how about two peltiers ? one on the bottom and one on the side. i guess alot depends on if you have to carry or move it much or not . or a bucket of dry ice and a chute on the side of your enclosure to drop some dry ice into it one bit at a time.
The extra peltier, by itself doesn't weigh much. But when you add in the extra heat sink, fan, etc etc, it adds up. It could be self-defeating to double up on those items. On the other hand, a second peltier, side-by-side with the other one could be made to share a common heat sink and fan. BUT another word of caution,. These peltiers draw between 3 and 4 amps each. By doubling up, you could cause a statewide brownout.
Well guys, the first results are in (and no, I'm not talking about the Qld election yet).
I've just done a full first trial run on my latest cooled box design and I'm thoroughly delighted with the results.
I started at an ambient of 20.75C and let the cooler run for 10 minutes without switching on the camera at all. After 10 minutes, the temp in the box was 4.8C and falling. The camera, by the way, was/is a 350D. I have not yet wired in any temperature controller so I just let the system run for 45 minutes to see what it would do.
I switched the camera on after ten minutes and started back-to-back 300 second subs at 1600. The temp sensor was velcro'd to the back of the camera beside the screen.
The crazy thing is that the general pattern of temperature decline did not seem to be significantly affected by the camera. The trend-line on the attached graph seems to have continued pretty much undisturbed.
It hit zero after 30 minutes, but continued to fall to -3.4 after 45 minutes. The 350D doesn't have a temperature sensor built-in so I can't say what the actual sensor temperature was.
This means, a drop of about 24C in real terms - no bull!
I know this means I'll have some condensation issues unless I regulate the temperature closely but even if I have to keep it to 3 or 4C on the positive side, I'll be delighted.
The design for this cooler box is built around a Custom Thermoelectrics air-to-air Peltier. The inside and outside of the box are both insulated. I'll post some more pics of the finished article if anyone want.
Well guys, the first results are in (and no, I'm not talking about the Qld election yet).
I've just done a full first trial run on my latest cooled box design and I'm thoroughly delighted with the results.
I started at an ambient of 20.75C and let the cooler run for 10 minutes without switching on the camera at all. After 10 minutes, the temp in the box was 4.8C and falling. The camera, by the way, was/is a 350D. I have not yet wired in any temperature controller so I just let the system run for 45 minutes to see what it would do.
I switched the camera on after ten minutes and started back-to-back 300 second subs at 1600. The temp sensor was velcro'd to the back of the camera beside the screen.
The crazy thing is that the general pattern of temperature decline did not seem to be significantly affected by the camera. The trend-line on the attached graph seems to have continued pretty much undisturbed.
It hit zero after 30 minutes, but continued to fall to -3.4 after 45 minutes. The 350D doesn't have a temperature sensor built-in so I can't say what the actual sensor temperature was.
This means, a drop of about 24C in real terms - no bull!
I know this means I'll have some condensation issues unless I regulate the temperature closely but even if I have to keep it that o 3 or 4C on the positive side, I'll be delighted.
The design for this cooler box is built around a Custom Thermoelectrics air-to-air Peltier. The inside and outside of the box are both insulated. I'll post some more pics of the finished article if anyone want.
It's sensor temp which is important. I could get the enclosure down to nearly zero even on basic test builds but getting that to the sensor was the difficult part. Best was 10* to start with but that climbed back up to a 5* difference from ambient once imaging began. Useful but not significant for a real noise reduction.
Yeah, I understand that. I'll never know what the actual sensor temp is though since it is not enabled in the 350D and will not show up in a FITS header. Bummer! I'll just have to stumble along in ignorance.
Peter
It will depend on the camera, but I've found my 1100D gives pretty clean darks around and below 5C ambient. I have no idea what the sensor temperature is, as I can't find it in the exif data.
It will depend on the camera, but I've found my 1100D gives pretty clean darks around and below 5C ambient. I have no idea what the sensor temperature is, as I can't find it in the exif data.
The 1100D is supposed to be one of the better suited astro Canons from what I understand. Low noise etc. I would have bought one if I could have found a cheapie or second hand.
Does it not return Sensor temp info in BYEos ?
The 450D I picked up real cheap because supposedly it has lens connection problems, error 99 or something. I don't have any Canon lenses so no problem. Keep meaning to pick up a cheapie lens to try it with, one day maybe. But it does report back sensor temp and works extremely well with BYE.
The noise is certainly temperature dependent, which is why we're having this thread, and for winter use it's fine without cooling, but once the ambient temperature exceeds 15C the noise becomes a snowstorm of exposures of 2 minutes or longer.
These are a mix of shots before and after fixing styrene and installation of TEC and DSLR. You'll get the sense of it I'm sure.
There were 3 main objectives here:
1. Reducing the gap between the back of the camera and the front of the cold-side fan assembly. It's about 4mm in this build;
2. Reducing the size/volume of the box itself; and
3. Using an air-to-air cooler and NOT making the cooling of the whole box the priority. Instead, cooling the body of the camera came first and the rest was an incidental (probably unavoidable) side-effect and did no harm anyway.
Hah ! I've gone along the same lines for my latest attempt. Went back to the first plastic enclosure which is a tight fit. Added a big wrap around alum plate internally with an extra pair of fins over the TEC area. Then added a copper contact finger to the tripod mount screw for some minor metal to metal contact. I haven't added an internal fan but that is an option.
Big change was the insulation method. Front area is internal styrofoam but the rest of the box is externally wrapped in multiple layers of bubble plastic, bubbles facing in to create an air cell insulation skin. I've seen\heard of it being used, single layer, to give window insulation when you can't afford double glazing. The back of the box also has an additional layer of styro but overall it has made for a smaller package and very light.
Testing is yet to happen due to other obligations at present. Stay tuned ...
Results: Better than previous attempts. I got down 9* below ambient (24 >15) after about 45 mins. Ran some exposures but as soon as started to do several longer exposures the temp came up 3*.
The insulation works well and an internal fan might help as well but the effect is limited at the sensor and uncontrolled. I'll find a miniature fan just to cover all angles.
But I went back to my earlier experiments with the ZWO AS! 120mc after having some workshop time with the software to understand it better. At 10 mins I had 12* cooler than ambient and eventually I achieved and maintained 20* below ambient at 4.2* C.
Figured out how to do longer exposures, the IR cut filter is gone and I can keep the front glass window clear of ice with my patented air injection system while the Silica keeps the sealed optical cavity dry. Sweet !
It's weird transient clouds at present but it's all up and running just in case it clears later as it has done so recently. Hopefully the clouds will think I have given up for the night and go away and annoy someone else.
Just a heads-up on this unit spotted on eBay - a double peltier unit with hot and cold side heat sinks and fans (i.e. air-to-air format). It draws 10 Amps though so it would be a sizeable whack out of your power supply but by crikey it would cool the cajones off your DSLR.
Latest iteration with most of the mods I've been planning.
Internal fan added, small pancake unit and a smooth alum plate for the TEC to mount to.
I've also spaced the plate clear of the bottom surface of the plastic box to minimise heat loss. I will add a bit more side insulation when I test it next although the bubble wrap system works very well with three layers.
First pic shows the fan and TEC mount plate with the rest of the alum bent inside the box.
Second is the camera inside and showing the front styro panel.
Last pic is closed up with TEC and radiator mounted. Wiring is a bit messy but it all works.
Weather is C#@*&%*! btw, rain and clouds for the next 3 days, ... figures
Peter, Slightly off topic, but do you think one of those would cool a chest-type freezer using 100mm insulation, for off-grid homes, rather than the normal compressor unit? Dunno which is the more efficient.
Last edited by mldee; 04-02-2015 at 07:57 PM.
Reason: Added Peter, grammar
Peter, Slightly off topic, but do you think one of those would cool a chest-type freezer using 100mm insulation, for off-grid homes, rather than the normal compressor unit? Dunno which is the more efficient.
G'day Mike.
Short answer is yes. Long answer is yes BUT it depends how big the chest-freezer is, what it's construction materials are and how much power you want to push into it.
These units (the double-peltier type) are now fairly commonly seen in cheap wine fridges and the like where the priotity is chilling not freezing. They are good at keeping temps between 10-15C but of the 3 I've owned over the past 5 years, none has lasted longer then 2 years. If you want to freeze stuff, I'm not sure these have got the grunt to do it.
Here's a first-blush response: I'm not yet convinced that the style of external (hot-side) heatsink you're using is optimal. I look at the sexy copper pipes coming out there and ask myself - is this a fast lane or a bottleneck? Would you get more heat into the aluminium fins if there was a direct-metal to ceramic contact as in a conventional TEC design? These copper tubes are seen fairly often with CPU coolers among overclockers and the like but they aren't dealing with the same amount of heat that a peltier can produce. Are they better than a flat-surface contact between the hot side of the peltier and the base of the heatsink. Not wishing to pour any water on the fire but I'd like to be convinced about that.
That neat little device is brilliant. I have ordered some Peltier units and a fan-Peltier combination. On top of this is a solid state fan/Peltier cooler. Even simple jobs take me months however as you have pointed out this is a very interesting project. If by some chance I get the thing up and running I will post pics. It was interesting to discover that there are off the shelf Peltier control boxes.
Here's a first-blush response: I'm not yet convinced that the style of external (hot-side) heatsink you're using is optimal. I look at the sexy copper pipes coming out there and ask myself - is this a fast lane or a bottleneck? Would you get more heat into the aluminium fins if there was a direct-metal to ceramic contact as in a conventional TEC design? These copper tubes are seen fairly often with CPU coolers among overclockers and the like but they aren't dealing with the same amount of heat that a peltier can produce. Are they better than a flat-surface contact between the hot side of the peltier and the base of the heatsink. Not wishing to pour any water on the fire but I'd like to be convinced about that.
Peter
Peter, the copper tubes are welded to a flat copper plate that is just marginally bigger than the TEC. A bit of Silicon paste and 12 volts and it shifts a lot of watts across. The plate is designed to sit flat on the CPU upper surface with thermal compound to shift the heat. It works.
When I first power it up the radiator assembly gets quite warm as does the ensuing airflow. It reduces as the internal temp drops. On the ZWO the response is extremely quick in the first 5 mins. I can drop about 12* C easily straight away.
I'll post a better pic of the TEC\Plate junction if you want convincing.
BTW: a multicore server CPU puts out a lot more heat than the Peltier. These heatsinks are from big servers.
Thanks Brent. It sounds formidable. As a result of that, I'm having a look at a CPU cooler assembly I have here and toying with different ways of attaching it. I'll watch your results with real interest and hope to profit from your experience.
Peter
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.
I'll find a miniature fan just to cover all angles.
Sweet !
