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.


Use of DSLR as SQM ?

This topic has been archived. This means that you cannot reply to this topic.
2 replies to this topic

#1 Christoph2


    Lift Off

  • -----
  • topic starter
  • Posts: 5
  • Joined: 15 Oct 2007

Posted 15 October 2007 - 09:58 AM


I know that the SQM here in US is more used than in germany where I am living. That’s the reason why I am posting here.
I am experimenting with my Canon 20D DSLR (unmodified) to use it as a sky quality meter like the SQM from unihedron.

The method is as follows:
First take a photo of the zenith using the following parameters:

Sensitivity: 3200 ASA
f-stop: 2.8
exposure time: 20 s
White Balance: automatic mode
contrast: lowest value (-2)
other Custom image parameters
are not relevant

The sky brightness can be estimated using the camera histogram on the TFT screen (from left to right):

22-23 mag/arcsec² left border of the histogram
21 mag/arcsec² 1st brightness level
20 mag/arcsec² 2nd brightness level
19 mag/arcsec² 3rd brightness level
18 mag/arcsec² 4th brightness level
17 mag/arcsec² right border of the histogram

See also the picture here

When using a modified DSLR (IR filter replaced) the values have to be corrected since the overall sensitivity of the DSLR is now higher. In this case I estimated that the sky brightness taken from the histogram should be increased by 0.5 mag/arcsec^2 (has to be verified).

In the expample above the night sky was slightly above 21 mag/arc². Lens with other f-stops can be used if the exposure time is changed accordingly:

For Canon 350D, 20D, 30D, 5D, 1Ds MkII (actual sensitivity is 4000ASA):
f1.8 t=8s
f2.0 t=10s
f2.5 t=16s
f2.8 t=20s
f4.0 t=40s

For Canon 10D, 400D, 40D? (actual sensitivity is 3200ASA as indicated):
f1.8 t=10s
f2.0 t=12.5s
f2.5 t=20s
f2.8 t=25s
f4.0 t=50s

I would not use the lowest possible f-stop of the lens since the histogram will be to broad due to lens vignetting.

So that’s the method. What’s missing is a calibration of my method with a SQM from unihedron. Since I am not owning a SQM it would be great if owners of a SQM and a canon DSLR would test this method and would tell me their results (I need the camera and lens type, the camera parameters as above, modified DSLR y/n, the position of histogram, poss. also jpg or raw-file of the picture).

Would do you think ?


#2 SleepIsWrong


    Viking 1

  • *****
  • Posts: 676
  • Joined: 07 Dec 2005

Posted 15 October 2007 - 10:28 AM


Should work but you need to subtract dark current and bias level as that will tend to shift your histogram towards the right to higher values. Maybe the camera does that automatically for longer exposures? I don't know what the spectral bandpass is for the SQM - but I'm guessing you'd want the camera's response to be somewhat close if you're going to try direct comparisons. I do sky quality assesments using standard stars and measured sky background from images but have never tried comparing those numbers with an SQM. I get something like 17.5 in V for my location, just NE of downtown Baltimore.


#3 s58y


    Fly Me to the Moon

  • -----
  • Posts: 6519
  • Joined: 12 Dec 2004

Posted 15 October 2007 - 11:44 PM

Samir Kharusi also has also posted here on CN about a different DSLR method of estimating the sky brightness. In his method you expose to get a centered histogram, and record how long it takes.

When I tried it, I found out that it's about 4 to 5 times brighter in my driveway than at a true dark sky site (22.0 mag sq arc-sec). This seems to agree pretty well with the typical SQM reading of 20.2 to 20.5 on reasonably good to great nights.

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