Is there some "PHD2 for AM5 (and similar)" info available?

[JS N]

Hi Bruce,

yesterday I got my ZWO AM5, not yet installed (and cloudy, anyway).
I thought I check the usual sources, including here, prior setting up PHD2 with the new mount.

However, there are quite a few threads here and elsewhere, pretty much either "yay!" or "OMG". But there is so much guessing, assuming and hearsay.
I didn't find anything specific for the AM5 and/or strain wave mounts to think about when using PHD2 with those.

Is there anything specific which should or must be done? Or do I, as usual, just run the new profile wizard, calibration, guiding assistant and then take it from there? 

I guess there is a growing number of users of this new type of mounts and since we all love PHD2, some basic guidance would be great.

Cheers,
Joe

[Mike Jerry]

Joe,

As an AM5 owner since the beginning of the year, I can point you in some directions.

First, is to get on the AM5 forums if you aren't already. This thread in particular will send you down a very deep rabbit hole but is excellent for its info:

https://bbs.astronomy-imaging-camera.com/d/15989-getting-the-best-performance-from-my-am5

Or if you are a Cloudy Nights member:

https://www.cloudynights.com/topic/880277-improving-am5-mount-guide-performance/

What I can tell you is that if you try and use the default PHD2 profile settings or the results from the GA, you will be very unlikely to have a pleasant experience with a strain wave mount. PHD2 is developed and tuned around worm and spur gear mounts. Strain wave mounts are completely different animals with a completely different rule book.

Start with those threads and feel free to come back and ask questions!

Mike

[Mike Jerry]

P.S. I should also add that PHD2 works great with strain wave mounts. Please don't read my post as trying to drive you away from it! It's just the default settings and suggestions being made by PHD2 need to be taken with a big grain of salt and seen in the light of the education you will receive by just reading the first thread I posted above.

Mike

[JS N]

Thanks Mike, that is very helpful, indeed!
I did the calculations for my mount, as suggested in the first thread you've linked, and it seems I'm on the "unlucky side" with my mount.
We'll see....

[Bruce Waddington]

Hi Joe.  I think you could help us get a better insight into the mount behavior if you could spend a little time doing some simple tests for us.  I'd like to see multiple back-back datasets like this:

1.  Data set 1

     a.  Set the mount guide speed as high as possible, at least 0.9x sidereal

     b.  Create a new PHD2 configuration profile using the new-profile-wizard.

     c.   Use the Calibration Assistant to get a clean calibration.

      d.  Make sure the polar alignment error is no worse than 10 arc-min.

     e.   Leave the RA guide algorithm as Hysteresis and the Dec guide algorithm as Resist-switch.

     f.    Set the guide exposure time to 0.5 - 1.0 second being sure Multi-star guiding has at least 3 candidate stars.

      g.   Run the Guiding Assistant for an extended time, at least as long as one worm period and accept its recommendation for Min-Moves.  Don't run the backlash test and don't change any other guiding parameters

       h.   Let PHD2 guide for 20-30 minutes without making any changes to settings

2.  Data sets 2 - N

        a.  Make whatever changes you want to various settings - but for each combination of settings, let PHD2 guide for at least 20 minutes before making other changes.

         b.  Keep the scope pointing in roughly the same part of the sky for all the tests, stopping and re-slewing if necessary.

         c.  Start and stop guiding between tests so the guiding sections in the log are clearly defined

Thanks,

Bruce

[JS N]

Hi Bruce,

unfortunately I've seen your post only after I did my first set of tests last night.
But I did very similar things to those you've requested.

I've learned a few things about the AM5 by reading through various sources.
On the contrary of what we're used with "traditional" mounts, the widely suggested starting parameters are:

Guide speed 0.5x sidereal or LOWER, if the mount allows it (there is quite a difference between AM5's and newer models have a much shorter worm period, I think 288 sec, mine is one with 432sec)
Many suggest going as low as 0.3x sidereal.
Ra and Dec Aggr. is suggested to set to 30 for a start and max guide pulse to 200ms. Fast guiding with 0.5 sec

So I've created a new PHD2 profile first. Had to do it anyway because I'm now using a mid range quality 60mm f/4 guide scope with the AM5 and no longer an OAG.
Next I did drift align->slew, so we're at 180° Az. and about 42° Alt.

I did a calibration run, then ran the guiding assistant for 2450 sec (which is a bit more than five worm periods). The graph is interesting.
After that, I've accepted the suggested changes and let guiding run for some 20min

Next I did a couple of shorter tests with different settings. It appears that fast guiding is the only way with that mount, going to 1 sec immediately was worse.

I reset the mount now and then to 180° Az/42° alt.

Eventually I did guiding on the Crescent nebula which is very high at this time, some 80°.

While testing, the mount was going nuts for a short time, wildly osciallating, no clue what happened there. It eventually stopped doing that.

Overall the guiding was OK, I have to admit I'd expected much better results. But seeing wasn't very good and maybe with different settings guiding can be improved.
Looking forward to hear what you think about it.

Find the logs here: https://openphdguiding.org/logs/dl/PHD2_logs_5BNg.zip

Regards,
Joe

[bw_m...@earthlink.net]

So I guess the question is whether you’re willing to now go back and just do the default-baseline test I asked for.  You can do that with a new test profile and not disrupt the settings that you have.  Given the poor mount mechanics, I understand the need to guide at a high cadence with 0.5s exposures but many of the other things don’t yet make sense to me.  For example, the only reason I can see for wanting to use a slow guiding speed is if the timer resolution in the mount controller or the driver is really low.  Otherwise, faster guide speeds are better because they introduce less intra-frame delay.  If you’d be willing to make a run using 0.5s exposures and the default PHD2 settings, I would be very appreciative.  By the way, you should upgrade to the PHD2 2.6.11dev6 release and start using the calibration wizard.

 

I’m quite willing to make changes in PHD2 to adjust the recommendations for these kinds of mounts, but I need to see the raw data to understand the underlying problems.

 

Thanks,

Bruce

--
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/e9de9000-d41b-4d2c-906f-0a7aa1ceb668n%40googlegroups.com.

[Mike Jerry]

Why was my post censored?

[Bruce Waddington]

It wasn't "censored" by anyone here and it doesn't show up in "pending" messages that need approval.  So I have no idea what happened.  If you sent a response via e-mail, perhaps you should try using the web interface in the forum and see what happens.

Bruce

[Mike Jerry]

Bruce,

I used the web interface to post the message. How bizarre.

All I did was post a message of encouragement to Joe that he is definitely on the right track and that the settings I am using are in alignment with his. Don't understand why any of that would have triggered some sort of Google algorithm, but I guess it did.

Now I'm gun shy. Let's see if this post goes through…

Mike

[JS N]

Sure Bruce, will do that. So all the bullet points you've originally asked for are still valid?
Will also update to the suggested version.

The tests though....will do whenever possible. The weather forecast for the next couple of days looks horrible...

[bw_m...@earthlink.net]

I’m talking about the “Data set 1” stuff only.  You can just do a shortened GA run because we have the long one from the tests you did.  Go ahead and use 0.5-sec guide exposures.  Since you’re familiar with all this, it shouldn’t take that long to do and it would help me a lot.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/ef4990b9-6d0a-4b8b-86a2-b9c8d15c2317n%40googlegroups.com.

[JS N]

So there was a very short window between clouds tonight, just enough to do the bare minimum of collecting data.
I doubt it will be of much use, will send more as soon as the sky conditions allow it.

https://openphdguiding.org/logs/dl/PHD2_logs_M2gN.zip

[JS N]

quite unexpectedly, the clouds dissappeared. I was able to collect quite some data.

1. with the new profile created, mount set to 0.9x sidereal, things got really weird. The mount started to oscillate wildly. But it didn't stop doing that even after stopping guiding in PHD2.
    I have the feeling it could be a firmware or Ascom driver releated issue.
    Eventually I gave up, created a new profile with 0.5x and the oscillation didn't happen again. Gut feeling is "it's related!" but that would require more testing to confirm or falsify.
2. Guiding generally was running pleasantly. Much better than anticipated after the first night with the AM5 for sure.
3. the last segment of guiding is interesting, it is azimuth towards northeast. It generally was much worse than south and west, lots of spikes and overshoots (at least it looks like that to me).

see here: https://openphdguiding.org/logs/dl/PHD2_logs_tTzF.zip

Regards,

Joe

[Bruce Waddington]

Thanks for doing this Joe.  I need to take a close look at these logs, something I plan to do this evening.

Bruce

[George Shoup]

Joe and Bruce

I have 2 cents I can add.  I have experienced worse guiding in some scope positions and tracked it to excessive gap between the worm and the main RA gear.  Adjusting and removing the slop solved the problem.  For example, near the counterweight down position east side heavy is not going to help.  Any wind will cause the RA to rattle within the gap between the gears.

A slight mechanical adjustment can make a big improvement.

George

Sent from my iPhone

On Aug 16, 2023, at 9:59 AM, Bruce Waddington <bw_m...@earthlink.net> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/a2ff4e3d-c816-45ae-967c-4255519d08c6n%40googlegroups.com.

[bw_m...@earthlink.net]

Yes, this can be a solution on conventional mount drive systems – but I don’t know about these strain wave systems.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/5915C377-CCD6-4944-8A24-60F98597A110%40gmail.com.

[JS N]

it's back, complete madness :-(

[bw_m...@earthlink.net]

Can you post the log file or do I already have it?

 

From: open-phd...@googlegroups.com <open-phd...@googlegroups.com> On Behalf Of JS N
Sent: Wednesday, August 16, 2023 2:59 PM
To: Open PHD Guiding <open-phd...@googlegroups.com>
Subject: Re: [open-phd-guiding] Re: Is there some "PHD2 for AM5 (and similar)" info available?

 

it's back, complete madness :-(

 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/1d4c7810-ce04-408b-94de-37ca6620d8a5n%40googlegroups.com.

[JS N]

not yet, it's a session currently running.
However, I'm not alone, example is https://bbs.astronomy-imaging-camera.com/d/16516-am5-ra-axis-wild-oscillation

[bw_m...@earthlink.net]

I don’t want to look at screen shots and other people’s interpretation of the problem.  I don’t doubt the problem exists, but I need to see the raw data.  Please post the log when your session is over.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/1707b80c-d1d6-458f-9ee0-17fdbd1a270dn%40googlegroups.com.

[JS N]

understood but the link posted above points to a problem with the AM5 mount which isn't PHD2 related at all.
I don't want to waste your time searching for issues not caused by PHD2.

As I've mentioned before, the oscillation continues even after PHD2 guiding is completely stopped, even the program itself completely stopped.
I assume there is something which starts this oscillation and the it's kind of self-propelling.

Of course there is the other part, making PHD2 and strain wave mounts a nicer pair. Happy to help with that for sure.

[sarg314]

What does it do when you run guiding assistant?  That's the most important thing to check. That will quantify the oscillation and tell you the frequency.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/2ab05fa9-0e55-4126-86c7-8bfe838fbac3n%40googlegroups.com.

--

Tom Sargent

[JS N]

next set of logs....had to stop because of bloody clouds....

https://openphdguiding.org/logs/dl/PHD2_logs_tRcS.zip

[Bruce Waddington]

Well, it's definitely a head-scratcher.  I found one section of the log to be quite amazing:

RA is shown in red.  So we can see an extended period of RA oscillation here that has gone on for 2 minutes or so.  At the point shown by the arrow, a dither is executed which results in a guide pulse of nearly 2 seconds in duration.  You would expect this to just make things worse - but it immediately resulted in stabilization:

Some of the comments I've seen talk about restricting the size of the guide pulses to avoid the problems yet here's a case where a large guide pulse settled things down.  In other sessions, large guide pulses were used for dithering and stabilization happened quickly:

This particular dither required an RA correction of 2.6 seconds and it worked fine.  So I don't think it's nearly as simple as avoiding large guide pulses.  I assume you can be near the mount when these problems happen - is there any possibility of camera/fan vibration when the problems are evident?  Like you, I'm starting to suspect a problem in either the mount controller firmware or the driver.  One thing you could do to eliminate the driver as a problem source would be to use a slightly different configuration: use ST-4 guiding ("on-camera" in PHD2) while defining an "aux-mount" connection in PHD2 that would be used to get pointing position info.  I don't know if the timing functions are being handled in the driver or the mount controller for your mount, but by using an ST-4 guiding connection, the timing will be handled in the camera. It might be worth giving this a try just to do some further problem isolation.  I really do think this is a ZWO problem that they should be working on with you.

Regards,

Bruce

[JS N]

Bruce, this fits what was reported elsewhere (not sure if it also is in the link I've posted above).

The oscillation does not stop after guiding is disabled. What does stop it is a short slew command (which a dither basically is).,
So if the dither causes enough movement, the mount leaves this oscillation state. I could consider a dither after every frame, so maybe I''d only lose one frame and not a series of them.

It also was suggested to add a small counterweight. It might also have to do with the load, I'm currently using an Askar FRA400 which is very lightweight.
I will try my 8" Newton, maybe that makes a difference.

From checking the logs, is there any general recommendation already for settings I could try out?

[JS N]

and another round.
Please let me know when it's enough.

https://openphdguiding.org/logs/dl/PHD2_logs_pX2Y.zip

[Bruce Waddington]

I think we've seen enough to know that there's something pretty substantially wrong with the mount - either the hardware or the firmware.  I really think the ZWO guys should know what this is and they're probably the only ones with the necessary hardware and firmware insight to identify a root cause.  We can all speculate about what it might be but who knows.  Frankly, since you just got the mount, I don't know why you're continuing to fight with it - it's clearly defective in some way and it seems to me that ZWO should provide a remedy.  In any case, I don't think we need to look at any more guide logs, I think our measurements are just too far away from the underlying problem.

Thanks and good luck,

Bruce