Back to forums

How to avoid guiding on reflections in the OAG within PHD2? (Gif of the issue included)

Open PHD Guiding Project PHD2 Guiding Troubleshooting Artifacts
15 replies•581 views
Ashraf AbuSara avatar
Ashraf AbuSara avatar
I have been having problems, especially when imaging on bright moon nights, where reflections on the OAG pickoff prism are causing significant deviations in guiding, where PHD2 locks on these reflections and the stars go 20-30 arc seconds in the opposite direction, completely destroying the exposure. 

I tried to single out these reflections by choosing a tighter selection criteria for HFD and SNR, but I still can't help but have some get selected in multistar guiding. Here is an example gif I made from the recorded frames in PHD2 when this happens to debug the issue. You can see the reflections in the top corner. The stars are flying by because PHD2 is trying to keep the top left reflections in the frame.



Any suggestions other than dismantling the remote equipment? (something I really don't want to do now).
Well Written Engaging
Paul Homer avatar
Paul Homer avatar
Ashraf,

Would it be possible to post an image or images of the guiding frames.

I have a QSI camera with an OAG front and I have never seen any reflections in the guide camera.

How close to the Moon are you pointing? I do not generally image close to the Moon due to the moonglow.

Paul
Concise Engaging Supportive
Logan Carpenter avatar
Logan Carpenter avatar
I'm not sure how to get rid of the reflections but if the reflections are only out in the corners/edges and you are using NINA, in the guiding settings in NINA you can select the percentage of the ROI to find a guide star and reduce that value so you are not including stars or reflections out towards the corners/edges. I personally haven't tried this myself but it is a setting in NINA.

Ashraf AbuSara avatar
Ashraf AbuSara avatar
Logan Carpenter:
I'm not sure how to get rid of the reflections but if the reflections are only out in the corners/edges and you are using NINA, in the guiding settings in NINA you can select the percentage of the ROI to find a guide star and reduce that value so you are not including stars or reflections out towards the corners/edges. I personally haven't tried this myself but it is a setting in NINA.


Thanks Logan. My understanding is ROI feature is not compatible with multistar guiding, which is a deal breaker.
Logan Carpenter avatar
Logan Carpenter avatar
Ah ok I didn’t realise that.
Ian Parr avatar
Ian Parr avatar
Had similar issues with a moon too close to my field so I made tube out of black cardboard to put over the end the guide scope and that worked. But the gradients exposed my folly.
There is also black paints (>99% non reflective) that like the unavailable Vanta Black can help.
For OAG not quite as easy but principal works the same so adding a black extension tube to the scope not only helps with OAG but dramatically reduces gradients from light pollution etc. Not elegant but very effective when needed and non-reflective black paint can work great but gently blow off any dust that accumulates on the paint.
andrea tasselli avatar
andrea tasselli avatar
Use a deep R filter on the OAG camera, that should get rid of most if not all reflections unless you're very close to the Moon, which you shouldn't?
Ashraf AbuSara avatar
Ashraf AbuSara avatar
Paul Homer:
Ashraf,

Would it be possible to post an image or images of the guiding frames.

I have a QSI camera with an OAG front and I have never seen any reflections in the guide camera.

How close to the Moon are you pointing? I do not generally image close to the Moon due to the moonglow.

Paul

Hi Paul. The animated gif I posted actually shows the frames from the guide camera as captured from PHD2. You can see the streaks of faux stars in the top left corner.

Not shooting close to the moon at all. In fact I realized when looking at the timestamps for the frames that this happened minutes before the moon was even over the horizon, so they are certainly not common from moonlight. Maybe a bright star out of the FOV.
Well Written Helpful Insightful Respectful
Ashraf AbuSara avatar
Ashraf AbuSara avatar
andrea tasselli:
Use a deep R filter on the OAG camera, that should get rid of most if not all reflections unless you're very close to the Moon, which you shouldn't?

Not shooting close to the moon at all. In fact I realized when looking at the timestamps for the frames that this happened minutes before the moon was even over the horizon, so they are certainly not coming from moonlight. Maybe a bright star out of the FOV. 

The R filter is a good idea though. Hopefully it doesn't hurt SNR too much at f/7.7.
Tony Gondola avatar
Tony Gondola avatar
andrea tasselli:
Use a deep R filter on the OAG camera, that should get rid of most if not all reflections unless you're very close to the Moon, which you shouldn't?

+1 or ir pass
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
The least intrusive and easy to do change would be to put a filter on the guide camera and refocus it and see if that helps. 

if you increase the guide exposure time does it change the reflections at all? Curious if doing that in concert with your HFD tweaks could eliminate them as candidates entirely?
Concise Engaging
Ashraf AbuSara avatar
Ashraf AbuSara avatar
Dark Matters Astrophotography:
The least intrusive and easy to do change would be to put a filter on the guide camera and refocus it and see if that helps. 

if you increase the guide exposure time does it change the reflections at all? Curious if doing that in concert with your HFD tweaks could eliminate them as candidates entirely?

That's a good idea. The problem with making those changes is it is really hard to get a direct feedback on whether it helped or not until it happens again. The issue is quite random. I''ll give the exposure changes a try as long as it is does not impact guiding.

PHD2 is obviously a great piece of free software. But I wish PHD2 was a bit smarter in this regard. If the locked star has significant diviation in movement compared to the majority of those other samples (something like > 2 standard diviations) then it should be rejected and a new different star should be selected.
andrea tasselli avatar
andrea tasselli avatar
You can chose which star is going to be the guiding one.
Ashraf AbuSara avatar
Ashraf AbuSara avatar
andrea tasselli:
You can chose which star is going to be the guiding one.

I get that, but I don't keep watching my remote setup the whole time. I just check on things every once in a while.. Everything is automated with multistar guiding implemented. This happened a few times around 2-4am when I looked the next day digging in the logs to troubleshoot my subs with significant streaks.
andrea tasselli avatar
andrea tasselli avatar
Oops, forgot about that tiny detail…
Paul Homer avatar
Paul Homer avatar
Ashraf,

How long have you had the OAG?

I would explain your issue to the manufacturer as I view it as a fault.

An alternative is to try and remove the cause of the reflections.

Paul