Chronos slewing problems (was: Pointing Plate Solve Problems)

[melhelm]

Hello Bob:Any time the scope slews to an object to image, the first image is always off center. For example, if I do a run of twelve 30-minute exposures, the first image is never centered and every 4th image is not centered, This is because I do a FocusMax run after every 90 minutes. For some reason, after a slew, the plate solve properly centers the mount only after one full exposure. I made a simple demo of short exposures of M27 WITHOUT focusing (see attached log file). It looks like I need to force a repeat plate solve after a slew. Any ideas?Mel

[Bob Denny]

HI Mel --

I see what's happening: Something is not going right with the first reslew. It is not slewing anywhere near to the requested (target) coordinates. This is a failure of the telescope control system. After the first data image, it sees the error is still there and does another reslew without syncing again. This proves that the sync was correctly done by the control system. It just took two slews for the mount to get the scope positioned to the coordinates. The requested coordinates in the two slews were identical, and were the target coordinates you specified in the Plan.

I am not in a position to add a feature that does multiple slews and pointing checks in case the mount doesn't complete a slew to the requested coordinates. The basic rule is "do it right or raise an error (telling us why you couldn't do it)". To avoid confusion, of course the mount may not be perfectly pointed to the requested coordinates in the sky... but when it is asked to slow to soandso, then it should move until its indicated coordinates are soandso. If you need me to explain further, I will be happy to do so. I can also talk with Chronos. Tim Cann is easy to work with and understands issues very quickly. Before doing so, thouogh, I wonder if there is a mechanical adjustment that may be causing this. Basically, the mount is not moving quite as far as it should when told to slew. I'm going to guess that he further it has to slew, the further off it is after the slew. It should not indicate that the slew has completed though, until it has moved the whole way.

05:03:10   Start slew to M 27...
05:03:12   Start rotate to PA 73.0 deg (raw rotator angle 163.2)
05:03:12   (wait for slew to complete)
05:03:36   (slew complete)
05:03:37   (request plan-start pointing update)
...
05:04:14   Plate-solve pointing image.
...
05:04:14   3702 image stars found
05:04:14   1092 catalog stars found
05:04:14   Solved! 192 stars matched.
05:04:15   Average residual is 0.4 arcsec.
05:04:15   Pointing error is 14.344 arcmin @ angle 139.73
05:04:15   True focal length is 1087.3 mm.
05:04:15   True binned plate scales (arcsec/pix): H = 3.07 V = 3.07
05:04:15   True image center (J2000): 20h 00m 16.0s   22d 32' 04.19"    <- It's 14.344 arcmin off
05:04:15   Imager sky position angle is 72.7 deg.
05:04:15   Rotator mech. position angle is 163.0 deg.
05:04:15   [sync] telescope synced
05:04:15   Re-slew to target.
05:04:15   Start slew to M 27...
05:04:16   (wait for slew to complete)
05:04:21   (slew complete)
05:04:21   Target is now centered.   <-- At this point it should be pointed perfectly (within the repeatability of the mount)
...
05:04:42   Imaging to M 27-S001-R001-C001-H
...
05:05:47   Image finished
05:05:49   Plate-solve final image
05:05:50   203 image stars found
05:05:51   1238 catalog stars found
05:05:52   Solved! 123 stars matched.
05:05:52   Average residual is 0.33 arcsec.
05:05:52   Pointing error is 13.629 arcmin @ angle 143.52  <-- Still has almost the same pointing error, even the angle![/HIGH]
05:05:52   True focal length is 1087.5 mm.
05:05:52   True binned plate scales (arcsec/pix): H = 1.02 V = 1.02
05:05:52   True image center (J2000): 20h 00m 11.0s   22d 32' 03.45"  <-- It didn't move at the 05:04:16 reslew anywhere far enough
05:05:52   Imager sky position angle is 72.7 deg.
05:05:52   Rotator mech. position angle is 163.0 deg.
05:05:52   Image FWHM is 1.8 arcsec (1.77 pixels)
05:05:52   (avg FWHM = 2.03 arcsec)
05:05:53   (stopping  the autoguider)
05:05:53   Re-slew to target.
05:05:53   Start slew to M 27...
05:05:54   (wait for slew to complete)
05:06:03   (slew complete)
05:06:03   Excessive pointing error, request pointing update for next target     <-- next TARGET not next EXPOSURE
...
05:06:17   Imaging to M 27-S001-R001-C002-H
...
05:07:23   Image finished
05:07:24   Plate-solve final image
05:07:26   201 image stars found
05:07:26   1238 catalog stars found
05:07:27   Solved! 144 stars matched.
05:07:28   Average residual is 0.33 arcsec.
05:07:28   Pointing error is 0.678 arcmin @ angle 258.17  <- This time it completed the move to the target coordinates 
05:07:28   True focal length is 1087.6 mm.
05:07:28   True binned plate scales (arcsec/pix): H = 1.02 V = 1.02
05:07:28   True image center (J2000): 19h 59m 33.0s   22d 42' 52.85"  <-- Within a half arcmin of the requested coordinates
05:07:28   Imager sky position angle is 72.7 deg.
05:07:28   Rotator mech. position angle is 163.0 deg.
05:07:28   Image FWHM is 2.0 arcsec (1.94 pixels)
05:07:28   (avg FWHM = 2.00 arcsec)
05:07:28   Within max error tolerance, no recenter needed
05:07:28   (autoguider still running...)

[Peter Prendergast]

Bob:

For some time now I have noticed a similiar problem with my Paramount. Not all the time, but typically when ACP asks TSXPro to slew to a target the first slew just isn't anywhere near correct. with NO sync or change the "re slew" is always, and I do mean always, dead on more or less. I think their has been some change in the way TSXPro interacts with ACP. Have others seen this????
Peter

[Bob Denny]

I hadn't heard of this with TheSky X. The interaction between TheSky and ACP is just via the ASCOM driver and that simply converts ASCOM calls to TSX scripting calls to slew. The logic is extremely simple. If a scripted slew call to TheSky results in an incomplete slew, and yet TheSky indicates that the slew completes, this is 100% certainly a bug in TheSky. Nothing has changed with ACP and telescopes, and ACP doesn't really know or care whether the telescope is a Paramount/TheSKy or an AP1200 or ??? It does care whether it's a GEM or simple equatorial of course.

Mel is using a completely different type of mount, TheSky is not involved at all.

It'a amazing how these device/control systems create a whirlwind, attracting all sorts of apparently similar problems which are actually unrelated, ha ha. Just blowing off steam here :-)

[melhelm]

I got this fixed. It turns out that the ASCOM driver software needed to be tweaked. Works great now.

[Bob Denny]

Ah ha! Thanks for letting us know. I'm moving this to the hardware/driver section and closing it.