ASI533MC Pro-Noise on red channel - Forums

Topic starter

Apr 9, 2022 · Edited: Apr 9, 2022

Hello everyone

In all of my latest captures noise is evident on the red channel of my integration even though i dither (20 pixels last night) with good frequencies (every 3-5 frames, RA only though).

This is an example of my last night's results of about 4 hours integration time ,90 secs exposure (Redcat51,Skywatcher SA 2i pro) of the mess on my red channel :

This pattern always(eliptical from upper left to right down corner) shows with or without filter on (L-Pro for that matter),although with the L-extreme there is a much lesser impact. I am imaging from B8..Is this an issue of my technique/software/hardare or this is something that I can not avoid from a light polluted area?

Here is the same thing with the L-pro and rosette:

Also, I use unity gain and offset of 10 , same thing has happened with 101 gain/ offset 45 and this was not an issue with my traditional DSLR camera even from my rooftop.

The processing becomes a pain when trying to DBE all these false data and I am wondering if this is normal...

I can not understand what I am doing wrong here, so help will be much appreciated!

Thanks in advance.

Helpful Respectful Engaging

andrea tasselli

Frequent contributor

Apr 9, 2022

Nothing wrong. It is just that your light pollution is probably broadband and some or a lot of it happens in the red spectrum which is relatively unaffected by the L-PRO filtering which is why the L-Extreme shows quite less. The question is really how good is your flat fielding technique as it appears to me that some of the banding might be due to incorrect flat fielding. Or light sources different from the sky alone.

Insightful

Dimitris Kavallieratos

Topic starter

Apr 9, 2022 · Edited: Apr 9, 2022

andrea tasselli:
Nothing wrong. It is just that your light pollution is probably broadband and some or a lot of it happens in the red spectrum which is relatively unaffected by the L-PRO filtering which is why the L-Extreme shows quite less. The question is really how good is your flat fielding technique as it appears to me that some of the banding might be due to incorrect flat fielding. Or light sources different from the sky alone.

Hmm I get what you are saying Andrea...My technique is that I use a tablet screen and a cloth ,capturing with NINA flat wizard.

So, LP is the only source for the excess red noise and the pattern is due to flat field wrong correction. Blue and green channels do not have this problem though(pattern and noise), is that possible?

andrea tasselli

Frequent contributor

Apr 9, 2022

I hold the strong suspicion that those Ipad panels (and not only those) aren't photometrically flat at all. My answer to this issue is to take loads of them (>30) and rotate the panels between each shot. Boring but required. Or get one of those specifically design for being flat panels, such as these (alternative is to take dusk flats but this might be tricky with very large FOVs):
https://www.gerdneumann.net/english/astrofotografie-parts-astrophotography/aurora-flatfield-panels/uebersicht-aurora-flatfield-panels-overview.html

Helpful Concise

Dimitris Kavallieratos

Topic starter

Apr 9, 2022 · Edited: Apr 9, 2022

andrea tasselli:
Boring but required.

Indeed it is! Thanks again Andrea for your help!

Jim Raskett

Apr 9, 2022

I don't know about the flats. Please try what Andrea suggested. It would be easy to rule that out and post your results.

I have used a 533 for a bit over a year now from B7 skies and have never noticed that issue. I shoot with a UV/IR, IDAS LPS D-1. and L-eNhance filters. Up until my last image I used an older Ipad to shoot flats. I just started using a cheap $20 tracing panel from Amazon with good results. I also use the N.I.N.A. Flat Wizard aimed at about 30k.

Have you tried rotating the camera to see if the band moves with the camera rotation?

Jim

Helpful Respectful Supportive

Dimitris Kavallieratos

Topic starter

May 21, 2022 · Edited: May 21, 2022

Just an update ,sky flats took care of all of that! Never going to take flats with my panel again for sure…

Dimitris Kavallieratos

Topic starter

Jun 20, 2022

@andrea tasselli Well...For some reason even though I am using sky flats and the problem had been eliminated it showed up again.From left to right,stretched with STR process(unlinked) : Master flat/Uncalibrated bebayered single frame/Calibrated single debayered frame, of M27 101 gain, 180secs exposure time from B8 with L-Extreme filter :

The issue happened during broadband imaging also with the L-Pro filter.

It seems like calibration makes the issue worse than without calibrating. Flats median ADU is about 30k and taken with Sharpcap (it means that I have one master with darks calibrated).
I also attached alight frame and my master flat below, if you would be so kind as to take a look.

It starts to get realy annoying putting so much effort only to not be able to integrate my data.

Thanks in advance!

andrea tasselli

Frequent contributor

Jun 20, 2022

Hi Dimitris,

Same old voes I see. I've looked at your files and I found the histogram from the l-extreme flat a bit puzzling. I'd have expected it to be split in two main distributions, one for the OIII response (blue and green pixels) and one for the Ha one (red), with one much less broad than the other as the sky is in fact rather blueish at dusk. Yours look like taken with a monochrome camera. As I don't have an l-extreme filter I can't compare it across with my results but the flats taken with l-enhanced (l-pro) are nothing like it. I therefore suspect a wrong calibration or wrong flat (it happens). If you have a previous flat (which worked for you) with the same filter it may be worth trying that one to see how it works on the lights. Additionally, as I can't see the processing hystory from the fits header I suspect this recent master flat wasn't processed in PI, right? What was the calibration procedure applied to it by SharpCap (I suspect). Do you still have the orginal raw flat frames? Did you take flat-darks after the flats. And what pre-processing was carried out on your M27 shots?

Helpful Insightful Engaging Supportive

Jim Raskett

Jun 20, 2022

This is what the light frame and masterflat look like when debayered (rggb) and stretched (unlinked) .

Dimitris Kavallieratos

Topic starter

Jun 20, 2022 · Edited: Jun 20, 2022

andrea tesselliSame old voes I see

Yeah unfortunately! I am doing something veeery wrong here and I cannot find it...

andrea tesselliAdditionally, as I can't see the processing hystory from the fits header I suspect this recent master flat wasn't processed in PI, right? What was the calibration procedure applied to it by SharpCap (I suspect). Do you still have the orginal raw flat frames? Did you take flat-darks after the flats. And what pre-processing was carried out on your M27 shots?

Correct Andrea, darks subtraction and averaging has been done from Sharpcap via the Capture flat tool, I will try to capture with NINA and calibrate/integrate the frames myself with dark-flats in PI and see what happens...What exposure time is recommended for 30k median ADU?1-10 secs I read that is the optimal interval.

Thanks again for your time!

Juno16This is what the light frame and masterflat look like when debayered (rggb) and stretched (unlinked) .

Yeap ,I know, but if you try to calibrate with the flat a very weird thing happens and that is my issue here (check my printscreen above)!

andrea tasselli

Frequent contributor

Jun 20, 2022

What exposure time is recommended for 30k median ADU?1-10 secs I read that is the optimal interval.

If you use NINA's flat wizard it will all handled by NINA. If you do it the old fashion way (me) I wouldn't get too hang up on the exact value of the median. I had good results with even 1/10th of the dynamic range. Off the bat I'd try to be at around 1/4th but no much greater then 1/2.

Dimitris Kavallieratos

Topic starter

Jun 20, 2022 · Edited: Jun 20, 2022

@andrea tasselli Well..Tried to retake flats, but still nothing.

Here is the master flat (30 flats/30 dark-flats from NINA), the final debayered and calibrated light frame is still like this :

Is it possible that the issue is hardware related??

andrea tasselli

Frequent contributor

Jun 20, 2022

The flat looks nothing like the previous one which is in a way puzzling but the hystogram looks more like what it should be. The flat itself is remarkably flat, which is suspicious. I don't see in the history that you subtracted the master dark-flat nor I see if the calibration files were taken at the same gain. Were they?

Mind you, you might still have odd results withthe l-extreme depending if the proper procedure is not followed, which consists in removing a master dark with no bias subtraction and no bias subtraction for the light frames calibration (just master dark and master flat).

Helpful

Dimitris Kavallieratos

Topic starter

Jun 20, 2022 · Edited: Jun 20, 2022

andrea tasselli:
The flat looks nothing like the previous one which is in a way puzzling but the hystogram looks more like what it should be. The flat itself is remarkably flat, which is suspicious. I don't see in the history that you subtracted the master dark-flat nor I see if the calibration files were taken at the same gain. Were they?

Mind you, you might still have odd results withthe l-extreme depending if the proper procedure is not followed, which consists in removing a master dark with no bias subtraction and no bias subtraction for the light frames calibration (just master dark and master flat).

I calibrated with WBPP ,maybe that is why you can not see the history?? I will try to do it manually ,although I did not use bias frames only darks and dark flats.

Edit: Same gain at 101 for flat/dark flats exactly as the light frames.

andrea tasselli

Frequent contributor

Jun 20, 2022

Dimitris,

Forgot to mention, try removing the pedestal when integrating the master flat.

jewzaam

Jun 21, 2022

I have a similar pattern with my ASI533MC and it's only a problem because of the cheap tracing tablet I use for flats. The red response is lower than green and blue. Ends up red isn't as well corrected. But it isn't horrible, just a bit more work to deal with. As for taking flats, I recommend trying NINA's flats wizard as mentioned in this thread, default settings. I have not tried sky flats. If I could automate it in NINA I would! Only thing I can't do with it as far as I know (as of my last looking into it months ago at least).

Helpful

Maxence Ouafik

Jun 21, 2022

It reminds me of an issue I had with the same camera a few months back. I was not able to find the cause but it happened without any filter and I did not encounter the same problem with my L-extreme.

Dimitris Kavallieratos

Topic starter

Jun 27, 2022

Well..Still the issue goes on, only difference is the drivers (ASI cameras the one night and the native ones the other, with NINA) :

The ASCOM ASI generic drivers has the ring effect/bad correction issue ,with the exact same calibration frames/workflow! Is this even possible??

Bad Frames :

Good Ones :

Dimitris Kavallieratos

Topic starter

Jun 27, 2022

Well..Still the issue goes on, only difference is the drivers (ASI cameras the one night and the native ones the other, with NINA) :

The ASCOM ASI generic drivers has the ring effect/bad correction issue ,with the exact same calibration frames/workflow! Is this even possible??

Bad Frames :
Good Ones :

And the red channels for both frames (good/bad)...

andrea tasselli

Frequent contributor

Jun 27, 2022

That doesn't suprise me one bit. I thought you had native drivers already. Well, at least it seems sorted, for once isn't the flat the culprit!

Dimitris Kavallieratos

Topic starter

Jun 27, 2022 · Edited: Jun 27, 2022

andrea tasselli:
That doesn't suprise me one bit. I thought you had native drivers already. Well, at least it seems sorted, for once isn't the flat the culprit!

OK...At least I think I found the culprit as you say!For some reason yesterday camera could not be found from NINA (or Sharpcap) and I chose to use the ASCOM ASI driver. I plugged and unplugged the usb of my camera, but it was still not discoverable all of the sudden! 5 minutes prior to that, I had polar aligned with Sharpcap without any problems with my camera connected.Anyway, next capture I am using the native drivers only!

Dimitris Kavallieratos

Topic starter

Jun 30, 2022

For future reference ,the culprit was indeed the ASCOM driver. I do not know if I have messed any settings, even though when it detected my camera I left it alone at the defaults..It almost messed half of my sessions and almost drove me away from the camera out of frustration, but thankfully I found it.

Thank you guys for the advice!

Helpful Respectful

Vince

Oct 3, 2022

Hello all,

I am experiencing a very similar issue. However, I’m using the 533mc pro with the asiair plus. After stacking I notice this pattern of blotchyness pattern on the edges of the frame with a shape towards the middle. If this is a driver issue, then why would it be present with using this camera with the asiair plus? Hoping to find a solution here.
Thanks

andrea tasselli

Frequent contributor

Oct 3, 2022

I suppose it is got to do with using the equivalent ALPACA driver, not a native one, under Linux.