Gap between the plates of a Thermal Electric Modules

[ramonegumpert]

Dear Experts

Does anyone know how big can we extend the gap between the 2 plates?

Thanks.

Regards
Ramone

 

[AlexLAV]

I'm not an expert in the field but I think you may extend the gap as far as you prefer. However you should keep in mind that the wider the gap the higher ohmic resistance of the conductors between the plates. The higher the resistance the higher heat deposition between the plates. Half of the total ohmic heat goes to the cold plate. Since the ohmic heat is proportional to square of the current while Peltie heat transfer is proportional to current, at very big gap overall cooling performance of the module is low. At the same time at very little gap there is considerable back transfer of heat from hot plate to the cold one.

 

[sophiecentaur]

Actually, more than half the resistively generated heat will go to the cold end because there is a greater temperature difference. (Bummer, eh?)

 

[AlexLAV]

Actually, more than half the resistively generated heat will go to the cold end because there is a greater temperature difference. (Bummer, eh?)

May be I’v said a bit vaguely. Let’s consider a thermo-electric module. What is power of a heat flux to the cold plate from the hot plate and the conductors between the plates? A solution of 1D heat transfer equation leads to (T_hot-T_cold)/L_gap*lambda_Q*S+1/2*P_ohm, where T_hot and T_cold – temperatures of the hot plate and the cold plate respectively, L_gap – distance between the plates, lambda_Q and S – specific heat conductivity and overall cross-section of the conductors, P_ohm – heat power deposition within the conductors due to ohmic loss. If the first term may be considered as a direct heat transfer from hot plate to the cold one, then addition due to ohmic heat deposition is equal to half of the total ohmic heat deposition.

 

[sophiecentaur]

I don't know much about the construction but there is no reason for the conductors to be anything but very low resistance, is there? And why should they not be long enough to make 'back flow' of heat insignificant?

 

[mender]

Theoretically, the shorter the TE element, the higher the possible heat throughput and the higher the power output. Getting peak efficiency of the energy conversion requires a high temp difference (delta T) across the module.

In practice the module is designed to work with a specific amount of heat to get the best delta T at that heat throughput. Since the heat conducted into the hot side also depends on the delta T across the hot interfaces, sometimes a TE module can put out more power at a lower efficiency by having shorter elements. This can lead to a higher device efficiency despite the lower module efficiency by increasing the heat flux at the expense of a lower delta T across the elements.

Sounds like short might be the way to go for more power but the lower limit on the space between the hot side and the cold side is a loss in module life and durability because of the stress from thermal expansion.

 

[AlexLAV]

The conductors, as far as I know, are made of semiconductor rather than metal, so their resistance is not very low. For example resistance of one of TE module I used to work with (40x40x4 mm in size and about 170 W of maximum cooling capacity) is about 1.7 Ohm. Taking into account that maximum operating current of the module is 11 A, the ohmic heat is far from low.

 

[sophiecentaur]

Oh, right. You live and learn. That could make a difference. And there would be a real limitation to the separation in manufacture.

 

[mender]

For example resistance of one of TE module I used to work with (40x40x4 mm in size and about 170 W of maximum cooling capacity) is about 1.7 Ohm.

I assumed the Op was talking about TE power generation.

 

[ramonegumpert]

Wow, I love this forum! There are scientists here!

wohooo!

thanks for your inputs. I am still digesting.

But please keep those useful comments coming in.

Yes, as some of you have guessed it. i am thinking of making a thicker TEG . I have only used a simple TEG for experiment and does find that while its easy to heat up the hot side, eg place on a stove just put out of fire, its difficult to get rid of the heat from the cold side and I would believe that more than 50% of heat is going to the cold side.

regards
Ramone

 

[mender]

Sounds like the problem is more of cooling than of thickness.

If feasible place the module in a vertical position on the side of the heat source with a finned aluminum heat sink on the cold side, with the fins aligned vertically as well. This will allow passive cooling via draft and will noticably increase the power output of the module by increasing both the temperature and the heat throughput across the module.

Depending on how well you design the cooling system and how much work/expense you want to go to, adding a fan to improve air flow across the heat sink can boost the net power output.

Increasing the thickness won't help if there is no reasonable provision for shedding heat.

 

[ramonegumpert]

Thanks bro!