SGP did not pause PHD2 during AF run


#1

Well, at least I’m full of helpful (?) error messages the last couple days. :slight_smile: SGP and PHD2 guide and debug logs are here. I started the sequence, SGP slewed to the target, platesolved, started PHD2, which found a star and started guiding, then SGP directed an AF run. All correct so far, but then PHD2 kept guiding, though I had checked the box to pause guiding during AF runs. I know SGP does not pause PHD2 outside of a running sequence, but the sequence was running here. Maybe because this was the first AF run of the session?

Kevin


#2

Isn’t there a checkbox for that? I know my copy pauses it on startup regardless of what I do b


#3

Kevin, I only looked at the PHD2 log, but according to that there was no pause request from SGP.
Andy


#4

Chris, there is a checkbox, and it was checked. I went back to be sure after I noticed it kept guiding.


#5

The option to pause was definitely on. You seem to have entered into some weird state in PHD2 during this time. We attempt to pause, but instead get a lock lost message back from PHD2. Rather than terminate the entire sequence, SGPro is programmed to roll with this and see if your focuser can come through on the other side OK (regardless of this failure). Maybe there is a better way to handle this… not sure. Maybe @Andy has recommendations based on how PHD2 gets into this state. Maybe there is even room for a type of mid-exposure / mid-sequence recovery here (similar to restarting the exposure if your guider goes off the rails).

[6/29/2015 11:33:25 PM] [DEBUG] [Camera Thread] Pause guiding during AF is active, pause command sent, waiting for guider to pause…
[6/29/2015 11:33:25 PM] [DEBUG] [Camera Thread] Checking PHD2 state…
[6/29/2015 11:33:25 PM] [DEBUG] [Camera Thread] PHD2 GetPhdStatus - Pre-Wait : Guiding
[6/29/2015 11:33:25 PM] [DEBUG] [Camera Thread] Sending to PHD2:
{“method”: “get_app_state”, “id”: 1001}

[6/29/2015 11:33:25 PM] [DEBUG] [Camera Thread] PHD2 GetPhdStatus - Post-Wait: LostLock
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] PHD2 GetPhdStatus - Post-Wait: LostLock
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] PHD2: Requested pause, but PHD2 reports it is not guiding
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] SGM_GUIDE_PAUSE complete…
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] SGM_GUIDE_PAUSE message received…
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] Pausing auto guider…
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] Checking PHD2 state…
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] PHD2 GetPhdStatus - Pre-Wait : LostLock
[6/29/2015 11:33:25 PM] [DEBUG] [Auto Guider Thread] Sending to PHD2:
{“method”: “get_app_state”, “id”: 1001}


#6

Here’s just a wild guess. When SGP sets a new filter (I use red for AF runs) the focus position is adjusted and guiding continues during that process. The guide star can move out of the search box during this process in my scope, since backlash comp kicks in and mirror shift pushes the star quite a bit. If the guide star is lost during this process, even temporarily, would that cause this problem to occur? If so, maybe guiding should be paused before the filter change?


#7

Kevin,

I think you are on to something there! The SGP log shows the filter changing at 11:33:17 with the associated focuser position change. Then, right after that PHD2 loses the guide star due to star mass change setting. The guide star did not leave the search region, but the star mass value change caused phd2 to drop the guide frames.

Although you could disable the star mass change option in PHD2, I think your idea of pausing PHD2 before the filter change would be a better fix for the problem.

Andy


#8

Kevin,

That is a good observation and removes a lot of the convenient “coincidence” that PHD2 suddenly, for unrelated reasons, just happened to act up exactly when AF started. I can’t verify now, but I will when I get home.


#9

I don’t recall seeing this issue with my OAG setup and TEC 140 refractor. If I remember correctly, I think I have seen the message at bottom of PHD2 (or is it SGP?) screen saying something like “guiding paused”, do you see this message?

I think my procedure may be different than yours. I usually do the following:

  1. Slew to a star near target DSO.
  2. Manually focus the star with Bahtinov mask.
  3. Center and sync on that star.
  4. Slew to DSO using SGP.
  5. Auto center DSO using SGP.
  6. Manually run AF to make sure it works.
  7. Start PHD2, manually pick a guide star and start guiding.
  8. Once guiding is stabilized, start SGP Sequence run.

I never let SGP Sequence start PHD2 to pick guide star and start guiding right after I start SGP Sequence run. The only time I let SGP to tell PHD2 to find guide star and start guiding is after Automatic Meridian Flip and Auto Center.

Peter


#10

Peter,

If we do pause after filter change (I think we might), it is likely that your filter change is very quick so you don’t notice. Kevin, on the other hand, might have a slower filter change coupled with a slower movement of focuser to get to that filter’s required position. Once all of that is done, PHD2 could have choked before we even send the pause command.


#11

FYI, my filter wheel is part of QSI camera and it’s 5 positions.

Peter


#12

The source of the issue for me isn’t the filter change per se, it’s the focuser offset with the filter change. When the filter changes to red from lum, the anti-backlash kicks in, pushing the focuser out 500 steps and then back in. And with my scope that means the mirror shifts, and I can lose the guide star. Of course you wouldn’t see this issue with your refractor, Peter. It’s times like this when you really miss your old C8, right? :smile:

Kevin


#13

Wrong. :smile:

Peter


#14

@astrovienna I have moved the guider pause in front of the code that changes filters and focuser pos. Hopefully this will help with your issue. It will be out with the first 2.4.2 beta.


#15

Thanks Ken! I’ll let you know how it works out.


www.mainsequencesoftware.com