2600MC Pro Flats Issue

[smashing]

Hello,

So hopefully someone can help me here. I've been using my 2600MC pro a few times now and mostly it has been fine until the other week when I decided to have a go without the L-extreme filter.

I set up everything like I usually would and started capture on M31 to see how it fared with my previous attempt and the DSLR. Managed to capture an hour and forty minutes worth of data but my flats seem to be messing up, I have re-taken them but it's still not right. I am using the APT flats aid which has given good results previously.

Below is the stacked (DSS) image with STF stretch and ABE in PixInsight (lights, flats and dark flats)

 

Below is a single light frame

Current flat used to create the image above

I am guessing I am doing something pretty stupid but can't seem to work it out. Any help would be appreciated.

 

[vlaiv]

Did you take other calibration files as well - darks and flat darks?

[smashing]

14 minutes ago, vlaiv said:

Did you take other calibration files as well - darks and flat darks?

Yep had dark flats but no darks for these...tbh darks on previous shots haven't really been needed 

[vlaiv]

1 minute ago, smashing said:

Yep had dark flats but no darks for these...tbh darks on previous shots haven't really been needed 

They are needed in order for flat calibration to work.

Dark current signal needs to be removed as it was not subject to vignetting and when you correct with flats - if there is still some signal left that was not subject to vignetting - you'll "counter" correct it - you'll make inverse of vignetting.

You can see this in your image - center is dark and dust shadows in that part of the image are bright.

Do follow proper calibration workflow in order to have good results, it is there for a reason.

[smashing]

37 minutes ago, vlaiv said:

They are needed in order for flat calibration to work.

Dark current signal needs to be removed as it was not subject to vignetting and when you correct with flats - if there is still some signal left that was not subject to vignetting - you'll "counter" correct it - you'll make inverse of vignetting.

You can see this in your image - center is dark and dust shadows in that part of the image are bright.

Do follow proper calibration workflow in order to have good results, it is there for a reason.

Thanks @vlaiv  i'll sort out some darks for that and take another stab at it

 

 

[smashing]

@vlaiv I did actually manage another target that night for which I do have some darks for (only 40 mins of data mind). Doing all of the above, STF stretch and ABE spits out the below

The correction seems better, below is without flats and dark flats. 

I'm wondering if i've caught some clouds in the stack somewhere

 

 

 

[vlaiv]

10 minutes ago, smashing said:

The correction seems better, below is without flats and dark flats. 

Correction does seem better - but not quite there yet.

How do you take your darks? Is the camera on the scope?

Is there any chance of light leak in your optical train?

[smashing]

26 minutes ago, vlaiv said:

Correction does seem better - but not quite there yet.

How do you take your darks? Is the camera on the scope?

Is there any chance of light leak in your optical train?

These were taken on the scope yep, refractor with the lens cap on...I'll bench it and take them again and see what I get.

[smashing]

@vlaiv so managed to get the camera off, lens cap screwed in and new darks taken.

Re-using the older flats

To my eyes the older flats now seem to work much better with the darks than the new set I took but still not quite there. So having a mess around and using DBE instead of ABE with a few points gets me the below, which looks MUCH better.

Cropped and a little finer placement gets

as always, processing is hard lol

[vlaiv]

14 minutes ago, smashing said:

So having a mess around and using DBE instead of ABE with a few points gets me the below, which looks MUCH better.

Try cropping first before you wipe background to remove edge stacking artifacts.

Yes, last image is big improvement, but I do wonder why there is issue with calibration.

[smashing]

5 minutes ago, vlaiv said:

Try cropping first before you wipe background to remove edge stacking artifacts.

Yes, last image is big improvement, but I do wonder why there is issue with calibration.

Not sure...I've taken darks for the M31 set as well so going to have a play with that and see what it does.

 

[ONIKKINEN]

Have you adjusted tilt on the camera? Actually not sure if the 2600MC has a tiltplate so might not apply...

But if you have and it has the same tiltplate i bought to use in another camera, you will have a gap that lets in light to the sensor. Might be just a few photons but thats enough to skew calibration frames out of whack. Did you take darks in a completely dark room?

[Jerry Barnes]

I have a filter drawer in front of my camera so just to be careful I wrap a scarf round it for taking darks.  I dont know if it helps, I should probably check, bright room vs dark room and see if I can see any difference.  Something to do while waiting for the clouds to clear. 

[smashing]

40 minutes ago, ONIKKINEN said:

Have you adjusted tilt on the camera? Actually not sure if the 2600MC has a tiltplate so might not apply...

But if you have and it has the same tiltplate i bought to use in another camera, you will have a gap that lets in light to the sensor. Might be just a few photons but thats enough to skew calibration frames out of whack. Did you take darks in a completely dark room?

Not in a completely dark room no but I may build a box to help next time

[Stuart1971]

There is deffo something wrong there, but it’s not light leaking into the imaging train, as the flat you used has not been subtracted form the image, as you can still see the dust bunnies, both in the flat, as you should do, but also still in the image after calibration…. I’m not sure what the issue is, but I’m pretty sure it’s not light getting in…

[david_taurus83]

You should always crop out the edges before doing background extraction. DBE is usually the preferred method as well. ABE with default settings will usually give me similar results to above. If your struggling with placing points you could try placing points in the 4 corners, and one on each edge as well and a couple near the centre that's not on nebulosity. Sometimes it won't remove all light pollution and gradients but it should flatten the image enough that you should be able to colour balance afterwards.

[vlaiv]

35 minutes ago, Stuart1971 said:

There is deffo something wrong there, but it’s not light leaking into the imaging train, as the flat you used has not been subtracted form the image, as you can still see the dust bunnies, both in the flat, as you should do, but also still in the image after calibration…. I’m not sure what the issue is, but I’m pretty sure it’s not light getting in…

Light leak can cause both under and over correction - depending when it happens.

It can happen during taking of darks, or maybe during taking of flats or perhaps during taking of lights.

Since first step is lights - darks, depending which one of them has extra signal - it can be either case of having extra positive signal before flat correction or extra "negative" signal before flat correction.

One leads to over correction and other to under correction (like flats haven't been fully applied).

Btw - flats divide rather than subtract

 

[smashing]

47 minutes ago, david_taurus83 said:

You should always crop out the edges before doing background extraction. DBE is usually the preferred method as well. ABE with default settings will usually give me similar results to above. If your struggling with placing points you could try placing points in the 4 corners, and one on each edge as well and a couple near the centre that's not on nebulosity. Sometimes it won't remove all light pollution and gradients but it should flatten the image enough that you should be able to colour balance afterwards.

Yes starting to see this now.

So below is what spat out after DBE playing around

And seems flats have done there job here, some light processing gets me the below which isn't too shabby....PixInsight is a little new to me but then, everyday is a school day

[wimvb]

On 20/12/2021 at 18:22, vlaiv said:

but I do wonder why there is issue with calibration.

I don't think there ever was an issue. The use of ABE (or DBE for that matter) on images that are not cropped, and where samples are not placed carefully, can have the effect visible in the original post. The absence of darks would add to that.

[smashing]

On 22/12/2021 at 13:24, wimvb said:

I don't think there ever was an issue. The use of ABE (or DBE for that matter) on images that are not cropped, and where samples are not placed carefully, can have the effect visible in the original post. The absence of darks would add to that.

Yep, after watching a number of videos people had similar things with ABE

[wimvb]

13 hours ago, smashing said:

Yep, after watching a number of videos people had similar things with ABE

I never use ABE, only DBE, where I have control over where I put the samples. Most of the time I only place about 12 samples; one in each corner, one further in along the diagonals from each corner, and one along each side.