Martha starts 08.12.12 31Jul09 - RG: problem with grids at 1452, 1500 and 1508 found by R. Koopmann, arising from bad wapp behavior, corected by flagging away chans surrounding galactic HI. See message to becky appended at bottom of this file. ** Observer notes intermitted power increases in Beam 0.1 ** similar to problems seen other nights. Again, it shows a very ** high Tsys and power jumps, so I toss it. ** The drift is 2p1 +002154 ** This is a spring drift.... ** The pos file name is therefore: pos_080501_+002154s.sav ** -- Working in: /home/rutados/data/idl/work/proc/08.05.01 cp /home/dorado13/galaxy/idlraw/08.05.01/*list* . gedit cal080501.list gedit drift080501.list >idl >@wasinit2 >@alfinit -- > callist='cal080501.list' > dir='/home/dorado13/galaxy/idlraw/08.05.01' > calib1,callist,dcalON,dcalOFF,dir=dir > help,dcalon,/st > calib2,dcalON,dcalOFF,ncalib > Any bad boards in system? y How many bad boards? [enter n] : 1 Enter nbeam, npol of bad board : 0,1 > help,ncalib,/st > save, ncalib,file='ncalib_080501.sav' ** There are 18 cal values for this run *** ** The calibration looks ok after Beam 0.1 dropped ** One high Tsys point, Beam 4, dropped -- Set up and run the BPD gui. > bpdgui Enter the new calsession: calsession_080501.sav Enter the new runpos: pos_080501_+002154s.sav Be sure to set CRAT to "fit", not "median" **The interp regions here are [3485 to 3523].** -- restore,'pos_080501_+002154s.sav' help,pos ** There are 18 drifts for this run ** -- restore,'pos_080501_+002154s.sav' -- restore,'d214400+294831.526180333.sav' -- flagbb,dred,cont_pt,mask,pos,GAUSAV=11,HAN=3,AGC=1,n1=0,n2=6 save,pos,file='pos_080501_+002154s.sav' -------------------------------------------------------------------------------- d134526+002155.812200174.sav q1 1 AGC some spikes 2.0, sw 1.0/1 d135532+002155.812200176.sav q1 nada GPS 81-322(dbl) d140539+002155.812200178.sav q1 nada d141549+002154.812300002.sav q1 nada GPS 561-599 d142555+002155.812300004.sav q1 3 AGC GPS 0-57 d143603+002155.812300006.sav q1 1 AGC d144610+002155.812300008.sav q1 1 AGC cont 4,3,5 d145617+002155.812300010.sav q1 1 AGC huge GPS 377-467; toss recs 397-453 d150625+002155.812300012.sav q1 1 AGC cont 4 d151631+002155.812300014.sav q1 1 AGC d152638+002155.812300016.sav q1 nada d153646+002155.812300018.sav q1 1 AGC cont 3,5 & 6,1 d154652+002155.812300020.sav q1 nada d155659+002155.812300022.sav q1 1 AGC cont 6,1 d160706+002155.812300024.sav q1 1 AGC some picket last 150 recs d161714+002155.812300026.sav q1 1 AGC some picket last 200 recs d162720+002155.812300028.sav q1 nada d163727+002155.812300030.sav q1 nada GPS 300-519 save,pos,file='pos_080501_+002154s.sav' strip_pv,dred,cont_pt,msmooth,n1=0,n2=6,showpol=3,gausav=11,han=5,rfimod=0,units=1 inspect,msmooth,showpol=3,nstrip=4,agc=1 reviewbb,dred,cont_pt,mask,pos,GAUSAV=11,HAN=3,AGC=1,n1=0,n2=6 > print,where(pos.badbox[0,0,0,2] eq 0) Overall this is a good run though Beam 1.1 is dropped because of power issues. Beam 2.0 shows a characteristic picket fence 1400-1430 MHz. There are some power issues elsewhere on occasion. And, of course, GPS. ________________________________________________________________________________ Message from Becky, Jul 30 2009: Hi Riccardo and Martha, Help! I'm having some issues with 3 of the +01 grids near our group NGC 5846 and need an expert opinion. I've attached screen shots showing the problems. In 1500+01, weird stuff is going on across the grid for decs of 0016 and 0026: (1) looks like rfi or low weight issues between channels 560-740, yet the weighting looks normal, except for a small area at high RA end. I checked out the drifts associated with those decs, and found a possible problem in d145627+002155. This drift (screenshot attached) has strong GPS and there is residual signal around the galaxy corresponding exactly to the channel range of the problem in the grid. (2) I don't understand the areas that seem to lack data, one on each side of the grid. The one at the high RA end corresponds to a low weight area in the weight window, but the other side is weighted normally?? I don't see any record of missing data in the flagbb logs, both of which were done by Martha. There is a note of problems with d151443+001437.814700133.sav : "** Note added after BPD: Everything ran fine, but it appears that the positions are screwy in some beams/records for the drift starting 1514. I am not sure what happened, but it seems the best approach is to drop this drift." I see this in 1516+01 - the weighting looks about half as it should be if that drift were taken out. The problems with 1500+01 seem to extend into 1452+01 and 1508+01 - same dec. Please see screenshots. What do you think? Was there some glitch in the drifts while gridding? Or do you think I did something wrong when bf-ing? I used the accelerator option (by hand in the galactic regions). Thank you! Thanks, too, Martha, for a useful telecon yesterday! :) Becky -------------------- RG responds on July 31: Hi becky, I think i found the problem with the +01 grids you mentioned. If you look at drift d145617+002155.812300010.sav, using flagbb, you'll notice that there is a broad "haze" around galactic HI, which is there in all the beams except 5,1. I don't know what caused it; it is not in the preceding or the following drifts, so it got turned on and off by the starting and stopping of taking data, so i presume it is due to a wapp glitch. The result of that was an unflagged, elevated plateau in the baselines, over 160 channels including galactic HI. When grid_prep combined that data with those from other drifts (N and S) overlapping the dec=002155 one, it resulted in an elevated baseline for all the channels of the plateau and all the RA's touched by the drift - except for the RA's blanked away when the GPS episode blew the whole spectrum away near ra=1503, which had been flagged bad. Hence that episode appears as a negative feature on all beams. Likewise, you get steps at the beginning and end of the bad drift, affecting the grids to E and W. I have flagged the "haze" away from d145617+002155.812300010.sav, created a new pos file and am now making a new "1500+01a" grid for you. I hope that will fix the problem. However, you'll have reduced sensitivity over that spectral region over 10min ra and the dec range of the drift. It's unfortunate that the data on that region is rather poor, but as Zarathustra oncce said, thus it crumbles, cookie-wise. I'll let you know about the new grid in a few hours. If the fix works, I'll redo the other two grids. - riccardo