Autofocus Problem - Planewave CDK + EFA

Sequence Generator Auto Focus
1

Hi,
I’ve been experiencing sequence aborts in the night as a result of the focuser not quite reaching the commanded position. The relevant section of the log is below. The problem seems to start, in this case, as autofocus commands 17247 and the focuser reports 17246. What is confusing me is the references, later on, that the focuser has been commanded to 15747 ‘but’ reports 15747. Is this a mistake, or am I missing something?

My OTA is a Planewave CDK14 with EFA connected via the PWI3 interface.

Any comments welcome. I have lost several nights’ imaging due to this problem and it’s starting to drive me crazy.

[08/14/20 00:48:13.215][DEBUG][Focuser Backlash Thread][SQ;AF;BC;] Moving focuser to original requested position (17247)…
[08/14/20 00:48:13.215][DEBUG][Focuser Backlash Thread][SQ;AF;BC;] Focuser backlash compensation needed. Steps: 100 Direction:IN
[08/14/20 00:48:13.215][DEBUG][Focuser Backlash Thread][SQ;AF;BC;] Focuser moving to 17247
[08/14/20 00:48:13.215][DEBUG][Focuser Backlash Thread][SQ;AF;BC;] Focuser move call complete
[08/14/20 00:48:13.215][DEBUG][Move Monitor Thread][SQ;AF;BC;] ASCOM Focuser: Move monitor thread is started…
[08/14/20 00:48:29.663][DEBUG][Main Thread][SQ;AF;BC;] PopulateDataModel: Transferring view to the data model…
[08/14/20 00:48:29.682][DEBUG][MF Update Thread][SQ;AF;BC;] Performing serialize…
[08/14/20 00:49:57.268][DEBUG][Camera Thread][SQ;AF;BC;] ASCOM Focuser: Timeout! Focuser trying to get to 17247, reports it is at 17246. Focuser move complete reports: False
[08/14/20 00:49:57.268][DEBUG][Camera Thread][SQ;AF;BC;] MoveFocuserAbs: End, move complete, at 17246…
[08/14/20 00:49:57.268][ERROR][Camera Thread][SQ;AF;BC;] Error in auto focus! Focuser failed to move to the requested position (17247). Focuser reports it is at 17246. : Error in auto focus! Focuser failed to move to the requested position (17247). Focuser reports it is at 17246.
bei qr.a(AutoFocusData A_0, sz& A_1, sz[] A_2)
[08/14/20 00:49:57.268][DEBUG][Main Thread][SQ;AF;BC;] Adding sequence level notification: Error in auto focus! Focuser failed to move to the requested position (17247). Focuser reports it is at 17246.
[08/14/20 00:49:57.305][DEBUG][Camera Thread][SQ;AF;BC;] Sending Notification: Error - Error in auto focus! Focuser failed to move to the requested position (17247). Focuser reports it is at 17246.
[08/14/20 00:49:57.305][DEBUG][Camera Thread][SQ;AF;BC;] MoveFocuserAbs: Start, move to 15747…
[08/14/20 00:49:57.305][DEBUG][Camera Thread][SQ;AF;BC;] ASCOM Focuser: MoveSync - start…
[08/14/20 00:49:57.305][DEBUG][Camera Thread][SQ;AF;BC;] Focuser moving to 15747
[08/14/20 00:49:57.306][DEBUG][Camera Thread][SQ;AF;BC;] Focuser move call complete
[08/14/20 00:49:57.306][DEBUG][Move Monitor Thread][SQ;AF;BC;] ASCOM Focuser: Move monitor thread is started…
[08/14/20 00:51:57.374][DEBUG][Camera Thread][SQ;AF;BC;] ASCOM Focuser: Timeout! Focuser trying to get to 15747, reports it is at 15747. Focuser move complete reports: False
[08/14/20 00:51:57.374][DEBUG][Camera Thread][SQ;AF;BC;] MoveFocuserAbs: End, move complete, at 15747…
[08/14/20 00:51:57.374][DEBUG][Camera Thread][SQ;AF;BC;] Auto focus (finally): closing dialog…
[08/14/20 00:52:08.282][DEBUG][Camera Thread][SQ;AF;BC;] Auto focus complete…
[08/14/20 00:52:08.290][DEBUG][Camera Thread][SQ;BC;] SGM_FOCUSER_AUTO_FOCUS complete…
[08/14/20 00:52:08.290][DEBUG][Camera Thread][SQ;BC;] Camera thread is IDLE…
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] BlockUntilFocuserMovementDone: Timeout! Focuser trying to get to 15747, but reports it is at 15747!
[08/14/20 00:53:08.363][ERROR][Sequence Thread][SQ;BC;] Focuser movement timeout detected…
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] EventMarker(4) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] Setting filter position 6…
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] Filter position 6 is already set. Skipping…
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] EventMarker(5) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] Checking for auto manual focus (pre)…
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] EventMarker(6) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.363][DEBUG][Sequence Thread][SQ;BC;] Running dither…
[08/14/20 00:53:08.380][DEBUG][Sequence Thread][SQ;BC;] mm - DM dither request sent: 500ms…
[08/14/20 00:53:08.600][DEBUG][Sequence Thread][SQ;BC;] mm - Pulse is async. Settle will be faked for 500ms
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] EventMarker(7) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] Resuming auto guiding…
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] EventMarker(170) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] EventMarker(171) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] EventMarker(172) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] Checking status of auto guider start…
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] EventMarker(173) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:53:08.603][DEBUG][Sequence Thread][SQ;BC;] User has elected to skip guider settling over time…
[08/14/20 00:53:08.656][DEBUG][Sequence Thread][SQ;BC;] Waiting for Auto Guider distance to fall below 0,5
[08/14/20 00:53:08.662][DEBUG][Sequence Thread][SQ;BC;] Distance is below 0,5, starting timer…
[08/14/20 00:53:09.170][DEBUG][Sequence Thread][SQ;BC;] Distance stayed below 0,5 for 0 seconds, done settling…
[08/14/20 00:53:09.171][DEBUG][Sequence Thread][SQ;BC;] Auto guider has settled…
[08/14/20 00:53:09.297][DEBUG][Sequence Thread][SQ;BC;] EventMarker(8) - t:NGC6992-3 (1); e:0; f:2; o:
[08/14/20 00:54:09.364][DEBUG][Sequence Thread][SQ;BC;] BlockUntilFocuserMovementDone: Timeout! Focuser trying to get to 15747, but reports it is at 15747!
[08/14/20 00:54:09.364][DEBUG][Sequence Thread][SQ;BC;] Capture event timeout…
[08/14/20 00:54:09.364][DEBUG][Sequence Thread][SQ;BC;] Run event reported timeout…
[08/14/20 00:54:09.364][DEBUG][Main Thread][SQ;BC;] Adding sequence level notification: Timeout! Sequence has been aborted.
[08/14/20 00:54:09.394][DEBUG][Sequence Thread][SQ;BC;] Sending Notification: Error - Timeout! Sequence has been aborted.

2

I’m a bit disappointed not to have had a response to this. Any of the devs care to comment?

3

This is a known issue with the Planewave focuser. SGP validates that the position is where we requested it to be after focus and since the PWI focuser uses an absolute encoder it cannot always “get there”.

We’ll have to adjust SGP to basically “not care” about the final position of the focuser to get around this. Or potentially there is some way in their ASCOM driver to fake this? I’m not sure.

Jared

4

Thanks Jared. I have read the other thread on the topic and realise it’s a known issue.

I have been using SGP with the Planewave CDK and 10Micron GM3000 for just over a month now and have had an overwhelmingly positive experience. The autofocus issue is the only thing that, literally, keeps me awake at night.

Is the issue common to other focusers that use absolute encoders? Would it be feasible, simply to disable the position check with a checkbox in the autofocus section of the equipment profile? I note that another user has offered to cover the costs of the work involved and I would also be prepared to make a contribution to the programing cost. (The alternative is to switch to Maxim DL, with the associated costs and downtime involved).

5

I forgot to mention this from my original question. What is the significance of the line below? Is this a bug in the code? If so, is it an SGP or ASCOM issue?

Andy

6

Has there been a somekind of solution to this problem?

8

Hi everybody,

This is Kevin, the software developer for PlaneWave. This thread was just brought to my attention.

@andychess: are you still seeing this problem? Based on my reading of the log file, it looks like at least a significant part of the problem is that the focuser interface never reported that the move was finished. Note the references to “Focuser move complete reports: False”. If PWI thought that the move was finished and it was just off by 1 micron from the requested position, this same general symptom might occur (SGP would time out after 1-2 minutes for the focuser to arrive at the final position and then report an error). But in this case it sounds like PWI was stuck reporting that the move was still in progress.

This might also explain some of the later messages where the focuser appears to have gotten to the correct position (15747), but still reports that the move is in progress so SGP keeps waiting until it gives up and throws an error.

I ran a few tests using the trial version of SGP, and it worked fine for my setup. However, I was able to trick the system into reproducing the problem in two ways:

  1. Click “STOP” while the move is in progress to end the move prematurely (so that the movement ends at a position that differs from what SGP requested)
  2. Unplug the motor while the move is in progress, so that the move never completes.

In any case, the behavior you are seeing is unexpected for normal operation, and I’d be happy to look into it. What version of PWI are you running? When you see this happen, does the PWI window still report “Focuser Moving”? And how frequently does this happen?

Best regards,
Kevin

9

Hi Kevin,

Thank you so much for getting in touch and taking up this issue.

Yes, I am still suffering from the problem which is affecting about 10 - 20% of my imaging nights. I run a remote setup and typically image one target and filter per night in order to minimise the number of autofocus runs and therefore the likelihood of failure. Just this week, I lost about 8 hours of time on two nights.

Now that the nights are becoming longer, I’m able to keep an eye on the session for the first couple of hours and have ‘caught’ the problem a few times before going to bed. In each case the symptoms are the same: SGP hangs during the autofocus routine and, if I go to the PWI interface, I always find the actual focus position to be 1 micron away from the commanded position. If I click on ‘GOTO’ on the PWI interface, the focuser corrects to the commanded position and SGP immediately continues with the autofocus run.

You may have read that I am in the process of migrating to Voyager which, as you may know, includes a workaround tab in the focus section that appears to address this issue. If the problem can be resolved within PWI then I would be delighted as, in all other respects, I am completely satisfied with SGP and know the software very well now.

I’m using PWI 3.4.0 with the EFA only (no rotator).

Regarding whether PWI reports ‘Focuser Moving’, I believe it does, but will verify this next time I catch the problem.

Please let me know if you require any further information, logs etc.

Kind regards,

Andy

10

Hi Andy,

Sorry for not following up sooner – I thought I would get an e-mail notification when there was new activity here, but it didn’t seem to come through and I didn’t remember to check back here until just now.

If possible, it might be worth arranging a time when I could log in remotely to look at your system and to exercise the focuser to see if it is working as expected. I have a script that can perform a series of random moves and then check to see if everything is working properly (for example, the focuser eventually reports that it has stopped and is at the correct position after a reasonable amount of time).

I’ll contact you directly to follow up on this.

Cheers,
Kevin

11

Hi,

I have this excatly same problem. Occasionally EFA is 1 micron off from commanded position, or PWI does not report focuser reached position to SGP properly.

If more logs needed, I can provide mine. But issue seems to be excatly same as what andychess is experimenting.

12

Hi Jussi,

Thank you for the report. If you could send some logs I would be happy to review them.

I’ll be in touch about possible remote testing.

Best regards,
Kevin

13

I just bought a CDK12.5 and have experienced the same exact concern. 1 digit off becomes a ‘failure’. Its no worries when I’m awake and simply accept the last good focus point, or re-run AF, but I lost a good 4 hours last night as I went to bed after a meridian flip and 30 mins later a failed AF routine.

14

I spoke with Kevin at Planewave and the latest update corrects this. V3.5.4 or higher. Wanted to put on this thread for anyone experiencing this in the future will know the ‘fix’.

15

Yes, since the release of the new version, I also haven’t had a single problem.

Andy

1 Like