Seismic Design/Richter scale 2

[spike]

We have an inquiry to furnish the design of a piece of equipment to withstand a 7.0 to 7.3 earthquake based on the Richter Scale.

Any ideas as to how the design forces could be calculated when a building code procedure is not the basis?

 

[Qshake]

In my opinion someone needs to educate the client on how seismic analysis and design proceeds and the development of seismic design critiera.

One may think that it's enough to tell the client we understand what you're asking and we'll sort through the design criteria to give you what you think you're asking for. But the reality of it is: they don't know what they're asking for and don't understand the risk involved.

Richter by the true definition isn't well defined outside of California or outside of a 100 mile radius of the actual epicenter and must be recorded on a wood-anderson seismograph. It's only thanks to the media that we have come to know the Richter scale as being the all-encompassing scale by which earthquakes are measured. There are probably 20 such scales in the US alone and several more in Japan.

That said, there are many magnitude relationships that help relate one scale to another. But that is really outside of what you need to do.

Another confusing issue is that a magnitude 7 is well within most life-safety design guidelines. That is most, if not all codes, state minimums for which a structure is expected to survive for reasons of life-safety. We can certainly design to a higher standard - and sometimes do (Nuclear facilities for example). For some areas of the US designing to an event with a 10% probability of exceedance in 50 years will result in a return period for earthquakes of 475 years. Although we understand that such an event can occur at any moment, the event would generally be a large earthquake say 6.8.....7.0.

Nowadays some codes use both life safey and Maximum Credible Event (MCE). The MCE may have a return period of anywhere from 1000 yrs to 2500 yrs. This would result in a very large event.

Also, it makes no sense to design for a specific magnitude but rather a range. this is a benfit from the use of Response Spectra. I highly recommend that it be used.

Go back to the client and educate them!

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.