Seismic Loading using COM624 1

In COM624 parlance, "cyclic" means repeated loading - but not the seismic variety. It has to do with statically applied repeated loading. This is covered in Matlock's paper covering p-y curves for soft clay (1972.)

If you are using cyclic criteria, then the p-y curves for cyclic loading are different than static curves. But I can't tell you how, exactly, since each researcher used a slightly different approach. You will need to read the technical paper(s) that formed the basis for the p-y curve(s) you are using in your analyses.

A word of caution: under some circumstances, the cyclic curves originally recommended - and incorporated in COM624 - may not be conservative. For example, the soft clay criteria proposed by Matlock (1972) includes a cyclic reduction factor of 0.72 for peak lateral resistance. However, tests performed by Shell Oil at Harvey, La. clearly show that more severe degradation can occur - for that site, the reduction should have been 0.29! That's a ratio of about 2.5...

P-y curves are a sophisticated tool - and not really recommended for novices. If you have a seismic problem, you need to hire a geotechnical consultant that is conversant in both seismic design and p-y curves. Tread carefully.

So are you saying the COM624 program is not a good program to use with seismic loading? If the equivalent static seismic force is given for design would the program be valid for analysis? Is it also not good for wind loading? Would it only be useful for something like crane loads?

Hmmm, you have hit the conundrum on the head.

You can run quasi-static analyses using COM624, but you have to be careful to use an "appropriate" number of load cycles at the current load level to get a reasonable response. COM624 is still a lot better at predicting the moment distribution than other approaches, such as elastic half-space modelling. But be careful with deflections - all of the currently available solution techniques do a relatively poor job of predicting deflections (including COM624 and LPILE - COM624's commercial successor.)

COM624 is better with wind loading than earthquake loading - load application is slower and more akin to the conditions assumed for "cyclic loading." It isn't a perfect modelling match, but it's pretty good. It's very similar to the wave loads that are experienced by offshore oil exploration platforms. Cyclic p-y curves were designed to model the soil response to repeated wave loads on pile foundations. COM624 resulted from decades of research sponsored by the oil industry - particularly Exxon and Shell.

I suggest that you purchase a copy of LPILE from Ensoft in Austin, Texas. Not only will you get a GUI interface, you will get technical support. If you aren't familiar with p-y curves, then the support alone is worth the price of the program. Talk to Bill Isenhower - a friend and classmate at The University of Texas. He's very knowledgeable in this subject area -