Large Jump After Star Lost Events

[Tim Stevenson]

First my equipment:  EQ6R Pro carrying an AT8RC with OAG and ASI220 Mini.  (Note that this is Pier 1 for me so not to get confused with issues with Pier 1.  BTW, Pier 1 was also up last night and functioned perfectly.)  Pier 2 was imaging just fine with reasonable guiding results (0.65" Total RMS).  Performed a GA after the flip.  It confirmed my Dec backlash of 2450ms and my polar alignment error of 0.6'.  Shortly after that, I started losing stars due to clouds.  Normally when it happens the guiding just stops and resumes when stars are reacquired.  I would have to just suffer with tracking vs guiding.  However, last night PHD2 issued an ~32" Dec command causing the mount to "jump".  Note that I didn't lose brightest star with the box around it but the dimmer stars with circles.

Segment 3 was the steady guiding period of 2h36m

Segment 4 was the GA

Segment 5 was fist lost star event and jump

You can see me struggling with trying to determine what's happening with multiple calibrations.  I took several 30 and 300 sec snapshots with tracking only and one with guiding showing the jump.  I measured the jump on that picture and it exactly matches the ~32" command that PHD2 gave the mount.

The second log attachment is from my Pier 1.  I of course had clouds during that run too and experienced lost star events; however, these events did not have the 30" command issued to the mount like Pier 2.

Also note that I have experienced lost star events many times before without the 30" command.  I just wait for the stars to be reacquired and guiding would resume.  I would throw out images if and when the clouds impacted the image quality.

Thank you for your support,

Tim Stevenson

https://openphdguiding.org/logs/dl/PHD2_logs_naNZ.zip

https://openphdguiding.org/logs/dl/PHD2_logs_nXRD.zip

[Bruce Waddington]

I don't think your mount is going to be well-served by Dec backlash compensation.  The event you're talking about wasn't caused by clouds.  What happened is that you set the backlash compensation value to 2450 and then the first direction reversal subsequent to that applied the 2450 ms correction and the star was lost.  I don't doubt there were clouds coming in but they didn't create this problem.  Here's the result of the GA backlash test:

Notice that there isn't really a flat top at the point the corrections switch from north to south (red->green).  In fact, the mount continued to move in the wrong direction for another 1.5 seconds which is suggestive of stiction.  This sort of thing isn't going to be handled well by backlash compensation. If you hadn't lost the guide star and then been plagued by clouds, the backlash compensation algorithm would have quickly reduced the size of the compensation value to avoid this problem - but it never got the chance.  Just bad luck, I guess.  I really think you should disable the Dec backlash compensation and just get back to what you were doing prior to the GA run.  The long guiding session at 20:49 showed a Dec tracking rms of 0.3 arc-sec so there's just nothing to worry about with that.  When you run the GA going forward, just un-check the box that triggers the backlash test. 

Regards,

Bruce

[Tim Stevenson]

Thank you Bruce ... so should I disable the backlash compensation in the Algorithm tab all together or keep it checked and let the backlash compensation algorithm take care of it?  It was enabled before with a pulse of 24ms.  Also, is it worth working to remove the stiction?

Tim

[Brian Valente]

Tim

Regarding the backlash compensation amount, i suggest your run the guiding assistant to measure your backlash and go with that as a starting point. .

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/dc93681a-1703-4a61-8b75-914d291bbe44n%40googlegroups.com.

--

Brian 

Brian Valente

astronomy portfolio https://www.astrobin.com/users/bvalente/ 

[Bruce Waddington]

Just disable the feature.  I don't know why you would want to fiddle around with the mechanics any further when the Dec tracking is already just fine.  If you're trying to move ahead and do high-quality imaging, I don't see why you'd want to spend time on something that isn't going to provide any obvious benefit.  

Bruce