From: Beverly LaMarr Message-Id: <199911231951.OAA21045@space.mit.edu> Subject: Re: new bad pixel? In-Reply-To: <199911222111.QAA07416@wiwaxia.mit.edu> from Dan Dewey at "Nov 22, 99 04:11:19 pm" To: dd (Dan Dewey) Date: Tue, 23 Nov 1999 14:51:23 -0500 (EST) Cc: fkb (Frederick K. Baganoff), gea (Glenn E. Allen), mwb (Mark W. Bautz), pgf (Peter G. Ford) X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Status: R Content-Length: 1299 Dan Dewey said: > Taking a look at the data from obsid's 336 and 1481 > carried out before and after the day 323 radiation pass > ( ~ 3 days ago ) I see a jump in the S5 count rate > once I've selected all "good" events ( that is > - eliminated grades 1,5,7 > - eliminate ENERGY below 100 , above 10,000. > - eliminate S4 streaky events ) > > The rate of good events in '336 in S5 is ~1.66/frame > and in '1481 2.04/frame. The arrival of the '1481 > events is not poisson either, it appears as if there > is almost always an event per frame... > It looks like a new bad pixel appeared, > around chipx, chipy ~ 700, 150 in S5 - > I see it in Obsid 1481 but not in 336 !?! > Okay, I see a pixel at 721, 110 (counting from 1) on s5 which had 16470 events in 16898 exposures. But, if I look at the bias for that science run I see that the overclock corrected value of that pixel is -121, so I would say that the problem is in the bias NOT the data. Oddly enough, this has happened before on s5. I haven't noticed it happening on any of the other devices. Peter reminded me that all you need is one very low conditioning exposure to make this happen, but doesn't -121adu seem awfully low? Bev