Seismosoft | Earthquake Engineering Solutions

Dear Sir/Madam,

I am analyzing characteristics of particular set of time histories.
If I look at the Acceleration and Velocity spectra (obtained from Seismosignal) for these histories, I have non zero values of spectral ordinates for periods in range 0.1 to 0.5.

But, Housner intensity, ASI and VSI is shown to be zero (Ground motion parameters, Seismosignal)

What could be the possible reason ?

Regards,

Hi amruti.

I have never observed this behavior before. I wonder if you could try loading the same accelerograms into SeismoSpect, which also can produce GMP's. It would be interesting to see if there is a difference between SeismoSignal and SeismoSpect.

Best of luck amruti.

My data is acquired at frequency 5000 HZ( 5000 acceleration readings per second). Does it have to do something with the problem ?

If you would like to send the accelerogram file to me, I will try to reproduce the error on my computer to either confirm or rule out the possibility that it is a processor issue.

I have sent you the mail

Hi amrutj.

I have loaded your accelerogram into both SeismoSignal and SeismoSpect. Housner intensity, ASI, and VSI were successfully computed in both programs. The discrepancy is in the Effective Design Acceleration, for which the two programs report differing results. This could be a bug in one or the other.

The appended data summarizes the ground motion parameters in question. Note that SeismoSignal cannot compute Fourier and Power Spectra, nor can it compute the mean period, for records with more than 32,768 points, as is the case with your accelerogram.

So, given that the results you question were computed successfully by SeismoSignal on my computer, I wonder if there is something simple that you need to do in your settings - perhaps increasing the number of decimal places to 10 (this is the value I use and it often helps to use a larger value than the default).

Best of luck amrutj.

Acc Spectrum Intensity (g*sec):
SeismoSpect 2.615190251
SeismoSignal 2.61518787
Ratio 1.00000

Vel Spectrum Intensity (cm):
SeismoSpect 450.1438021
SeismoSignal 450.145213
Ratio 1.00000

Housner Intensity (cm):
SeismoSpect 250.4167476
SeismoSignal 250.4146638
Ratio 0.99999

Effective Design Acceleration (g)
SeismoSpect 2.931545647
SeismoSignal 6.390391003
Ratio 2.17987 (There does appear to be a problem here)

Time History of acceleration from file: SP51.txt obtained from SeismoSoft forum user amrutj

Thanks a ton for your time

I have tried increase number of decimals to 10 before I submitted query. It did not work in my case. I will try to find what could be possible reason.
Effective Design Acceleration discrepancy is the new problem that has arisen [:(].. Please let me know, If you find the bug


Thanks once again

Interesting observation

I tried to filter input time history with lowpass filter (butterworth)
If order is 4 we get same ans as Seismo Spect for EDA.
If order is 1 EDA by seismosignal is 6.46

Note that the filtering and baseline correction parameters have to be equal in both programs (they are set in the first page in SeismoSignal and in the settings in SeismoSpect).

If you suspect that there is a bug in one of the two programs, can you please send us the input file(s) at support@seismosoft.com, so that we fix it?
Thanks in advance

SeismoSoft Support

I have sent the mail