Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Keeping a variable load at X temp using TEC's
New Posts  All Forums:Forum Nav:

Keeping a variable load at X temp using TEC's

post #1 of 6
Thread Starter 
Now i believe most peoples answer to this question would be to adjust the voltage/ amps supplies to the TEC's to keep the load at X temp.

Up until recently this is how i thought i should be doing it too.

i have since realised the best way of keeping a variable load at x temp is to leave the voltage / amps unchanged but turn unneeded TEC's off. this drastically decreases the electricity used.eg

Numbers below are based on 226 watt tec's and a load of 200 or 50 watts and a ambient of 25c with hot and cold side C/W's of 0

4 TEC's cooling a 200 watt load to x temp uses 509 watts at y volts
4 TEC's cooling a 50 watt load to x temp uses 142 watts at z volts
2 TEC's cooling a 50 watt load to x temp uses 92 watts at y volts
1 TEC's cooling a 50 watt laod to x temp uses 74 watts at y volts

this is all well and good but im not sure how i could turn the tec's off and on based on load in this way.
post #2 of 6
Not sure how to do it either.

Plus, won't the tec respond to slow to load changes? Say, you increase load suddenly, then the tec needs to catch up cooling the CPU. Even a few second delay between the load increase and the tec reacting will be detrimental for the CPU (there will be a spike in temps before the tec catches up cooling it down again).
    
CPUMotherboardGraphicsRAM
E8400 @ 3.8 GHz DFI LANPARTY DK X38-T2R eVGA 8400GS 2GB Mushkin DDR2 800 (single stick) 
Hard DriveOptical DriveOSMonitor
250GB WD SATA None Windows XP Pro 32-bit HP Pavilion f1703 
KeyboardPowerCaseMouse
N/A Corsair HX520 Coolermaster Cavalier N/A 
Mouse Pad
N/A 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
E8400 @ 3.8 GHz DFI LANPARTY DK X38-T2R eVGA 8400GS 2GB Mushkin DDR2 800 (single stick) 
Hard DriveOptical DriveOSMonitor
250GB WD SATA None Windows XP Pro 32-bit HP Pavilion f1703 
KeyboardPowerCaseMouse
N/A Corsair HX520 Coolermaster Cavalier N/A 
Mouse Pad
N/A 
  hide details  
Reply
post #3 of 6
Hmm, I agree that with a given load/temperature/cooling there is an optimum number of TECs and having too many is detrimental.

I'm just wondering how the reduced heat transfer on each side affects this.

It's certainly something I'd like to look into and simulate. Its hard though because every new equation I find for my software gives me different and clearly wrong results...

Also, Its actually an important thing to consider in my opinion. There's no way I'm running a chiller that isn't power efficient (I care too much about waste and bills). And like most, my system will be running more or less idle for 90% of the time so this is where efficiency is most important.
Edited by Scarlet Infidel - 6/2/08 at 2:51am
post #4 of 6
Thread Starter 
Quote:
Originally Posted by Chozart View Post
Not sure how to do it either.

Plus, won't the tec respond to slow to load changes? Say, you increase load suddenly, then the tec needs to catch up cooling the CPU. Even a few second delay between the load increase and the tec reacting will be detrimental for the CPU (there will be a spike in temps before the tec catches up cooling it down again).
um yes ... however if we use a chiller and cool the water to xx temp this would resolve this problem . however this wouldn't be keeping the cpu at x temp any more
post #5 of 6
Thread Starter 
Quote:
Originally Posted by Scarlet Infidel View Post
Hmm, I agree that with a given load/temperature/cooling there is an optimum number of TECs and having too many is detrimental.

I'm just wondering how the reduced heat transfer on each side affects this.

It's certainly something I'd like to look into and simulate. Its hard though because every new equation I find for my software gives me different and clearly wrong results...

Also, Its actually an important thing to consider in my opinion. There's no way I'm running a chiller that isn't power efficient (I care too much about waste and bills). And like most, my system will be running more or less idle for 90% of the time so this is where efficiency is most important.
i second that .. the only way i can think of it working is... (in a very basic form)
have a TEC come on say every 0.5c this would require a multi channel controller however and would result in a variance in x over the range.


having a system like above would mean you could have a 1000 tec's and it would always be running at it's most efficient. If you controlled it via voltage it would be a disaster.

heheh cooling 50 watts using 1000 TEC's to x temp would result in 26,100 watts of electricity being used !!! lol
(this is using the same stuff as in my first post)

humm turning TEC's on and off based on load is the way to go i think
post #6 of 6
Quote:
Originally Posted by Ultrasonic2 (muffy) View Post
i second that .. the only way i can think of it working is... (in a very basic form)
have a TEC come on say every 0.5c this would require a multi channel controller however and would result in a variance in x over the range.


having a system like above would mean you could have a 1000 tec's and it would always be running at it's most efficient. If you controlled it via voltage it would be a disaster.

heheh cooling 50 watts using 1000 TEC's to x temp would result in 26,100 watts of electricity being used !!! lol
(this is using the same stuff as in my first post)

humm turning TEC's on and off based on load is the way to go i think
what are you planning to do? a chiller?
if so, id just use a temp probe in the water, once temp exceeds X, have a relay activite the TEC PSU, and below said temp shut it off.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Keeping a variable load at X temp using TEC's