NEEShub services are beginning to be phased out. Please begin using DesignSafe ( Please submit a support ticket ( if you need any assistance.


Support Options

Submit a Support Ticket

You are here: Home

NEES-CFS building test update

NOTE: Project PI Ben Schafer summarizes the initial test results from the August 16 shake table tests, below.


Last week we tested the CFS-NEES building (completely outfitted with essentially all exterior and interior non-structural in place, we call this Phase 2e) on two Northridge EQ records: Canoga Park and Rinaldi. Canoga Park's response spectra lines up well with DBE levels and Rinaldi's with much higher MCE levels. In fact, tuning the shake table proved challenging, and we even overshot the Rinaldi record which means we had peak horizontal accelerations of 1.1g and vertical accelerations of 1.1g!. The video does not do it justice, it is quite a sight to see a building get pushed around that hard that fast.

Initial videos at and more details and videos to be posted there later.

Performance was again far better than expected. After Rinaldi, and all the noise and excitement of the test, the building returned to straight. Interior non-structural and exterior gravity walls around openings all saw cracking in the gypsum board panels. The exterior DensGlass showed some minor damage in some corners and at seam lines. We removed the gypsum board for visual inspection of the hold down that saw the greatest demands and observed no signs of distress. We have lots of data to process and the period of the building shifted post test, so damage was indeed imparted, but nothing near failure.

To put some numbers to this:

For Rinaldi (MCE level, i.e. "the big test") on the Phase 2e CFS-NEES building we measured a peak story drift along the West wall (short direction for the building) of 0.95%. Call it 1% drift if you like.

For Canoga Park (DBE level) on the Phase 2e CFS-NEES building we measured a peak story drift along the East wall (short direction for the building) of 0.52%. Call it 1/2% drift if you like.

For comparison when we ran Canoga Park (DBE level) on the Phase 1 CFS-NEES building, i.e. the building that only had the shear walls and the floor sheathed, but otherwise all-steel on the gravity walls and no interior structure at all we had a peak story drift measured along the first floor of the south wall line of 1.4%.

So, for the same record, between Phase 1 (shear walls only) and Phase 2e (completed building) the drift reduced from 1.5% down to 1/2%, obviously there is a lot more going on in the completed building than is captured is the engineering model focused on the shear walls.

Lots more stories to tell from the data, for sure the building diaphragms are experiencing some interesting shear, perhaps evidence of a "flexible diaphragm" but the other data tells another story. Some sleuthing remains here...

Also, we have closely monitored the change in stiffness (period) of the building through construction and before and after all our tests. This data is fascinating too.

Further, we continue to work on our computational models of the building. We have introduced the gravity walls into one of our lateral models because without it the predicted period is far too long when compared with our tests. Yes, this means gravity walls, even all-steel without sheathing, see lateral loads - it seems a real building does not always care what the engineer assumes. We have a lot more work to do in the modeling phase, but we are seeing interesting results, we are on the right path, and working hard to distill them into something with practical impact.

We pushed the UB shake tables to very near their maximums, so there is no more pushing our building. We begin deconstruction now and wrap up the experimental testing phase of the project.

We have a lot more to learn as we process all the data, but overall I am left humbled by how well the building performed. Nothing magic, but we are now working hard to sharpen our pencils a great deal more to decipher best how the building engages this superior performance and what its limits are.

I would like to pass on significant thanks to the UB team, especially Mark Pitman. Also, to my graduate student Kara Peterman who has done a stellar job on this project, particularly, these last few months in Buffalo. Amazing.

We have a lot more to learn from the test data, and I look forward to sharing the results, in-depth with the Industry Advisory Board this fall. For now, congratulations to all, and back to work.