rather than focusing all attention on the danger of collapse, which has never occurred for infill frame buildings in the United States. For example, one risk totally ignored by the codes which had a profoundly damaging effect in Oakland as well as Mexico City, was the pounding of adjacent structures. It would be much more cost effective to implement an ordinance which directly addresses the danger of pounding, than it is to require full code upgrades on buildings where damage was caused by pounding. It would also provide for a higher degree of life safety.

  2. Mitigation ordinances should also address the specific parts of buildings seen to be at risk, rather than force the complete upgrade of entire structures which have stood the test of time. The requirement of a complete building retrofit of buildings damaged in an earthquake should only be done if the cause of damage is clearly a product of the inadequacy of the whole system and not just a portion of it. For example, for infill wall buildings, directly addressing the potential problem of falling debris hazards with attachments and canopies would be more cost effective than requiring full internal structural upgrades to current code.

  3. Retrofit design should avoid the removal of some heavy and stiff non-structural materials from existing buildings, such as the interior terra cotta walls, when as a result, the remaining stiff materials, such as the cladding, will take proportionally more load as a consequence. (The fact that the weight is reduced may be misleading.) The Oakland Hotel provides a good example of this kind of problem.

  4. Until a specially developed static equivalent code is available for infill frame buildings, the use of dynamic analysis based design should be encouraged by specific reference in city ordinances. Proper design for these buildings should focus on limitation of drift to avoid excessive damage to the masonry, but it is essential that this drift limit not be so stringent as to make the retrofit unrealistic. Initial cracking of the masonry in shear or in flexure must be accepted as part of the design. The guidelines for this procedure have been recently issued by the Structural Engineers Association of Southern California. (Appendix)

  5. The retrofit of infill frame buildings must assume that some cracking of the masonry will take place in an earthquake. By designing to limit the story drift of the structure, unacceptable levels of panel degradation can be avoided. The prevention of cracks is infeasible, and would impose a higher standard of performance on older buildings than that which exists for new buildings. City officials, owners and the public need to be educated to expect a certain level of cracking as part of the safe and expected response of these buildings to earthquake shaking, and that the repair of such cracks would be necessary following an earthquake.

  6. It is also important that Engineers use research models and results for justifying their designs, and that building officials accept this until a specific code procedure is developed. There has been extensive research work completed world wide on the subject of the performance of masonry.

  7. A solution to the problem of professional liability needs to be developed which specifically deals with the problems faced by professionals working on existing buildings. (City and county governments have been granted immunity from liability by State law.)[xxii] > >Since the performance of existing buildings cannot be fully predicted, and when that performance cannot be fully dependent on the retrofit engineer or architect's work, then there needs to be some way to limit that professional's liability in order to encourage affordable upgrade solutions. Hazard mitigation is only achieved if the cost of upgrading is made reasonable enough to be widely carried out, rather than so costly as to destroy the economy of the building at risk. It is also essential that this problem be addressed for historic structures, where the choice is between limiting damage to historic fabric versus destroying that fabric during the upgrade (or worse, causing the demolition of the building altogether.) A state wide system of insurance, which serves to spread the risk, may be an essential part of the solution to this problem.

  8. City and state governments in earthquake areas should develop a post-earthquake inspection, evaluation, and hazard mitigation program designed not only to prevent hasty demolitions, but also the unnecessary closure of buildings. In Oakland, where immediate demolition was not the problem it was in Santa Cruz, the experience following the Loma Prieta Earthquake dramatically illustrates the devastating economic and social impact of long term closure of earthquake damaged buildings. Most of the now vandalized and vacant downtown buildings, including the Woodrow Hotel, could have been quickly stabilized and safely kept in service, at least on a limited basis until permanent repairs could be carried out. Long term closure guarantied the ruination of the downtown economy, and failed to bring about the redevelopment envisioned by the starry-eyed planners in the weeks immediately following the earthquake.[xxiii]

There is now a crisis in downtown Oakland. Few of the major buildings damaged by the earthquake have been reopened, and the City is preparing an ordinance to fine owners if they neither demolish their buildings or upgrade them. Such a proposal only further serves to obscure the fact that the crisis has been brought on in part by the lack of knowledge of the behavior of the infill frame buildings, combined with the years of misinformation about their risk of collapse by engineers who analytically separated the masonry performance from their performance as complete structural systems.[xxiv] At risk is nothing less than the long term economic health of the downtown. This problem gets to the very core of the purpose of historic preservation because it extends beyond the often arbitrary boundaries of already designated landmarks to encompass all older buildings -- the substantial stock of buildings with hundreds of thousands of square feet of office space on which the life and health of the downtown of most American cities still depends.


[xiii] Memorandum from Sig Freeman, Wiss, Janney, and Elstner Associates to Joe Wong, City of Oakland, March 13, 1990, rev. April 4, 1990.

[xiv] For example, in the Case of Oakland City Hall, even with the proposed base isolation retrofit for the building, computed stresses on the exterior masonry are reported to be in the range of 200psi (ultimate) in some places.

[xv] This observation was made in August, 1991 by James Hill, Engineer, Long Beach, in reference to the seismic retrofit of the Oakland Medical Building in Oakland.

[xvi] R. E. Englekirk & T. A. Sabol, Strengthening Buildings to a Life Safety Criterion, Earthquake Spectra, Vol. 7, No. 1, 1991.

[xvii] This should not be confused with the situation which was frequently observed in Mexico City where the existence of partially infilled frames caused short column failures. In the Mexico examples, the windows where this occurred were often ribbon windows extending from column to column of a small enough vertical dimension to prevent the columns from bending. In the early 20th Century United States examples, the windows are taller, located in the center of the bays, and usually surrounded with masonry. If yielding does occur, the interaction of the masonry around the window with the bending frame is much less likely to lead to a short column failure of the columns.

[xviii] Ibid, p86.

[xix] Ibid, p84.

[xx] It is important to note that the hairline cracking most often seen on interior and exterior walls following an earthquake is not the same as displacement of the masonry. As the masonry cracks, the loads shift to other panels. It is only when all the panels crack that the frame becomes engaged, and the masonry begins to be shifted. It is much less costly to engineer a retrofit which can limit the significant deformation of the masonry, than it is to avoid the onset of cracking and minor movement in the masonry.

[xxi]Letter from Stanley Mendes to Nels Roselund, May 20, 1989. Quoted by permission.

[xxii]cf California Health and Safety Code Sect. 19167 (from SB 445)

[xxiii]In several newspaper articles, including one in the New York Times, planners in Oakland, and at the University of California, Berkeley, were reported as predicting that the damage would spur Oakland's renaissance. In the two years since, this has been proved wrong.

[xxiv]For example, A report by David Messinger & Associates prepared in 1978 found the building seriously deficient in lateral resistance based on an erroneous analysis of the structure based on the Uniform Building Code. In the intervening ten years, the occupants became increasingly frightened of the possibility of collapse. When the earthquake occurred, this only served to increase the fear of the building, leading to its near demolition, and now to an extraordinarily expensive proposed retrofit.



Website by

© Randolph Langenbach

M-Arch (Harvard), Dipl.Conservation (York, England)

FAX: 801-730-1616

Writings and photographs found on this website is ©Randolph Langenbach and other authors. For use permission, write to the address above.

Please send your reactions, comments or questions regarding this website to the Email address above