×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Exciting new tool in Ekos for Mount Modelling

  • Posts: 2876
  • Thank you received: 809
The target appears very tiny because you are very far off target. You can delete any points that you don't want to see anymore.
7 years 1 week ago #15196

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
patdut,

here is a demonstration of what I meant about the built in error in the telescope simulator, a series of 20 solutions spread across the sky using the simulator and the mount modeling tool using a wizard generated grid pattern. I just ran this a few minutes ago. Interestingly, the pattern is slightly different than one I ran the other day. and one point is even outside of my circle, which is interesting too. Since I made this tool, I have run several variations of patterns like this with the simulator and it produces interesting patterns. What I don't actually know is whether the error I am seeing is error in the simulators, error in the astrometry.net files, or if it is a periodic error simulation in INDI CCD, which Jasem suggested it may be.

7 years 1 week ago #15197
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
here's a couple from two days ago

Last edit: 7 years 1 week ago by Rob Lancaster.
7 years 1 week ago #15198
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809

If you see any pattern to when it happens, please let me know. So far, I haven't seen it and I have been running it quite a bit. I have only been using the simulators recently due to a lack of good weather.
7 years 1 week ago #15199

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Oh, I forgot to post the more scientific and interesting ones I did. Keeping a constant DEC and changing the RA and then keeping a constant RA and changing the DEC. The patterns are much more clear. Note that the wizard cannot generate these points. I had to enter them manually. The reason is because for the constant RA version, you would not really want that because I think most mounts want a change in RA for building their pointing models. For the constant DEC version, you could do part of the circle but not the whole thing due to the fact that most of these points are below the horizon. The simulator is the only thing I know that can image below the horizon.

7 years 1 week ago #15204
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 472
  • Thank you received: 165

Now it seems to happen pretty much every time after the first star, log shows rows like this:
2017-03-16T09:32:11 Capturing image...
2017-03-16T09:32:09 Slew complete. Solving Alignment Point. . .
2017-03-16T09:32:09 Slewing to target coordinates: RA (01h 20m 22s) DEC ( 27° 21' 09").

ie. it starts slewing and immediately thinks it's complete even though the telescope simulator has just started slewing taking a few seconds to complete. All I did was connect simulators (telescope and ccd), go to mount model dialog and press generate and start.

For the first star it works correctly:

2017-03-16T09:32:02 Capturing image...
2017-03-16T09:32:00 Slew complete. Solving Alignment Point. . .
2017-03-16T09:31:24 Slewing to target coordinates: RA (15h 43m 27s) DEC ( 26° 14' 36").
2017-03-16T09:31:24 The Mount Model Tool is Starting.

This is Intel i5-6500 quad core machine with indi and kstars compiled from current git, running Ubuntu 16.04.

Edit: I also tried with EQMod simulator instead of telescope simulator and the behavior is the same except EQMod also complains as the mount model is trying to sync the mount while it's still slewing:
2017-03-16T09:48:08 Solver aborted after 4 seconds
2017-03-16T09:48:08 Syncing failed!
2017-03-16T09:48:08 Syncing to RA (17h 38m 51s) DEC ( 16° 54' 19")
2017-03-16T09:48:08 Syncing failed!
2017-03-16T09:48:08 Syncing to RA (17h 38m 51s) DEC ( 16° 54' 19")
2017-03-16T09:48:08 Slewing failed!
2017-03-16T09:48:08 Slewing to target coordinates: RA (15h 43m 27s) DEC ( 26° 14' 36").
2017-03-16T09:48:08 Syncing to RA (17h 38m 51s) DEC ( 16° 54' 19")
2017-03-16T09:48:08 Solution coordinates: RA (17h 38m 51s) DEC ( 16° 54' 19") Telescope Coordinates: RA (00h 38m 16s) DEC ( 87° 55' 37")
2017-03-16T09:48:08 Solver completed in 4 seconds.
2017-03-16T09:48:04 Starting solver...
2017-03-16T09:48:04 Image received.
2017-03-16T09:48:01 Capturing image...
2017-03-16T09:48:00 Slew complete. Solving Alignment Point. . .
2017-03-16T09:48:00 Slewing to target coordinates: RA (00h 00m 09s) DEC ( 6° 57' 28").
Last edit: 7 years 1 week ago by Jarno Paananen.
7 years 1 week ago #15222

Please Log in or Create an account to join the conversation.

  • Posts: 472
  • Thank you received: 165
Funnily enough I tried increasing logging verbosity to get better picture of what is going on and now it works fine, probably the logging slowed it enough so that the status has time to change to busy before first polling.
7 years 1 week ago #15227

Please Log in or Create an account to join the conversation.

  • Posts: 472
  • Thank you received: 165
Now that the model tool works for me I noticed a weird thing with the accuracy calculation. For one point the tool got stuck iterating even though the solution is within couple of seconds from the target and accuracy is set to 30. From log:

2017-03-16T13:13:06 Slewing to target coordinates: RA (00h 01m 31s) DEC ( 62° 16' 16").
2017-03-16T13:13:06 Syncing to RA (00h 01m 29s) DEC ( 62° 16' 14")
2017-03-16T13:13:06 Target is within 0° 00' 33" degrees of solution coordinates.
2017-03-16T13:13:06 Solution coordinates: RA (00h 01m 29s) DEC ( 62° 16' 14") Telescope Coordinates: RA (00h 01m 31s) DEC ( 62° 16' 16")
2017-03-16T13:13:06 raDiff -33.376 decDiff -1.7653 targetDiff 33.423

The last log print is printing the raw diff values from align.cpp after line 2425 where targetDiff is calculated. For some reason raDiff is more than 33 seconds even though the difference is only two seconds and dRA field in UI also shows "-00h 00m 02s". I've noticed this earlier with regular slew to target as well that sometimes it just keeps iterating essentially the same position until giving up.

Edit: adding the magic divide of raDiff by 15.0 makes it work and targetDiff is sane as well.
Last edit: 7 years 1 week ago by Jarno Paananen.
7 years 1 week ago #15228

Please Log in or Create an account to join the conversation.

No that makes it wrong. "-00h 00m 02s" is 2 seconds, not 2 arcseconds, but is equal to 30 arcseconds. It is 33 because the dRA display rounds it off, it is in fact "-00h 00m 02.2s".
7 years 1 week ago #15229

Please Log in or Create an account to join the conversation.

  • Posts: 472
  • Thank you received: 165

Ok, then it's just that the RA in simulator has much larger errors as the code calculates both in the same way with degrees and the / 15 made the error more symmetric. EQMod simulator seems to have slightly smaller error so it passes with defaults. As said this is only with the simulator but I've had the similar issue on my scope as well that the slew to target routine tries to make miniscule changes even though the target is spot on in the center already. I increased accuracy tolerance to 50 and it seems to do it less often. Could be my mount or something, but I'd think HEQ5 is pretty accurate in that sense. Then again I'm at fairly high latitude (61.5) and quite often imaging near NCP where plate solving seems to be more challenging.
7 years 1 week ago #15230

Please Log in or Create an account to join the conversation.

Time to create page: 2.882 seconds