View Full Version : Pelts on AC? possible?
Is it possible to find a pelt I can run straight from the AC mains by using a bridge rectifier? Are there any pelts ou there that kan take 240 or 120 volts?
Plz help :) I'm trying to solve the PSU problem here...
JaY_III
08-13-01, 06:13 AM
Originally posted by Quaky
Is it possible to find a pelt I can run straight from the AC mains by using a bridge rectifier? Are there any pelts ou there that kan take 240 or 120 volts?
Plz help :) I'm trying to solve the PSU problem here...
sorry pelts are DC only.....
if you had AC, it would try to warm and cool both sides of a pelt at the same time(actualy switch very fast)
Well, lets say you use a rectifier - then you get 110V DC.
Now, lets say a peltier is rated for 72Watts @12V DC (yes, this is on the low end) - then P=I*V -> I = 6 amps. V=I*R -> R=2 ohms.
Take that 2 ohm peltier and plug it into 110V DC, and you get V = I*R -> I = 55 amps! (that should trip your breaker/blow your fuse, but let's pretend that it doesn't) P = I^2*R -> P= 6050 Watts. Good luck removing that kind of heat, even with some extreme water cooling.
Conclusion - running peltiers directly off 110/220V power is a bad idea.
You may consider getting a second old AT or ATX power supply. I've seen directions somewhere on the net on how to fool these power supplies into thinking there is a mobo attached. Then you can plug your peltiers into these power supplies.
Hey Vimal, I just asked if there are any pelts that operate on high voltages. You don't have to wave around all that formulae to prove how stupid my question was. I am gathering info on this subject to run some tests with it. It is possible that there are TEC's that can handle several hundred volts, like diodes eg.
And for the record, when converting AC to DC with a bridge rectifier your voltage multiplies by 1.4 (2 times square root of two) as the buffered output voltage is equal to the peak voltage of the sine curve. But you allready knew that off course. :D
Don't be ****ed, you pointed at my mistake, I pointed at yours. Now we both learn, that's how it goes :)
Please accept my apologies if I sounded obnoxious.
I (incorrectly) interpreted your question as to whether a regular peltier could be powered of 110V.
Good point on the rectifier. As far as I'm concerned, electrical theory is useful only down to an order of magnitude for what I need. This also explains why I fry 9/10 parts that I work with ;)
I'm not sure if there are special high-voltage TEC's, but I would doubt it. Peltiers are essentially a series of P-N junctions made out of some sort of semiconductor. The only way for it to handle a really high voltage is to be bigger (same effect as having many peltiers in series) or have a dedicated resistor somewhere in there. I guess that is why most large-scale coolers use A/C type systems instead of TEC.
Thx Vimal, I took it as a habit on this forum ro react immediately if I felt attacked, so I don't have te keep hard feelings for anybody in this community. It looks like I overreacted a little. You wrote something interesting though:
Peltiers are essentially a series of P-N junctions made out of some sort of semiconductor. The only way for it to handle a really high voltage is to be bigger (same effect as having many peltiers in series) or have a dedicated resistor somewhere in there. I guess that is why most large-scale coolers use A/C type systems instead of TEC.
I got a few questions:
How big is bigger? :)
What do you mean by series? Do you mean that a pelt is built from small components that each dissipate some power? Thus the more of these components the more powerfull the pelt.
And the dedicated resistor ... How about that?
For all the pelt experts out there, is there a website which describes the working of a pelt?
TIA, Quaky
I was just looking at this site:
http://www.coolworks.simplenet.com/thermote.htm
I guess 10x surface area should take care of it. Increasing density should have the same effect (that is probably how they have both 72W and 156W in a 40mmx40mm package)
Peltiers look to be a bunch of diodes, all lined up in parallel but wired in series. If you took 10 peltiers and strung them along positive to negative it could work. As to what you could do with 10 peltiers...maybe cool processor, chipset, video card, system ram, network card, and even a couple sodas while you're at it. Come to think of it, that would be a really funny case mod if you had a soda cooler in there.
How exaxtly would you wire them in series?
AC, if run through a rectifier and buffered as quaky suggested, should be converted to DC. (Unless I missed something in my little green Radio Shack Electronics-for-dummies book)
As to wiring peltiers in series, connect the negative lead of one to the positive lead of the next. Then you are left with one positive lead and one negative lead which should be able to handle a higher voltage without toasting. Same way the old christmas tree lights were wired. Note that wiring peltiers in series should work IN THEORY. I have not tested this, and would like a real electrician's opinion first.
You can use two pelts in series, however they need to be the exact same type. (technically not 100% correct, but they must have the same DC resistance) Hooking two pelts in series has the advantage you can use a PSU with less amp's but a higher voltage. You are also more likely to find a PSU of a high wattage at 30 - 35 Volts.
Not sure if I understand this correctly, but... well..... it doesnt take a 13 year-old to know that putting them in series will make them weaker.....
Try paralell...
if you hook up 2 pelts that drain eg. 10 amps at 12 volts, they use 120 wats. hook them in parallel, and the current will double to 20 amps, set them in series and double the voltage and the current through the two pelts is 10 amps, and they divide the voltage accoring to their resistance. Let's say they are exactly equal, they each get 12 volts (PSU is at 24volts) thus you have the exact same ammount of power at half the current.
Hope this explains my idea.
dimmreaper
08-30-01, 04:04 PM
Originally posted by Oright
Not sure if I understand this correctly, but... well..... it doesnt take a 13 year-old to know that putting them in series will make them weaker.....
Try paralell... Running them in a series at there rated voltage makes them weaker.
However if you run two in a series and double the voltage, the electrical resistence of each peltier will reduce the voltage for the other peltier to the rated voltage, and both peltiers will be running in spec at full power.
I know this is a little late, but you could do what you wanted to
by putting many of the same type of TECs in series. So long as
you stay within the TECs voltage and current specs. However
one danger is the high DC voltage in your system. One problem
is that DC from AC has too much ripple in it, this will make your
string of TECs less efficient. You could filter it, but now the cost
of your PS just increased by a factor of ten. Better to find an old
working computer PS and use that.
I saw the recifier idea. I'd NOT advise that.
Peltier modules need -clean-, -constant-, -consistent- power. They do not like the half-wave result of running AC through a rectifier bridge or diode. It's also been discovered that they don't like PWM either, which is why som companies that make temperature regulatros for Peltier systems have been re-engineering them to always supply -some- power to the Peltier.
If you're powering a Peltier, use clean, well-regulated and smooth DC ONLY.
oO, who runs a water/Peltier cooling system - his Peltier is powered by a 250-watt industrial switching supply.
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.