In two parts

GO TO PART 2

 

---

 

PART 1:

MASONRY INFILL FRAME BUILDINGS IN EARTHQUAKE AREAS-EARTHQUAKE DAMAGE AND THE BUILDING CODES

 

On the day following the Loma Prieta Earthquake, a visit to downtown Oakland and San Francisco seemed anticlimactic. Except in places like the San Francisco Marina and the collapsed freeway in West Oakland, the effect of the earthquake on the cities was subtle - a pile of broken glass here, a thin crack in the masonry above, a fragment of broken masonry there. A few walls collapsed, but when compared to the views of Mexico City in 1985 or those of San Francisco in 1906, the damage did not seem the least bit catastrophic, despite the news reports to the contrary. In fact, you could even feel the ironic disappointment in the comments of engineers and researchers who had rushed to study what they had expected would be wide spread devastation.

 

What has happened during the two years since has changed all of that. While the scene in downtown Oakland on October 18th was unduly quiet and unimpressive, the aftermath of the Loma Prieta earthquake has indeed been catastrophic for the city. It is now time to begin to explore some of the reasons why.

 

The experience in Oakland is significant in a number of ways:

 

1.  The effect of this earthquake on the bay area were focused on only a few areas where the soil responded to the shaking, one of which included downtown Oakland.

2.  In Oakland, the most significant damage was concentrated on the major downtown steel and concrete frame buildings constructed in the early part of the 20th century, instead of those hitherto thought to be far more vulnerable, the unreinforced masonry bearing wall buildings.

3.  The economy of the city, especially of the downtown, was not strong enough before the earthquake to provide the reinvestment necessary to repair and strengthen the damaged buildings after the earthquake, or to replace them.

Several of the damaged buildings are historic monuments which define the character of downtown Oakland, whose loss would remove a significant part of what historic character the city has.

 

4.  This is the first time that any city in the United States has been subjected, following an earthquake, to the closure of such a large percentage of buildings of frame and infill wall construction. Because of their size and location, the temporary or permanent loss of such a significant number of these buildings has had a major negative impact on the city's economy.

 

The question of why the City of Oakland suffered this loss is not, however, a simple one. The fact that these buildings were damaged in the earthquake is evident, but the necessity for them to have remained closed for such an extended period is not so obvious. The main issue has surrounded the fact that Oakland adopted an Ordinance which requires these buildings be brought into conformance with a slightly modified version of the 1988 Uniform Building Code, before they can be reoccupied. Owners were prevented by the Ordinance from simply repairing their buildings to pre-earthquake strength. To reoccupy them, a full code upgrade is required as well. As a result of this Ordinance, any structure which has exhibited significant cracking of the masonry has been prevented from being repaired unless it is retrofitted to meet the standards of the 1988 UBC.

 

THE DAMAGED BUILDING ORDINANCE AND THE EFFECTS OF THE BUILDING CODES ON THE PRESERVATION OF DAMAGED BUILDINGS

 

The Damaged Buildings Ordinance which Oakland adopted following the earthquake specified that all buildings:

 

"shall be made to substantially comply with current code. . . if as a result of the earthquake damage, the pre-earthquake lateral capacity of the structure has been reduced by 10% or more. . . "

 

The city issued guidelines on how to determine the 10% loss of capacity. [i] When the City enacted the Damaged Buildings Repairs Ordinance,[ii] it seemed to most people to be eminently reasonable that these buildings should be upgraded. There have been instances in other earthquakes around the world when buildings, damaged in earlier earthquakes have collapsed. One recent example claimed the largest number of lives in the San Salvador earthquake of 1986.

 

However, it eventually has become apparent that the UBC is incompatible with the infill wall masonry structural system. Because of the huge costs of bringing these buildings into conformance with this Code, even with the use of the variance which allows a base shear of 75% of code, most of the building owners have balked, and some even have been forced to abandon their property. By forcing a major upgrade at the time of the repair, the City has placed what has turned out to be an insurmountable burden on the owners of these buildings, as the cash flow from the rental income, the capital resources for such major improvements are simply not available. In addition, because of this loss of use the equity on many properties has been wiped out, because the underlying land values cannot support the pre-earthquake values which were based on income. Thus, unable to get loans and without financial assistance from the city, many of the owners of damaged buildings are not in a position where they can repair their structures. As a result of this impasse, few buildings have yet to be repaired.

 

Of the approximately 30 frame buildings closed because of earthquake damage, by the second anniversary of that event only one has been reopened. [iii] Since the original objective of the Ordinance was to improve the level of public safety, it can be seen that this objective has not been achieved. While the delays may not all be attributable to the Ordinance, there is no question that the Ordinance has played a significant role.

 

The problems which have emerged with the codes are due to the existence of the masonry in these buildings. In discussing the example of one building, the Woodrow Hotel, one is forced to ask whether masonry deserves the "bad rap" that it is getting in California at this time in history, especially in the case of its use as the infill and cladding on early modern infill frame structures.

 

 

FRAME AND INFILL WALL BUILDINGS: AN HISTORIC BUILDING TYPE

 

Infill frame buildings are not a specialized or unusual building type. Most of the early twentieth century mid-rise building in the central business districts are of this type. With the invention of the "skyscraper" in Chicago during the 1880's, the shift from bearing wall masonry construction to the use of a steel (and, later, concrete) frame meant that the vertical load carrying structure of these larger commercial buildings was shifted from massive bearing walls to a light weight network of steel columns and beams. But despite the folklore to the contrary, this did not mean that all of the masonry ceased to play a significant structural role. For many frame buildings constructed during the first half of the 20th century, the masonry cladding was consciously relied on to stabilize and stiffen the structure against lateral loads.

 

For taller buildings such as the Woolworth Tower in New York City, the steel frame was braced with either diagonal struts or portal braces (the precursor to the moment frame). For smaller buildings, little of this was necessary if the masonry, which by necessity had to bear upon the frame, was also built up to fill the frame, except for the openings for windows and doors. The view of New York City's Flatiron Building under construction (figure 1), and of 343 Sansome Street in San Francisco under renovation (figure 2), shows how the masonry cladding forms a solid wall bearing on the frame.

In recent years, construction practice and codes have changed. The cladding on midrise and highrise buildings is now made structurally independent from the frame, so that it is incapable of being loaded by lateral forces acting on the building. This practice can be seen on many contemporary buildings by the flexible sealant joints which are visible on the facades. This shift in construction practice can also be seen when one compares the frame of an early steel building, when revealed by demolition (figure 3), with that of a new one of the same height under construction. While the new building is most likely much lighter in weight, the older one will usually be found to have the much lighter frame.

 

Does this lighter frame mean that the older buildings will be more likely to collapse in earthquakes? While it is risky to make blanket statements about future earthquake risk, history has shown that collapse of the frame and infill wall buildings has not been a problem in earthquakes in the United States![iv] Why then is this type of construction perceived to be a potential problem in Oakland?

 

The answer lies in a combination of reality and public perception. The reality is that because of its greater than 1 second period (unusual for the Bay Area), the earthquake was of a character which focused the most intense shaking on midrise buildings. The perception is that, had the earthquake been the "big one," that is, a magnitude8. 0 on the Hayward or San Andreas Faults, these same buildings may have collapsed. This seemed credible to many, including the engineers involved in the development of city policy, because of the extent of the observed damage which did result from a 7. 1earthquake with its epicenter over 75 miles distant. In addition, for those who were in the buildings at that time of the earthquake, the shaking was particularly frightening. It is this human experience which binds reality together with perceptions. Some who were in the City Hall for example said that they "will never enter that building again. "Because of the rarity and unpredictable nature of earthquakes, the belief that collapse almost happened is something which is very difficult to refute. [v]

 

This earthquake has taught a major lesson: that people tend to blame the buildings for the trauma of the experience of the earthquake itself. Many were genuinely frightened, which is, of course, a reasonable reaction. The effect of this experience, however, on the making and enforcement of public policy has been profound and lasting, and to my mind, excessive. Once people have come to identify particular buildings as dangerous, it is very hard to change their minds even though the risk may be relatively quite small.

 

 

FRAME AND INFILL WALL BUILDINGS IN OAKLAND:

THE DAMAGE CAUSED BY THE LOMA PRIETA EARTHQUAKE

 

The damage to be found in frame with masonry infill buildings in Oakland was almost completely limited to cracks in the masonry infill and cladding. On the interiors of the building, where partition walls were frequently constructed of hollow clay tile with a plaster finish, cracks were common. In some cases, individual hollow clay tile blocks were sheared off, causing small sections of the wall to collapse. It was very rare, however, for large sections of the interior hollow clay tile walls to fall, or for walls literally to tip over out of plane. The folklore that terra cotta "explodes" in earthquakes was not born out.

 

On the exteriors, the cracks are still visible in the masonry cladding on many downtown buildings. In some cases, pieces of masonry fell to the sidewalk, although no one was struck by falling masonry. Of the masonry falling off of the buildings, the most frequent occurrence was the shearing off of decorative terra cotta.

 

The Oakland City Hall (figure 4) had cracks on the interior, with the most severe damage limited to the steel frame and terra cotta clad clock tower. This multi-story tower swayed on its base causing the near failure of some supporting steel beams. Across the street, the historic flat-iron shaped Broadway Building (figure 5) suffered damage to its exterior skin of terra cotta when it swayed against its neighbors, causing it to bend at the 3rd and 4th story levels (figure 6). The architect of the Broadway Building, Llewellyn B. Dutton, had worked for Danial Burnham, the designer of Flatiron Building in New York (figure 1), now a National Landmark. The Broadway Building has since been the subject of a protracted preservation battle with opposing sides fighting over the interpretation of the superficial engineering data which has been produced to date. None of the engineering reports addressed what may have been the principle cause of the damage - the pounding. The solution for this problem is very different than arguments over the structural competence of the building on its own.

For many of the damaged buildings, the pounding with adjacent structures appeared to be a principal factor, yet this is not addressed in any of the codes. While it is conceivable that damage to these buildings would have occurred in any case, pounding often caused the greatest life safety risk by focusing the damage in one area high up on the building. On one building the stairway was caused to collapse, and a large section of exterior wall was almost dislodged onto a building below by pounding.

 

Many other buildings suffered similar damage. In two buildings, including that mentioned above, the interior stairways were partially destroyed, posing a severe risk to the occupants, some of whom were stranded. Rarely, however, was there damage be found to the underlying structural frames. Columns damaged on two concrete infill frame buildings caused by short column shear failure are believed to be the only examples of frame damage. In no case did any building suffer the collapse of any portion of its load bearing structure.

 

In one case, the Oakland Hotel (figure 7 and 8), so much of one facade fell that all of the remaining bricks had to be removed. This one cladding and infill failure is notable because the Hotel Oakland had been partially strengthened during its conversion to housing for the elderly. This provided an anomaly. The only building to lose large pieces of its upper story masonry was the only damaged building which previously had been strengthened (although not fully to code). The building as a whole did not have to be evacuated. [vi]

 

In discussing the damage on the west wing of the Oakland Hotel with Loring Wyllie of Degenkolb Engineers, the designers of the retrofit, an interesting fact emerged. When the hotel was constructed, the partitions were all made of hollow clay tile. During the retrofit, this was all removed. While the damage to the facade could be explained by the fact that the new shear wall in the damaged wing was narrower than the one in the other wing, this did not explain why none of the other infill frame buildings lacking any new shear walls were damaged on the exterior to the degree that this one was. Perhaps the lack of the hollow clay tile partitions holds the clue. Whereas these partitions would have contributed to the lateral resistance of the building, the new stud and plaster board partitions could not. Without the stiff and brittle tile partitions, the full earthquake force was shared between the new shear wall and the exterior brick facade. Because of its relative lateral flexibility, the shear wall could not experience much lateral force from the earthquake until the brick infill facade was shaken almost completely off of the frame. [vii]

While this explanation can only be presented here as an hypothesis, it does illustrate one of the principal dilemmas of historic preservation retrofit design. Often both architects and engineers are quick to specify removal of the hollow clay tile partitions, especially if they do not support interior decoration of historical value. While addressing the fear that falling pieces may injure an occupant, such removal can have consequences to the performance of the building as a whole that the engineer may not have anticipated. While the bare frame, devoid of its terra cotta infill, is much easier for an engineer to analyze, it may be the post-elastic behavior of redundant elements in the building, like that of the terra cotta, which really determines the safety of the masonry on the facade, and thus of the fate of people standing on the sidewalk below.

 

This facade failure of a partially upgraded building poses a dilemma: Is a more stringent structural upgrade requirement sufficient to take the loads off of the masonry justified, or does the comparatively low damage found in all of the other buildings underscore the need for a greater understanding and respect for their original structural integrity? To neglect the strength and ductility already inherent in the infill frame buildings risks retrofit designs which require them to be either gutted and rebuilt at great expense. If this remains the only alternative, most historic buildings will be condemned and torn down in the name of seismic safety.

---

 

PART 2:

THE WOODROW HOTEL - THE SEARCH FOR AN AFFORDABLE ALTERNATIVE ENGINEERING SOLUTION

 

 

At the time of the earthquake, the Woodrow Hotel (figure 9) was a low rent single room occupancy (SRO) hotel with 72 rooms. It is a 7 story steel frame building with brick infill exterior walls and timber floors. When constructed in 1912, the Woodrow was an attractive small hotel similar to many constructed in San Francisco and the Bay Area at the time. While of marginal historic significance in its own right, the building is a good example of a period building type. It contributes to the architectural character of its neighborhood, which has a number of similar hotels constructed at about the same period. It is a good subject of study for the seismic retrofit of historic buildings, because it shares many characteristics with other significant historic buildings with similar vulnerability to earthquake damage throughout the United States and in other countries as well.

Over the course of the last year and a half, two engineering designs for the repair and retrofit of the Woodrow Hotel have been carried through to working drawings, a third alternative has been proposed, and the building has been the subject of a lengthy National Science Foundation research project. It is not yet been determined which retrofit approach will be constructed. However, what has transpired so far does provide insight into both the opportunities for the use of a significant alternative approach to seismic retrofit, and the problems which the pursuit of such an idea can encounter along the way.

THE EFFECTS OF THE LOMA PRIETA EARTHQUAKE

 

The Woodrow Hotel was damaged by cracking of the brick masonry infill on the rear of the building (figure 10). The front facades were undamaged, despite the existence of much larger windows than found on the rear (figure 11). At the time of the earthquake, the building was occupied. At first, the post earthquake inspectors said that occupancy could continue. One week after the earthquake, a new team of inspectors "red tagged" the property and evacuated the building.

 

For many of the low income residents, this meant a move into the street. In fact, many of the residents left their belongings behind only to later find that they had been stolen and vandalized. Eventually, while the earthquake had caused damage which a good mason could have repaired quickly and relatively inexpensively, vandals caused over half a million dollars of damage to the building and its contents. Most of the occupants had thought that the closure was going to be temporary, as the damage did not seem severe and all of the interiors were intact. The elevator and all of the plumbing continued to work. Except for the cracks on the back wall, nothing appeared to be wrong with the building.

 

The Woodrow Hotel was constructed in 1912, the same year as the impressive new City Hall located only 3 blocks up the street, the first high-rise city hall in the country. Like the City Hall, the Woodrow is a steel frame building, but unlike the City Hall, its floor and partitions are all constructed of timber. The steel frame is comparatively light (figure 3 shows a cross section of a building of similar construction to the Woodrow). [viii] The only braces are small corner braces only at the level of the 2nd floor. The rest of the frame is effectively only pin connected: the beams rest on riveted seats, or are connected only at the web.

 

The exterior walls are brick, 2 wythes thick. On the street front, the 2 wythes are fully contained within the steel frame, with a 3rd wythe forming pilasters around the columns (figure12). On the rear facades above the second floor, only one wythe lies within the frame, with the other placed outside the frame to form a continuous weather surface. This outer wythe is connected to the inside wythe with the bond bricks of the common bond (every 6th course). At the first floor level in the rear facades, the masonry is 3 wythes thick, with two wythes engaged within the frame (figure 11).

 

An unusual feature of the construction of this building is that between the two wythes of brick are 1/2" smooth steel reinforcing bars. These run vertically between the spandrel beams, and are placed 2. 5' apart. They were undoubtedly installed to resist any tendency for the panels of brick to fall out in an earthquake, but this kind of reinforcing is not commonly found in other buildings of the period. These bars may be a result of the fact that the City Hall, which was under construction at the same time, has steel bars in the wall. [ix] Perhaps the engineer in 1912, (W. W. Breite, of San Francisco), was influenced by what he saw being done on that much more substantial and prestigious building nearby.

 

These reinforcing bars are shown on the engineer's original drawings which survive, and their existence has been confirmed in the field. It is not known how common this practice may have been. They do not exist in the Hotel Touraine, another nearby brick and steel frame hotel of the same age which is also currently under repair and retrofit. Their discovery in the Woodrow has precipitated a debate over their effectiveness. Some feel that they may create a plane of weakness in the brick, making it more prone to in-plane failure. All agree that they help avoid out-of-plane collapse of the infill walls. These bars provide evidence that engineers forthrightly addressed the earthquake problem at the time that the building was constructed.

 

The damage to the Woodrow from the Loma Prieta earthquake was limited to the rear walls, concentrated at the 2nd and 3rd floor levels. The worst area of damage was at the 2nd floor on the narrow rear facade (figure 12). Approximately 50 bricks fell from one inside corner of the light well, and bricks on the three other corners were loosened from the frame but did not fall. "X" cracks with some displacement of the brick were caused across the panels of the rear wall at the 2nd floor level. The first floor wall, which was constructed of 3 wythes of brick without openings, was undamaged. The long rear facade showed several "X" shear cracks and some other cracks, but no brick displacement except at the rear corner. The front (street) facades were completely undamaged. The only evidence of the earthquake on the interior was minor cracking of the paint film and plaster along the edges of the corridor walls where the walls meet the ceiling. No other damage of consequence was found.

 

 

THE REPAIRS ORDINANCE AND THE REQUIREMENT FOR A FULL SEISMIC UPGRADE

 

In the months following the earthquake, the owner of the Woodrow Hotel, Kulbushan Gupta, engaged a local engineer, Owen O'Neil, to prepare a design for the repair of the building. Because of the enactment of the Repairs Ordinance, the building first had to be evaluated as to whether it had lost over 10% of its lateral capacity. The method for arriving at a conclusion as to its loss of capacity was based on a set of guidelines developed in March of 1990 by a consultant to the City of Oakland, Sigmund Freeman of the engineering firm of Wiss, Janney & Elstner.

 

These Guidelines are based on a variant of standard static equivalent code procedures for determining the base shear on the building and measuring it against the ultimate capacities of each element resisting lateral forces. A table of allowable stresses, ultimate strengths, and story drift limits (stiffness) for steel reinforced concrete, plaster & lath, and masonry is included in the Guidelines. For masonry, the figure to be used for ultimate strength is 50 psi for unreinforced masonry and 75 psi for masonry infilling a frame.

 

This document provides no means to include a value for masonry after it has cracked. One is directed to "apply 90% of the base shear. . . to the post-earthquake condition of the building. "For the "post-earthquake condition" (ie. the damaged condition) one is supposed to create a new mathematical model. Herein lies the problem. There is no guidance or figure provided for the evaluation of the masonry after it has cracked in the earthquake. It is assumed that if an element is cracked, its contribution to the lateral force resisting system is nil. The guidelines conclude by stating that "If the loads on any element exceed their [sic] strength, it will be concluded that the structure has lost more than 10% of its pre-earthquake capacity. "In other words, with the infill masonry buildings, this means that if the engineer's model predicted that if an earthquake force equal to 90% of the Loma Prieta Earthquake were to occur and cause previously uncracked brick panels to crack, the structure would be subject to a full upgrade before it could be used again.

 

 

1) THE CONVENTIONAL RETROFIT DESIGN APPROACH

 

It was determined that the Woodrow Hotel fit the city's definition of a building with an over 10% loss of lateral capacity. It was then necessary to proceed with a design for a seismic upgrade which met the requirements of the Uniform Building Code. The first design which was submitted and actually approved by the city was a design by Owen O'Neil for the replacement of 2 wythes of brick over a substantial portion of all facades of the building with reinforced concrete. This design was based on the precedent set by the retrofit design for the Emporium Capwell building, which had at that time just been completed by EQE Engineers of San Francisco. This 1920's Beaux Arts building was the first building to be repaired following the earthquake. The EQE retrofit design included the installation of concrete shear walls on the exterior of the building. These new shear walls covered most of the windows and ornamental brick and terra cotta facade. The justification for this was the need to carry out as much work as possible on the exterior of the building so as not to disturb the store inside.

 

In the case of the Woodrow the windows were necessary, which meant that the new gunite surface had to extend over most of the facade in order to gain the stiffness which the calculations indicated was necessary. While this design meant that the rooms would not be disturbed by the retrofit, it did mean that, like the Emporium, the original architecture of the building would be altered beyond recognition. The Woodrow has an attractive, heavily articulated brick exterior, and this would have been destroyed.

 

This first design proved to be infeasible because it was predicated on removing 2 wythes of brick and replacing them with concrete "gunite" on what the engineer had thought to be a 3 wythe thick brick wall. There would have been no wall left to gunite against. In addition, the Department of Housing and Urban Development (HUD) expressed concern over the loss of any historic value the building may have.

 

It was then decided that an alternative needed to be explored. It was at that time that the author (Randolph Langenbach) became involved with the project with Lerner & Nathan, Architects, under a grant from the California Preservation Foundation. We began investigating the concept of the energy dissipating design and the beneficial use of the soft story as a means of avoiding the destruction of the facades.

 

At the same time, the engineer, Owen O'Neil developed another conventional proposal using steel braces on the inside of the exterior walls. Because of the overturning forces, the existing columns were not considered to be adequate, so these braces were complete unto themselves set behind the existing frame of the building. This frame was originally designed to be two bays wide on each side of the building, but after the city expressed concern that its flexibility would still allow the brick to crack, the design was expanded to 3 bays. (The City has since agreed to allow 2 bays. ) This interior solution was certainly more respectful of the existing architecture of the building. However, with a total of 8 to 12 bays of steel braced frames running from basement to the roof, this solution is certainly not inexpensive (approximately $1. 5 million), and not consistent with the use of the building for low income housing without massive government support. [x]

 

 

2) A PROPOSAL TO TAKE ADVANTAGE OF THE "SOFT STORY" IN THE RETROFIT DESIGN

 

What was immediately notable about the Woodrow Hotel following the earthquake was the complete lack of damage to the street front walls. On these two walls the windows are larger than in the rear, leaving very little shear strength in these walls. One would expect that the deformations would have caused some visible damage around the windows. If the floor diaphragms had been stiff, the forces would have been transferred to the rear walls and the elevator shaft, but with the flexible floors which simply rest on the steel frame, it is likely that the deformations in the front would have not been fully coupled with the rear walls. In that case one has to look to the front walls themselves for an explanation as to why they exhibited no damage.

 

There are two explanations for this lack of damage. 1) Since the masonry of the front walls stops at the second floor, with open store fronts below, the resulting "soft story" would help to isolate the masonry above from the earthquake forces. 2) The large openings above serve to break the shear walls of masonry into a spandrel and pier system capable of larger story drifts with inelastic flexural movement of the piers capable of leaving no visible marks. It is likely that a combination of both of these explanations served to reduce the loads experienced within the plane of the walls, and also transfer some of the remaining loads into the interior of the structure where the forces were dissipated in the racking of the timber floors and walls without much visible effect.

 

 

3) A PROPOSAL TO USE 3M VISCO-ELASTIC DAMPERS

 

Visco-elastic dampers are manufactured by 3M corporation and have been in use for over 25 years in high-rise buildings, to provide damping to resist the vibrations caused by high winds. They consist of steel plates coated with a propriatary glue like material (not unlike the substance used on Scotch Tape) which has viscous as well as elastic behavior when stressed in shear.

The visco-elastic dampers have never before been used for seismic resistance in buildings. In this proposed installation, instead of being used to enhance the comfort level, these dampers are intended to be used in order to reduce the transmission of the forces up into the superstructure of the building, by utilizing the "soft story" behavior to decouple the superstructure from the ground motion (figure 13 and 14).

 

It was the comparison of the damage to the rear walls of this building with the lack of damage in the front which led us to the idea of developing a retrofit scheme based on flexibility and energy dissipation rather than on strength. It was at the time that this was beginning to be explored that a chance meeting between the author and Professor James Kelly led to the idea of using the 3M visco-elastic dampers in the design. The logic of it is simple: if the street fronts of the building were protected from damage during the Loma Prieta Earthquake by the soft story, we should try to avoid defeating this protective feature in any subsequent earthquake while overcoming any possibility that the frame in the front could become unstable from excessive sway (known as the P-∆ effect).

 

Prior to the proposal to use the dampers, we explored a solution to the problem by proposing that the new shear wall which Owen O'Neil had designed for the street fronts of the building (scheme #1 above) be located on the inside corridor wall instead (steel diagonal braces could be used instead of concrete. ) The purpose of moving it back was two fold: 1) it would avoid spoiling the architectural integrity of the facade, and 2) it would separate the rigid shear walls from the front brick walls, coupling them together through the floor diaphragms. The floors would serve as an energy dissipating damper buffering the exterior brick wall from the ground motion in much the same way as the "cross walls" work in the ABK methodology for bearing wall buildings. This design was explored with the assistance of Nels Roselund, Vice President of Kariotis and Associates. Although this design was never completed, the design with the dampers seems to hold more promise, offering a promising solution for buildings which have flexible wooden diaphragms above open storefronts.

 

With the discovery of the existence of the dampers with their as yet unexploited potential for use in earthquake design, the alternative study shifted away from the shear wall idea to that of using the dampers. With the dampers, it seemed possible to reduce the effects of the ground motion on the upper floors of the building as a whole. The first proposal was to cut away the ground floor wall in the rear, and thus break the rigid coupling of the building to the ground. This immediately raised some thorny problems, including an interesting legal dilemma: If the design of the retrofit explicitly required the building to sway over the property line, would the owner have to purchase an easement from the neighbor? (One must remember that it would likely sway over the property line in an earthquake even without the dampers. ) Of the several experienced real estate lawyers contacted, all agreed that the problem was unprecedented, and thus without an obvious legal solution.

 

This issue seemed uncertain enough to threaten the viability of the whole idea. As a result, the author then proposed that the 1st floor rear wall be retained, with the dampers used only to protect the street front facades. This solved the legal problem because the structure would not be rendered more flexible than it is already, but it also meant that the design of the building with its potentially torsional behavior would be more complicated. In addition, the back wall would have to be repaired and strengthened to resist the full force of an earthquake without the help of the dampers, returning us in part to the dilemma with the codes described above.

 

In terms of engineering performance, the dampers provide an important advantage over the use of the recessed shear wall and the flexible floor diaphragm scheme first explored because they are precisely engineered, and their dynamic performance is known. This allows for their easy inclusion in an analysis of the performance of the building as a whole.

 

The dampers introduce a known amount of damping at the onset of shaking, without large deformations within the elastic range before the plastic behavior begins. Also, the maximum limit is set, and the full support of the steel braces come into play if the dampers exceed their designed limit of deformation. Should the damping material fail to perform as designed, the braces are still secure. The only variable is the amount of damping available within the designed amount of story drift. The damper material is not relied on to hold the braces together or to prevent further drift.

 

The dampers are advantageous because they introduce a significant amount of damping within very small lateral movements. Not only do they avoid a significant elastic phase in their deformation, but the plastic yielding is fairly constant across their deformation range. The dampers are manufactured specifically for each usage. Their characteristics, such as the amount of damping per unit deformation, and the limits of the movement allowed, are based on the engineering analysis of each installation. (See Appendix, by James Kelly for the technical details on the sizing and performance of the 3M Viscoelastic Dampers)

 

 

PROBLEMS ENCOUNTERED IN PROCEEDING WITH THE INSTALLATION OF THE DAMPERS AT THE WOODROW HOTEL

 

The proposal to use the visco-elastic dampers has raised some difficult issues which have little to do with their actual engineering value. The principal problems fall within the realms of finance, law, and city approval.

 

(1) The owner is reluctant to spend the funds necessary to carry out the dynamic analysis and engineering design which dampers would require. In this case, having paid for the two separate designs based on the standard code static equivalent approach, the owner is quite reluctant to pay for having it designed again, without the assurance that the city would approve the project in a expeditious manner. While the saving of a considerable amount of money in construction cost is likely with the damper scheme, this saving was less important in this project because of the government funding of the project with grants and low interest loans. The owner's main objective was to obtain a building permit so that the funding could be obtained and the project proceed. Further design only serves to delay the access to these funds. [xi]

 

 (2) There are several legal impediments to the use of the dampers.  

 

(2a) The first emerged with the issue of the drift over the property line. This was an unprecedented problem, and the several experienced attorneys who examined it could not make a firm determination as to how best to solve it, short of actually purchasing an easement from the neighbors. It was solved in this case by redesigning the project to employ the dampers on only the front two walls. The city accepted the explanation that if it could be established that we were not making the building any less stiff than it was before the earthquake, then the property line issue would not be an obstacle.

 

In the end, altering the building to suit the dampers is probably not as good an idea as using the dampers to improve the building's existing deformation characteristics. If the building is too stiff to activate the dampers, they are probably not suitable for that application.

 

(2b) Another legal issue recently has emerged - raised by 3M Company itself. 3M has become concerned that the proposed use of the dampers represents a departure from their previous use over the past 25 years for high-rise building subject to wind vibrations. In that use, the buildings are designed to code and the dampers are installed to aid in the comfort of the occupants.

 

In the proposed earthquake usage, the dampers are to be used to reduce the computed loads on the superstructure of the building, and thus reduce the strengthening work necessary. In the case of the dampers installed around the whole building, the reduction in forces applied to the 2nd floor and above was about half. As such, 3M is concerned that if an earthquake occurred where any one was injured or killed, they would become the "deep pocket" subject to enormous potential losses in a law suit.

 

This problem has been raised in particular with a proposal to use the dampers on the Golden Gate Bridge, where the scale of any failure, remote as it may seem, would be catastrophic, even if the dampers had nothing actually to do with any failure which might occur.

 

(3) Another difficulty is in the approval process. While the Building Department has expressed interest in the damper technology, the approval of a scheme with the dampers is potentially a long and expensive process. The city requires a peer review, and with little guidance from the code on the use of these devices, the selection of peers and the preparation of a scope of work for their services is difficult.

 

 

In the long term, these kinds of problems proved surmountable. While reviewing the problems which have emerged in the Woodrow Hotel project, it is important to not lose sight of the objectives in proposing the use of the dampers. The dampers are one solution, but not the only solution based on taking a "soft" approach to the retrofit design for frame and infill wall buildings. Whether they are used as part of the primary minimum strengthening after the legal problems are overcome, or as an added element to reduce the level of property damage, they have the potential of greatly improving the performance of these buildings in an economical and historically sensitive way.

 

In the end, The principal conceptual hurdle to overcome in the introduction of a soft approach in the retrofit of buildings with masonry is the acceptance of a certain amount of cracking in the masonry during earthquakes. This requires changes not only in the applicable codes, but also major change in the way people, including most engineers,perceive and understand the problem of the effects of earthquakes on buildings.

 

In fact, by adding dampers into conventionally designed braces, the risk of damage which comes from introducing rigid elements into existing buildings is reduced. The dampers may substantially reduce the forces which could be attracted to the new rigid elements. If damping is provided by elements which are intentionally designed to yield, there is always a danger of their being too weak, and thus not limiting the drift, or too strong, and not yielding at all. Because of the ability to manufacture the dampers to a known drift limit control, the risk of having too weak an element is avoided.

 

 

 

CONCLUSION:  GO TO PART 2

 

---

[i] p4 sect 18-1. 05B

[ii] First enacted in October, 1989, and revised in May, 1990.

[iii] Statistics provided by Oakland Public Works, Seismic Safety Div. , August, 1991.

[iv] It is important to note that infill wall buildings have collapsed in the Mexico City earthquake of 1985, and in other earthquakes. However, there are many important differences between the construction of the Mexico City buildings and that of those found in the United States, which make the comparison inappropriate.

[v] In all of my thought on the subject of hazard mitigation, I was unprepared for the psychological impact which the shaking had on the public, including the city policy makers, and the consequential impact which this has had on decisions subsequent to the earthquake.

[vi] Of all of the damaged infill frame buildings, the Oakland Hotel may have been the only one with falling debris from the upper stories other than from damage caused by pounding. (The Broadway Building and the Grant Building were both affected by pounding. ) This is important because the risk from pounding is not addressed by code conformance in retrofit design, which can mean that the repairs Ordinance and the proposed SB547 Mitigation Ordinance may not even address the one problem which may be the most significant life safety risk.

[vii] The Sheraton Palace Hotel in San Francisco is similar in construction to the Oakland Hotel, and also was damaged in the Loma Prieta Earthquake. While it suffered from extensive cracking in its interior terra cotta partitions, the Palace Hotel did not have bricks fall to the street. Cracks were found on the exterior masonry, but often they revealed underlying rusting of the buried steel columns.

[viii] The steel beams running in the direction of the joists are only 6" deep (6"I 12. 5#), supporting the joists, they are 10" deep (10" I 25#) and the columns are 8" square, built up out of 3/8 thick steel plates and angles. (ref: original drawings by W. W. Breite, Engineer)

[ix] The original specifications state: "All exterior walls throughout shall be reinforced horizontally at the level of each stone or terra cotta course with 1/4" deformed medium steel bars. . . securely held in position and to the steel columns by bending it completely around the column. . . "The exterior wall reinforcing on the Woodrow Hotel consists of 1/2" smooth steel bars set vertically in the brick collar joint, and fastened through holes in the beams.

[x] At the present time (August, 1991) this scheme has been submitted to the Building Department for a permit. It is currently going through plan check. Formal estimates and bids have yet to be carried out for it. The visco-elastic damper scheme has been developed in concept, but has not been pursued further because the owner cannot spend the funds for the engineering required before the project is funded, despite the fact that it promises to save an immense amount in construction costs.

[xi] The professional fees for engineering are significantly higher for a dynamic analysis based design, compared to those for a static equivalent design.

[xii] In the case of the 1960's bare concrete frame buildings described by Englekirk and Hart, the use of visco-elastic dampers may prove to be ideal in that they can be fabricated to add ductility to the system at very small displacements. If ductility can only be added by yielding of structural members, then considerable damage to the original brittle concrete frame may have occurred before this ductility is effective in damping the earthquake forces.