It's one thing to build a model building and test its quake resistance on a laboratory shake table. It's quite another to do the experiment in the field and shake an actual building in the real world to the point of partial collapse.
That is what a bi-coastal research team did to a red-tagged warehouse in the Imperial County city of El Centro, a hundred miles east of San Diego.
The 2-story structure was equipped with multiple cameras and more than a hundred sensors to record movement and acceleration.
"Based on a worst case scenario, we might bring the entire building down," said Andreas Stavridis, PhD, Asst. Professor of Engineering at the University of Buffalo and principal investigator of the study. "That's why we have to be careful."
In learning more about how buildings fail, the goal is to gain insight into how best to retrofit similar designs to make them more quake resistant, Stavridis said.
"Our models can predict very well up to a point," he said. Shaking a real building gives researchers the opportunity to measure the real world effect of structural and foundational variations that can be difficult to model in a laboratory or computer software.
His research focus is a type of construction widely used a century ago for commercial buildings and large apartment houses. Hundreds remain in Hollywood, downtown and elsewhere in metropolitan Los Angeles. It is a concrete/brick hybrid known as "reinforced concrete with masonry in-fills," still used for new construction inLatin America and Europe.
It represented a step beyond traditional brick structures by adding a frame made of steel-reinforced concrete, with bricks filling in the wall space in between the concrete columns. Unfortunately, the unreinforced brick wall segments still proved vulnerable to breaking apart and falling during quakes, and even the concrete columns--when lacking sufficient reinforcement--proved
vulnerable to cracking and failing.
The 1933 Long Beach Quake demonstrated the vulnerability of brick buildings, which led to limits on new brick construction. After more collapsed in the 1971 Sylmar Quake, Los Angeles required retrofitting remaining brick structures, though that is still a work in progress in many areas. In fact, it was brick buildings in Napa that took the brunt of the magnitude 6.0 quake this past August.
Sylmar and the 1994 Northridge quake revealed the vulnerability of buildings framed with concrete deemed "non-ductile"--in other words, brittle--and prone to cracking under seismic vibrations. Such buildings, without reinforcement in the form of "shear walls," can pancake when they fail.
"These are the most vulnerable buildings in Los Angeles, and the most vulnerable people living in them," said Mitchell Englander, member of the Los Angeles City Council."
Englander and others in LA City Hall are now looking into requiring retrofitting of non-ductile concrete buildings, but see the need to develop a funding mechanism to cover the enormous cost.
"How do we incentivize? That's what we have to do--not just an unfunded mandate, Englander said.
More than 14-hundred potentially vulnerable non-ductille concrete buildings were identified in LA by a survey released earlier this year by the Pacific Earthquake Engineering Research Center (PEER). The Los Angeles Dept. of Building and Safety is now doing followup. "They're promising by year's end they'll have some kind of inventory," Englander said.
But what exactly happens in such buildings as they are stressed by earthquake to the breaking point?
While working for his Masters and PhD at UC San Diego, Greek-born Stavridis was involved with testing brick walls on a shake table. But he yearned to go the next step and test a real building.
It was while doing field surveys after the 7.2 magnitude Easter Sunday Mexicali Quake of 2010 shook the Imperial Valley that Stavridis came across what he was looking for: a 1920's era two story concrete frame with brick infill. Once used for grocery wholesale, and more recently as a cabinet shop, its second story suffered significant structural damage, and the then-owner considered tearing it down till the demo bid came in at $300,000.
Demo contractor Armando Samaniego ended up acquiring ownership of the building, and offered the prospect of an experiment doing part of the demolition, it was an offer he could not pass up.
"I feel good to be part of it," said the enterprising Samaniego, owner of Imperial Valley Scrap Masters, among other endeavors.
For Stavridis, putting together the project would take another four years, working in collaboration
with another UCSD alumnus, Babak Moaveni, now at Tufts University in Massachusetts as an Associate Prof. of Civil and Environmental Engineering. Moaveni is the co-principal invetigator and will be involved in analyzing the digital data gleaned from the shake experiments. The University of Buffalo provided much of the instrumentation to measure three-dimensional motion.
Stavridis obtained a grant from the National Science Foundation (NSF), and has been working with the NSF-sponsored George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES). The shake table at UC San Diego is part of the Nees network. So is UCLA's mobile field laboratory developed for full scale shake experiments in the field on structures, foundations, once even a levee. Its tools include a heavy duty shake machine, an "eccentric mass shaker," that can generate an oscillating force of up to 10 tons.
The shaker was developed four decades ago by Steve Keowen, currently Senior Development Engineer at UCLA in the Department of Civil and Environmental Engineering. Keowen is within weeks of retirement, but agreed to run the shaker for the El Centro project.
Last February the team first tried out its methodology on a ten story building in Utica, New York, something of a rehearsal for the El Centro testing, Stavridis called it.
Set-up alone took nearly a week. in order to get the shake machine to the second floor, Samaniego's demo crew knocked out one of the brick in-fill walls on the building front facing Commercial Ave, then used a crane to hoist it in.
Once inside, the shaker was bolted to the concrete slab floor, the bolts having to be re-tightened with a foot-long box-end wrench after every four-minute run of the shaker.
The shaker has two off-center compartments that can be filled with lead weights and rotated to cause the vibration to simulate a quake. By varying the rotation speed, the shaker can send
vibrations at different frequencies.
Like a guitar string or a drum, a building has a natural frequency at which it vibrates when agitated. The motion is amplified when the incoming vibration--from the shaker, or an earthquake--
occurs at the building's natural frequency.
From textbook formulas, the natural frequency of the El Centro building should have been around 2.5 Hertz (cycles per second). But Stavridis knew it would be lower on account of the damage the building had suffered from the 2010 quake, and the frequency at which the shaker produced the biggest amplitudes turned out to be around 1.7 Hertz.
The testing began slowly, with minimal lead loaded into the shaker's compartments. After 18 pound bricks were added by hand, the instruments detected large accelerations and displacements with increasing weight. Also as expected, the building behaved differently when the orientation of the vibration was changed from north-south to east-west.
By the second day of testing, bricks had begun tumbling. By the last run, four sections of brick infill had collapsed. The team stopped short of inflicting enough vibration to cause the concrete columns to fail.
The first floor, which had previously been reinforced with concrete shear walls, appeared to be undamaged, which was good news for Samaniego. His plan all along has been to remove the damaged second floor, and turn the structure into a one story building for its re-opening.
Stavridis did not get all the data for which he had hoped. But he believes he has enough to provide new understanding of the motions and failure mechanisms of a type of building still in wide use from Los Angeles to Athens.