DSLR Cooling to Near Ambient with Fan Cooler

[Chris]

Well done Gina, a very nice illustration of the benefits of cooling the chip:)

[Gina]

Thank you

[Gina]

I think I've proved the difference between cooled and uncooled cameras so tonight I'm going to run an extended darks sequence with the two cooled cameras. I didn't really take the camera with the blue cooler to the extreme and with this seeming to have a better cold finger (or sensor) I'm planning to add 5000 and 7200 secs to the darks sequence run overnight. It would have been nice to have set it up in the obsy where it's cooler but it's been raining all afternoon and evening and I don't want to get stuff wet. It's supposed to be fine tomorrow, so I'll re-install the WF rig on the dovetail ready for tomorrow night. It is fine and clear I'll do some long awaited imaging but otherwise I can do a lower temperature darks run.

[Gina]

Results from last night :-

Somehow I failed to add the 5000s to the list in APT so the sequence goes straight from 3600s to 7200s (2 hours). The latter really is TOO MUCH

1. Histogram - 3600s - Silver cooler
   
2. Histogram - 3600s - Blue cooler
   
3. Image - D_2012-10-26_01-24-07_9703_ISO3200_3600s__28C.jpg
   
4. Image - D_2012-10-26_01-25-45_7173_ISO3200_3600s__24C.jpg
   
5. Image - D_2012-10-26_03-24-47_9704_ISO3200_7200s__28C.jpg
   
6. Image - D_2012-10-26_03-26-29_7174_ISO3200_7200s__24C.jpg
   

[Gina]

Results from overnight in obsy last night. Ambient temperature went down to freezing outdoors. I took darks sequence up to 2 hours and here are the histograms and images for 2 hours at ISO 3200.

1. Histogram_Silver_D_2012-10-27_04-43-34_9732_ISO3200_7200s__7C
   
2. Histogram_Blue_D_2012-10-27_05-02-04_7204_ISO3200_7200s__3C
   
3. Image - D_2012-10-27_04-43-34_9732_ISO3200_7200s__7C.jpg
   
4. Image - D_2012-10-27_05-02-04_7204_ISO3200_7200s__3C.jpg
   

Looks like I should have taken the exposure up a couple more notches or taken some at ISO 6400. If it's cloudy again I might do that tonight. Previous experience has shown that the noise increases more than pro rata at ISO 6400 (the maximum for the 1100D). However, if 6400 looks viable at around 0C it would be very handy.

[Gina]

I've set up a darks run on the WF cameras with ISO 6400 up to 3600s and with both ISO 3200 and ISO 6400 for exposures of 5000s, 7200s and 10,000s. That lot will run until breakfast toime tomorrow morning

BTW I found where the light leak was into the Silver cooler camera - the boby cap - fixed now with duct tape So any sensor output will not be due to light.

[Gina]

Now have results from last night's darks sequence run. The computer on the Blue cooler camera crashed before the end of the run but I have full results for the Silver one. Here is an image and histogram from the last dark - 10,000s at ISO 6400. That's 2.77 hours or over 2 hours 45 mins. EXIF T showed 8C but I think this is reading about 3C high - ambient estimated at about 2C.

I don't think CMOS chips suffer from reciprocity failure but sometime I'd like to try one of these extremely long exposures on a very faint object but I don't think the background illumination will allow it. I could do with a very narrow NB filter - best I have ATM is a Baader 7nm Ha. Maybe on a clear night without moon either on the scope or on a well stopped down lens... Don't hold your breath

- ->

[Gina]

For comparison here's the ISO 3200 version.

- ->

[Chris]

I think looking at the results you could perhaps max out the ISO @ 6400 for 5 minute subs and you'd have little noise to show for it on your final stacked image, especially if you consider natural dither and darks. Well this is what I think if thats all the noise you get for 10,000 seconds!!

[Gina]

Yes, I agree I was amazed

[russellhq]

Gina, I was reading the following post and thought of this one. It's a software project by one of SGLs members (Ags) for quantatively measuring sensor noise on Canon DSLRs:

http://stargazerslounge.com/topic/166291-phototool-2012/

[Gina]

Gina, I was reading the following post and thought of this one. It's a software project by one of SGLs members (Ags) for quantatively measuring sensor noise on Canon DSLRs:

http://stargazerslou...phototool-2012/

Thank you - I'll take a look at that

[rcheshire]

Great work Gina. I like the reduced complexity, compared to a TEC. There is a lot to be said for this approach.

[Gina]

Great work Gina. I like the reduced complexity, compared to a TEC. There is a lot to be said for this approach.

Thank you It's certainly worked a lot better than I expected and I'm thinking that when I get the Bayer matrix removed from one of the sensors giving a lot better sensitivity, I might just use this cooling method for a mono camera on the scope. I'll keep the ultra cooling system for OSC.

At this rate I could have 4 1100Ds - the OSC TEC cooled with full set point cooling, a mono with filter wheel for the scope and a pair of wide-field with lenses and Ha/OIII clip filters. The latter three being just fan cooled cold finger.

[Chris]

I take it the bayer matrix needs to be removed professionally?

Edit: lol literally a minute after writing the above I found this thread:D

Both amazing and brave!

[Gina]

I haven't got past the outer glass cover plate yet I don't want to destroy the sensor before I even start to remove the Bayer matrix! An annoying setback

[Chris]

Chisel knife?

Once you in though I'm not sure about using a dremmel after looking at the resulting flat field image, not that I'm taking anything away from RAC what he's done is pretty amazing! I'm thinking acetone is worth a go.

[Gina]

Chisel knife?

Yes, I was thinking of something like that - hammer and chisel but only little ones. I don't think I'd be too upset if I destroyed one sensor if I could see a way ahead to do the mod.

Once you in though I'm not sure about using a dremmel after looking at the resulting flat field image, not that I'm taking anything away from RAC what he's done is pretty amazing! I'm thinking acetone is worth a go.

Could be Various solvents are possible and also paint stripper has been suggested. If only I can get in...