Maker Pro
Maker Pro

Trying to TEG cooler to reach 5w

Nate1368

May 7, 2017
3
Joined
May 7, 2017
Messages
3
I’m trying to get a TEG cooler (Peltier module) to generate approximately 5volts, 1amp, 5 watts, currently, at around 60oC I get around 0.9mV, and 0.8w how should I go about achieving this. I have an SP1848 TEG cooler with specs of 23.6w, 9.24A at 50oC but currently, it is not achieving this. This then was connected with a USB DC-DC step up the module and it is only stepping up to around 1W, 1V. How would I go correcting this and what would the recommendation to achieve 5 volts 1amp 5 watts?
 

davenn

Moderator
Sep 5, 2009
14,262
Joined
Sep 5, 2009
Messages
14,262
I have an SP1848 TEG cooler with specs of 23.6w, 9.24A at 50oC but currently,

those are substantially higher ratings than the ones I found for that part number

at 60C, 2.4V and 469mA

link to the datasheet you got your info from please


Dave
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,880
Joined
Jun 21, 2012
Messages
4,880
I have an SP1848 TEG cooler with specs of 23.6w, 9.24A at 50oC but currently, it is not achieving this.
Surely you don't expect this device to produce 23.6 watts of power at 9.24 A when the hot side is 50C warmer than the cold side?

The specs you provide, which are also included on the datasheet, are the performance you can expect when the device is operated as a heat pump, moving heat from the cold side to the hot side, The device is not very efficient at generating electricity when operated in reverse, where you heat the hot side and cool the cold side and attempt to extract electrical power from the two wire leads. You might want to consider using a Thermopile instead.
 

davenn

Moderator
Sep 5, 2009
14,262
Joined
Sep 5, 2009
Messages
14,262
Surely you don't expect this device to produce 23.6 watts of power at 9.24 A when the hot side is 50C warmer than the cold side?

That was my inference as well ... hence why I asked for his datasheet to see the variation from the data I found


OK confirming my initial thought, that you don't understand peltier coolers .... this unit is a cooler
it uses a lot of power to generate a cool side for cooling down whatever you want ... electronics, beer etc etc

they are not a power generator .... there may be some variation of peltiers that do generate power, but this isn't one of them


Dave
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,880
Joined
Jun 21, 2012
Messages
4,880
there may be some variation of peltiers that do generate power, but this isn't one of them
AFAIK, all Peltier diode stacks do generate power, but not very efficiently. Their biggest deficiency, other than the expense, is a very limited range of temperatures through which they can operate. For more information please visit some of the links on this Google results page concerning Peltier generator efficiency.

A Thermopile isn't very efficient at converting a temperature difference into useful electrical power either, but they are simple and inexpensive to produce. More important, they can operate over a wide range of temperature between the hot junctions and the cold junctions. Hundreds of thermo-couples can be connected in series, with the hot junctions pointed inward and the cold junctions pointed outward, to obtain practical voltages at useful currents.

Try using iron wire (such as florist wire) twisted together with bare copper wire to form a thermo-couple (TC). You can then arrange series-connected TCs as a multi-pointed "star" with the outer, cold TC junctions being clamped to a cylindrical heat sink between thin mica insulation to prevent shorting. You might want to stack several of these to form a kind of "chimney" to conduct heat up through the hot TC junctions pointing inward.

Iron and constantan might be better materials for TC junctions compared to iron and copper because the two metals are easily welded to form a TC bead on the end. The two materials form a J-type thermo-couple and this type of TC wire is readily available. It requires about a hundred series-connected J-type TC junctions to produce a one volt potential if the cold junctions are held at 80F and hot junctions are heated to 411F. Use this table to find the Thermopile potential for other cold-junction/hot-junction temperatures.
 

davenn

Moderator
Sep 5, 2009
14,262
Joined
Sep 5, 2009
Messages
14,262
AFAIK, all Peltier diode stacks do generate power, but not very efficiently. Their biggest deficiency, other than the expense, is a very limited range of temperatures through which they can operate. For more information please visit some of the links on this Google results page concerning Peltier generator efficiency.

OK

but the figures the OP is quoting are NOT for generation rather they are stating power requirements to do specific amounts of cooling, aye
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,880
Joined
Jun 21, 2012
Messages
4,880
the figures the OP is quoting are NOT for generation
Yessiree. The OP is FOS for thinking the power specs that apply to using the Peltier device as a cooler (heat pump) apply in any way to the power produced when the Peltier device is used as a TEG or Thermo-Electric Generator.

Unfortunately, most folks "experimenting" with alternative methods of electrical power generation have zero experience and training in thermodynamics, and don't have the foggiest idea what Carnot efficiency means, or how it limits the maximum efficiency of any heat engine, which a Peltier TEG certainly is. Every Peltier TEG I have ever seen doesn't even come close to the maximum efficiency possible for the temperature difference between the hot and the cold surfaces of the device. And, because of the materials used in a typical Peltier device, the operating temperature difference between hot and cold surfaces, as well as the absolute temperature of the device, are both quite limited. The temperature difference directly affects the conversion efficiency of heat energy to electrical energy.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,880
Joined
Jun 21, 2012
Messages
4,880
Last edited:
Top