Date: Wed, 12 Dec 2007 12:24:08 -0500 (EST) From: Jun-Hui Zhao To: Jon Swift Cc: Peter Teuben , smamiriad@cfa Subject: Re: using gmakes and gfiddle on SMA data Hi Jon: I haven't tried 'gmakes' and 'gfiddle' for handling SMA data. SMA data do not supply wide band. We may be able to generate one from averaging the spectral chunks' data. Peter T. might give us suggestions on how to convert line data to wide band data so that one can follow the same procedure of handling hatcreek data for SMA. Please send me your script and I will look into the problems that you encountered. Jun-Hui On Tue, 11 Dec 2007, Jon Swift wrote: > Hi Jun-Hui: > > I am trying to use gmakes and gfiddle to calibrate the amps and > phases of my SMA data. This works smoothly for the LSB data. But for > the USB data (with no data in line=wide,1,1), I'm running into some > problems. Gmakes places the USB gains into the first wide band > channel in the output gains file. Gfiddle can read these gains with > the line=wide,1,1 tag added, and subsequently outputs the fits into > the first wide band channel. When I use gapply to apply these fits to > the data, it doesn't apply them to the USB. I've tried the "single" > and "extra" options with different combinations of line=wide,1,1 or > wide,1,2. > > > I also tried simply combining the LSB and USB calibration data with > uvcal, uvcat and smachunkglue to make them more compatible with the > gmakes/gfiddle/gapply process. Uvcat and uvcal both produce datasets > that look ok with smauvplt, but gmakes complains about the channel > widths differing by more than 10% and refuses to find solutions. > Smachunkglue complains that the variable ischan is not found and > returns with no output. > > Any help or guidance you could provide here with either combining LSB > and USB data into a single dataset with correctly defined wideband > channels, or with applying the LSB fits from gfiddle to the USB data, > would be greatly appreciated. > > Thank you in advance, > Jon > >