FocusMax hangs on some filter change focus changes

[Jon Holtzman]

I am having intermittent issues with FocusMax hanging up on ACP commanding focus changes with filter changes. The ACP console says, e.g. : "Switching from SI to R filter for imaging, Focus change of 290 steps required" (but it's not always these filters or this focus change), FocusMax reports Moving, but then never returns. This does not happen every time, but it happens frequently, such that I never get Scheduler to run more than 30 minutes or so.

I recognize that it certainly seems more likely to be a hardware/communication issue than a software one, but I can't seem to make it fail with repeated commanded focus changes from FocusMax itself or on focus runs. I have also replaced cables and changed USB connections, but the problem persists. It also seems that the issue started a few months ago when I upgraded ACP.

So just checking to see if anyone has had any similar experience and/or whether somehow the focus changes on filter changes are commanded differently somehow? Any troubleshooting advice would be appreciated.

[Dick Berg]

Hi Jon,

Annoying, isn't it, when something fails only sporadically.

Here's what I read in your message - You are not having trouble with focusing per se. It's only when changing filters for a target image, when the focuser has to move from one position to another, that it fails, sometimes.

In the sequence of things, ACP first commands MaxIm to change filters, then waits for the change to complete. ACP then commands FocusMax to move the focuser's position. Make sure you do not have MaxIm itself connected to the (telescope or) focuser. Otherewise there could be possible/potential mixed signals going to the focuser when ACP tell MaxIm to change filters, especially if you have filter offsets set up in MaxIm, too.

As I mentioned, ACP tells FocusMax to move the focuser. FocusMax works through the ASCOM focuser driver to tell the hardware to move. Then FocusMax polls the hardware frequently through the driver to wait for the signal that the focuser has stopped moving. Then ACP gets that signal and moves on. This is the way it's supposed to work. Grossly oversimplified, the ACP code looks like this:

(focuser).Move(somePosition) <-- ACP tell the focuser to start moving; this works - you see Moving
(focuser).IsMoving <-- ACP loops here and listens for the completion signal from FocusMax; True while moving, False when finished
(focuser).Position <-- ACP saves the new position value for use later

In the error cases, FocusMax does not get the signal from the hardware, and thus neither does ACP.

What kind of focuser you have?
What version of ACP were you running, then and now? An upgrade to ACP should not have changed this code. None the less, ACP is sometimes not getting that completion signal, and so hangs up. Did you also upgrade MaxIm?

I'm sorry I haven't really given you a clue about what to do. I'm just asking questions for the sake of provoking thought. It certainly does seem like a hardware or comm problem, especially as it's intermittent.

It will be useful to see an ACP run log showing an event like you describe. Also it would be useful to see the MaxIm log for the same time. You can find that log here: C:\Users\XXXX\Documents\MaxIm DL 6\Log\. There is no FocusMax log for the specific case of changing filters and focus positions.

[Jon Holtzman]

Hi Dick,

Thanks very much for the reply with details included. In fact, it's possible (I hope?) that you may have suggested the solution, since MaxIm has been in fact connected to the focuser, although I have all focus offsets set to zero since ACP takes care of them. I think I've always had things set up this way, but it also occurs to me that around the time the problem started, I also installed a new filter wheel, so perhaps that has some different timing/communication issues that what I was using previously without issue. I will give that a try and hope that resolves the issue!

Would you by any chance have some sort of template script that I might modify to perform a series of these operations during the daytime (or is there some ACP script that I could draw from)? I have not yet delved into Visual Basic.

To complete my incomplete message, the focuser is an Optec TCF-S, which I have been using for some time. The new filter wheel is an FLI Centerline. The current ACP is 8.3 (build 1, release 8.3). MaxImDL version 6.26.

[Bob Denny]

Hi Jon —

I would recommend against trying to program your way around the intermittent problem you are having. DIck’s info was meant to help you understand why things behave as they do. As Dick explains, ACP is not seeing the the combination of the filter change and focus change complete. So this could be a misbehaving filter wheel or focuser. Given the vital information about your recent filter wheel change being correlated with the beginning of this problem, I would “focus” there. Maybe you will see the filter move but the the software signal may still not occur.

You can set ACP up to take images during the day (ACP Settings &gt; General &gt; Simulated Images). You can set FocusMax up to run in a simulation mode as well. As used by ACP, FocusMax does not touch the filter wheel. Leave the Scheduler out of the picture since it is just giving ACP work to do, with ACP doing the actual work. So make up a multi-filter observing plan and run it in ACP. It will run to completion including plate solving and pointing updates, etc. To separate the wheel from the focuser, you could create a temporary FilterInfo.txt with 0 offsets so ACP will only turn the wheel. I hope I’m making sense here. This is a hardware problem and I’m guessing the new filter wheel is involved.

[Jon Holtzman]

Thanks Bob, I was not looking for code to get around the issue, just looking for code that I could use to test Dick's suggestion about whether not having MaxIm connect to focuser solves the issue. So I just want to issue a bunch of repeated commands from ACP to change filter wheel and focus. If there turns out to be another issue, like hardware, it would also be useful to have this to try if I change anything. So just looking to be able to quickly command 100 or so filter/focus changes to see if I trigger errors under different conditions.

The daytime simulator looks like it could do the trick, but includes lots of other stuff as well.

Hi Jon —

I would recommend against trying to program your way around the intermittent problem you are having. DIck’s info was meant to help you understand why things behave as they do. As Dick explains, ACP is not seeing the the combination of the filter change and focus change complete. So this could be a misbehaving filter wheel or focuser. Given the vital information about your recent filter wheel change being correlated with the beginning of this problem, I would “focus” there. Maybe you will see the filter move but the the software signal may still not occur.

You can set ACP up to take images during the day (ACP Settings &gt; General &gt; Simulated Images). You can set FocusMax up to run in a simulation mode as well. As used by ACP, FocusMax does not touch the filter wheel. Leave the Scheduler out of the picture since it is just giving ACP work to do, with ACP doing the actual work. So make up a multi-filter observing plan and run it in ACP. It will run to completion including plate solving and pointing updates, etc. To separate the wheel from the focuser, you could create a temporary FilterInfo.txt with 0 offsets so ACP will only turn the wheel. I hope I’m making sense here. This is a hardware problem and I’m guessing the new filter wheel is involved.

[Bob Denny]

Ah, OK. But why do you have MaxIm connected to the focuser? Normally ACP talks to the focuser through FocusMax (or PWI). Just curious. If this is for manual usage, maybe you could save a MaxIm Configuration for that and another for ACP automation?

[Jon Holtzman]

Yes, I think that was just an incorrect configuration on my part, but one that didn't cause any problems for a long time (perhaps because of the hardware configuration/combination). I'm hoping that this will be a quick solution, but would love to just test it quickly with a simple script!

Ah, OK. But why do you have MaxIm connected to the focuser? Normally ACP talks to the focuser through FocusMax (or PWI). Just curious. If this is for manual usage, maybe you could save a MaxIm Configuration for that and another for ACP automation?

[Eric Dose]

one note: For manual focuser changes during an ACP run, the FocusMax "Jog" button is your best friend. Allows both absolute and relative focuser positioning, even during a camera exposure.

[Bob Denny]

Jon -- PS:

The daytime simulator looks like it could do the trick, but includes lots of other stuff as well.

That's why I recommended it. It replicates the CPU load, timing of commands, traffic on USB... in short it's the "real world". YOu can gthen look right at it and maybe see what's going on.

[Jon Holtzman]

Hi Dick,

Just wanted to let you know that I've run for several nights now without any focuser hangups, thanks to your insight and description. I appreciate the constructive comments.