Figure C-5 Spectral Acceleration in the High Frequency Region

DOE-STD-1020-2002

using constant response reduction coefficients, R or inelastic energy absorption factors, F; (2)

to account for the fact that stiff structures may not be as stiff as idealized in dynamic models; and

(3) earthquakes in the eastern U.S. may have amplification extending to lower periods or higher

frequencies than standard median design response spectra. Constant factors permit the

elastically computed demand to exceed the capacity the same amount at all periods. Studies of

inelastic response spectra such as those by Riddell and Newmark (Ref. C-12), indicate that the

elastically computed demand cannot safely exceed the capacity as much in the low period region

as compared to larger periods. This means that lower inelastic energy absorption factors must be

used for low period response if the actual spectra are used (i.e., the inelastic energy absorption

factors are frequency dependent). Since constant inelastic energy absorption factors are used

herein, increased spectra must be used in the low period response region. Another reason for

using increased spectral amplification at low periods is to assure conservatism for stiff structures.

Due to factors such as soil-structure interaction, basemat flexibility, and concrete cracking,

structures may not be as stiff as assumed. Thus, for stiff structures at natural periods below that

corresponding to maximum spectral amplification, greater spectral amplification may be more

realistic than that corresponding to the calculated natural period from the actual spectra. In

addition, stiff structures that undergo inelastic behavior during earthquake ground motion soften

(i.e., effectively respond at increased natural period) such that seismic response may be driven

into regions of increased dynamic amplification compared to elastic response.

Figure C-5 Spectral Acceleration in the High Frequency Region

for an Example Design/Evaluation Ground Response Spectrum

C-17