Backlash compensation: Advanced Settings vs. Guiding Assistant values

[George Hart]

I'm investigating star elongation, present at all exposure durations and increases with duration.

Mount is an HAE69C EC, polar aligned with Sharpcap to "Excellent" (with which PHD2 vehemently disagrees, typically saying  I'm 5-7 arc minutes off).

Flexure is one thing I'm looking but given the issue is presenting with 2 different scopes (480mm and 910mm) with all other components the same, I'm looking for other causes as well.

I have some questions around backlash compensation settings seen in my various tests, including an Oct 2 session whose logs I've uploaded.

https://openphdguiding.org/logs/dl/PHD2_logs_SVn3.zip

1. What causes Guiding Assistant to report differing values of backlash compensation needed from night to night? Over multiple nights, I've seen values from "no compensation needed" to 900ms. Is this seeing driven or is something else at work?
   
   
2. 2 Oct was a session where G.A. reported no backlash compensation needed. When I looked at the Advanced Settings/Guiding tab, I noticed that DEC backlash compensation was still enabled and set to 90ms. This seemed contradictory and I disabled it to see if results would improve (they did not). Are those different controls or am I misinterpreting something?

Appreciate any thoughts and as always, thanks for the help.

George

[richar...@bigpond.com]

Hi George

 

If you took all your images from one night and turned them into a “video”, eg play them in PixInsight Blink, do the stars end up drifting across the screen ?

 

If so, then will say its Flexure. I was sceptical of this for years then finally tightened, stiffened my setup, plus most importantly braced my rings back to the mounting plate and even to each other. Changed my world !

 

Thanks Brian & Bruce

 

Cheers Richard

--
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/b7156ac2-7578-4107-acf3-0e64742c2d66n%40googlegroups.com.

[George Hart]

I will do the video drift experiment and report back.

I do want to "bump" the second set of questions around backlash compensation settings (repeated here for convenience).

1. What causes Guiding Assistant to report differing values of backlash compensation needed from night to night? Over multiple nights, I've seen values from "no compensation needed" to 900ms. Is this seeing driven or is something else at work?
   
   
2. 2 Oct was a session where G.A. reported no backlash compensation needed. When I looked at the Advanced Settings/Guiding tab, I noticed that DEC backlash compensation was still enabled and set to 90ms. This seemed contradictory and I disabled it to see if results would improve (they did not). Are those different controls or am I misinterpreting something?

Thanks, George

[Bruce Waddington]

The backlash measurement - actually a measure of Dec reversal delay - can be affected by many things, especially pointing position, variable amounts of stiction, and payload balance in Dec.  It's not uncommon to see variations like what you're talking about and those differences aren't really interesting because the compensation algorithm is adaptive.  So your example of 900ms might have quickly been reduced to a lower value by the behavior of the algorithm, and compensation values below 100ms don't generally have much effect.  I guess there is no 'apply' button when the recommendation is that no backlash compensation is required so I suppose we could add that.  Overall, this isn't anything worth worrying about, the algorithm pretty much takes care of itself.

Bruce

[richar...@bigpond.com]

Hi  Geoge

 

90ms is not much at all and I’d be very happy with that, mine is a lot more but I can say that PHD2 Backlash Comp does a great job.

 

You can just look at your first, middle and last light, has a bright star moved across the FOV, if so back to flexing.

 

Regards Richard

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/d977d5d3-d119-4ef4-98dd-bb7be6fee9aan%40googlegroups.com.