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.

Photo

Are these columns of dead pixels normal? QHY165C

  • Please log in to reply
6 replies to this topic

#1 bblindsey

bblindsey

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 25 Mar 2018
  • Loc: Stillwater, Ok

Posted 09 June 2019 - 10:50 AM

Just bought my first cooled CMOS camera, the QHY165c. Transitioning from the DSLR hasn't been as bad as I thought it would be. However, I am getting several rows of dead pixels near the border of my test images. There are hot pixels throughout the columns which makes me think it's a setting of some kind, not bad pixels...

 

 

Bad columns.png

 

This frame was shot at:

Gain- 1206... automatically suggested by APT

Offset- 256.... which put my minimum ADU in my bias frames at 954

300 second exposure which put my mean ADU in the above image at 1506

 

 

Initially I assumed it was an offset issue, but I have adjusted the USB traffic, offset, and gain in multiple configurations and still get the issue. I have been shooting exposures inside the house and can't get the bad columns to produce any data.

 

95% of the time, this area will be cropped out. But after dithering, and for mosaics, it will be annoying enough to force me to figure out the issue.

 

Thanks for the help

 

 



#2 OldManSky

OldManSky

    Fly Me to the Moon

  • *****
  • Posts: 5,737
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 09 June 2019 - 11:04 AM

Is that a stacked set of images? If so, it’s probably not dead columns, it’s due to drift and/or dithering - after alignment and stacking.

Since I see some dim stars in the “dead” area, that seems likely.

 


Edited by OldManSky, 09 June 2019 - 11:05 AM.

  • bblindsey likes this

#3 pyrasanth

pyrasanth

    Aurora

  • *****
  • Posts: 4,893
  • Joined: 08 Jan 2016

Posted 09 June 2019 - 11:05 AM

Unless you have a class 1 sensor you will have column defects in cameras used for astronomy as well as dead pixels. These are dealt with in post processing using dark & bias frames as well as cosmetic calibration which will remove the defects. Whilst it looks like a dramatic defect the number of actual pixels affected is normally a very tiny fraction of the ccd or cmos overall amount and is of little consequence provided calibration is carried out. On a colour array these hot pixels are like multi coloured speckles and initially before calibration look like quite a mess but are easily corrected.

 

Dithering, as OldManSky states can also help mitigate these issues.

 

If you have a substantial number of column defects it might be an issue but otherwise do not worry. It is worth checking the manufacturers specifications to see if your sensor conforms to their standards and if so don't worry about it as mostly everybody using class 2 sensors has to deal with this in some form. Cooling the sensor can help mitigate these effects.


Edited by pyrasanth, 09 June 2019 - 11:08 AM.

  • bblindsey likes this

#4 bblindsey

bblindsey

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 25 Mar 2018
  • Loc: Stillwater, Ok

Posted 09 June 2019 - 11:08 AM

Is that a stacked set of images? If so, it’s probably not dead columns, it’s due to drift and/or dithering - after alignment and stacking.

No, this is a single exposure with a quick auto stretch applied in PI. I know the intensity of the issue will dissipate as I stack frames, I'm just being picky I think lol.gif



#5 bblindsey

bblindsey

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 25 Mar 2018
  • Loc: Stillwater, Ok

Posted 09 June 2019 - 11:12 AM

Unless you have a class 1 sensor you will have column defects in cameras used for astronomy as well as dead pixels. These are dealt with in post processing using dark & bias frames as well as cosmetic calibration which will remove the defects. Whilst it looks like a dramatic defect the number of actual pixels affected is normally a very tiny fraction of the ccd or cmos overall amount and is of little consequence provided calibration is carried out. On a colour array these hot pixels are like multi coloured speckles and initially before calibration look like quite a mess but are easily corrected.

 

Dithering, as OldManSky states can also help mitigate these issues.

 

If you have a substantial number of column defects it might be an issue but otherwise do not worry. It is worth checking the manufacturers specifications to see if your sensor conforms to their standards and if so don't worry about it as mostly everybody using class 2 sensors has to deal with this in some form. Cooling the sensor can help mitigate these effects.

Thanks for the input. I have seen some column defects on other sensors, but do you think this one being on the border is significant in any way? There was a slight gradient on the image that made that side of the image slightly dimmer, which is why I was suspicious it might have something to do with the offset value and clipping pixels or something.



#6 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,994
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 09 June 2019 - 11:16 AM

This is the overscan area. I do not know why, but QHY drivers by default seem to include the overscan area during readout, and you must go into the driver and disable overscan to get just the active pixel area. All sensors, including CCDs, have an overscan area, which is an area of non-active masked pixels, often spanning two or even all four sides of the sensor, that are used for calibration purposes. Check your driver, find the overscan options and disable the readout of the overscan.

 

If you cannot for some reason find this setting, PI's ImageCalibration process has some features to eliminate overscan areas as well. 


  • jdupton, darkstar3d, bblindsey and 1 other like this

#7 bblindsey

bblindsey

    Explorer 1

  • *****
  • topic starter
  • Posts: 57
  • Joined: 25 Mar 2018
  • Loc: Stillwater, Ok

Posted 09 June 2019 - 11:23 AM

This is the overscan area. I do not know why, but QHY drivers by default seem to include the overscan area during readout, and you must go into the driver and disable overscan to get just the active pixel area. All sensors, including CCDs, have an overscan area, which is an area of non-active masked pixels, often spanning two or even all four sides of the sensor, that are used for calibration purposes. Check your driver, find the overscan options and disable the readout of the overscan.

 

If you cannot for some reason find this setting, PI's ImageCalibration process has some features to eliminate overscan areas as well. 

You're a boss, Jon bow.gif  thanks! I have come across many replies of yours that have made this DSLR to QHY transition a lot more forgiving than it would have been otherwise. 




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