×

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

Bi-monthly release with minor bug fixes and improvements

Setting up for Meridian Flip

  • Posts: 181
  • Thank you received: 13
Still getting to grips with Meridian flip using EQ8 mount and 12" RC telescope with ASI mini guide camera....
Imaging run last night with Meridian Flip expected at 2.06 which is 2h after target crossed Meridian as defined in Mount setting. Indeed this happened and stats show it occurred as expected with the necessary steps: flip plan: guide suspend, mount slew: mount track: re- align: guide calibrate(I think) : guide suspend: refocus: start guiding: next capture.
Problem was post flip images exhibit wild guiding. Please see pre and post flip images attached. I would appreciate if anyone can identify the issue and confirm which settings control Meridian flip and guiding as I may have missed something?
I have noticed possible bugs:
1) If guiding (internal guider) fails for some reason it can leave the mount tracking at a non-sidereal rate. Is there a safety section of code that puts the mount to sidereal rate after a guide failure?
2) I find re-calibration of guiding fails with little or no movement of RA and DEC axes. I found if I re-booted the raspberry PI and re-started everything, with the telescope more or less in the same location, guiding calibration succeeds. It seems previous calibration in the same session messes up the second calibration?

Tonight I attempted to image the same target and watched the flip from the observatory. It looked good however the first image post flip showed the same wild guiding problem. I notice guide calibration after the flip align is quite short around 8 secs and looked wrong on the graph. This time I manually intervened and did a guide calibration which worked did an align and started guiding and then re-started the image queue and all was well again.
I now know the problem is guide calibration fails after the flip! I now suspect this is bug 2 I mentioned above. Help?
Last edit: 1 year 7 months ago by David Bennett.
1 year 7 months ago #85273
Attachments:

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

  • Posts: 1185
  • Thank you received: 370
Hi David,
it would be great if you post your log files. It is quite odd that the tracking rate changed. I‘m almost 100% sure there is no code changing it.

Please take a look at the guiding options (you can reach them from the guiding tab with the options button). As long as you have guiding problems after the flip, set „Reset Guiding Calibration after Mount Slew“.

Please also check whether you have limits set on the mount tab.

Cheers
Wolfgang
1 year 7 months ago #85280

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

  • Posts: 181
  • Thank you received: 13
Hi Wolfgang,
Here's my log files for this mornings image run 12th August (log_22_31_32.txt - flip process started at T01:03:30.786) and for morning before 11th August (log_23-30-35.txt - flip process started at T02:06:12.598 ). On the 11th it ran unattended and on the 12th I got up to witness the meridian flip and intervened to stop and re-start guiding, did an align and re-started imaging queue which completed fine so post-flip images are good this time.
Here's my guiding options I am using:

Here's Mount Tab:

This is the guide TAB after I manually re-started guiding which looks good to me.

I am running Astroberry V2.0.4 on a raspberry PI3 at the telescope (which is up to date) and running Kstars Version 3.6.0 Stable on my Windows 10 desktop connected to remote PI across my LAN (wired not WIFI)

I don't have anything set for guider in the profile but when I connect I obviously connect to the guide scope. Not sure to use CCD1 and CCD2 instead of CCD which I read somewhere will handle both ZWO CMOS cameras.

I have been working on setting up Astroberry and equipment for reliable unattended imaging since the beginning of the year so will be really happy with Astoberry if you can help me solve this meridian flip issue. This will then leave me with experimenting with focussing to get the best results.
[2022-08-11T23:38:50.078 GMT Summer Time WARN ][ org.kde.kstars.ekos.capture] - "{Device: ZWO CCD ASI2600MM Pro Property: CCD1 Element: CCD1 Chip: 0}" Ignoring Received FITS as it has the wrong capture mode 1
[2022-08-12T00:38:07.250 GMT Summer Time WARN ][ org.kde.kstars.ekos.focus] - Focus GSL error: invalid argument supplied by user

Thanks for responding. David
Last edit: 1 year 7 months ago by David Bennett.
1 year 7 months ago #85288
Attachments:

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

  • Posts: 1185
  • Thank you received: 370
Hi David,
please set your logging to "Verbose" and activate all modules. It might be the combination of start focusing during guide calibration. That's at least the only explanation I have.
Wolfgang
1 year 7 months ago #85308

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

  • Posts: 1185
  • Thank you received: 370
By the way: is there a reason that you have "store and reuse guide calibration" disabled? I would expect this is not necessary with your EQMod mount. Try to enable this and disable "calibrating after each mount slew". For most mounts it's not necessary to recalibrate after a flip.

HTH
Wolfgang
1 year 7 months ago #85309

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

  • Posts: 181
  • Thank you received: 13
Will enable full logging and run a sequences with meridian flip as soon as I can.
I thought when EKOS does an auto focus it suspends guiding until focussing has completed? I presume it does this because guiding might introduce unwanted movement during focussing. However unguided might be just as bad I suppose if mount is poorly aligned. For me unguided mount just tracking is pretty good for more than 30 secs so well long enough to capture a focus image of only a few seconds.
Does EKOS wait for guiding calibration to finish if running before suspending guiding and focussing?
1 year 7 months ago #85310

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

  • Posts: 181
  • Thank you received: 13
I think it was because I was getting guiding calibration failing sometimes with insufficient movement in RA and/or DEC so decided always to re-calibrate. I will put re-use on to test again but no reason not to re-calibrate each time even if it reduces imaging time.
1 year 7 months ago #85311

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

  • Posts: 1185
  • Thank you received: 370
No, it does not wait for guiding calibration to finish. start focusing suspends guiding, i.e. the calibration is stopped. I'm not sure whether calibration continues after focusing finished, I could not test this combination yet.
Last edit: 1 year 7 months ago by Wolfgang Reissenberger.
1 year 7 months ago #85312

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

  • Posts: 1185
  • Thank you received: 370
OK, I think I found the explanation. It seems like suspending guiding during calibration, focusing and resuming afterwards does not continue the calibration. Instead, guiding continues. If this is true, we should call it a bug.
Last edit: 1 year 7 months ago by Wolfgang Reissenberger.
1 year 7 months ago #85316

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

  • Posts: 181
  • Thank you received: 13
OK. I hope to test tonight with re-use calibration enabled to see if this circumvents this problem.

Is a state machine approach appropriate for EKOS jobs? Maybe this is how it is coded already? I am thinking it would be easier to understand with a state diagram what we expect to happen on a meridian flip and indeed at other times during an imaging run.
1 year 7 months ago #85326

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

  • Posts: 181
  • Thank you received: 13
Hi Wolfgang. I need some more time to come to a conclusion on re-using guiding calibration. I tried several short image runs on different targets approaching the meridian and yes guiding did not recalibrate however this didn't entirely fix the problem. The drift graphics on the guiding TAB indicate it takes a minute or two to get the drift back to pre flip low levels but there is no time delay and the next capture after flip shows drift in the image but subsequent images are OK.
To add complications I am trying to work out why focussing on filter change and post flip doesn't always happen. Same with alignment not happening after flip. I think it depends if or how I have aligned and auto focussed before starting the sequence queue. If auto focus and alignment happens after the flip then it seems guiding has had time to settle down before capturing the next image. Weird because guiding is suspended during these operations.

Will test further.
1 year 7 months ago #85330

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

  • Posts: 181
  • Thank you received: 13
Had a better image run last night. I ran it as a job this time and not straight from the Sequence Queue.
Stumbled across a guiding settle time under Configure Kstars > EKOS > Capture and set it to 60 seconds. The description says this is applied after guiding resumes so that would be after an auto-focus or alignment but I didn't see this happen. I need to puzzle this out but what I saw was no delay after an auto focus following a filter change but an 8 minute delay after the auto focus for the meridian flip constraint. Work in progress ..
1 year 7 months ago #85368
Attachments:

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

Time to create page: 0.403 seconds