×

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

Bi-monthly release with minor bug fixes and improvements

indi_celestron_aux

  • Posts: 37
  • Thank you received: 0

Replied by Maciek on topic indi_celestron_aux

On my last session under stars, I had similar issue. There was bad alignment calculation that gives me average of ~400arcsec error on goto and solve new image. After running "slew to target" in ekos alignment tab, function was not able to get error small enough jumping around and finally gave up. There was one time when error was not decreasing but increasing with every one try.
I believe this behavior is related to alignment issue that developers are aware, and trying to resolve it. Lets just give them time.
3 years 1 week ago #68807
The topic has been locked.
  • Posts: 200
  • Thank you received: 57

Replied by Paweł on topic indi_celestron_aux

I have noticed this problem. This is indeed problem with alignment subsystem. Workaround for now is increasing the accuracy setting to realistic value (few arcmin). It starts to be usable - currently the driver is simply unable to get 30 arcsec goto accuracy. We are working on it. The other way is to switch to "sync on solution". Note that 1arcmin pointing accuracy is the error of the size of the Jupiter disc, so fairly small.
3 years 1 week ago #68808
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

I spoke to soon Jochym, 10 minutes ago I did a normal GoTo and the scope is still hunting up and down.

Could it be something to do with tracking? as it is "slewing with tracking"

The hunting starts as soon as tracking turns back on after the slew. If one turns off tracking while it is hunting the hunting stops. If you turn tracking off and then quickly on again the hunting starts again. If you turn tracking off and wait about 5 seconds before turning it back on the hunting stops.

This doesn't happen for every goto but if I do goto's between Mizar and Alioth (in Ursa Major) it happens about 70% of the time. Nothing special about these two target except that they are close together and further testing doing goto's between other close targets shows that this seems to be the problem. This is probably why "slew to target" in the aligment tool shows the problem because it is just making small goto moves. A bit further testing and I can get the hunting virtually 100% percent of the time if the angular distance between the start and finish target is less than about 2 degrees.

Is there a setting for the settle time before tracking starts after a slew?

I hope this helps, if you want me to try anything just ask, I am a software engineer so can poke around in code if required.

Ken

P.S. I have tried putting a 10 second delay in the code (sleep(10) after the goto before turning tracking on but it doesn't help, as soon as the tracking kicks in the hunting starts. Manually stopping tracking for 10 seconds and re-starting tracking fixes it.



 
Last edit: 3 years 1 week ago by Ken Kelso.
3 years 1 week ago #68809
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

So as they say ASSUME makes an ass out of u and me.

AstroGene1000 was correct it is backlash settings on the scope that is causing the problem, I zeroed all the backlash settings and now I can't get it to hunt on a goto.

So I think this issue is closed but a note needs to go in the driver details that the backlash settings on the scope can cause this behaviour.

Ken
3 years 1 week ago #68836
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

You were right Astrogene1000, I zeroed the backlash values and now it is OK, thanks you.
3 years 1 week ago #68837
The topic has been locked.
  • Posts: 276
  • Thank you received: 52

Replied by Gene N on topic indi_celestron_aux

Good to hear your seeking is resolved.

Question though, in AltAz mode does it ever get to the actual requested position -exactly- ?
I found in EQ mode it would always be hunting in ALT, from a post I did earlier in this thread the ALT rate never went to 0 and the reported DEC never matched exactly the requested and would change over time
09:53:05.081: [INFO] ReadScopeStatus - RA 20.812913 hours DEC -9.423441 degrees
09:53:06.168: [INFO] Tracking - Calculated Alt -15.151356 deg ; Az 156.548387 deg
09:53:06.169: [INFO] Target (AltAz): -15.151356 156.548387 Scope (AltAz) -15.100665 156.304250
09:53:06.169: [INFO] TimerHit - Tracking AltRate -3040 AzRate 14647 ; Pos diff (deg): Alt: -0.052828 Az: 0.240476
3 years 1 week ago #68840
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

Where are you getting this info from?
 
3 years 1 week ago #68844
The topic has been locked.
  • Posts: 276
  • Thank you received: 52

Replied by Gene N on topic indi_celestron_aux

I took the existing Driver DEBUG outputs and put on INFO if (TraceThisTick) LOGF_DEBUG("TimerHit - Tracking AltRate %d AzRate %d ; Pos diff (deg): Alt: %f Az: %f", altRate, azRate, AltAz.alt - AAzero.alt, anglediff(AltAz.az, AAzero.az));
  github.com/indilib/indi-3rdparty/blob/ma...aux/celestronaux.cpp
3 years 1 week ago #68846
The topic has been locked.
  • Posts: 200
  • Thank you received: 57

Replied by Paweł on topic indi_celestron_aux

Are you testing the main branch or the recent development branch (github.com/indilib/indi-3rdparty/pull/348)?

If possible, run the development branch - we have re-worked the tracking code. The old one is going to be replaced anyway - so your testing will be more useful with new code (and it seems to work better in my tests).
3 years 1 week ago #68847
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

If I do a GoTo in Alt/Az mode and then look at the co-ordinates on the hand controller they match and do not change.
3 years 1 week ago #68876
The topic has been locked.
  • Posts: 33
  • Thank you received: 3

Replied by Ken Kelso on topic indi_celestron_aux

Hi Jochym, I have been testing the main branch, I will now clone your repo and test with that.

Ken
Last edit: 3 years 1 week ago by Ken Kelso.
3 years 1 week ago #68877
The topic has been locked.
  • Posts: 200
  • Thank you received: 57

Replied by Paweł on topic indi_celestron_aux

Sure:
For one time testing you can just get:
<code>
github.com/jochym/indi-3rdparty/archive/...ads/celestronAUX.zip
</code>
and unpack it always points to the head of the branch. Be warned, it is over 100MB and you will need to re-download it for every change you would like to test.
So it is *much* better to check out the branch with git:<code>
git clone github.com/jochym/indi-3rdparty.git
cd indi-3rdparty
git checkout celestronAUX
git pull
</code>
Each time you need to refresh the repo you just go to the directory and execute <code>git pull</code> the initial download is large (536MB at the moment) but each update is tiny. And you can switch to different branches and remote sources without downloading it again. Chose your poison - as they say ;)[/code][/code]
Last edit: 3 years 1 week ago by Paweł. Reason: Typo
3 years 1 week ago #68879
The topic has been locked.
Time to create page: 1.441 seconds