seismic design of steel water reservoir

[Cherie]

Hi,

I'm designing a 32-foot high, 0.5 MG reservoir in Seismic Zone 4 in California, and the geotech engineer has supplied us with three different design spectral graphs by 1) Campbell & Bozorgnia, 2) Boore, and 3) Sadigh. I've been instructed to use the most conservative of the three. Sadigh results in the most conservative zero period acceleration, but Boore results in more conservative freeboard and shear anchorage requirements.

In past projects, I see that our company has been given a graph containing the average of the three spectral graphs by the geotech engineer, and therefore used that.

Are there any clear requirements by AWWA, or is it up to us as engineers to make a judgement call? Can we just stick with the Sadigh graph to make it simple?

Thanks.

 

[Qshake]

Well, for one thing, using the zero period acceleration is similar to using a prescriptive code limit on the elastic response spectrum. So it's odd to me that you recieved the three spectra for something you could have found yourself in the applicable code.

Also, using the zero period ignores the displacement controlled part of the response spectrum. Or, in other words, there will be a displacement due to ground excitation and you'd like to know what that is. Then you can also ascertain the stability of the tower under loading at that maximum displacement.

Regards,
Qshake

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

 

[JedClampett]

If this is a steel reservoir (you didn't say), there's no reason to use seismic spectra at all. AWWA D100-96 has a defined seismic design procedure that is applicable for all seismic zones.
Unless there is something in the California code that over-rides AWWA D100, I don't see a reason not to use it.

 

[JedClampett]

Whoops, you did say steel in the title.

 

[JStephen]

Jed, AWWA D100 does have standardized seismic design, but also allows for use of response spectra, and perhaps half the tanks in the state are designed that way. Design using the response spectra can give seismic loads double the D100 standard, or higher- it's not a trivial difference.