Strange vertical lines showing up after DBE

[Peje]

When processing my images from Saturday night I noticed vertical lines appearing on my stacked image, they become extremely noticable after ABE/DBE. I have recently moved to ISO200, it  doesnt happen with every image but it did happen with both targets I imaged on the same night. The pattern I'm seeing looks like it's related to my Super-Bias file, stretched screenshot attached. It doesn't seem to happen on every image, but my latest two show it very badly (screenshots attached of Betelguese & Iris Nebula).

I created the Super-Bias using Pixinsight, the master file was created from 100 bias images. The only data point I can see at the minute is that the two images which show the lines weremade from 1 or 3 minute exposures, image with 4 or 8 minute exposures don't have it. (screenshots attached of M42 & M101)

I tried stacking my Betelguese images with the master-bias instead of the super and the result was definitely reduced lines but they do seem to be present. The only thing I could think of was that something has changed in terms of read noise and that I need to update my Bias library files, these were created in February.

Any help with this would be greatly apprecited,
Pete.

[zicklurky]

This looks like banding to me, it's sometimes present on Canon cameras. Pixinsight has a "CanonBandingReduction" script which may help. From what I've read it's more prevalent when stacking shorter length subs.

[Peje]

3 minutes ago, zicklurky said:

This looks like banding to me, it's sometimes present on Canon cameras. Pixinsight has a "CanonBandingReduction" script which may help. From what I've read it's more prevalent when stacking shorter length subs.

Thanks, this gives me something to investigate. It seems strange that it's only just started happening.

[Peje]

Some more data, I manually calibrated and stacked the files; firstly without flats and secondly without bias.

Without flats looked the same, without bias the lines were gone.

[Peje]

Took 200 bias images and created a new master bias then another superbias, the attached screenshot shows there is very little difference between old and new.

Haven't tried the canon banding script but from what I've read it will only be masking the problem when I'd rather prevent it happening in the first place.

Right now the two alternatives I am considering are manually pre-processing without bias or going back to shooting using ISO400.

[harry page]

Hi

You have to carefull using a superbias as it can cause a lot of defects , In fact I recomend you do not use it

simply create a master bias say out of 100 images and all will be fine

Harry

[Peje]

1 minute ago, harry page said:

Hi

You have to carefull using a superbias as it can cause a lot of defects , In fact I recomend you do not use it

simply create a master bias say out of 100 images and all will be fine

Harry

Thanks Harry. I did notice that some of the lines did remain even with just using the master bias, have you seen this before?

Is the theory to not use bias calibration a bad idea?

[harry page]

Hi

Yes use a bias /dark at the same temp as you took the images ( difficult I know )

temp will make a big differance and have seen it a lot

Harry

[Peje]

7 minutes ago, harry page said:

Hi

Yes use a bias /dark at the same temp as you took the images ( difficult I know )

temp will make a big differance and have seen it a lot

Harry

I take it the temperature relates to the dark more than the bias? I generally cool my camera in a powered cool box to simulate a cool night.

[harry page]

Hi

Temp will affect the bias as well , but as you say darks more

Harry

[wxsatuser]

You don't need to subtract the bias as its subtracted with the dark frames.

Bias is needed to calibrate flats but if you subtract darks from your flats then bias is not needed again.

[frugal]

I have this same problem with my Canon 60D. The trouble is that aligning the darks temperature to the lights temperature is a nightmare with an uncooled DSLR. The best way I found to get rid of this banding was to change from ISO400 to ISO800. Strange I know, buy the banding is far worse at lower ISOs.

[Peje]

I did previously shoot with ISO400 so I'll have to run through my calculations again to see what the idea ADU is for that, I'm a little hesitant to move back up as I like shooting at the lower ISO with low sensor gain.

[Peje]

Ok after running the numbers again it seems that moving back to ISO400 not only removes the lines issue, I'm confident of this as I've shot extensively with ISO400 across a wide variety of exposure lengths, but it will also allow me to achieve the optimum ADU with half the exposure length of ISO200. (Obvious I know)

Compared to my old f/6 scope needing 8mins @ ISO400, my new f/4 scope needing 8mins @ ISO200; the new scope will only require 4mins @ ISO400 meaning I can grab twice as many exposures allowing me to further improve SNR.

This lines problem has turned out to work in my favour, thanks everyone!