Back to forums

Alignment fails in WBPP only when using dark files

Pleiades Astrophoto PixInsight Stacking Dark Frames
17 replies•201 views
Craig Dixon avatar
Craig Dixon avatar
I'm trying to stack some data acquired with a 6nm Dual narrowband filter on a OSC camera and the data stacks well if I don't use dark frames. However, when I add any darks or master darks, alignment fails. I've been trying to figure this out all day but can't find an answer. I'm using the same master dark as I've been using for months and have also tried the individual darks.
Well Written
andrea tasselli avatar
andrea tasselli avatar
Check that gain/offset/duration are exactly the same between lights and darks. And you don't do any double counting (e.g. using bias subtraction twice).
Helpful Concise
Craig Dixon avatar
Craig Dixon avatar
gain, offset and temp is all the same but I'm not sure what you mean by "double counting" so I'll google that. I'm using the default settings on WBPP and have been using FBPP on defaults to try and find the cause.
Craig Dixon avatar
Craig Dixon avatar
I think I've found it. For some reason, WBPP & FBPP are showing different focal lengths for some of the subs. I'm not sure why this would be as the scope has a fixed focal length so that's the next thing to investigate.
andrea tasselli avatar
andrea tasselli avatar
focal length isn't a reason for not aligning frames. I can align frames with vastly different focal lengths and never fails. Even less so when the difference is less than 0.1% Failure of alignment is due to difference in background or out of focus stars or very high image scale with faint stars.
Craig Dixon avatar
Craig Dixon avatar
@andrea tasselli Would you mind taking a quick look for me? I can't figure this out.

https://drive.google.com/drive/folders/1FeFIULY_IWn-G8EjsqHDVhMiJGuTaP8X?usp=sharing

I've included the lights, flats, darks and dark flats (The dark flats don't match the exposure time exactly but this is the closest I have. WBPP is usually okay with this if it's close enough). I've also added a folder for the stacking attempts I'e made. As you can see, the data is aligned properly if I use no calibration files. If I add darks, alignment fails and if I just use flats, the flat correction is all wrong. I've also included the log form FBPP when all calibration frames are included.

This is 6nm dual narrowband data from a 533MC Pro to capture the Ha regions in M63.

Thanks for your help with this!
andrea tasselli avatar
andrea tasselli avatar
I'll give it a go. Let you know.
andrea tasselli avatar
andrea tasselli avatar
Apparently you are calibrating the light frames twice. Also, I'm not sure how, the light frames have been optimized ahead of being debayered (in fact ahead of everything) which can only lead to wrong results. This is a step I didn't include in the final integration, just relaying on the SNR as quality estimator. All frames calibrated and aligned as expected. Results below. I can send you the final integration, should you wish so.

Color Corrected:
Craig Dixon avatar
Craig Dixon avatar
That's very odd. Why on earth would that happen? I'm just using the default settings in WBPP and FBPP. This stack was actually just a trial run as it's the first time I've used the filter and I wanted to check everything was good before collecting more data.
andrea tasselli avatar
andrea tasselli avatar
I'm not sure. But the only thing obvious was the automatic dark matching which should never be used. But there is no issue with any of the steps, if followed in the proper order. Apart from optimization which was never carried out as it couldn't align the frames.

Again and again, don't use WBPP.
Craig Dixon avatar
Craig Dixon avatar
Oaky, thanks, I really appreciate you taking the time to help me again. If you don't use WBPP, which method do you use to stack?
andrea tasselli avatar
andrea tasselli avatar
Also, set to overwrite, avoids confusion.
andrea tasselli avatar
andrea tasselli avatar
These:

These are the steps that WBPP, as a macro. would carry out for you. But now you have a much tighter control on each and every single outcome.
Craig Dixon avatar
Craig Dixon avatar
Thanks. I'll have a read up on all of those processes and have a go at doing this manually. The problem seems to be specific to this data as everything else I stack comes out as expected. Very odd.
Well Written
Craig Dixon avatar
Craig Dixon avatar
I felt like I needed to get to the bottom of this so I did some experimenting. the subs stack perfectly on their own and with flats (Albeit the flats aren't dark-corrected). Any time I introduced either a master dark from my darks library or a series of dark subs, the alignment failed. I quickly shot 10 new darks and the stark worked perfectly. For some reason, my darks suddenly stopped working. Needless to say, my next job will be to put the camera in the fridge and shoot a new set of darks to create a new dark library.
Helpful
andrea tasselli avatar
andrea tasselli avatar
Why in the fridge?
Craig Dixon avatar
Craig Dixon avatar
It's currently 25 degrees here so keeping the camera at -10 for several hours is difficult. If you put the camera in the fridge (which is dark when you close the door) with the cap on, the cooler doesn't need to work so hard to keep the sensor at -10
Helpful Concise
andrea tasselli avatar
andrea tasselli avatar
Go figure. Mine didn't seem to have any problem with that. Now is dead, however.