Designing For Geotechnical Acceleration vs UBC

[PEStructural]

I am designing a 125 foot diameter steel flat bottom water tank. The geotech report included a probabilistic seismic risk evaluation which they gave as a tripartite response spectra.

My question is, when comparing the acceleration from the geotech to the acceleration computed from the UBC, are both accelerations apples and apples or is one based on working stress and one based on strength design? (i.e. the 1.4 factor comes into play).

Any help is greatly appreciated.

 

[Focht3]

Hmmm,

You should ask your geotechnical engineer that question. My gut says the two values should have the same basis for use - but this may not be right. Ask.



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora. See faq158-922 for recommendations regarding the question, "How Do You Evaluate Fill Settlement Beneath Structures?"

 

[doejohn]

I agree with Foch3, you'll have to ask your Geotechnical Engineer. The following is only general info...

The UBC, as typical with structural design codes and standards, defines most loads (D=Dead, L=Live, etc.) at "service level" (meaning roughly a maximum value that may be experienced in typical use), the load level at which ASD provisions have been "calibrated". Alternately, strength design provisions are calibrated against load levels judged to have a sufficiently low probability of being exceeded during extreme situations, load levels which calculated by factoring the service loads D, L, etc. But you knew all that...

The point is that earthquake loads E are the exception to the rule, in that they are explicitly defined at the "sufficiently low probability of being exceeded" level in the UBC. Per 1631.2, this is a 10% exceedence probability in 50 years [the code-assumed design life]--a 475-year earthquake (for std Poisson temporal distribution), like one thinks of a 500-yr flood. Although this code section is for dynamic analysis, this exceedence level is the intent behind all the UBC provisions. For service-level forces, E is reduced by 1/1.4.

Now the standard procedures for conducting a probabilistic hazard analysis result in ground motions that have some exceedence probability explicitly targeted by the investigator. So for what probability level did your Geotech provide results? 10% in 50 years (475-year return period)? If so, it corresponds directly with UBC provisions for E, a "strength" force level. If not, they don't.