[KoolGuy]

Im planning on getting a cheap one, attacching a heat sink to the bottom with a fan and placing a drink on it.

But im curious what happens if i over volt it? Will it cool better?

And would it be stable by it self? Basically if i were to not add a heat sink. And not add something to cool would it balance it self?

 

[flak-spammer]

Overvolting a peltier would be akin to over-volting something really sensitive. It will most likely result in a burnt out peltier. Also even a small peltier can still produce a massive amount of heat, since it pumps heat from the cold side to the hot side as well as producing it's own heat. That means you'll need a hefty heatsink for your beverage cooler. No it will not operate without a heatsink, well actually, I'll state that better. It will operate but for a very very very short period of time then it will burn out. So you must cool the hot side. Remember it's a heat pump, it's not magic it works just like your fridge. Except there's no moving parts.

 

[KoolGuy]

Hm... I think i got a pretty big heat sink and im a strap two fan on it then over volt the fans to moar pwr!

But can i over volt it if i can keep it cool enough?

Should i use thermal paste on the hot side?

 

[flak-spammer]

No, don't over-volt it, unless you want premature failure. Definitely use thermal paste on the hot side. Also try and make sure you get some compressing action between the heatsink and whatever you are cooling. We usually put a cold plate, a piece of copper, between the the cold side and the object to be cooled. This is so that we can compress the TEC between the heatsink and the cold plate. They operate most effectively at between 150-300PSI of pressure on them. If you don't use any pressure it will probably result in premature failure.

 

[killabytes]

When it comes down to it, they work fairly poorly. They're really only about 10% efficient at the most.

Why not just buy a pre-fab drink cooler?

 

[flak-spammer]

Quote:

Originally Posted by killabytes When it comes down to it, they work fairly poorly. They're really only about 10% efficient at the most. Why not just buy a pre-fab drink cooler?

Hmm this is a problem. They can quite easily be more than 10% efficient. In the world of heat pumps efficiency is not typically defined in percentages. It's defined in terms of COP. While yes generally TECs are not extremely efficient when run at Vmax. However manufacturers suggest you run them between 75-80% of Vmax which gives a COP of about 1. 1 unit of energy is consumed for each unit of energy moved. TECs are just quite nice because they are so small, and when treated properly effectively last forever...some companies suggest that MTBF could be as high as 200,000 hours. They also have no moving parts so are quite durable.

 

[KoolGuy]

hmm pressure eghh? I dont think any cup i have is heavy enought. Ill work something out. Im sure i can figure something out... What if i put it inbetween my CPU and scythe. Would it help?

 

[Ultrasonic2 (muffy)]

yes use thermal paste.

The question on over volting is hard to answer exactly becuase you haven't given us a link to the TEC you wan to use or told us by now much you want to hver volt it by

However you Can over volt TEC's providing you can keep them cool. however their performance will be WORSE than at Umax at the rated Th( hot side temp ) ( assuming you are keeping the th at the rated value all the time )

I have manufactures charts backing up my statements that you can over volt but you dont want to .

It's also important to remember if your Th is greater than the rated value then Umax will have increased and therefore even though you may be applying more volts than Umax is it may not actually be greater than Umax at the th your running your TEC at

Now if you can understand all of that your doing really well

 

[Ultrasonic2 (muffy)]

Quote:

Originally Posted by killabytes When it comes down to it, they work fairly poorly. They're really only about 10% efficient at the most. Why not just buy a pre-fab drink cooler?

yes TEC and phase change have low efficiency but can have VERY high COP . C.O.P is actually what you mean. Run at very low input voltages a TEC can move many many times more heat than electricity you apply to it

 

[KoolGuy]

i assume the umax is the point of greatest efficiency?

Like i said this is intended to be a experiment to cool my drink / keep it cool while i game..

http://cgi.ebay.com/ws/eBayISAPI.dll...STRK:MEBIDX:IT

LOL i know ebay..

 

[Ultrasonic2 (muffy)]

Quote:

Originally Posted by KoolGuy i assume the umax is the point of greatest efficiency? Like i said this is intended to be a experiment to cool my drink / keep it cool while i game.. http://cgi.ebay.com/ws/eBayISAPI.dll...STRK:MEBIDX:IT LOL i know ebay..

umax is the applied voltage where Qmax is achieved for a rated hot side temp normally 25ish C ( Qmax is heat moved )

you should read my stick all this stuff is covered in there and why you need to apply pressure to a tec and why you need to have a hot and cold plate and what not

 

[killabytes]

Quote:

Originally Posted by flak-spammer Hmm this is a problem. They can quite easily be more than 10% efficient. In the world of heat pumps efficiency is not typically defined in percentages. It's defined in terms of COP. While yes generally TECs are not extremely efficient when run at Vmax. However manufacturers suggest you run them between 75-80% of Vmax which gives a COP of about 1. 1 unit of energy is consumed for each unit of energy moved. TECs are just quite nice because they are so small, and when treated properly effectively last forever...some companies suggest that MTBF could be as high as 200,000 hours. They also have no moving parts so are quite durable.

Those are two totally different devices. I'm stating that if you put for example 100W into a TEC on average you'll get 10W of cooling/heating power. Again those numbers are just examples.

 

[Ultrasonic2 (muffy)]

Quote:

Originally Posted by killabytes Those are two totally different devices. I'm stating that if you put for example 100W into a TEC on average you'll get 10W of cooling/heating power. Again those numbers are just examples.

that is just made up rubbish that is not based on any facts

 

[Ultrasonic2 (muffy)]

i have coved efficiency and COP in my sicky

You originally talked about TEC efficiency but your second post of 100/10 is an example of COP not efficiency

if you bothered to actually go to a manufactures web site they will clearly state while having a low efficiency JUST LIKE phase change a TEC can have a COP far greater than one or EVEN phase change units (when run at the correct input voltages)

efficiency is a measure of input power relative to work done. ( which can never be greater than one )

while COP is a measure of input power in watts V heat moved in watts which can be far greater than one

Efficiency and COP are completely different

 

[KoolGuy]

So do i have a really weak TEC coming? Whats the average price for a good one?

I looked at another with 92 Qmax. Can a air cooler handle that? Would it help reduce temperatures further?

 

[Ultrasonic2 (muffy)]

Quote:

Originally Posted by KoolGuy So do i have a really weak TEC coming? Whats the average price for a good one?

sorry we got side tracked

that TEC has the PMAX listed not It's Qmax it's Qmax will be about 45Qmax

But that is fine if your just cooling a drink .

What voltage do you want to run it at ?

 

[KoolGuy]

Quote:

Originally Posted by Ultrasonic2 (muffy) sorry we got side tracked that TEC has the PMAX listed not It's Qmax it's Qmax will be about 45Qmax But that is fine if your just cooling a drink . What voltage do you want to run it at ?

Well i guess 12v since all i have are PC power supplies...

what about this one

http://cgi.ebay.com/12706-TEC-Thermo...item2a0a023a38

Have more q max (92W)
Its Umax is 12v
and i have a extra pwr supply that can power it NP imax 6A (+ the molex can handle it!)

Your summary really summed things up (Just finished reading) and now i pretty much under stand how ever that makes me want to experiment to.

You think i can still find a way to put this on my CPU?

 

[Ultrasonic2 (muffy)]

Quote:

Originally Posted by KoolGuy Well i guess 12v since all i have are PC power supplies... what about this one http://cgi.ebay.com/12706-TEC-Thermo...item2a0a023a38 Have more q max (92W) Its Umax is 12v and i have a extra pwr supply that can power it NP imax 6A (+ the molex can handle it!) Your summary really summed things up (Just finished reading) and now i pretty much under stand how ever that makes me want to experiment to. You think i can still find a way to put this on my CPU?

you OP said you wanted to over volt it and you want it for a drinks cooler

but you wont be over volting it if your running it at 12v and ther is no way in the world such a low wattage TEC could cool a cpu on it's own. Norally if you want to call a load with any kind of desent COP you would want at least twice the Qmax of the load your trying to cool then you run the TEC's at a low ish input voltage

O and im glad my sticky helped

and the TEC you just posted DOES NOT have a Qmax of 92 it has a Pmax of 92

 

[KoolGuy]

Quote:

Originally Posted by Ultrasonic2 (muffy) you OP said you wanted to over volt it and you want it for a drinks cooler but you wont be over volting it if your running it at 12v and ther is no way in the world such a low wattage TEC could cool a cpu on it's own. Norally if you want to call a load with any kind of desent COP you would want at least twice the Qmax of the load your trying to cool then you run the TEC's at a low ish input voltage O and im glad my sticky helped

Yea i did change my mind. When i started reading your sig LOL..

So you saying that even if i did some kinda CPU|Pelter|Heatsink|fan order it wont cut it?

If thats the case back to the drink LOL..

Since the u max is already 12v i see no reason to OV it so ill keep it there.

Now will 92W cool a drink?

 

[Ultrasonic2 (muffy)]

Those two TEC's wont beable to cool a modern CPU but a more powerful one(s) can