Garage Ruins Examined for Quake Lessons

As they sift for lessons in the rubble of the massive Northridge earthquake, experts have been alarmed by the spectacular failures of several large parking structures which could have entombed scores of victims had they collapsed during business hours.

Of all modern buildings, parking structures “appear to have suffered the largest incidence of partial or total collapse,” said a preliminary report by UC Berkeley’s Earthquake Engineering Research Center.

The most troubling aspect of the failures, experts said, is that several new parking structures took the heaviest hits, while many older garages--built at a time of weaker building codes and less engineering know-how--came through with less damage. They also noted that structures assembled from precast concrete sections seemed most vulnerable, and that the garages may have been subjected to more vertical quake movement than they were designed to withstand.

At least eight public parking structures at heavily used shopping centers, colleges and hospitals either partly or completely collapsed. Four were built in the last six years, and the others were built between 1965 and 1982. Cracked or crumbling columns and beams caused building inspectors to close at least 20 others in Los Angeles, Santa Monica and Glendale. Many of those will take months to repair at a cost of millions of dollars.

Although experts are not unanimous about the causes of the failures, or how to prevent future ones, several said they signify a dangerous combination of inadequacies in building codes and an increasing drive to cut costs by designing for the minimum code requirements.

“In Southern California, the construction market is very cutthroat,” said Neil Hawkins, chairman of the civil engineering department at the University of Illinois, who inspected more than a dozen parking structures damaged by the quake.

If “you can still satisfy the words of the code and the system is cheaper, everybody will buy that one,” Hawkins said.

“The pressure is very high to cut the building down to the absolute minimum requirements of the code,” said structural engineer Bill Holmes, part of a quake study team sent by the Earthquake Engineering Research Institute in Oakland. “Because of the highly competitive nature . . . the costly details of tying together precast structures or the extensive engineering that would be required is perhaps not there.”

Structural and seismic engineers who came from around the country to study the effects of the quake generally agreed that current practices in the design and construction of parking structures need to be revised to require more resistance to seismic forces--particularly the unexpectedly large vertical thrust of the type produced by the Northridge temblor. Some also suggested that more diligent inspections and review of construction plans are needed to protect against future disasters.

Most of the attention of the engineers focused on today’s prevalent building method of stitching together large precast concrete slabs and beams rather than pouring the concrete into forms to harden as a single unit--or building with steel, which is more expensive but is the most quake-resistant material.

Experts say precast structures--relatively cheap to build because they can be put up quickly--require special care to strengthen the connections between the structural elements. “It’s like a card house and if you don’t tie your cards together carefully, it will all collapse,” said Helmut Krawinkler, co-director of Stanford University’s John A. Blume Earthquake Engineering Center.

In effect, it is more difficult to make precast structures as sturdy as other buildings. The more engineering and materials put into the ties, the more costly the building--defeating the purpose for which precast concrete was selected.

Most experts said precast structures proved most vulnerable to the intense shaking of the 6.6-magnitude Northridge temblor.

At the Northridge Fashion Center, where two large precast parking structures built in 1988 partially collapsed, two much older poured-in-place garages built in 1971 suffered cracking but did not fall.

At Cal State Northridge, an $11.5-million parking structure, built mainly of precast elements just two years ago, also partially collapsed while a cast-in-place garage just three blocks north, also completed in 1991, came through intact.

And a portion of the upper deck of the Glendale Civic Center garage, constructed in 1988 by pouring a slab over precast beams, caved in, breaking through two decks below as it fell.

Several experts interviewed by The Times said such failures exposed the limitations of the building code in ensuring the safety of the bare-bones modern parking structure.

Parking structures typically feature long spans and open architecture, both to save money and park the largest number of cars. Interior partitions and walls--which would dissipate part of a quake’s destructive energy--are kept to a minimum. The exterior walls also have large openings--saving money and helping to light and ventilate the structure, but again providing fewer structural elements to absorb the shock.

Since the code upgradings after the 1971 Sylmar earthquake, advances in engineering know-how have allowed designers to widen the spans and reduce other internal support and still get their plans approved by building departments.

In the past, experts said, engineers were more apt to design buildings well in excess of code requirements--in part because of uncertainty over how much structural strength it would take to meet the minimums. Ironically, advancing knowledge may be tempting them to refine their seismic calculations to a thin margin.

“The engineer probably has too much confidence in the code being correct,” said Hawkins, of the University of Illinois. “Unfortunately, he always has to work to the minimum of the code--otherwise, he won’t get the job.”

According to the public perception, the code is, “ ‘God protect us all,’ and it isn’t that way,” said Fred Herman, chief building official for the city of Palo Alto and a board member of the International Conference of Building Officials.

“It is a minimum standard, and building departments fight all the time to get designs to meet the minimum,” Herman said.

Said Peter Yanev, chairman of EQE International, a leading earthquake engineering firm, “If you (just) meet the code, you’re barely legal . . . . If there’s any error, you’re definitely underneath--and people make mistakes.”

In fact, the Uniform Building Code, which is used by all California cities and counties, does not dictate specific methods of construction, but rather sets seismic strength guidelines, allowing the designers great leeway in deciding how to meet expected stresses.

In high-risk seismic areas, parking structures, as well as other buildings, are supposed to resist the sideways movement generated by a large quake. This is done through the use of stiff and strong “shear walls,” which resist the movement, or with “ductile” building frames designed to bend, but not break, under pressure.

Experts cited several possible explanations for garage collapses.

Some structures apparently failed because internal columns that should have flexed with the rest of the building instead snapped under the sideways force.

Such internal columns were not designed to help resist the sideways movement, but merely to carry the weight of the building. Nonetheless, they were supposed to “ride along” with the parts of the building designed to resist the sideways force.

“One way or another, it is absolutely clear that the (internal columns) could not follow the movement of the lateral load resisting system,” said S.K. Ghosh, director of codes, standards and engineering services for the Portland Cement Assn., a trade group based in Skokie, Ill.

In other cases, experts said, it appeared that key connections between structural elements pulled apart under the force of the quake, causing the buildings to collapse. This may have contributed to the parking structure failure at the Glendale Civic Center, said an engineer who visited the site.

Nelson Behrend, architect for the civic center garage addition that collapsed in the quake, said he had visited the site and did not yet know why the structure failed, but he suspected vertical movement was to blame.

Indeed, experts said some failures may have resulted from the crushing of columns by the strong vertical thrust of this particular earthquake.

Most quakes generate primarily sideways movement, and the building code sets no special requirement for sustaining vertical pressures. However, the Northridge quake caused strong vertical as well as horizontal thrust, and some experts say that may have been a factor in parking structure failures.

David Anderson, vice president for operations of A.T. Curd Builders Inc. of Glendale, which built the collapsed Cal State Northridge garage, said the internal columns were designed only to handle the dead load of the building, and “that’s all they were required to handle.”

The “earthquake did something that was not anticipated by the code.”

Robert Engelkirk, a structural engineer who designed the Northridge Fashion Center garages that collapsed in the temblor, also blamed a “failure in the vertical load carrying system.”

Regardless of the cause of specific collapses, some critics contend that the failures show that more can go wrong with precast buildings.

“I’ve been telling my wife for years not to park in (precast) garages like that,” remarked Peter Yanev. “It’s almost a joke in the family,” he said.

Yanev, who joined a U.S. study team in touring quake-ravaged Armenia in 1988, warned of problems with precast concrete in his report on that disaster.

“Of the building systems affected by the earthquake, seismically under-designed precast-concrete frames . . . performed most poorly,” he wrote in a report on the Armenian quake. Similar designs are “also popular in the United States,” he wrote.

“While the system is probably adequate in zones with very low seismicity, the earthquake demonstrated that it should be eliminated or radically modified in higher-risk zones in both countries.”

Regarding the Northridge earthquake, Yanev said, “We can come up with a million reasons why the code was inadequate or the engineers made mistakes or the earthquake was stronger than anything we’ve ever seen before. . . .

“There is nothing new here we are talking about,” he said. “Theoretically, everything can be made to work. . . . What we need to do is beef up the codes and to penalize the poor performance.”

Yanev said he believes the code should be changed to impose a higher standard for precast than other types of construction in earthquake zones.

Other experts, however, said precast construction can do the job if building departments are more rigorous in reviewing plans and conducting inspections.

“I think it is more a matter of enforcement than a failure of the code,” Krawinkler said. “From a building department perspective, it’s difficult, because ensuring compliance with code requires much larger manpower. Yes, it can be done, but not the way building departments are staffed today.”

Still others predicted that the building code will be changed to address vertical pressures such as those that came with the Northridge quake.

“I think there will definitely be some harder looks taken at this vertical acceleration aspect,” said Ray McCann, a member of the Structural Engineers Assn. of California.

Even if the experts reach a consensus on the lessons of the Jan. 17 earthquake, their recommendations cannot be written into the Uniform Building Code until 1997 when it will next be revised by the International Conference of Building Officials.

However, the California Building Standards Commission, a state agency, could require local governments to upgrade their standards in the meantime.

Until then, local officials must decide whether older parking structures that are repaired must be brought up to current standards, however inadequate some critics may think them to be. In Los Angeles, any portions of parking garages that suffered structural damage will have to be brought to current code, said Nick Delli Quadri, senior structural engineer with the city’s Department of Building and Safety.

Given the financial pressures and the short half-life of even pressing issues, some experts are concerned that the lessons of the garage failures will go unheeded.

“That is always a financial story,” Krawinkler said. “Garages have little impact compared to freeway overpasses. It may fizzle out. Hopefully it will not.”

Times staff writer Julio Moran, correspondent Jill Leovy and Richard O’Reilly, director of computer analysis, contributed to this story.

* RELATED QUAKE COVERAGE: A3, A20, B1

Potential Failures of Cast-in-Place Concrete Structures

Floor or roof: damage

Infill wall: with cracks extending through concrete frame

Joint: cracking

Beam concrete: cracking, exposure of reinforcement

Columns: out of plumb (at any level)

Potential Failures of Precast Concrete Structures

Separation leading to possible loss of vertical support

Corbel: damage leading to possible loss of vertical support

Story: out of plumb (at any level)

Joint: damage

Floor or roof: damage

Column: failure

Beam-column: joint distress or separation