Can adding a second peltier cooler help lower temperatures even further?

[fsonnichsen]

I am using a small peltier cooler (fan on cold side, water cooled on hot side) to lower the temperature of an insulated box. I use this for testing the temperature drift of optical devices. I can get temperatures down to about 10deg C with this but was hoping for around 2deg.

My question is--would adding a second cooler in parallel to the box help? At first glance I would say yes--I am doubling the amount of heat energy I can drawing from the box. But the coolers themselves operate at finite temperatures. The temperature can go no cooler than the cold side of the peltier chip. In that case I would just reach temperature faster--but not lower.

The temperature of the chips-on paper-is something like -50deg. But that is on their surface and this quickly dissipates. It is hard to get an accurate reading of the chip, mounted with fins etc, with the means I have (IR gun, thermocouples). Like a lot of things in the lab, empirical always seems to be the bottom line.

Anyone have some experience with this? I am using tec1-12710 chips.

Thanks
Fritz

 

[russ_watters]

Have a look at the data sheets for the peltier. You'll see that as temperature difference gets larger, heat transfer drops. So yes, adding a second in parallel can help.

Also, how is the water cooling performance - is it doing a good enough job to remove heat?

 

[fsonnichsen]

Thanks Russ.

yes-the water cooling seems to do a good job. I have an air cooled unit with a pair of fans - looks like a jet turbo engine. My water cooled one is small-the cooling cell of course hardly bigger than the chip. It seems to perform a bit better than the air cooled. It is obviously best to use a LOT of water.

I was cooling it the a 70ml / min peristaltic pump and it worked-but the case got hot. I added ice to the cooling water and that helped. I since bought a much faster pump which is on its way here. (I don't have a lab sink but that would be the best coolant).

When a 2nd power supply arrives here I will put a 2nd unit in my chamber and hopefully this will get me down near 2deg.

cheers
fritz

 

[russ_watters]

Cool.

FYI, I did a peltier retrofit on an astronomy camera and also turbocharged a tiny peltier fridge I bought online. The fridge gets the cold side (passive/no fan) down below freezing but it needed a lot of insulation to keep the sodas cold. The cooling power at those temps is really small, which is why dual coolers could help. But make sure your insulation/container is solid.

 

[fsonnichsen]

Yes-insulation is important. I originally used a plastic junction box for the chamber and it performed poorly. The equilibrium box temperature was about 3 deg lower than room--I added 1/2" and then 1" insulation and that finally worked. Measuring the temperature of the box it is now close to room temperature.

I used to use an Andor ICCD with a peltier cooled CCD. The temperature was reported in the software at around -40C. Very sensitive spectro for $200,000.

I got into my lab this evening and redid the air cooled version of the box. I better seated the cooler with only very small gaps for he wires. The box is 15L and I used a 10A 12710 chip--I got down to 5 degC and about 4deg if I wait an extra 15 minutes. So I am home free as far as my present needs but when my parts are in I will look at comparing single and double units of the water cooled versions.

The one thing that is a devil is when mounting the unit there is a small amount of the "cool" side exposed to the air and you see condensation. So I am probably gaining heat there--the chip is of course, so thin that it is not practical to insure that the cool side completely sits in the box. I could spray insulating foam around this area but then I risk trapping heat--so probably just leave it as is. I insulated the clamping screws that enter the box using standoffs-probably nylon screws would be better but these are under tension (to clamp the chip/heat sinks) and I have seen a lot of nylon screws fail.

thanks
Fritz