processing SAC Ground Motions

ca493 · Post by **ca493** » Fri Dec 28, 2007 2:13 am

Hi Silvia,

you dont happen to have any .tcl script that converts the SAC ground motions into a format that can be read by OpenSees do you?

I can change the heading of the SAC GMs so that they are like the PEER GMs (so they can be read by ReadSMDFile.tcl), however, the SAC GMs have 6 columns of data as opposed to 5 columns in the PEER GMs. OpenSees seems to be able to read it and apply the accelerations but i am unsure if they are correct. Do the 6 columns make a difference?

Thanks!

silvia · Post by **silvia** » Sun Dec 30, 2007 11:14 am

the script reads line by line, so you should be okey.

dobi3 · Post by **dobi3** » Tue Mar 10, 2009 11:14 am

Dr. Silvia,

I am also trying to run SAC GMs, but OpenSees gives me a convergence error at the beginning of the record (3.45 sec). The time step for input ground motion is 0.02 sec, and for the time-step Dt for lateral analysis I select a small value (as little as 0.00005 sec). I tried scaling down the record as well, but nothing helped.

If OpenSees reads the record line by line (and having 6 columns instead of 5 is really not the problem I encounter), do you have any suggestions for the origin of my error massage?

Thank you very much!

dobi3 · Post by **dobi3** » Tue Mar 10, 2009 11:18 am

I forgot to mention that the model runs just fine when PEER GMs records are used. So, I know the problem is in applying the SAC GMs.

silvia · Post by **silvia** » Tue Mar 10, 2009 11:21 am

the peer gm's are scaled by g by the script, as they are specified as a factor of g.
are the sac ground motions already scaled by g? if so, remove the scaling in the script..... it may be 10x too big.

dobi3 · Post by **dobi3** » Tue Mar 10, 2009 11:58 am

That's it! They are not scaled by g and the units are in metric.

Thank you very much, Dr. Silvia!