LRGB Processing Problem

[Veej]

ooooh, its my first post.... Hello! ( long time lurker, regular reader, hoarder of good things from FLO )

So I have dived into monochrome imaging - I'm shooting on a Qhy268 m with Antlia V-Pro LRGB filters and in this case im using a Starfield 102.

I have stacked and processed the L,R,G and B channels separately in Pixinsight using the Weighted Batch Preprocessing script. I'm getting no rejected frames and each channel has its own set of calibration frames. L,G and B each take around 15 mins to process, but R takes closer to 2 hours.

The following image shows the result of these stacks with a single STF  applied to make them visible. As you can see, the R channel ( top right )  is really oversaturated which obviously doesn't work well when I combine the LRGB. I've tried rerunning WBPP but the result is always the same.

Any thought or suggestions? Am I missing something obvious in Pixi or could it be something more fundamental like a back to front filer ? ( they are 36mm unmounted )

[tomato]

On the face of it the red stack looks like it has got swapped for the Lum stack, but if you processed each channel separately it’s hard to see how this could happen. Why is the red stack taking much longer to process, are the integration times the same for the RGB channels?

[Veej]

you are not the first person today to suggest that maybe the L and R are swapped - im about to give that a test. ( i have confirmed the filters are in the right order in the FW and NINA labelled them so its not human error )

the integration times and amount of subs are the same across all channels, with each having their own set of calibration frames.

it seems that for some reason it took 2 hours to platesolve - im assuming because of the over saturation. ( no idea why though )

 

Edited November 14, 2023 by Veej

[Veej]

ive taken a few subs tonight with the L and R filters. It looks like they are behaving as expected - Images taken with L are brighter than with R given the same exposure time. 

If the weather gods allow, I will redo the R and L subs + calibration frames tonight and try a fresh set of processing. I think this is starting to look like the problem is in software land - maybe a bad cache or incorrectly labelled images.

[Budgie1]

9 hours ago, Veej said:

I have stacked and processed the L,R,G and B channels separately in Pixinsight using the Weighted Batch Preprocessing script.

Just so I can get it clear in my head did you:

1. Run a separate instance of WBPP for each channel - Run WBPP for Lum, let it finish then reset the lights with the Red subs & calibration frames and run WBPP again, continuing on for Green & Blue?
2. Or did you load all the channel subs into the Lights folder in WBPP, with their corresponding Flats & Flat-Darks and just run WBPP once to create all four stacks at the same time? 

If you did #1, have you tried #2 to see if that makes a difference?

[Veej]

I did 1. 

is 2 a possibility with WBPP ? ( I'll give it a try tomorrow and find out )

i did find the clear cache button tonight, so i gave that a press just in case.

[Budgie1]

Yes, I only use #2. it saves time and gets the job done in one pass. If you use the auto-crop then it lines up all the channels together and crops them, so you don't have to do that in post-processing either.

When you load all the lights into WBPP it will recognise the different filters from the FITS header or the file names and group them accordingly. Likewise with the Flats, it will group these by exposure time or filter name.