×

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

Bi-monthly release with minor bug fixes and improvements

Spikes on guiding graph after dithering

  • Posts: 152
  • Thank you received: 20
I am struggling with some weird behavior of internal guider when SEP-Multistar algorithm is selected. Guiding is going well, but after dithering, it seems that Ekos is not giving the mount enough time to restore position and starts exposure too quickly. I have experimented with 2 delay settings (one global and second in Guiding settings) but it doesn't helped me. These spikes are marked with red arrows on my screenshot. PHD2 is working correctly, also with SEP without multistar is working well.
Last edit: 3 years 4 months ago by hades.
3 years 4 months ago #63157
Attachments:

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

  • Posts: 1208
  • Thank you received: 559
Thanks, I have noticed this too--mostly after focus instead of dithering, though--but didn't realize it was just with SEP MultiStar.
Are you sure it's only with that? I can see from your screenshot that it doesn't happen all the time.

I've tried previously (unsuccessfully) to find this same issue, and I'll try again.
I had assumed it was my mount. I'll try again and see if I can figure it out.
If/when I do find the issue, the fix won't be part of 3.5.0, though.

FWIW, my plan had been to add a condition that didn't start capturing until the guiding drift was small.

Hy

PS I have found that this hasn't been adversely affecting my image quality for the most part, but clearly something I should try to address.
3 years 4 months ago #63196

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

  • Posts: 152
  • Thank you received: 20
I have set
Guide->Options->Guide -> Settle: 1
KStars->Settings->EKOS->Capture -> Settle: 10
and the guider has enough time to settle before next exposure started.
But it should be much better if the behavior will be dynamic instead of static waiting. I mean the plan you mentioned is better solution for this issue.
3 years 4 months ago #63197

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

  • Posts: 535
  • Thank you received: 109
Is the guider properly respecting the dither, or is it noticing the guide stars jump after guiding is enabled again, and the spike is correcting the mount back to pre-dither tracking?
3 years 4 months ago #63212

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

  • Posts: 1208
  • Thank you received: 559
I do not believe it's that issue (that multistar isn't respecting the new target). That is easy to check for in a debug log, though.
[For instance the similar issue I'm seeing happens after a focus.]

In any event, with the debug guiding logging on, observe the problem. then look at the debug guide output .
You should see the lines below right at the start of each guide iteration:
[2020-11-03T20:10:12.114 CST DEBG ][     org.kde.kstars.ekos.guide] - ################## BEGIN PROCESSING ##################
[2020-11-03T20:10:12.114 CST DEBG ][     org.kde.kstars.ekos.guide] - Star    X :  227.41  Y  :  502.865
[2020-11-03T20:10:12.114 CST DEBG ][     org.kde.kstars.ekos.guide] - Reticle X :  227.614  Y  : 502.586
[2020-11-03T20:10:12.115 CST DEBG ][     org.kde.kstars.ekos.guide] - Star    RA:  1133.58  DEC:  2506.65
[2020-11-03T20:10:12.115 CST DEBG ][     org.kde.kstars.ekos.guide] - Reticle RA:  1134.6  DEC:  2505.25
[2020-11-03T20:10:12.115 CST DEBG ][     org.kde.kstars.ekos.guide] - -------> AFTER ROTATION  Diff RA:  -1.14606  DEC:  -1.28798
The "reticle" is the target position. The top 2 lines are the position and target in image coordinates for the main guide star.

You could look at the guide iteration before the dither starts, and the one after the dither completes and see that this is changed.
In fact, any/all of the iterations after the dither until the next dither (or other reason for change of target) should show the same "reticle" target.

BTW, the top 2 lines are pixel positions,. The error is graphed in RA and DEC arc-seconds which is shown in the "AFTER ROTATION" line.

Hy
The following user(s) said Thank You: Alfred
3 years 4 months ago #63217

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

  • Posts: 1208
  • Thank you received: 559
I did finally implement what I described further up in this thread. Check out
indilib.org/forum/general/8357-new-captu...ng-analyze-plot.html
and let me know if it help with your issues. (You'd need to enable the new checkbox in capture.)
Hy
The following user(s) said Thank You: Rick Bassham, Alfred, hades
3 years 3 months ago #64189

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

  • Posts: 152
  • Thank you received: 20
Thank you Hy, I will update to the improved version and would check it in field, as soon as the local weather allows.
3 years 3 months ago #64205

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

  • Posts: 152
  • Thank you received: 20
So here is my testing done yesterday:

Left side of the graph is "Only Start if Guide Deviation" setting OFF
Right side of the graph is "Only Start if Guide Deviation" setting ON
RMS on right side is twice better than the left side, dithering duration is also twice better if the setting is ON.
hy, thank you for such huge improvement
The following user(s) said Thank You: Jasem Mutlaq, Hy Murveit
3 years 3 months ago #65064
Attachments:

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

  • Posts: 1208
  • Thank you received: 559
Thanks for testing. FWIW, try out the new rms--the one in the 3rd row (2 rows directly below the one you're using) instead of the one in the first row.
That "capture rms" is a statistic that only measures guider drift when images are being captured. This better reflects the "important drift"--the drift that affects our images.
It will likely show less of an improvement, though ;)
3 years 3 months ago #65068

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

Time to create page: 1.157 seconds