Jump to content


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Mitigation Methods For ASI294 Uneven Cooling?

  • Please log in to reply
2 replies to this topic

#1 jdupton



  • *****
  • topic starter
  • Posts: 2,205
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 28 May 2018 - 10:31 PM

Camera Gurus,

   I have recently been thinking of getting a larger format OSC camera to complement my SXVR-H694 mono camera. I am attracted to both the ASI071MC-Pro and the ASI294MC-Pro cameras. From what I have read here on Cloudy Nights, the ASI294MC-Pro appears to be the more sensitive camera but is smaller than the ASI071MC-Pro (Micro-4/3 vs. APS-C). I expect either would fulfill my objectives. (I sort of wish there was something in the APS-C size class with the sensitivity of the ASI294MC-Pro.)


Background Reading:

CN Thread: Help me with this artifact - ZWO ASI294



   I am debating whether to look seriously at the ASI071MC-Pro over the ASI294MC-Pro due to the uneven cooling due to the undersized cold finger in the latter camera. As I understand from reading threads here, there are some things that can be done to mitigate the effects of the undersized cold finger in the ASI294MC-Pro. They are:

  • Don't cool the camera to the max: This would probably not be a problem for me as I am accustomed to only cooling my SXVR-H694 to only 0° C. I mostly shoot in the summer and my average nighttime temperatures often range from 90°+ F early in the evening to 80° F just before dawn. I use a 0° setting on my mono camera to ensure I can shoot with the same settings year round. (I keep a set of Master Bias and Master Darks for processing.)
  • Shoot with shorter exposures: It appears that the recommended sub-exposure length for the ASI294MC-Pro is about 2 minutes or less. Under my best case backyard conditions in a red / orange zone, I don't see that as a huge problem. At my dark sky site 90 minutes away in near a blue / brown boundary, there should still be no problem.
  • Always take darks with the lights for each imaging session: I see this as problematic for me using portable set-ups. By then end of the night, I am ready to pack up and would not want to wait for taking many darks. In addition, I have been spoiled by the noise reduction gains of taking many Bias and Darks. My Master Bias and Master Dark libraries for use with my SX camera contain integrations of 700 - 1,000 (bias) and 140 - 200 (10 minute or 5 minute darks) frames depending on how ambitious I am when I update my library. I really like the smoothness of these Master calibration files and how they calibrate my images.

   My question here is whether there are other mitigation methods that have not been explored for calibrating the ASI294MC-Pro images. Hopefully, some folks with the cameras may have tried something already or would be willing to try some experiments. If we think outside of the box, are there better / more effective ways to help with image calibration?

   Since one key recommendation is to take darks during each session to lessen the gradients due to uneven cooling, we lose the advantages of having a Master Library Dark made up of many hours of sub-exposures. What if we could do both? Two oddball things come to mind.

  • Make a Library Master Dark from many frames (250+?) taken at the same set point temperature, gain, offset, etc. During the imaging session, take one to four dark frames only. This would not be a big deal and would not lengthen the session unduly. In processing, stack the session dark frames to a Session Master Dark. Next, could we then Linear Fit our 250+ frame Library Master Dark to match the Session Master Dark and then use the result to calibrate our images? Would that give us something with the basic noise characteristics of 250+ integrated frames but have it more closely match the darks taken with the lights during the session?
  • What about a three step calibration process? Suppose we first do a "dummy calibration" with dark scaling turned on using one or more of our lights with the Session Master Dark, then repeat with the Library Master Dark while noting the scale factors for each that PixInsight reports. Then we using the scaling factors to adjust and manually scale our Library Master Dark and do a final calibration of the lights with the manually scaled Library Master Dark. Would this be any different than just doing a linear fit of the Library Master Dark?

   If you were to play "what if" with other wild processing ideas, what other possible methods could be used to help mitigate uneven cooling of a sensor? Surely, with all the image processing power PixInsight brings to the table, it seems there must be other completely left-field approaches to solving the problem in software.

   OK, it's time for the PixInsight and camera characteristics Wizards to relax, have a few too many cold ones, and then weigh in with other off-the-wall ideas. Brainpower and software can solve nearly any problem most of the time. Is this one of those cases?


Edited by jdupton, 28 May 2018 - 10:56 PM.

#2 AhBok



  • *****
  • Posts: 2,526
  • Joined: 02 Dec 2010
  • Loc: Lakeland, TN

Posted 28 May 2018 - 11:13 PM

See the other thread on the ASI294MC Pro. Some of us with previous issues have solved our problems. My issue was solved by learning how to take better calibration frames with this camera. I’ve found that the slightest light leak affects my darks and flats adversely. This forced me to deal with the problem. As far as temp is concerned, I always cool my camera to 30-35c below ambient without issue. I’m learning to love my camera, but it is different for sure from my DSLR and mono CMOS cameras. There is a learning curve of its own.

#3 Jon Rista

Jon Rista


  • *****
  • Posts: 24,082
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 29 May 2018 - 12:27 AM

If the TEC is indeed smaller than the sensor, then it may well be that you can never properly match the full field of the sensor with the darks, as the border could always be changing (especially due to all the mild thermal hot spots that tend to exist around the periphery.) I suspect ZWO may have been trying to minimize costs by reusing a standardized TEC design for 4/3 sized sensors, but not all of those sensors are exactly the same size. The real solution would be for them to use a properly sized TEC, especially if there are other factors about this particular sensor that warrant a larger TEC area.


The only other idea I have would be to simply crop out the offending peripheral region and just keep the properly cooled center. Not sure exactly how much you might have to toss to get good calibration results, or whether having the full field of the sensor is more important than having the high dynamic range and low noise.

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Recent Topics

Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics