More problems in Paradise

103 views
Skip to first unread message

Alan “Big Al” C

unread,
Jan 31, 2024, 9:23:02 AM1/31/24
to Open PHD Guiding
Good morning,

I am having more problems in paradise. The guiding last night went off the rails again. I made sure PHD2 was selecting the correct camera for guiding, it was on the native ZWO driver, and I calibrated at the meridian and equator as always, ran the guiding assistant, and applied the suggested settings.

But it will not guide as if it is calibrated. It will guide on my first target of the night, but as soon as it slews to the second target for the night, it goes off the rails. It was windy last night, but I have imaged before in worse conditions and it stayed calibrated. I finally got it to sort of work by doing a complete reboot. When I plate-solved on M45, PHD2 calibrated itself (I didn't request the calibration). So it calibrated on target at 20:31 and worked until it switched to the next target at 00:15:00. Presumably because it was calibrated on M45 and not M51.

So short of reinstalling PHD2 and starting from scratch, is there a way I can force PHD2 to calibrate at the start of every target since it won't calibrate at the meridian? Or is this more hardware problematic? This all started with my new ZWO ASI2600MM. Before this camera, my guide camera was connected directly to my PC via USB and it worked flawlessly. Now the guide camera is connected to one of the two USB 3.0 ports on the 2600mm camera and the camera is connected to my computer. It hasn't truly worked properly since this seemingly minor change. So perhaps I should go back to the original connection???Capture 1.JPG

Capture 2.JPG
PHD2_GuideLog_2024-01-30_190053.txt

Alan “Big Al” C

unread,
Jan 31, 2024, 12:10:08 PM1/31/24
to Open PHD Guiding
So I am looking at this log file in a text editor, and I notice this message on every start to guiding, "no dark, defect map in use". I have built a dark library and bad pixel map multiple times now. Does this line mean that PHD is not using the dark library and bad pixel map that I have built? If that is the case, where do I go in the settings to apply them?

Brian Valente

unread,
Jan 31, 2024, 12:12:29 PM1/31/24
to open-phd...@googlegroups.com
Alan

part of the problem is you aren't being terribly methodical about problem solving.

You have a profile named "bin2" and you set it to bin1. That will cause things like dark libraries etc. to become invalid (you probably got a message from PHD on this as well)

If you are changing Bin level, you need to set up a separate profile, and when you want to move between them, you change the chosen profile, not just change the binning



--
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/81c03ae8-73a2-4670-84d1-abaadc37e1dfn%40googlegroups.com.


--

Alan “Big Al” C

unread,
Jan 31, 2024, 12:55:29 PM1/31/24
to Open PHD Guiding
" you aren't being terribly methodical about problem solving " Well, I am here asking questions about what could be wrong. Kind of part of the problem solving process... ask someone who knows more.

" You have a profile named "bin2" and you set it to bin1. That will cause things like dark libraries etc. to become invalid (you probably got a message from PHD on this as well)" I am not sure where you see that is set to bin1. It is true that I have a profile named bin2 and it was set to bin1. And I did change it and I did get a message for doing so. I created the darks and BMP and went on my way because the messages went away. It did not occur to me that I should have created a new profile. I will try that this afternoon and go through the new profile wizard.

This is why I asked if the camera connection could be to blame. I know USB can be a bottle neck and I am only further creating a problem by inserting the image camera between the guiding and the imaging. Either way, I will start a new profile using the profile wizard and start at default state.

Brian Valente

unread,
Jan 31, 2024, 1:50:54 PM1/31/24
to open-phd...@googlegroups.com
Alan

>>> Well, I am here asking questions about what could be wrong. Kind of part of the problem solving process... ask someone who knows more.

*methodical* was the key point here, not about whether to ask questions. 

Change one thing at a time, evaluate, and if needed, make one more change, repeat. If you change binning, you need to create a new profile and recreate your calibration frames. they don't transfer to different binning. 

The main thing I see here is you are trying to problem solve your guiding while also trying to get a full night of imaging. imo that's a recipe for frustration and not getting it fixed. 

You might take a night and just focus just on getting guiding working correctly. Otherwise you are rushing to solve it to get to imaging (we are all prone to do this, myself included).  One and done right?

As far as I can see, there are no camera connection issues here. Have you seen any complaints about camera disconnecting or timing out? You do need to be aware of the dual zwo cameras and the driver possibly connecting to the wrong camera, but I think we've covered that

You have poor calibrations that result in poor guiding. Until you solve the calibration problems, you aren't going to get good guiding. I would create a new profile for bin 1 and create the necessary settings and ark frames. Then stick with the calibration assistant and focus on getting a good result there. Once you have that, do a guiding assistant run and follow those recommendations. 

Once you have a good calibration, leave it alone. 

the issues I do see:
- changing equipment profile details
- lost star events
- improvements for guide camera focus
- erratic mount performance (a guiding assistant run of 10-15 minutes can help identify issues here for you and for us) and/or
- poor sky conditions





Alan “Big Al” C

unread,
Feb 1, 2024, 10:55:24 AM2/1/24
to Open PHD Guiding
Well I think I may have gotten it. It didn't go off the rails last night so that's good. The guiding is still lousy but I am chalking it up to bad seeing. The wind was blowing about 30 mph... anyway, I took the suggestions offered. Below is summary of them

I turned everything off, reassessed the cables, the balance, the "home position", and the polar alignment. I adjusted the balance of the RA by a fraction but everything else is good. I ensure the camera was securely fastened and no visible movement. The guide and OAG are securely fastened and as far as I can tell, the OAG is securely connected to the OAG. I polar aligned using an electronic polemaster camera and only needed a very minor adjustment in azimuth.

Issue: Changing equipment profile details.
Recommendation: Create a new profile for the gear.

I created a new profile using the new profile wizard. There are three different drivers shown on my screen for ZWO cameras. ASI ZWO Camera, ASI ASCOM (1), and ASI ASCOM (2). I selected each driver to see which pixel size would auto populate - only one would and that is ASI ASCOM (1). I have it set to Bin2 with 2x2 mean noise reduction, the calibration does not auto restore, and a minimum HFD of 1.6 with maximum of 10, and minimum SNR for autoselect of 30. (the default is 6 but that has never worked for my guide camera on a OAG although it did work flawlessly with my guide scope so maybe it's a pixel scale thing?) The only thing about this profile is the gain settings for the camera are locked. I can't change them and its set to 100. I don't have an answer yet as to if this is a good thing although it did work last night.


Issue: Lost star events
Recommendation: ??

This is why I pumped up the minimum SNR originally. Setting it at 30 solved all those problems with my last OAG. It worked again last night as well.


Issue: Improvements to guide camera focus
Recommendation: Per PHD2 user manual, use the FWHM profile to fine tune the focus.

Done and done. Although it might need further adjustment on a calmer night. The turbulent air was making some awful bloated stars.


Issue: Poor sky conditions
Recommendation: don't image and don't guide. Have a beer instead

I didn't image, but did do the above and I did have an adult beverage.


Issue: erratic mount performance
Recommendation: a guiding assistant run of 10-15 minutes can help identify issues here for you and for us

I calibrated as close to the meridian and 0° DEC as I could get. I then ran a calibration assistant and applied its suggestions. It slewed to 0° and +5° of the meridian point west. Ran the calibration a second time, I got a notification telling me the new calibration was significantly different the previous calibration. I did not get any dialogue boxes during the sanity-check. The CA called the new calibration acceptable so I accepted the new calibration. Next I slewed to a different target on the opposite side of the meridian to see if it would maintain the calibration and it did. But the RA was bad... 2.37" bad. So I ran a guiding assistant and let it run for 573 seconds. The GA said to try using 2-3 second exposures, and applied some adjustments to the RA and DEC min-move settings. Resumed guiding for another couple of hours or so after that with no lost stars, it performed a meridian flip and resumed guiding again with no lost stars. The DEC error was good, if I remember it was 0.85-ish. The RA was still bad, 1.73 after guiding for unattended for a couple of hours.

At this point, I am short on ideas of what to try next. I will try again on a calmer night - perhaps the bad seeing was the culprit. Either way, I took a 3 minute sub and 5 minute sub during the guiding to ensure the camera were staying connected (that was issue #1 previously, the image camera and guide camera were on the same driver). I still don't understand the ASI ASCOM (1) driver - I really wish I could rename these drivers to something more specific but at least it is evidently specific to the ASI290mm guide camera. I just can't figure out why the gain settings are off limits to me now when before they weren't. But I digress. The 3 minute and 5 minute sub both had a good star profile albeit fat - presumably from the bad seeing conditions.

Brian Valente

unread,
Feb 1, 2024, 11:00:17 AM2/1/24
to open-phd...@googlegroups.com
Hi Alan

That sounds like good progress

can you share the logs? If your Dec is okay and your RA is significantly worse, it could point to performance issues in your mount



>>>This is why I pumped up the minimum SNR originally. Setting it at 30 solved all those problems with my last OAG. It worked again last night as well.

I also have a higher min SNR, it's a good idea imo



Alan “Big Al” C

unread,
Feb 2, 2024, 11:57:50 PM2/2/24
to Open PHD Guiding
You bet. Sorry about the delay. I have had a busy couple of days.
PHD2_GuideLog_2024-01-31_182052.txt
Reply all
Reply to author
Forward
0 new messages