×

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

Bi-monthly release with minor bug fixes and improvements

Meridian Flip does not complete on iOptron Mount

  • Posts: 1119
  • Thank you received: 182
I can only update the handset, not the mount. Communication with the mount only works through the handset and that was last updated over a year ago.

How can I update the mount? Directly connecting the mount through serial does not work.
4 years 4 months ago #45378

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

  • Posts: 1009
  • Thank you received: 133
Ooooops!
My bad. I'm too much living in my CEM60 world, and overlooked what your mount actually is :blush:
4 years 4 months ago #45381

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

  • Posts: 1185
  • Thank you received: 370

Let's take a look to the log file:
[2019-11-02T02:36:48.875 CDT INFO ][     org.kde.kstars.ekos.mount] - Meridian flip: slewing to RA= "03h 48m 10s" DEC= " 24° 10' 34\""
...
[2019-11-02T02:36:49.320 CDT DEBG ][     org.kde.kstars.ekos.mount] - Setting meridian flip status to  5
[2019-11-02T02:36:49.321 CDT INFO ][   org.kde.kstars.ekos.capture] - "Telescope completed the meridian flip."
[2019-11-02T02:36:49.324 CDT INFO ][   org.kde.kstars.ekos.capture] - "Performing post flip re-alignment..."
...
[2019-11-02T02:36:49.399 CDT INFO ][           org.kde.kstars.indi] - iOptron CEM60 :  "[INFO] Slewing to RA:  3:48:10 - DEC: 24:10:35 "
[2019-11-02T02:36:50.322 CDT DEBG ][     org.kde.kstars.ekos.mount] - Mount status changed from  3  to  2
[2019-11-02T02:36:50.326 CDT DEBG ][ org.kde.kstars.ekos.scheduler] - Mount State changed to 2
You can see that the mount starts the slew at 02:36:48.875, but it takes until 02:36:50.322 that the mount receives the information that the mount is actually slewing (= mount status 2). At 02:36:49.320, the mount module checks the mount status and receives the information that it is tracking - although half a second ago a slew command is issued. That's why the mount module draws the false conclusion that the slew already has completed and declares the meridian flip as completed.

What I do not understand is, why it takes 1,5 secs until the mount reports the status change. One possible explanation is, that the Kstars INDI listener polls the mount status with the given frequency. If it is too slow, it recognizes too late that the mount is already slewing. That's why I proposed to increase the polling frequency.

But not sure whether that helps.

Wolfgang
The following user(s) said Thank You: Jasem Mutlaq, Jose Corazon
4 years 4 months ago #45383

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

  • Posts: 1119
  • Thank you received: 182
Looks like shortening the polling period did the trick.

Here a screen record of what was going on tonight during the flip:

www.dropbox.com/s/n61dcxrj71h7z6h/Flip%20Test.mkv?dl=0

And here the link to the log of what was going on during the flip (occurred around 19.50:38)

www.dropbox.com/s/1l5d5em5gln02sz/log_19-03-07.txt?dl=0

Thanks, Wolfgang, I hope that solved the problem!

Jo
Last edit: 4 years 4 months ago by Jose Corazon.
4 years 4 months ago #45433

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

This needs to be fixed in the driver instead of the timer hack. I'll think of something.
4 years 4 months ago #45480

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

So indeed the mount was responding with TRACKING after the SLEW command was accepted as shown in the logs. Right now, I added a check in GOTO to check for the state and wait until it officially changes to SLEW.

Changes are pushed to GIT, please test and report back.
The following user(s) said Thank You: Wolfgang Reissenberger
4 years 4 months ago #45482

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

  • Posts: 1185
  • Thank you received: 370
Great, I’ll check it asap. Is it straight on the way to git master or in Phabricator?

If we fix this, I think it will also resolve the issue with star trails during alignment.
4 years 4 months ago #45484

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

It's in INDI git since it is a driver issue.
4 years 4 months ago #45485

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

  • Posts: 1185
  • Thank you received: 370
OK, understood. But I think this is a general issue, not only of iOptron mounts. Wouldn't it be easier if a GOTO immediately sets the status to IPS_BUSY?

And by the way, the problem with early start of align capturing might have the same underlying problem.
4 years 4 months ago #45487

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

That wouldn't solve the problem, because let's say it goes like this:

1. Slew commanded.
2. EQNP set to IPS_BUSY
3. ReadScopeState is called, we called #GLS and lo and behold, the mount reports it is TRACKING. So let's set EQNP back to IPS_OK.

So actually this was indeed a case of a mount driver issue. It is technically (I suspect) some race condition in the firmware itself, because if slew is accepted, but the state should change to SLEWING.. but maybe it was accepted, but the actual motion was not started until few 10s of milliseconds later when the motor kicked in? I don't know this is all educated guess, but the fact is mount firmware report TRACKING after it accepted SLEW command.
The following user(s) said Thank You: Jose Corazon
4 years 4 months ago #45501

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

  • Posts: 1119
  • Thank you received: 182
OK, I installed the drivers from the indi github site. Will try them out as soon as the skies clear again here, but according to the forecast, that's not supposed to happen for the next 10 days.

:-(

I have one other unrelated question regarding polar alignment, where something else changed over the last few months, but I'll open a separate topic for that.
4 years 4 months ago #45503

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

  • Posts: 1185
  • Thank you received: 370
Well, I'm not really familiar with this low level driver stuff. I would not be surprised if it is build in an asynchronous matter. That would mean there is a certain delay between the acceptance of a slew command and its execution. It's really tricky!
4 years 4 months ago #45504

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

Time to create page: 0.454 seconds