Skip to content
Queensferry

Cumulative tutorial 12

What a spectrum analysis does not tell you

The same building, the same record, two methods, and the gap between them.

The three-storey building has been analysed twice against the same record: once by response spectrum, once by time history. The two do not agree exactly, and the gap between them is not an error in either — it is a consequence of what the spectrum method throws away.

Quantify what the spectrum gives, what it discards, and how large the discarding is on this building. Then decide whether the extra cost of the time history bought anything worth having.

What you are given

  • Modal base shears: 5.598, 0.429 and 0.059 MN
  • SRSS total 5.615 MN; absolute sum 6.087 MN
  • Seismic weight 11.77 MN

Tutorial 12

Work the problem

4 questions

  1. Question 1

    What fraction of the SRSS base shear comes from mode 1 alone?

  2. Question 2

    By what factor would adding the modal peaks in absolute terms over-estimate the answer?

  3. Question 3

    What is the base shear coefficient from the spectrum analysis?

  4. Question 4

    A time history of the same record gives a peak base shear of 5.34 MN. By what percentage does the spectrum analysis exceed it?

What this establishes

What the spectrum discards is time. It cannot tell you when the peak occurred, in what order members reached their worst, how many large cycles there were, or what the structure was doing at the instant a particular column peaked. If none of those matter, the spectrum is the cheaper and better-founded choice.