Damping Justification for Seismic Analysis
Best practice for justifying the damping ratio used in seismic analysis, covering typical values, the basis for selection, and the consequences of over- or under-estimating damping.
Why damping justification matters
Damping has a significant effect on the seismic response in the amplified region of the response spectrum. Higher damping reduces the spectral acceleration and hence the seismic demand. Overestimating damping produces non-conservative results; underestimating damping produces overly conservative results. The damping ratio must be justified for the specific structure or equipment, not selected generically.
Basis for damping selection
The damping ratio should be selected based on the construction type, the connection detail, and any test data. Welded structures have lower damping than bolted structures. Concrete structures have higher damping than steel. Equipment with internal components may have higher damping than bare frames. If the standard specifies damping values, those values should be used. If the standard allows the engineer to select, the selection should be justified with reference to the construction type and any available test data.
Consequences of incorrect damping
- Overestimating damping reduces the spectral acceleration in the amplified region, producing non-conservative seismic demand
- Underestimating damping increases the spectral acceleration, producing overly conservative demand and potentially over-designing the anchorage or structure
- The effect of damping is most significant in the amplified region — in the rigid region, damping has minimal effect because the spectral acceleration equals the ZPA regardless of damping
- For equipment near the peak of the floor response spectrum, a small change in damping can change the demand significantly
Damping in the response spectrum
The response spectrum is always associated with a specific damping ratio. If the structure damping differs from the spectrum damping, the spectrum must be scaled or a new spectrum generated at the correct damping. Damping scaling is frequency-dependent — the scaling factor is larger in the amplified region and approaches unity in the rigid region. Using a spectrum at the wrong damping without scaling is a common error.
Modal damping
Some solvers allow per-mode damping, which is useful for structures with mixed materials (e.g. a steel frame supporting a concrete mass). The damping for each mode is computed as a weighted average of the material damping based on the modal strain energy in each material. If per-mode damping is not available, a single conservative (low) damping value should be used.
Documentation
- State the damping ratio used and the basis for selection
- If damping differs from the standard value, provide justification
- If the spectrum was scaled for damping, document the scaling method
- If per-mode damping is used, tabulate the damping per mode