Seismosoft | Earthquake Engineering Solutions

Hi all,
I'm trying to add individual modal responses to get full response. But all response histories don't have same time-step.

mode no. T(sec) time-step in response history
1 0.48 0.00333
2 0.16 0.00143
3 0.096 0.001

how do I get all time-histories with same time step (say 0.001 s)?

thanks!

Open the Help System of the program, and search for "time-step".

Seismosoft Support

Thank you for reply. It seems that "time-step" function changes no. of datapoints in accelerogram. We can reduce/double the time-step as required. But my question is about response time-history of SDOF systems and not the accelerogram.

In the program: Elastic Response Spectra > Response Time-Histories > Period of SDOF system > Values > Refresh

How to control time-step in this tabular result of time-history?

thanks!

Hi awanish. In the Settings, you do have control over the parameter "max dT/T". This might be used to achieve your desired result depending upon the modes you are looking to solve.

Note that, at least on my computer, changing this parameter after the input time history has been analyzed does NOT have any efect on the time step used. You MUST set this parameter when you first open SeismoSignal. So you'll have to re-start the program to get the changes to take effect unless your system behaves differently than mine.

Say, for example, that you are interested in periods of 0.2 and 0.6 seconds, and that your input time history has a time step of 0.005 seconds. With dT/T set equal to 0.01, the software will use dT = 0.005 seconds for the T = 0.6 case (because 0.005/0.6 = 0.0083 < 0.01 ->ok). With dT/T set equal to 0.01, the software will use dT = 0.00166667 (because 0.005/0.2 = 0.025 > 0.01 -> no good). The software uses the first fraction of the input time history time step which meets the dT/T criteria.

0.005/2 = 0.00250, 0.00250/0.2 = 0.0125 > 0.01, no good
0.005/3 = 0.00167, 0.00167/0.2 = 0.0083 < 0.01, ok.

So, for this example, you would need to run SeismoSignal twice, with two different setting of the dT/T parameter.

(1) Run with dT/T = 0.01. Use the results for a period of 0.2 seconds. The output time step will be 0.00167 seconds.

(2) Run with dT/T = 0.0028 (since you want to force the program to use a time step of 0.00167 seconds to get results for 0.6 seconds to have the same time step as for a period of 0.2 seconds - 0.00167/0.6 = 0.0028). Use the resuts from this run only for the 0.6 second period. The output time step will be 0.00167 seconds (the time step for T = 0.2 wil be musch smaller now, hence the need for the run in case 1).

I hope this makes a littel sense, awanish. If you try it a few times I think you'll start to see what's going on.

Best of luck.

Thank you very much[:)] You took out so much time to explain!