Walter's LogBook
	September 2005

Overview

This month has had it all -- clear skies, foggy skies, rain, and even a full sky aurora! I've been concentrating on refining variable star coordinates when I can so that all outbursts get detected, as well as a tweak to AcquireImages.wsc to minimize inter-exposure downtime. For September, my outburst detector saved 1018 minutes of exposures (9/1..9/23) or an average of 73 minutes per night!

The SXV-H9 is the imaging camera, with exposures for variable stars of 120 seconds at 4x4 binning (unguided). This month's session notes are below the overview section.

Focuser Temperatures, September 2005

Session Notes

Thu/Fri September 1/2, 2005

The Clear Sky Clock called for deteriorating skies starting around 3am. Sure enough, I awoke to the sound of music playing on the PC speakers (first time this has happened in many months!). Clouds started causing plate solve failures at ~02:10.

I was late realizing that the skies would be usable tonight, so I only got my bias frames plus a single dark frame done at the beginning of the session. So I did dark frames (a complete set of 20 since I threw out the initial one I took earlier) after it clouded over. I closed the dome by software while the darks ran which put the temperature up about 0.5C. Darks were taken within 1.5C of the temperature at the beginning of the session.

I processed the images before going to bed, and fine-tuned the position of V344 Lyrae (which was visible in tonight's stacked image!) taking the position error from (x,y) = (1,1) to (0,0).

Fri/Sat September 2/3, 2005

I was going to open the dome at 19:30, but when I looked out the living room window it was raining! I was able to open the dome just after I started doing dark frames (the rain showers had passed and the sky had finally cleared).

I tweaked the coordinates for TU Tri.

Sun/Mon September 4/5, 2005

V1008 Her was in outburst tonight! (I think this is the first time I've seen it!) I am dropping EY Cyg from the plan since it has dropped into the top of the tree. That tree has grown quite a bit this year and needs to be trimmed back significantly!

Mon/Tue September 5/6, 2005

I walked up Victoria Street to where the new subdivision starts and saw Venus and Jupiter together low in the NNW just after 8pm. Very nice!

Usually there is ~10sec between exposures when running an ACP plan, but tonight that gap is only ~2sec, so more stars were covered tonight. With the outbursts detected, ACP came within 4 minutes of finishing the plan tonight! Wow!

I fine tuned the coordinates for: V1008 Her, AY Lyr, V368 Peg, TW Tri, WY Tri, AF Cam, FT Cam, SS Aur, FS Aur. They now show zero error in my nightly Quick Look file (except for V368 Peg which I'll try again with a better image from a different night).

Wed/Thu September 7/8, 2005

I walked up Victoria Street to the new subdivision to see the crescent moon, Venus, and Jupiter lined up horizontally low in the west sky. Very nice. A lady walked by on the street (headed west) and didn't even notice them!

ACP woke me up at 2am -- the sky had clouded over, so I closed the dome from the control room. I sleep better when the dome is closed (30% POP tonight, 60% by morning).

Thu/Fri September 8/9, 2005

The night started clear (after a day of rain) but got foggier as the night went on. About 4am ACP woke me up since the scope got lost, so I shut everything down (and closed the dome by remote control).

Fri/Sat September 9/10, 2005

In addition to the usual all-night ACP imaging run, I also did some visual deep sky & variable star observing with the C8 and 17½" in the back yard from just after 2am until 5am.

Sat/Sun September 10/11, 2005

I accidently left the World Community Grid agent running all night and its demand on the CPU reduced tonight's exposure time by 45 minutes (the night's efficiency dropped from ~88% to 82%). D'oh!

Sun/Mon September 11/12, 2005

With the current solar activity (even featuring a naked eye sunspot that I've observed with my #14 welders glass) it was bound to happen: an all-sky aurora. Fortunately the CCD was able to pump through it.

Tue/Wed September 13/14, 2005

My outburst detector failed to detect FO And, so I did a plate solve and tweaked the coordinates in wVariables.wsc. After re-running the offline version of the detector FO And was properly detected. This would have saved 14-16 minutes of exposure time last night!

TY Psc Outburst, 2005 Aug-Sep

TY Psc Movie (97k AVI file)
 

Image Dates				9	2005 Sep 06
1	2005 Aug 22	5	2005 Aug 28	10	2005 Sep 08
2	2005 Aug 24	6	2005 Sep 02	11	2005 Sep 09
3	2005 Aug 25	7	2005 Sep 04	12	2005 Sep 10
4	2005 Aug 26	8	2005 Sep 05	13	2005 Sep 13

Sun/Mon September 18/19, 2005

It's been 5 nights since the sky has been clear enough for imaging. V1028 Cyg was reported to be in outburst during that time (d'oh!) but the outburst only lasted three days (based on AAVSO Quick Look data) so I missed it. V1363 Cyg is dropping into the tree now, so I'm dropping it from the plan.

Fri/Sat September 23/24, 2005

I decided to look at how much time the outburst detector is saving (see data below). The total is 1018 minutes over 14 nights for an average of 73 minutes per night!

I've tightened the ACP max pointing error from 3 arcmin to 2 arcmin. As well I've put some print statements into AcquireImages to see if I can determine where the 25 seconds per image is going (this adds up to almost an hour a night). My suspicion is that it is the HFD calculation that is the culprit (in which case I could disable that since it is not needed) but it could be the saving of the image to hard disk (not much I can do about that).

Night	Exposures Skipped	Time Saved		Night	Exposures Skipped	Time Saved
2005/09/01	12	24 min		2005/09/09	55	110 min
2005/09/02	31	62 min		2005/09/10	46	92 min
2005/09/04	43	86 min		2005/09/11	25	100 min
2005/09/05	50	100 min		2005/09/13	31	62 min
2005/09/06	49	98 min		2005/09/18	28	56 min
2005/09/07	35	70 min		2005/09/21	22	44 min
2005/09/08	39	78 min		2005/09/23	18	36 min

Thu/Fri September 29/30, 2005

I started running NISTime all night with it set to update the computer clock every hour. This no longer screws up MaxIm's exposure counter (as it did years ago under Win98), so now my computer clock is always within 2 seconds of the correct time.

V368 Peg Outburst, 2005 Sep

V1008 Her Outburst, 2005 Sep