I am experimenting with stacked Peltier cells trying to build a simple diffusion cloud chamber. I noticed that when simply stacking two peltier cells the upper side reaches a temperature (around -15 C) higher than the temperature obtained (around -25 C) if I put a small aluminium plate between the two, that is cooled down by the lower peltier and dissipates heat of the upper peltier. I was wondering whether there is any physical explanation for this (e.g. the aluminium block being larger than the lower peltier is able to dissipate better the parasitic heat produced by the upper peltier) or whether this is just a random result due to poor insulation/tightening of the system.

Some more data:

  • lower peltier is a 12710 module operating @12V
  • upper peltier is a 12706 module operating @5V
  • lower peltier is backed by a 10mm dissipator made up of 4 heat pipes surrounded by 2 aluminum radiators cooled by 3 10 cm fans .the aluminum block is 10x10 cm and thick twice the peltier

Setup 1: dissipator-->12710->10706 @ -15 C

Setup 2: dissipator->12710->aluminum block->12706 @-25

Styrofoam insulation around the thing.

In both the setups there is no thermal load but room air at the top and the temperatures achieved vary some C from one experiment to the other but stay pretty constant (10mim) during the same experiment.

any clues? Thanks!

  • $\begingroup$ Could you try the setup dissipator — aluminium block — 12710 — 12706? I suspect the dissipator is not doing enough of a good job and that in setup 2 you have a better dissipation due to the Al block. I suspect the setup I propose to yield even lower temperatures. Please report back. $\endgroup$ Jan 28, 2020 at 12:06
  • $\begingroup$ Ideally put a bigger aluminium block on top of the two Peltier modules and make sure to get a very good dissipator all around the block. This way the Peltier modules will able to dump much more energy into the aluminium, removing also a lot more heat at the cold side. $\endgroup$ Jan 28, 2020 at 12:36
  • $\begingroup$ thanks a lot for your suggestion. actually the dissipator is the main suspect here.. However I did not understand your second comment. How many new setups are you suggesting? the first comment seems to suggest dissipator-->aluminium block-->12710-->12706. but what about the second comment? where should I put the "bigger aluminum block+dissipator"(b.a.b.d)? is it like fan dissipator-->b.a.b.d.-->12710-->12706 or fan dissipator-->12710-->12706-->b.a.b.d ("on top"). thanks again $\endgroup$ Jan 28, 2020 at 18:16

1 Answer 1


I suspect the dissipator isn't able to remove enough energy out of the system, and that the aluminium you placed is able to absorb some energy, thereby helping the Peltier modules to dump some of the heat they extracted from the cold side, to the aluminium (and eventually some of it to the dissipator).

I suggest the following setup to get a noticeable improvement in cooling:

dissipator --> big aluminium chunk (the more massive the better) --> Peltier 12710 --> Peltier 12706.

I would maybe even swap the Peltier modules. I am guessing from their names that they have 127 pairs of legs (n and p legs) and are 6 mm and 10 mm thick respectively. The thicker they are, the higher their electrical resistance and the higher their thermal resistance. I don't know how much of a difference it could make. But my suggestion to use a much bigger aluminium block should be the biggest short term improvement.

  • $\begingroup$ thanks, I'll give this a try. I am not sure about swapping the two peltier as the 06 and 10 stands for the nominal current hence the 10 is generically more powerful, both in terms of potential heat transfer and joule dissipation. usually stacked peltier combinations scale down the power as you rise the stack. $\endgroup$ Jan 28, 2020 at 19:29
  • $\begingroup$ @l4teLearner So how did it go? $\endgroup$ Feb 1, 2020 at 8:54
  • 1
    $\begingroup$ I did not see an improvement but I suspect I should improve the dissipator I am using.. and the peltier! I will try with better dissipator, maybe liquid cooler. thanks anyways $\endgroup$ Feb 2, 2020 at 8:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.