Wild Oscillating Behavior

[Carl-Erik Svensson]

Hi Everyone,

I'm hoping someone can help identify what's going on.  In the middle of my imaging session (without meridian flip or physical obstruction), PHD2 started guiding in large pulses in alternating directions on both RA and DEC.  The pattern looked like diamonds on the graph.  It caused large tracking errors in my images.  When I closed PHD2 and took images unguided, the issue went away.  It's only when PHD2 is enabled that the issue presents.  It came on suddenly after 3 hours of good functionality, and now has stuck around across multiple nights.  Here's the logs:

https://openphdguiding.org/logs/dl/PHD2_logs_hP2a.zip

Attached are a couple images of the error that it seems to generate.

[Carl-Erik Svensson]

Another bit of information: I did try to recalibrate, and the behavior came back right away, and then subsided.  I'm not convinced it's gone, since it showed up right away, but has been going for a few minutes alright after that.  It's almost like it reaches the 2" threshold and then just massively overcorrects, and then tried to recover and just gets in a loop of massive overcorrection.  Sometimes it is able to wiggle its way out of that loop and sometimes not.  Maybe the algorithm is faulty, or too aggressive?

[Brian Valente]

Hi Carl-Erik

No, it's unlikely the software is a problem

 

The primary issue seems to be a very poor calibration

Your other calibrations show similarly poor results. Without a good calibration nothing else will work

In addition, i don't see you have done any guiding assistant runs? that would have brought calibration issues to your attention

The oscillations may be caused by backlash compensation being enabled inside of your mount software. Can you doublecheck and see about disabling that

Overall I'd suggest starting next with a baseline guidelog by following these steps, which should help you get to a solid calibration

https://www.dropbox.com/s/zrf0au7jaisvt85/How%20to%20create%20a%20baseline%20for%20guiding%20results%20using%20OpenPHD2.pdf?dl=0

--
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 on the web visit https://groups.google.com/d/msgid/open-phd-guiding/13625f44-754d-472e-a633-d1f02e923593n%40googlegroups.com.

--

Brian 

Brian Valente

astro portfolio https://www.brianvalentephotography.com/astrophotography/

portfolio brianvalentephotography.com 

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

[Raya Johansson]

Hi Carl-Erik,

I remember I had something similar before, and it was Stellarium causing the issue so I normally disconnect it when i'm already guiding ! I enable the PEC and it helped me alot 

Raymond

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i6kUYBVp92ZBZdBU%3DxQJFB18m%2B6pZLUSjj5kBNA56Jknw%40mail.gmail.com.

[Bruce Waddington]

Wow, this is pretty amazing.  The logs show that something is constantly changing the mount guide speeds.  I doubt you’re doing it manually and since EQMOD is used by several thousand people, I doubt it’s the source.  Maybe there’s another app changing these settings in the driver – PHD2 never sets them. The pattern I see is that you calibrate with a particular guide speed setting then start guiding with a different one.  That, of course, is a disaster and will lead to the wild results you’re talking about.  Here’s a summary of that behavior from the two logs you sent – the lines in bold are calibrations.

 

C:\Temp\PHD2_Support\CarlSvensson\PHD2_logs_hP2a\PHD2_GuideLog_2022-10-28_192912.txt

RA Guide Speed = 1.5 a-s/s, Dec Guide Speed = 1.5 a-s/s, Cal Dec = 30.7, Last Cal Issue = Rates, Timestamp = 10/27/2022 7:54:26 PM
RA Guide Speed = 1.5 a-s/s, Dec Guide Speed = 1.5 a-s/s, Cal Dec = 30.7, Last Cal Issue = Rates, Timestamp = 10/27/2022 7:54:26 PM
RA Guide Speed = 1.5 a-s/s, Dec Guide Speed = 1.5 a-s/s
RA Guide Speed = 1.5 a-s/s, Dec Guide Speed = 1.5 a-s/s, Cal Dec = 30.7, Last Cal Issue = Rates, Timestamp = 10/27/2022 7:54:26 PM
RA Guide Speed = 1.5 a-s/s, Dec Guide Speed = 1.5 a-s/s
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM


C:\Temp\PHD2_Support\CarlSvensson\PHD2_logs_hP2a\PHD2_GuideLog_2022-10-30_192105.txt

RA Guide Speed = 4.5 a-s/s, Dec Guide Speed = 4.5 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 3.0 a-s/s, Dec Guide Speed = 3.0 a-s/s
RA Guide Speed = 3.0 a-s/s, Dec Guide Speed = 3.0 a-s/s
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM
RA Guide Speed = 9.0 a-s/s, Dec Guide Speed = 9.0 a-s/s, Cal Dec = 56.6, Last Cal Issue = Orthogonality, Timestamp = 10/28/2022 11:22:39 PM

 

Obviously, it’s impossible for PHD2 to guide correctly when the mount guide speed is different than the one used for calibration.

 

Good luck,

Bruce

--

[Carl-Erik Svensson]

Thank you everyone for the feedback!

I think you are right about guiding with one speed and calibrating with another. I was trying to adjust some settings and did change the guiding speed via the EQMod settings at some point after an initial calibration. It is strange to me that even after I recalibrated, it seemed to rear its head again, though only once and not as consistently. 

That said, it sounds like there is a much more detailed calibration sequence that I can try, beyond the “push here dummy” :)

One question about this step of the calibration procedure:

“ 5. Fix calibration errors if any are reported. If the calibration process results in any errors, make sure to address them and re-calibrate before continuing.”

What’s the best way to address the various issues? I have often times seen an error with not being able to return to the initial position. How would you fix that, for example?

Thanks again for all the feedback and taking the time to look at my issue. 

Best,

Carl

You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/ObRZ3OCkIcg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/003301d8ed48%24cd391c90%2467ab55b0%24%40earthlink.net.

--

--
Carl

[Brian Valente]

Hi Carl

>>>What’s the best way to address the various issues? I have often times seen an error with not being able to return to the initial position. How would you fix that, for example?

There's a troubleshooting section in the phd documentation, you should definitely check that out

https://openphdguiding.org/man/Trouble_shooting.htm

Regarding your specific "not being able to return" - that may just be a warning instead of an error? you should end up with a usable calibration. that sometimes has to do with backlash, but again it depends on the specifics such as which axis this happens on 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CANEDzNpFqtnrmMYMYFoPXHCLYQQsFgOVJXw8-1fTRb_i-Koz_A%40mail.gmail.com.

[Carl-Erik Svensson]

Thank you, Brian. I’ll take all this information to the next clear night and take notes if I hit any specific issues. 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i4pM8d2tG02_myoRcfqdnHYA-ME7b3mfjD3UWVyQ%2BsAEw%40mail.gmail.com.

--

--
Carl

[Bruce Waddington]

I think you’ll need to do a lot more prep before you try to get past all this.  To begin, you really need to understand the basics of what calibration is doing and how to react when things don’t go well.  Try watching this video:

 

https://www.youtube.com/watch?v=MzjyjEpDW90

 

If you scroll ahead to the 7:00 minute mark, you can get right to the start of the talk.  The whole second half of the talk is focused on the different calibration alerts and what to do about them.

 

When you’re ready to try again under a night sky, you must first follow these instructions – carefully – to configure EQMOD properly:

 

https://github.com/OpenPHDGuiding/phd2/wiki/EQASCOM-Settings

 

Don’t use a mount guide speed lower than 0.5x sidereal, higher is better.  Once set, Leave. Them. Alone.

 

Then you need to start over in PHD2 with a new configuration profile because the one you have now is messed up from all the earlier mistakes.  Run the new-profile-wizard to start fresh and don’t fiddle around with any of the parameters. Do a new calibration, then run the Guiding Assistant to get recommendations for the Min-Move settings.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CANEDzNq_Ww%3DOZYbZtSEbyC4QuxpW55R%2B%2BJYEn4aQX45v3q4MXg%40mail.gmail.com.

[Carl-Erik Svensson]

Yeah, I’ll give it a go. I haven’t had quite so many issues with PHD2 or other guiding software in the past. I guess I just got lucky. 

It would be super helpful if all this calibration complexity were addressed more in the app. It is pretty tempting to just hit “guide” and let PHD2 do it’s thing, and the fact that it usually works without intervention definitely hides this underlying complexity, so when things go wrong it’s a bit more surprising, given the deceptively simple setup. Maybe as part of that new profile wizard (which is really great btw) you could add a calibration step that walks through some of these finer points. 

Have you used the guiding in KStars before? For some reason that seems to work with a lot less configuration. Maybe they are using some kind of shortcut or heuristic?

Anyways, thanks for all the help!

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/005601d8ed68%2429605370%247c20fa50%24%40earthlink.net.

--

--
Carl

[Carl-Erik Svensson]

I just wanted to follow up with my results and ultimate resolution.  I watched the presentation linked earlier and learned a bit about what PHD2 is doing under the hood, which was very helpful context.  To solve my specific issue, I deleted the profile I was using and then used the profile wizard to create a new profile and then calibrated on that new profile.  All the wonkiness went away.  The key was creating a new profile.

Also, now that I know where all the settings are in EQMod, ASCOM, and PHD2, I will know that when they change, I will refresh my profile and calibration.

Thanks so much for all the help!  Attaching my latest calibration result, which as I understand it, looks pretty good.

[bw_m...@earthlink.net]

Calibration looks good, glad you got it sorted out…

 

Regards,

Bruce

 

From: open-phd...@googlegroups.com <open-phd...@googlegroups.com> On Behalf Of Carl-Erik Svensson
Sent: Wednesday, November 16, 2022 7:00 PM
To: Open PHD Guiding <open-phd...@googlegroups.com>
Subject: Re: [open-phd-guiding] Re: Wild Oscillating Behavior

 

I just wanted to follow up with my results and ultimate resolution.  I watched the presentation linked earlier and learned a bit about what PHD2 is doing under the hood, which was very helpful context.  To solve my specific issue, I deleted the profile I was using and then used the profile wizard to create a new profile and then calibrated on that new profile.  All the wonkiness went away.  The key was creating a new profile.

 

Also, now that I know where all the settings are in EQMod, ASCOM, and PHD2, I will know that when they change, I will refresh my profile and calibration.

 

Thanks so much for all the help!  Attaching my latest calibration result, which as I understand it, looks pretty good.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/4b7e788e-173d-48cd-8028-9233544f493cn%40googlegroups.com.