From bec@haystack.mit.edu Mon Jul 12 18:32 MET 1999 Received: from ivscc.gsfc.nasa.gov by picasso.geod.uni-bonn.de with SMTP (1.38.193.4/16.2) id AA12441; Mon, 12 Jul 1999 18:31:47 +0200 Return-Path: Received: from dopey.haystack.edu by ivscc.gsfc.nasa.gov; Mon, 12 Jul 1999 12:30:54 -0400 Received: (from bec@localhost) by dopey.haystack.edu (8.8.6 (sendmail_886_v2)/8.8.6) id MAA14764; Mon, 12 Jul 1999 12:30:41 -0400 (EDT) From: Brian Corey Message-Id: <199907121630.MAA14764@dopey.haystack.edu> Subject: Re: Spurs at MATERA at 8512.99MHz (fwd) To: weh@ivscc.gsfc.nasa.gov Date: Mon, 12 Jul 1999 12:30:41 EDT Cc: lpetrov@ivscc.gsfc.nasa.gov, dgg@leo.gsfc.nasa.gov, dbs@leo.gsfc.nasa.gov, weh@leo.gsfc.nasa.gov, cma@leo.gsfc.nasa.gov, nrv@leo.gsfc.nasa.gov, tac@leo.gsfc.nasa.gov, ckodak@leo.gsfc.nasa.gov, bec@ivscc.gsfc.nasa.gov, colucci@asi.it In-Reply-To: <378A0B2F.E342DCE2@ivscc.gsfc.nasa.gov>; from "Ed Himwich" at Jul 12, 99 11:35 am X-Mailer: Elm [revision: 109.18] Status: R I've been looking at the pcal and fringe data for Matera from CA063 and CA062 this morning, hoping to find hints of what's wrong. No answers yet, but here's what I've found so far... X channel 3 is nearly as bad as channel 4. The pcal data for the other 6 channels look OK. The pcal amplitude variations in channels 3 and 4 are in fact remarkably similar. The one thing that those channels have in common, and that makes them different from all others, is that they use the IF1 high output. Just a coincidence? There are short-time-scale (10-30 minutes) variations in the pcal phase differences between channels 3 and 2, and between 4 and 5, that look suspicious. Nothing comparable happens for other adjacent channel pairs. So I suspect the pcal phases really are being corrupted. (No surprise.) I tried looking at the fringe phase channel differences for confirmation, but there are too few high-SNR points to draw any conclusion. The fringe amp ratios between channels tend not to vary as the pcal amp ratios do, which again points to a pcal-only problem. I noticed in the logs for the 2 experiments that there appears to be a rolloff in the X-band IF input level of ~20 dB between the low and high frequency ends. (In fact, all VCs above #3 are being operated with 0 dB video attenuation.) The relative weakness of the signal at the VC03 and VC04 IF frequencies could be responsible for the pcal problem -- e.g., spurious signals generated from the low-frequency pcal tones could be overpowering the pcal at freqs 3 and 4. If that's what's happening, then the usual test of looking for spurious signals by turning off phase cal won't find anything, because the spurs are being derived from the phase cal itself. Unlocking the receiver LO should reveal them, however. Does anyone know how long this problem has been present? My only access to Matera pcal data is through the fringe data at WACO (I don't do data bases), so my time horizon extends only a very short way into the past. Whatever the outcome of this problem, Matera should add a filter to flatten their X-band IF passband. --Brian