Peer-to-Peer Standards Exchange

 Scaling of MRSA Drifts

Jump to Best Answer
  • ASCE Standards
  • ASCE7-16
  • ASCE7-22
  • Seismic
  • Earthquake and Earthquake Loads
Matthew Bosch-Willett's profile image
Matthew Bosch-Willett posted 03-08-2023 12:38 PM

ASCE 7-16 (and 7-22) section 12.9.1.4.2 states that, if the MRSA drift base shear is < Cs*W where Cs is determined in accordance with Eq. 12.8-6 (or 12.8-7 in ASCE 7-22), then the MRSA drifts are scaled up by Cs*W / Vt.  There seem to be two interpretations of this provision: (a) scale up to Eq. 12.8-6 even if S1 < 0.6, and (b) scale up to Eq. 12.8-6 only if the ELFP is controlled by Eq. 12.8-6 (which would not happen when S1 < 0.6).  In scenario (a), for long-period buildings in sites with S1 < 0.6, Eq. 12.8-6 might produce a larger Cs coefficient than the ELFP results.  In scenario (b), no scaling is required when S1 < 0.6.  The commentary for this section seems to support scenario (b), but many engineers prefer scenario (a).  Which interpretation is in accordance with the code's original intent?

Ronald Hamburger's profile image
Ronald Hamburger posted 03-08-2023 02:12 PM Best Answer

The reason ASCE 7 Section 12.9 requires scaling up modal response forces is because engineers may model a structure using assumptions that results in an overly flexible model.  Since use of an overly flexible model will result in a period T that is excessively large, it will also result in computation of seismic forces that are too small assuming the computed period T is not in the constant acceleration portion of the spectrum.  However, while forces computed from response spectrum analysis decrease with period, deflection (and drift) increase with period.  Therefore if an engineer develops a model that is overly flexible, the predicted drift from the analysis is already conservative and there is not need to scale it up further.  The one exception to this is for structures located close to major active faults (where S1 is greater than or equal to 0.6g).  On such sites, ground motion can have large velocity pulses, that create very large drift.  On these sites, and on these sites only, it is necessary to scale the drift from modal analysis to the drift obtained using the base shear computed from Eq. 12.8-7.