What happened to TEC cooler?

Because they are very inefficient. You have to cool off the heated side so that it can take the heat from the processor. They are also very unforgiving if a component fails.
 
yeah, cool off the heated side like you cool off the CPU and it gets sandwiched in between for 50 more bucks and 50 less degrees

i have this cool idea for a nice plan with this i will post tomorrow
 
Don't even think about stacking Peltiers :P You end up getting a lot of heat you have to dissipate if you do that. Also they don't work very well if you cool the hot side to near 0C
 
who said stacking?

what about this:

CPU->Copper cold plate->peltier->copper cold plate-> cooling solution

much better temps with this, and you can apply any cooling solution, of course water is the only thing that would work
 
Actually, with TECs, if you get the wrong type and don't do your math, they can actually increase temperatures. Not what you want. Besides, they are really power hungry. Running my computer at 450++ watts takes up enough electricity, but if you add in another 200+ for a good TEC (not some cheap 40 watt one that can raise temps), then that hurts with the power bill.
 
who said stacking?
I did :P It's a good idea in theory but it breaks down when you actually look at the numbers for the heat it will end up generating.

For a CPU with a TDP of 69W you'll probably want at least a 250W peltier. You don't want to run them at full power either, they are even less efficient then.
 
oh yeah, i am not stacking , if i ever do this then just single. and a separate cheap PSU is a good solution for the TEC. What is the TDP for a C2D on load?
 
Last edited:
how come no one sticks onto this idea?? i mean its phase change cooling for a small fraction of the cost if done right, i mean am i missing something? how come no one likes it?
 
how come no one sticks onto this idea?? i mean its phase change cooling for a small fraction of the cost if done right, i mean am i missing something? how come no one likes it?
Because, as mentioned before, TECs are quite inefficient. If you have a CPU generating 50 W you will need a cooler to remove on the order of 500 W from the hot side of the TEC. This really can only be done well using a good water cooler. For a reliable TEC system you’d probably end up spending just as much as a vapor compression chiller.
 
wait, so how do you determine the inefficeincy?

say the CPU gives off 80w and you are using a 437w peltier, how do the numbers work and correlate?
 
i mean its phase change cooling for a small fraction of the cost
Is nothing like phase change cooling.... for astart phase change as the name suggests changes the state of a substance continuously in order to create a cold plate. A TEC or a peltier, what ever you want to call it uses thermocouples and basic rules of physics to operate
 
wait, so how do you determine the inefficeincy?

say the CPU gives off 80w and you are using a 437w peltier, how do the numbers work and correlate?
The efficiency depends on several factors and is generally in the range of 5-10% of Carnot (ideal) efficiency. The point is that even with the most efficient TEC you’ll be producing several times the amount of heat that you would have with the CPU alone. This is more than most heat sinks can handle and thus requires water cooling (even with air cooling you’d need really good case ventilation).

Is nothing like phase change cooling.... for astart phase change as the name suggests changes the state of a substance continuously in order to create a cold plate. A TEC or a peltier, what ever you want to call it uses thermocouples and basic rules of physics to operate
I think he meant the performance of phase cooling for a fraction of the cost.
 
I think he meant the performance of phase cooling for a fraction of the cost.

thats exactly what i meant


how would i determine the efficiency of a certain peltier. what variables and equations are accounted for. any good links to help me?

EDIT: excellent link cromewell!!!
 
Back
Top