Wow, I had not checked this thread since a while after I started it! Looks like it has taken a life of it's own and there is some amazing building of stuff going on!
My plan now is to send my DSLR O/S and for 350 Euros get is cooled with a cold finger and anti dewing element. Will be doing that next month.
Clear skies,
John K.
G'day John. Yes, I suppose I'm one of the guilty parties who hijacked your thread. But, as you have said, there's some good stuff coming out of it. I'm really enjoying the spirit of adventure almost that comes from being free to experiment and try stuff.
By the way, would you mind telling us (me) where you're sending your DSLR to be modd'd, and what type it is? I'm toying with the same idea - despite the fun I'm having building little mini-fridges
Here's a close up pic of the copper plate at the bottom of that radiator array. You can just see the TEC and it's two wires sandwiched beneath it. The copper thermo-siphon tubes are sweated onto the back of the plate.
Righto you blokes. It's raining and the forecast is for more and more and more. So I have no choice but to tinker.
That's my only excuse. That being said, here's an idea I want to throw up for your comments.
I was looking back over this thread and trying to see stuff I'd missed - brushed over - and there was one theme I kinda liked - an aluminium bar underneath the DSLR, fastened in the usual way by the screw under the camera and cooled directly from the base of the peltier.
To make matters worse, I had left my disc grinder out on the bench overnight and also had a piece of 80x200x20mm bar that was just begging me to do something.
So I cut a shape that would allow me to fasten it to the base of a TEC while maintaining electrical isolation of the top and bottom of the peltier itself. The bar then narrows to about 50mm and runs the length of the camera.
I then buried the whole assembly in some polystyrene rescued from a recent purchase of a new UPS.
In the attached images, I've positioned a heatsink and a small SLR where the proper ones would go and I've also put a thermocouple into the body of the al bar directly under the center of the camera.
Here's the plan - run the TEC and cool the bar. Because the DSLR is sitting on top of the bar and in direct contact with it, there ought to be a reasonable amount of heat transfer. Because the body of the bar is well insulated, there ought to be minimal loss of cooling that way.
As usual with such rushes of invention, I have been pulled to a halt by not having some screws the right length to attach the TEC to the bar. A trip to the hardware on Saturday will fix that and then a test of the theory.
Cheers
Peter
UPDATED:
Running it with the TEC in place but no camera yet.
Ambient 22.5. After 30 minutes, the internal temp in the cooling bar is down to 2.3C and stable. Here's a graph of the first 25 minutes.
So that's a drop of 20C in 30 minutes. How much of that would translate into the sensor I don't know of course.
Hi Peter, I tried an alum bar across the bottom but it only had air conection. I also tried a copper finger with metal to metal contact but I was unable to measure any better sensor response with either setup.
I think the sensor temp managed to stay a little lower for longer comapred to earlier attempts but that may have been my optimistic opinion at the time.
You really need to get a camera with sensor temp displayed at EOS to obtain valid results. The big issue I had with cooling a DSLR is that as soon as you start running a series of longer (30 sec eg) exposures the sensor temp rapidly climbs to wthin 5*C of your first measurement. Any darks taken are invalid as the noise values are not consistent.
The little ZWO I can cool down to -2.5 and hold it there for long imaging sessions. I'll keep playing with the DSLR stuff but to get consistent results the real answer is a cold finger mod. I may look into that if I go ahead with the IR cut mod although the ZWO does that already.
You really need to get a camera with sensor temp displayed at EOS to obtain valid results.
Yep.
Both my DSLRs are in cooler boxes ATM and I don't want to disassemble them. So I had a try with my Canon S3 IS which has an internal temp sensor. It also has limited bulb settings and the max I could do is 60x15 secs subs. Still, I ran it for an hour with the camera sitting on but not screwed into the bar.
Bar temp fell pretty much in line with the data in the previous graph - down to 2.5 or so within 30 minutes and stabilised there.
The body of the DSLR must be well insulated - the internal temperature never went below 24 and climbed to 29 by the end of the sequence. As far as I could tell, the cooling of the bar had no appreciable impact on the internal temps of the camera. Perhaps if I had screwed it down so there was a positive pressure between bar and camera and some sort of metal-to-metal contact via the securing screw it might have made some difference. But I doubt it would have been significant. I think I can discard this experiment as a failure. I could always use it as a stubby cooler I suppose!!
Hahaha, yeah, a stubby cooler seems far more effective and useful.
You're right about the camera body's insulating effects. I guess they made it that way for the 'normal' user to protect all the innards.
My copper finger had a metal to metal connect courtsey of a rough drill bit deforming the hole edges but it was minimal at best. I hoped it would at least give a faster path for all them hot bits and maybe it did but insufficient to be of any use.
I'll continue to use my current setup as is till mod time comes. It works well enough for my imaging experiments and I love BYEos. It may tempt me one day to get a newer better Canon just for Astro.
Back to the drawing board for now
By the way, would you mind telling us (me) where you're sending your DSLR to be modd'd, and what type it is? I'm toying with the same idea - despite the fun I'm having building little mini-fridges
Peter
Peter - I am going to use Luis Campos - if you want his email address PM me.
Yep.
Both my DSLRs are in cooler boxes ATM and I don't want to disassemble them. So I had a try with my Canon S3 IS which has an internal temp sensor. It also has limited bulb settings and the max I could do is 60x15 secs subs. Still, I ran it for an hour with the camera sitting on but not screwed into the bar.
Bar temp fell pretty much in line with the data in the previous graph - down to 2.5 or so within 30 minutes and stabilised there.
The body of the DSLR must be well insulated - the internal temperature never went below 24 and climbed to 29 by the end of the sequence. As far as I could tell, the cooling of the bar had no appreciable impact on the internal temps of the camera. Perhaps if I had screwed it down so there was a positive pressure between bar and camera and some sort of metal-to-metal contact via the securing screw it might have made some difference. But I doubt it would have been significant. I think I can discard this experiment as a failure. I could always use it as a stubby cooler I suppose!!
Peter
So much for bright ideas. The most successful is the copper finger. Stay tuned, mod #6 on the way, with an intriguing twist - which is classified - just in case another bright idea bites the dust.
I confess the bar works if you remove the plastic bottom cover and fit a metal shoe directly in contact with the camera chassis. Brutal, and I think the electronics should first be coated in PCB lacquer or some such thing. I do not recommend this approach. The camera and bar require insulation to minimize competition with surrounding air mass, if you do try it.
EDIT: Bar temp -13C
A fully controllable coldfinger with sensor de-icing, as posted some years ago (#5) has been the most successful of all my hairbrained attempts at cooling.
Ok I will wait. I have my copper finger and heat sink from my old laptop and my TEC - ready for a design that works. Thanks for doing all the hard yards on this development guys.
Peter - I am going to use Luis Campos - if you want his email address PM me.
Thanks John,
I had a look at his site and wondered about the filter choices he offers. I have a 40D that was modified by Spencers in the states and the filter they installed is one they called "visual plus Ha" but I couldn't figure out which (if any) of Campos'choices was an equivalent. Any idea?
I stripped down the cooler containing my 350D this morning after having run it a few hours last night. That sesson ended prematurely when the camera operation became flakey and BEOS didn't behave.
Not sure if this was the problem but when I opened it up this morning there was at least 2 tablespoons of liquid sloshing around inside it - dyed bright green (I suspect that's from the polystyrene dyes). I hadn't built in any drainage holes. I can see now that this might be a darned good idea.
I stripped down the cooler containing my 350D this morning after having run it a few hours last night. That sesson ended prematurely when the camera operation became flakey and BEOS didn't behave.
Not sure if this was the problem but when I opened it up this morning there was at least 2 tablespoons of liquid sloshing around inside it - dyed bright green (I suspect that's from the polystyrene dyes). I hadn't built in any drainage holes. I can see now that this might be a darned good idea.
Peter
Yeah, I had a few drops of excess humidity in one of my attempts. But I also had a few excess holes I hadn't noticed so it didn't get to floating camera time. But you're right, it's a concern for any cold finger modification as has been noted elsewhere.
Yeah, I had a few drops of excess humidity in one of my attempts. But I also had a few excess holes I hadn't noticed so it didn't get to floating camera time. But you're right, it's a concern for any cold finger modification as has been noted elsewhere.
I think it depends on the design. Dew can be managed by exposing a small area of the cold finger to the air. I read this somewhere a few years ago and found it works. If you let ice/dew form on a square inch of metal it seems not to form on ones camera electronics. The explanaton for this escapes me - hence, folk law - but it works - with my set up, anyway?
In a sealed enclosure design you should be able to use Silica Dessicant. Once the air inside is dry then there is nothing there to form dew or ice.
I'm using that method on the ZWO cooler system. The 1.25" extender has 4 holes in the side which are exposed to another sealed cavity containing 3-4 small gel packs. Works extremely well. Also makes sense to keep the air volume in the enclosure as small as possible, less water molecules floating around to form ice etc and makes the dessicant last longer before you have to cook it for a few hours to dry it out again.
I have plenty of silica packs and keep a small sealed jar of them ready just in case. All the gear at work arrives with several include so I just collect them all.
Just bumping this up to see what is happening on the Cooler Box Design project?
And also to post up a source of Copper sheet material, whether your building a box system or cold finger, these guys can supply in the small quantity we need:
No further progress here although I'm planning to use the ASI 120 in cooled mode once I complete my Canon exposure experiments. If I can acquire another 450D or even an 1100D I'll take the plunge and do the IR mod on one and investigate a cold finger mod.
The dew problem seems to be the big bugbear. I have it resolved on the ZWO 120 with it's sealed front cavity and desiccant system so for now I'll experiment with that.
Frustrating this is. I keep coming back to Gary Honis' Cooler box design Mark III which yielded a 43.7F (or 24.27C) drop in temperature. No cold finger. His Dark Frame tests show substantial improvement, but without a temp sensor near the sensor chip its hard to know what real drop was realised.
Page down on that link page to see his dark test results and images.
I noticed that there was another prototype that experimented with gas phase change cooling, where he was running gas into the sensor area and had constructed a little manifold cold finger.
If the camera cooled by only half the total differential, dark noise reduction would be ~4x less with those figures - which is a substantial reduction.
Here's some real-world figures from last night.
The cooler is my air-to-air design box.
Camera EOS 40D which has a temp sensor on the chip.
Using a TEC controller with temp fixed in range 5-7C.
Ambient 22-24C throughout.
Camera sensor was at 7C before starting imaging run.
Chamber temp was 6.8C at the same time (after 60 minutes pre-imaging running).
Camera sensor was at 16C throughout imaging run - after an initial warmup perod) - taking 10-minute subs.
That tells me the sensor added 10C when working hard.
That also tells me the temp of the sensor was 16C below what it would OTHERWISE have been.
That tells me that we achieved significant reduction in dark noise.
That's impressive Peter. I might try the cooler box before I install a cold finger. Do you have any desicant or dew prevention measures inside the box? Condensation mangement is a major issue with the cold finger and likely requires a low pass filter heater.
