John,

I have noticed twice now that when the scope performs a meridian flip it says that it is turning the Optec rotator, but the first image is flipped as if it was unflipped. If I abort and immediately restart the run it then does the correct flip without any problems. I noticed that I have the rotate box checked but I also hav ethe box checked for automatically select guide star. Is the second box causing this problem?

Chris,

Please post the log for this session.

Here are the log files the first feiled flip.

ccdap20080102_190934.log (12 KB, 60 downloads) first filed flip

Here is the successful flip

ccdap20080102_200225.log (13 KB, 59 downloads)

Hi Chris,

Your second log was not that of a successful flip but a run after the target had crossed the meridian.

The first log indicates the rotator was initially at 142 and moved to 322 after the meridian flip, as expected. Thus, the rotator was told to move and it reported back that it moved. Whether it did, or whether the camera did is another question.

If you can repeat the test, set up a random target in the east near the meridian with a PA of 228 or so. Initially move the rotator away from 142. Watch to see if the rotator moves properly as the run starts and later as the target crosses the meridian. If the rotator doesn't move as expected, I would check for loose fittings on the camera, insufficient cable slack, etc.

BTW, the current CCDAP version is 3.42.1.

John,

While I can and will do what you requested, There is no problem with the actual rotator-camera complex. As you can see from the second log, the camera assumed the correct angle flawlessly from a cold start. What you cannot see is the images that it took. After the first log file where the meridian flip was done while imaging, the images were upside down, despite what the softwre reported. In the second file where I started the run with the target on the west, the images were in the same orientation as they had been when I imaged on the East of the meridian. This tells me the camera can rotate fine, and it has rotated several other times before and since. This is the second time that the reported rotation on meridian flip did not actually occur. On both occasions I was able to restart on the west with the correct rotation occurring immediately.

Again, I come back to the check box about automatically acquire guide star after meridian flip. Should that be checked? Is is possible that this is somehow messing up the software to think that it shouldn't perform the rotation?

I am going to uncheck that tonight or as soon as I can and see if the problem is solved.

Hi Chris,

The "auto guide star select" will only stop CCDAP3 and allow you to select the guidestar before it continues after the meridian flip. It doesn't affect the rotation option. Make sure you have saved your profile with that option checked. I thought I had once and all my images were inverted after the flip. Turned out I had not saved my profile with the check in the box.

Thanks, Frank. The box is definitely checked, but it still reports a phony rotation. I would prefer that it rotate so it reacquires the originally intended guide star and doesn't wait for me.

Chris,

The second log you supplied was taken after the meridian flip, i.e. was taken after the target had flipped the meridian. Are you saying the second log was run immediately after the first the rotator moved at the start of the second run then but it didn't as part of the first run? I have no reports from the many other Pyxis users indicating a rotator command failure.

That is why I suggested a test while watching.

Auto guide star select should have no impact on rotator operation. To clarify, if autoguide star select is not checked, then you must manually define the guide star. The preferred operation for most users is of course checked.

John,

Yes what I didn't clarify is that the first log was the meridian flip that didn't rotate. The second was the restart immediately after that previous failed rotation. Interestingly, that restart resulted in a successful rotation, judged on the basis of the visual inspection of the frame following the meridian flip being inverted and the frame after the restart being in the same orientation as the East of meridian images.

Last night I "rediscovered the wheel" as you described about the automatically select guide star. It did ask me to select the guide star after the flip and it did not change the problem as again, the image was inverted immediately followint the meridian flip. And again, abort followed by an immediate "run" to restart resulted in a correct rotation on the West side with the proper image orientation.
I believe you that you haven't seen this before, but it is happening pretty consistently.
When it isn't 14 degrees outside I will try to recreate this and observe the rotator's behaviour. Since it works every time that I restart the run, I am doubtful that it is a mechanical failure such as the camera slipping in relation to the rotator so the rotator turns, but the camera remains stationary. That would seem to cause failure to rotate to some extent on all occasions. Think if there is any way that there could be a miscommunication and I will try to inspect the problem.

For the moment, I am working around it as I only do semi-unattended imaging.

Chris A

Hi Chris,

One thing you might try is to manually rotate your pyxis with CCDAP3 and watch what happens. If all your connections are screw fittings, then make sure that camera is not "unscrewing itself" when you rotate it in a CCW direction. I had this happen to me and it took me a while to figure out what was happening. I finally turned on my observatory webcam and watched the camera move as the rotator went from 0 to 90 to 180, etc. What I saw was the camera start to move and then "slip" (I have a STL-11K with FW8, and that is a heavy offset package). After I saw it happen, I knew what I needed to do and had someone tighten up the camera rotator connection for me. It has been working just fine now since I got all my connections tightened. Hope you get this resolved quickly.

Chris,

Are you using a real serial port or a USB-serial converter? If the latter, what brand of USB-serial converter are you using and how many hubs are there between the PC and the Pyxis?

Has this always happened since day one or is this a recent occurrence? If recent, what has changed between when it was working and when is stopped working - either hardware or software?

For anyone reading this thread using a Pyxis rotator, have you seen this problem? If you have not, are you suing a real serial port or a USB-serial converter? If the latter, what brand of serial-USB converter are you using?

Thanks Frank. Since it is cloudy tonight I will try to do some trouble-shooting on the connections etc.

Chris