instrument response, evalresp, phase sign

Users Mailing List Archive
#1

Hi Obspy users,

I have a question about obspy.signal.invsim.evalresp
http://docs.obspy.org/packages/autogen/obspy.signal.invsim.evalresp.html
that returns the full transfer function from a RESP file (from dataless SEED files)

I compared the output transfer function with the output of evalresp software from Iris
http://www.iris.edu/manuals/evalresp.htm
The obspy’s transfer function appears to be the complex conjugate of the iris’ one (no effect on the amplitude but the phase is multiplied by -1).

There is a note on the obspy documentation for obspy.signal.invsim.pazToFreqResp
http://docs.obspy.org/packages/autogen/obspy.signal.invsim.pazToFreqResp.html
specifying that we need to conjugate the output frequency response to get the actual phase function.

Is this convention also true for obspy.signal.invsim.evalresp?
Does any one already observe this?

Regards,

M.L

#2

Hi,

Thanks for pointing this out. It looks like the complex conjugation was never needed and I therefore removed it. You can get the new version of *obspy.signal.invsim.evalresp* by updating your installation to the developers version.

Let me know if you encounter further problems.
Cheers,
Yannik