From george at gly.bris.ac.uk Thu May 7 09:26:26 2009 From: george at gly.bris.ac.uk (George Helffrich) Date: Thu, 7 May 2009 17:26:26 +0100 Subject: [sac-dev] Polezero rewrite, features, and demo macros to replace manual figures In-Reply-To: References: Message-ID: Dear All - Here is my rewrite of the transfer documentation, including the points about responses that Renate Hartog made about units and SAC's indifference to them (in general). I've made changes to MacSAC (sorry, Brian) to fix the problem that Arthur reported about responses in the polezero directory (a typo in a string constant and an incorrect subroutine call), and added a SUBTYPE to the NONE response to document resulting physical units. It is described in the documentation, but, briefly, if your pole-zero response yields velocity or acceleration, then you can say something like transfer from polezero subtype funny.pz to none subtype vel which will label the output units in the trace as being velocity (or acceleration or displacement depending on whether SUBTYPE respectively is VEL ACC or DIS). This is motivated by the EVALRESP options and the VEL and ACC transfer types (but without the implicit multiplication by omega). -------------- next part -------------- A non-text attachment was scrubbed... Name: transfer Type: application/octet-stream Size: 22884 bytes Desc: not available URL: -------------- next part -------------- Something surprising that emerged from this change is that apparently SAC's preferred acceleration unit is cm/sec**2. Odd: displacement in nm, velocity in nm/sec and acceleration in cm/sec**2. Perhaps 981 is a sacred number rather than 9.81e9. I've also defined a demo subdirectory in SAC's default macro directory and am starting to place pointers to it in the documentation. This replaces those hard-to-reproduce figures in the old paper SAC documentation. For example, here's the demo macro for the TAPER command: -------------- next part -------------- A non-text attachment was scrubbed... Name: demo Type: application/octet-stream Size: 76 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: demo-taper Type: application/octet-stream Size: 452 bytes Desc: not available URL: -------------- next part -------------- (Beware that it uses some MacSAC-specific features to restore user graphics settings to their values prior to the demo, and to check for an active graphics device.) Scanning the documentation, there are what look to be SAC PCF- or PCM-generated figures for BANDPASS, BEGINFRAME, CONTOUR, GTEXT, HILBERT, PLOT1, PLOT2, PLOTC, SYMBOL, TAPER, WINDOW, and XVPORT. So with this TAPER offering, we have 1 down, 11 to go. Any code packrats have old PCF files available, or are there volunteers to back-engineer them? On 30 Apr 2009, at 16:24, Arthur Snoke wrote: > In preparing an update for the polezero section of the transfer help > page, I decided to try it (I use polezero files, but not within SAC). > I found that the sample given in the current help message does not > work, and would appreciate comments from those who have used it > (successfully). > > Also, the documentation claims that the search will look for a > polezero file in ${SACHOME}/polezero/. I tried that, but it did not > work. > > The sample given is > > EXAMPLE: suppose the file was named sro.pz and you want to remove the > instrument response from station ABC.Z. > > u: READ ABC.Z > u: TRANSFER FROM POLEZERO SUBTYPE SRO.PZ TO NONE > > Because the displacement response for any seismometer I have worked > with is zero at zero frequencies, the polezero transfer function is > zero at zero frequency. Taking out an instrument means dividing by > the transfer function in the frequency domain, so the above expression > will either blow up or just give an output waveform that is a sinusoid > at the fundamental frequency. > > I got results that compared well with my home-grown method within the > passband of interest by using the line > > transfer from polezero subtype M06C.pz to NONE FREQ 0.001 0.007 0.1 > 0.2 > > The seismometer is an STS-2 and is the BHZ channel at 40 sps. I had > already used decimate (three times) to reduce the sampling rate to 1 > sps. > > I typically run the equivalent to SAC rtr before doing the instrument > correction. Should that be included in the sample? > > I then moved M06C.pz to my ${SACAUX}/polezero directory, I got the > following: > > SAC> transfer from polezero subtype M06C.pz to NONE FREQ 0.001 0.007 > 0.1 0.2 ERROR 108: File does not exist: > SAC> ls /usr/local/sac/aux/polezero > M06C.pz llsn > SAC> > > If it does not work as advertised, we should either remove the > polezero subdirectory or make it work. Comments? > > Arthur Snoke > _______________________________________________ > sac-help mailing list > sac-help at iris.washington.edu > http://www.iris.washington.edu/mailman/listinfo/sac-help > George