Help ASI 1600mm PRO- Darks' mean ADU unstable - Forums

Topic starter

Apr 18, 2021 · Edited: Apr 18, 2021

Hello, I've recently had some issues with my ASI 1600mmPRO that I noticed while taking 10 minute darks at gain 200. The camera is not 1 year old yet. Before I explain what I did, all the test were done in exactly the same conditions, same laptop etc... I tried using different USB cables, different capture software and a different PC with the same results. My main issue is that the mean ADU of my darks are not stable. They start at ADU 920 for example, go up slowly till about 970 and then drop back to 915 or so. This does not happen at gain 75, only gain 200 (did not test other gains yet). So unless I need to test more on gain 75 and give it more time, so far the issue appears only in high gain but that could be inaccurate. I have plotted a graph showing the mean value of each frame for 5 minute and 10 minute darks at gain 200. The x_axis is the frame number. The y_axis is the mean ADU value. I compared these darks with my older darks, and the old darks' mean was extremely stable (+- 1 ADU difference between frames). The cooler power seems to be stable with no big fluctuations (just 1 or 2% max and it rarely changes). I am currently in the process of taking 3 minute darks at gain 200 to see if the same trend happens. So far the mean appears to be stable. I will update this info once capture has finished. Still waiting for ZWO support to reply so I'd appreciate if anyone has any insights or if someone else had this issue just so I save myself some time and trouble. Cheers,

Anthony

Helpful Engaging

matthew.maclean

Apr 19, 2021

I certainly have no idea, but I think I would open a couple of the brightest and darkest frames, stretch them, and see if there was any pattern to the brightness differences (e.g. one particular side, one corner, some repeatable pattern, etc.)? Any chance lighting nearby was changing and could have leaked into the camera during that particular run? I just happened to have taken a bunch of darks on my camera last winter and realized after I thought I was done that my filter tray doesn't seal perfectly tight and all my darks had a bright band along the edge where the tray comes apart…lesson learned there….

Helpful Engaging Supportive

Anthony Saab

Topic starter

Apr 19, 2021

I certainly have no idea, but I think I would open a couple of the brightest and darkest frames, stretch them, and see if there was any pattern to the brightness differences (e.g. one particular side, one corner, some repeatable pattern, etc.)? Any chance lighting nearby was changing and could have leaked into the camera during that particular run? I just happened to have taken a bunch of darks on my camera last winter and realized after I thought I was done that my filter tray doesn't seal perfectly tight and all my darks had a bright band along the edge where the tray comes apart...lesson learned there....

Hey Mathew, I'm currently repeating the tests in a completely dark room with the AC on just to make sure it's not light leaks not drastic ambient temp changes. I will get back with the results.

Padraic Moran

Apr 19, 2021

Hi Anthony, I had a similar issue just last week, with asi1600mm darks at gain 200 coming out with varying ADUs. Needless to say, they didn't calibrate well!
I re-did the run, in a fully-darkened room, and all worked well. I can't say that it wasn't 'user error'.

I don't remember ever having this issue with darks at gain 139. Interesting to hear how you get on with your next batch.

Well Written Respectful Concise Engaging Supportive

Anthony Saab

Topic starter

Apr 19, 2021

Padraic Moran:
Hi Anthony, I had a similar issue just last week, with asi1600mm darks at gain 200 coming out with varying ADUs. Needless to say, they didn't calibrate well!
I re-did the run, in a fully-darkened room, and all worked well. I can't say that it wasn't 'user error'.

I don't remember ever having this issue with darks at gain 139. Interesting to hear how you get on with your next batch.

Hello! Yes I'm currently taking the darks in a completely dark room with the AC on. The mean seems stable (between 912 and 916) but I think I'm going to have to give it some more time to make sure the mean changes are not happening randomly. Hopefully it was user error related to light leaks or ambient temp changes.

Anthony Saab

Topic starter

Apr 20, 2021

So I repeated the test today, and so far it's not going so well. Yesterday the mean was stable at around 912 (around 35 frames). Today, with the exact same conditions (same room, completely dark, AC on same temperature, same cooler power %), I'm getting a mean of 933 and slowly rising. I'm not sure what to do now…

Well Written

matthew.maclean

Apr 20, 2021

Then that starts to sound like a camera problem to me unfortunately…..

the only other thing I could think to ask would be is this near the maximum limit of the cooling (~100% TEC power)? Is it seeming to maintain the temperature you set easily?

Anthony Saab

Topic starter

Apr 20, 2021

Then that starts to sound like a camera problem to me unfortunately.....

the only other thing I could think to ask would be is this near the maximum limit of the cooling (~100% TEC power)? Is it seeming to maintain the temperature you set easily?

The cooler is running at around 58%. The mean appears to have stabilized at 939. So yesterday it was stable at 913, today at 939. I know that CMOS chips' electronic signature changes pretty rapidly, but I'm not sure if that's what's happening in my case.

Still no answer from ZWO...

matthew.maclean

Apr 20, 2021

You definitely need to hear from them as 26ADU does sound non-trivial. Only last obvious thing to check is the “offset” didn’t get changed between the two days somehow? (Hopefully by 26 counts...)

Also, they are good about providing a dark-current plot for a lot of their cameras:

https://astronomy-imaging-camera.com/product/asi1600mm-cool

I think that is your camera so about half-way down the page gives the dark e/s/px (log scale) and you could deduce approximately what temperature change would be needed to account for the increased 26ADU. It doesn’t solve your root issue obviously, but it might tell you something.

Helpful

Anthony Saab

Topic starter

Apr 20, 2021

No I'm using the same offset of 50

I do need to find the root of the issue as you said since it might be hiding some defect that might cause some other worse issues down the line, especially since the camera is still under warranty.

Markus A. R. Langlotz

Apr 22, 2021

Hi Anthony,

what do your plots show? Is it the mean value of ADUs in a darkframe over the number of darkframes? So this would show a certain time-dependency.
Maybe, you just see an overshoot/undershoot in the TEC regulation, think about 6 degrees doubling your thermal signal, so 5% more could translate to
small thermal instabilities below a tenth of a degree.
Starting the exposure batch even after stabilzing the sensor thermally, will bring in fluctuations, as you now start the sensor work and read it out on a regular basis, so the chip will become warmer and there the TEC regulation has to counteract.

To get real stable darks, let the TEC stabilize, start a batch of lets say 100 darks and withdraw the first 50 darks.

CS

Markus

Helpful Insightful Respectful

Anthony Saab

Topic starter

Apr 22, 2021

Markus A. R. Langlotz:
Hi Anthony,

what do your plots show? Is it the mean value of ADUs in a darkframe over the number of darkframes? So this would show a certain time-dependency.
Maybe, you just see an overshoot/undershoot in the TEC regulation, think about 6 degrees doubling your thermal signal, so 5% more could translate to
small thermal instabilities below a tenth of a degree.
Starting the exposure batch even after stabilzing the sensor thermally, will bring in fluctuations, as you now start the sensor work and read it out on a regular basis, so the chip will become warmer and there the TEC regulation has to counteract.

To get real stable darks, let the TEC stabilize, start a batch of lets say 100 darks and withdraw the first 50 darks.

CS

Markus

Hi Markus,

I thought about removing the "bad" darks and just calibrating with the stable darks, however, this unstable behavior will also translate to my lights especially if it's an electronic issue changing the offset. The result would probably be bad; one issue might be that the flats would over or undercorrect if I'm not mistaken.

Helpful Insightful Respectful