Snap, Crackle, Pop! A Tour of Dr. DesRoches's Lab
May 2006
Georgia Tech CEISMC Gazette

Dr. Reggie DesRoches of the School of Civil and Environmental Engineering took us on a tour of his lab in order to explain just how one tests structures for their resistance to earthquakes.

In Dr. DesRoches's own words:

This is a unique facility. It's probably one of the top five in the country in terms of size, capabilities, being able to test things at full scale. One of the reasons we have such a high door is so that we can bring full sized trucks in here with girders. We typically test the girders under a million pound loading machine. These things normally take on the order of 600 to 800,000 pounds of load before they fail.



The way we normally do things in earthquake engineering, there are a couple ways of testing things. You can put a structure on a big "shake table" that actually shakes to simulates an earthquake. We do not have that here. The other way of doing it, which is just as effective, is to actually displace the structure as it would be displaced during an earthquake by using these actuators. Actuators can put on the order of 200,000 pounds of load onto a structure in order to deform it. Once it deforms we can see how it would be damaged and look at ways to repair it to minimize damage.

So you can see here the cracks occurring in the joint area, which is normally where most of the damage would happen. We'll go back through and patch up the joint area and see how well it would work if they were to have a stronger joint versus a weaker joint.

Certainly one of the challenges we face in earthquake engineering is that these are all parts of a building. There really is no facility in the world where you can do full-scale tests. Unlike other fields, such as the aeronautical engineering field, they build a prototype plane and they test it. For us, we build a building and then we may have to wait thirty years until an earthquake hits it to know that we did it right. So we really don't have a chance to test things other than some of the things we do in the lab to know whether or not we're doing it right.

This is called a strongwall. It's on the order of ten feet thick. We can test buildings as tall as two stories high. So each of these points in the strongwall is where we would put our actuators to push off against the wall onto a building. The wall is very stiff because obviously when you push you're pushing back on the wall. We don't want the wall to be damaged, so each of these spots can take a couple hundred thousand pounds of force.

This represents a section of a bridge. Georgia and many other states like to raise the elevation of bridges so that it limits the probability of a truck not being able to pass under the bridge. Trucks are getting taller and taller, so what you'll find is that periodically you'll get impact between the truck and the bridge that it's going under. So Georgia goes through periodically and increases the actual height of a bridge. They'll raise the bridge in some cases as high as three feet so that larger trucks can go under the bridge without the likelihood of impact. What we're doing here is testing this method against low level earthquakes. They've been doing this for years, and now that the new codes have come out for earthquakes they realize that they really didn't consider what would happen if this type of elevated bridge were subjected to a low level earthquake. We constructed the bridge, we put the right loads on there, we put a pedestal on there, we instrumented it, we're pushing it back and forth, which represents the type of loading that the structure might get under an earthquake.

This is the actuator. So this just pushes back and forth. You can do it rapidly, you can do it slowly, but it pushes back and forth to represent the motion that you would have during an earthquake.

You can go out through the "graveyard," where I'll show you some of the tests that we've previously done. We keep specimens out there both as an education for the students as well as to see how things behave after they've been under the weather for a while.

This is the "graveyard." This is where we keep a lot of specimens that have been previously tested and have failed. Here we see a steel frame from a building that we've tested to failure. See the failure across the bracing elements there?





We're doing many different tests. Pre-stressed concrete girders as shown here. This is the typical girder you would see on a bridge with a deck. We look at different types of concrete and how lightweight concrete might behave versus normal weight concrete, in order to try to get the spans a little bit longer.




Stressed to failure.

We use some of these pieces when we have another test. We will cut undamaged portions from the specimen and use it again.






Steel sections. These are actual bearings that you may see in a very large bridge, what we call "pin bearings." The bridge would just sit on this. At one point we were testing these as part of a project we have with Tennessee DOT.







So this is a curved steel girder. We did a project for the Federal Highway Administration where we were testing how these curved steel girders would behave under deadloads. So you can see here the buckling that occurred in this particular section between where the two stiffeners are. One of the interesting aspects of this is we actually did a model of this same section and loaded it up, and we almost replicated the failure mode exactly using a computer model.

That's why testing is so important; we need to continually validate that our models are replicating what would actually happen out in the field. So facilities like this are important to keep us on track to make sure that we're modeling things well.

So that's our facility; this is what we do!