Runaway guiding with AP Mach2 and one specific guide star

[luca...@gmail.com]

Hello,

I am a bit baffled by this issue I have encountered and hope that you guys can shed some light on what is happening.

My setup is FSQ106 with OAG on a AP Mach2 mount. I have framed a wide-field view of IC410 and when PHD2 selects one specific star that is a bit oblong (possibly an unresolved double star) there is dramatic runaway guiding as can be seen in the logs. I thought it may be due to bad calibration, so I calibrated again but it didn't solve the problem. Last night, after stopping and restarting PHD2, it selected a different guide star and all was well.

Do you have any thoughts on what might be going on?

Link to guidelogs: https://openphdguiding.org/logs/dl/PHD2_logs_MzXV.zip

Thanks,

Luca

[luca...@gmail.com]

To add a bit of information about the guide logs:

2-15-2022

Log 1,2 Sn2-223 (before and after meridian flip). No problems, guiding is seeing limited.

Log 3-8 IC410, problem guide star, runaway guiding

Log 9, 10 Calibration and guiding assistant

Log 12 IC410, different guide star, all good

Log 13, IC443, all good

Log 14-16, M97, all good

Log 17, M81, all good

Same story on 2-16-2022:

Log 2 and 5, IC410 same guide star as on 2-15-2022, runaway behavior

Log 6, IC410, different guide star, no problem

[Bruce Waddington]

Hi Luca.  It looks to me like the problem “star” is really sensor noise.  I recommend that you try using a bad-pixel map for the guide camera, excluding that area of the sensor if you can.  The LodeStar sensors are somewhat prone to this sort of thing and that’s actually the reason we implemented bad-pixel maps in the first place.  The problem area may come and go over time, perhaps as a function of temperature, but that shouldn’t cause a problem if you can get the bad pixel map to cover that area of the sensor.  For reference, the problem area is located around X=983, Y=407.  Another thing you could try is using the native driver for the camera.  The two drivers handle the image downloads slightly differently and perhaps the native driver will change the appearance of the problem spot.

 

Let us know how it goes,

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/5c6f626d-06dc-4daa-8deb-43d8b5d0a2e1n%40googlegroups.com.

[luca...@gmail.com]

Hi Bruce,

Thank you for the quick response. Your suggestion makes sense and I will give it a try. I have generated a Bad Pixel Map and have switched to the native SXV driver. I am still scratching my head a bit on this one because if it truly was a bad pixel issue, one would imagine that the problem would arise with every field of view. Nevertheless, if I can erase the problem star by modifying the bad pixel map, sensor defect or not, this should take care of the problem.

I'll update the thread once I have a chance to test under the stars.

Thank you,

Luca

[Bruce Waddington]

Hi Luca.  I share your lack of confidence on this and what I’ve suggested may not fix the problem – it’s just a comparatively easy thing to try and a good thing to do in any case.  Another possibility that comes to mind is that you may have an internal reflection in the guider field of view, something that a few others have encountered at times.  That would explain why it’s such a transient problem.  When you do your testing, it would be a good idea to capture a guide camera image if you see the problem recur.  You can just use the File/Save Image menu item to do that.  Once the image is captured, we can look at it with full-scale imaging software to get a better idea of what the problem is.

 

Good luck,

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/b82d074a-a22b-41c3-9d93-a54c0a7826f3n%40googlegroups.com.

[luca...@gmail.com]

Unfortunately, the Bad Pixel Math did not fix the problem. New guide log here:

https://openphdguiding.org/logs/dl/PHD2_logs_RM7x.zip

Runaway guiding is bad again at the same location and only on IC410 target. After it settles down having moved in Dec a large but very repeatable distance, guiding is perfect but at the wrong location (shifted FOV). I think you have an interesting idea regarding the reflection from the OAG. I know that if the OAG is a bit lower, it will put star like structures in the imaging camera FOV, so it's not impossible that it's projecting a reflected star on the guide camera, especially given the crowded star field in Auriga. Unfortunately, I did not see your message until Auriga set, so I'll have to wait for the next session to collect an image.

Thanks again,

Luca

[luca...@gmail.com]

I collected more data of the runaway guiding behavior. The new log is here:

https://openphdguiding.org/logs/dl/PHD2_logs_bA7D.zip

Same target, same problematic star. I am starting to think that you may be right that it is a reflection from the OAG prism. As you requested, I collected two images, one from IC410 and one from M106.

The links are below:

https://www.dropbox.com/s/z5z39c433prea0o/IC410.fit?dl=0

https://www.dropbox.com/s/6nvwib2etzy9hth/M106.fit?dl=0

Thank you!

Luca

[Bruce Waddington]

Hi Luca.  Yes, these definitely look like internal reflections, even more than one in a frame:

 

 

Is this a high-quality OAG?  Have you looked inside to see if there are any unpainted surfaces or fittings?  I am thinking about ways we can try to avoid these things in star-selection but it would be a safer solution if you could eliminate the source of the problems. 

 

Regards,

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/9f47d7ea-b3e8-4b56-afad-0b6faf42072an%40googlegroups.com.

[luca...@gmail.com]

Yes, it is a high quality OAG (Optec Sagitta) but it has a very large prism and it has created reflections in the imaging FOV before. I'll check with Optec as well to see if by any chance my unit could have a manufacturing defect.

thanks!

Luca