ASI1600 pro, all calibration frames match, flats seems to over-correct. Why?

Shlomi

Active member
Hi,

I have been astropicturing for a while now, usually narrowband since I am imaging from inner city. When doing broadband, it seems that flats are not behaving very nicely.

All frames are newly captured, all same temp, gain and offset:
Code:
lights:   60 seconds, temp -10c, gain 76, offset 30, x120
darks:    60 seconds, temp -10c, gain 76, offset 30, x100
flats:     1 seconds, temp -10c, gain 76, offset 30, x20
darkflats: 1 seconds, temp -10c, gain 76, offset 30, x20

I used WBPP2 to do calibration. Flats are taken with a flat panel (tracing pad from amazon).
For the master flat, wbpp picked LinearFit, equalize fluxes, and the median is at 25,361 (16bits). (tried with lower, ~15k as well..)

However, the result seems to be over-corrected: If you blink these images the vignetting seems to match:
master_flat.jpg master_light.jpg

Any clues?
 
Last edited:
I'm afraid I can't help directly, but I understand you perfectly, I have the same issue with my ASI294MM. Narrowband works fine, but if I try some galaxy (broadband) the results are pretty much the same, the flats seem to over correct.

Adam Block has a couple of videos correcting the flats and his technique works quite well:


The problem seems to be the darks, which should be taken more often. But at least in my case these are recent and do not seem to be the cause of the problem. According to me, high light pollution plus imperfect flats cause a lot of problems in the images.
 
Thank you @Marcelofig, that is exactly what I was looking for! I watched most of the fundamental series, somehow I missed this part ?‍♂️

In my case it looks like the root cause may be different, since I took all frames (lights and calibration) at the same time, so no old darks in my case.

Off to attempt to tune my flats, thank you again for pointing me back at Adam's videos!
 
Just to update, after watching the three videos, I was equipped with enough knowledge to go and look for the root cause myself. I indeed found it! This is my first broadband image of the season, so I was experimenting with some gain/offset combos until I settled on the values I wanted (g76, o30). I took all calibration frames together, making out-of-date darks extremely unlikely. However, I looked at the stats of the master dark and master flatdark (I feel compelled to say Electronic Signature here ;)), and indeed they were substantially brighter! After further investigation, it turns out I left a few flatdarks with a higher offset (90) in the same folder, so my flat darks were a mix of 30 and 90 offsets. Recalibrating now, will see how this comes out tomorrow. Thanks!
 
Back
Top