×

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

Bi-monthly release with minor bug fixes and improvements

EQMOD Mount Drift in Kstars?

  • Posts: 112
  • Thank you received: 6
This issue is a bit confusing and just started. The mount seems to drift in Kstars, but it is not really drifting? I have verified from the images the drift is only a few arc seconds over the course of 20 minutes, but the mount thinks it drifted in DEC over 20 arc minutes? The first images have the correct RA DEC location info in their headers, but the last are way off. This is causing some major issues for me doing plate solves. I have verified times are correct, and the location is correct in Kstars and in the EQMOD driver - tracking sidereal. Any other ideas as to what may be causing this?

This is also very disconcerting:

I had to reboot the RP3 (INDI) because of a camera disconnect. Before the last update, this would not be an issue. Everything would come back up and the mount would be in the place it was last pointed. Now when I reconnect the devices in INDI, the mount started to slew WAY further west from where it was and bottom'd out on my pier. Any idea why this would happen???

2017-12-18T04:55:03: [INFO] updateLocation: long = -111.669 lat = 37.0831
2017-12-18T04:55:03: [INFO] Setting UTC Time to 2017-12-18T04:54:58, Offset -7
2017-12-18T04:55:03: [INFO] Device configuration saved.
2017-12-18T04:55:03: [INFO] Saving device configuration...
2017-12-18T04:55:03: [INFO] Device configuration applied.
2017-12-18T04:55:03: [INFO] Setting Slew rates - RA=64.00x DE=64.00x
2017-12-18T04:55:03: [INFO] Setting Custom Tracking Rates - RA=0.5 arcsec/s DE=0.5 arcsec/s
2017-12-18T04:55:03: [INFO] Use Backlash : RA: False, DE: False
2017-12-18T04:55:03: [INFO] Setting Backlash compensation - RA=1 microsteps DE=1 microsteps
2017-12-18T04:55:03: [INFO] updateLocation: long = -111.669 lat = 37.0831
2017-12-18T04:55:03: [INFO] Loading device configuration...
2017-12-18T04:55:03: [WARNING] Can not load HorizonLimits Data File ~/.indi/HorizonData.txt: No such file or directory
2017-12-18T04:55:03: [INFO] Mount is unparked.
2017-12-18T04:55:03: [WARNING] Init() : Setting both ST4 guide rates to 0.5x (2)
2017-12-18T04:55:03: [WARNING] Init() : First Initialization for this driver instance
2017-12-18T04:55:03: [WARNING] Init() : Setting default Init steps -- RAInit=8388608 DEInit = 8388608
2017-12-18T04:55:03: [WARNING] Init() : Motors already initialized
2017-12-18T04:55:02: [INFO] Mount has PPEC.
2017-12-18T04:55:02: [INFO] Mount has auxiliary encoders. Turning them off.
2017-12-18T04:55:02: [INFO] Mount has home indexers. Enabling Autohome.
2017-12-18T04:55:02: [INFO] EQMod Mount is online.
2017-12-18T04:55:02: [INFO] Successfully connected to EQMod Mount.

Thanks!
Last edit: 6 years 3 months ago by David Rankin.
6 years 3 months ago #21720
Attachments:

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

I noticed this as well. I believe it might be due to the new Alignment Subsystem.
6 years 3 months ago #21722

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

  • Posts: 112
  • Thank you received: 6
Is there a way to roll it back to the previous alignment system? I'm a little concerned it almost destroyed my EQ8.
Last edit: 6 years 3 months ago by David Rankin.
6 years 3 months ago #21733

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

not via PPA. You can compile an older version of INDI yourself. If you checkout from GIT, try to checkout from this commit: 142837901f8f4a47674cf9440f2e13fa4bded529

I already emailed Jean-Luc to see if he can shed light on this issue.
Last edit: 6 years 3 months ago by Jasem Mutlaq.
6 years 3 months ago #21736

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

  • Posts: 112
  • Thank you received: 6
Thanks, I guess I'll just wait for a fix.

Cheers
6 years 3 months ago #21739

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

  • Posts: 226
  • Thank you received: 88
Hi,
Having a drift is normal:
- guide commands are not sync commands, thus a part of the drift comes from those guide commands;
- the alignment model is not perfect, and it depends on the way you built it. Furthermore it is not continuous, there may be small jumps when you leave a triangle to another one. And maybe a bigger jump when you leave the convex hull, as the alignment system uses a nearest point algorithm outside the convex hull.
Concerning the reconnection problem, did you stop the mount before rebooting ? Normally the driver does not make any moves on startup so that's strange.
6 years 3 months ago #21772

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

  • Posts: 112
  • Thank you received: 6
1. The mount is not actually drifting. I plate solve my first and last images against Gaia and the difference is a few arc seconds.
2. I do accurate polar alignment before starting.
3. The drift error showing in Kstars is on the order of dozens of arc minutes!
4. That was not happening at all before the last update. The RA DEC info being written to the headers was accurate from start to finish, now it is dozens of arc minutes off by the last image.
Something has changed.

Cheers
Last edit: 6 years 3 months ago by David Rankin.
6 years 3 months ago #21773

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

  • Posts: 226
  • Thank you received: 88
Ok, the best is to revert the commits, that would also solve the other issue with altaz mounts. I ask to Jasem if he can do so.
6 years 3 months ago #21777

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

  • Posts: 112
  • Thank you received: 6
Hopefully that will resolve it. Would you post here when they are rolled back? I was going to try a run tonight for a few hours with the latest QHY release.

Thanks for the awesome support :)
6 years 3 months ago #21778

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

  • Posts: 1309
  • Thank you received: 226

Replied by Andrew on topic EQMOD Mount Drift in Kstars?

Any news on the status of this bug? I am hesitant to update to Kstars 2.8.9 until it is.
6 years 3 months ago #21830

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

This bug is not related to KStars, it's in INDI. And no not yet, it's not simple, but hopefully I can work with Jean-Luc to figure out a solution that does not require a complete revert of the IAS system. If anyone wants to step in, be my guest.
6 years 3 months ago #21848

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

  • Posts: 112
  • Thank you received: 6
knro,
As a developer, I certainly understand the frustration of hard-to-deal with bugs. Definitely appreciate what you guys have done here with INDI. Its awesome. I'm sure you'll get this sorted out.
In the meantime, could you post some instructions on how to roll back to the previous IAS system? Is it possible to do this and use the latest QHY drivers?

Thanks!
Last edit: 6 years 3 months ago by David Rankin.
6 years 3 months ago #21859

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

Time to create page: 0.531 seconds