Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Peltier Module help

Status
Not open for further replies.

fred666777

New Member
im trying to make a small refrigerator i am planning on getting 2 of these
would it be better to stack the two on one side

(FAN/HEAT SINK)
Cold
=======
Hot
Cold
=======
Hot
(FAN/HEAT SINK)

or put them on opposite sides of the box?
and what kind of power supply would i need for this?
would something as simple as this work?
 
im trying to make a small refrigerator i am planning on getting 2 of these
would it be better to stack the two on one side

(FAN/HEAT SINK)
Cold
=======
Hot
Cold
=======
Hot
(FAN/HEAT SINK)

or put them on opposite sides of the box?
and what kind of power supply would i need for this?
would something as simple as this work?

Those power supplies wot work
unless modified to have the control loop on 12V instead of the main3.3V
as it is, they would work for 12V only if you load the 3.3V up to half the defined load at least. The issue is common with any of the multi- output supplies.
 
Peltier elements are built for optimum temperature difference between plates and depending on the heating/cooling power they require different currents and voltages.

Two examples: element size 30X30X4.7mm Umax=8.6V, Imax=3.9A; element size 40X40X3.9mm Umax=15.9V, Imax=6.1A

With these parameters they produce a Δtmax of 67deg/C.

Boncuk
 
One issue is the efficiency of a peltier cooler. As cooling devices they are not as nearly as efficient as the regular phase change mechanical systems. A small refrigerator is going to use more power than a large home unit.
Assuming that the ones you have chosen have the more typical 35% efficiency when cooling you will get about 34 BTU of cooling capacity. A typical kitchen refrigerator has around a 600 - 900+ BTU cooling capacity.

A BTU is the energy of one pound of water changed one degree F.
 
will this **broken link removed** work for this **broken link removed**?
and if i were to get 2 peltiers would i need multiple/more powerful power supply?
 
Last edited:
The Peltier element has 120W of power. You might operate it with a supply delivering 7.5A, but you must take care to limit the current to 7.5A (120W require 10A at 12V) not to fry the power supply.

Limiting the current produces heat!

Furthermore it reduces cooling power of the Peltier considerably.

Boncuk
 
Peltiers are really not all that fussy about power quality. As long as you stay under their maximum voltage ratings they will work on about any DC power source.
Even a cheap battery charger will work just fine for powering them.
 
I'm afraid it won't work well using a 12V/7.5A power supply. The 120W peltier will try to achieve it's rated current and overload the power supply.

If nothing is done to limit the current my guess is a burned out power supply.
 
so should i look for a supply with 10A at 12V?

will 10A be too much for the peltier?

why does it say 80.3W if its a 120W Peltier?

Voltage: 12V
Couples: 127
Umax (V): 15.2
Imax (A): 8
ΔTmax(℃): 67
(W):80.3 W
Resistance ( Ω): 1.5 ~ 1.6
Dimensions: 40x 40x 3.5(mm )
 
15.2V * 8A = 121.6 W

12v / 1.6Ω = 7.5 A

7.5A * 12V = 90 W


They likely rounded off a bit on the numbers.
 
Last edited:
holy moses....

peltiers are not a common thing on here are they ?

Not much said so far is quite right...

Small peltier refridgerators often only contain one small peltier unit and a very large heatsink to disperse the heat given off by the hotside. Small peltiers like the ones usually used in fridges can generally be chilled "passive" with just a heatsink but any like the ones you have chosen will need a fan too. Any small peltier will benefit from a fan. Generally peltiers with a Qmax over 100w need water cooling on the hotside. Peltier fridges are so cheap generally it barely seems worth the effort to make one. unless you understand peltiers and how they work it will probably be more effort than you imagine and if not built correctly will be a waste of time.

Note
Umax is maximum voltage not necessarily what you need to run it at.
Imax is the maximum current draw if you were to power it 100%.
The wattage of 80.3w is probably quoted as Qmax or Qcmax and it is the maximum wattage moved at maximum power when the delta T is at 0ºC. (i.e. no cooling .)
The delta T max of 67ºC is the maximum difference in temp between the hot and coldsides of the peltier you will rarely obtain this in real life because of thermal heat transfer inefficiencies. the delta T is affected by numerous specs but in the main it goes down with power input. The coldside temp is obtained by taking the delta T from the hotside temp.

The resistance you can ignore it is not necessary in calcs. Peltiers are semiconductors and do not rigidly follow Ohm's Law. It is much better to use a manufacturers performance chart.

Peltiers are horribly inefficient at more than 75% power because at 75% power they are generally cooling to 85% of the max. they can achieve.
The reason why 127 couple peltiers (15.2v.) are so common place is because they can be nicely run with a 12v supply which is roughly 75% (near as, actually 78% but who's argueing.) Obviously you will not get 80.3w cooling more like 70w but that is only assuming you use excellent thermal interface material and correctly torque the peltiers. They need a torque of between 150- 300 p.s.i. At 12v the peltier you have chosen will probably draw somewhere in the region of 6 amps probably slightly over.

What ever the cooling power at the operating point you have chosen you will have roughly double that wattage of heat coming off the hotside and you will need to disperse this for efficient operation. So if you have 70w cooling you will 140w coming off the hotside.

Stacking the peltiers will not work as you hope it will not increase the cooling.
Peltiers are merely electrical heatpumps pumping heat from one side of the peltier to the other. No matter how many peltiers you stack up no more heat can be transferred than that which can be transferred by the peltier at the bottom.

Undervolted to less than 50% power pelteirs become very efficient average units can easily obtain a COP of 5 (better than phase change.) but obviously they have a low wattage of little use. So you use large peltiers and more than one of them. Chillers built up like this are very efficient and can be built for virtually any power, within reason. Where phase change champions is that at the same COP as a TEC the phase change will attain a much lower temp.

Virtually any peltier with very little load (as in a refridgerator.)and over 50% power will attain a coldside temperature approaching or below 0ºC assuming a room ambient in the low 20's.
 
Last edited:
Nice! You explained more in one post Than I have been able to decipher in about 6 months of reading up on them.

I understand the stacking cooling limits well enough but as for stacking them to achieve greater temperature differentials that still works right? That is assuming the layer above the first one has sufficient pumping capacity to make up for the inefficiencies of the first stage.

Any tips on the practicalities of running them in reverse as thermal to electric generators other than dont exceed the thermal limits of the internal solder junctions on the hot side?
 
Nice! You explained more in one post Than I have been able to decipher in about 6 months of reading up on them.

Yes they can be a bit daunting, the only real way to learn about them is to get one and play with it. A useful bit of side reading is this free to download handbook from the manufacturers, Melcor. :- **broken link removed**

I understand the stacking cooling limits well enough but as for stacking them to achieve greater temperature differentials that still works right? That is assuming the layer above the first one has sufficient pumping capacity to make up for the inefficiencies of the first stage.

Yes stacking just increases the delta T max. You get about roughly 8ºC per layer added to the delta T max. Each extra layer has to be able to transfer all the heat transfered by the lower layers plus all the heat generated internally so the upper layers get big quickly...but at the end of the day the whole stack can only transfer the heat equivelent to just the bottom peltier.
Manu's rarely stack peltiers past 4 layers which gets a delta T in the region of 100ºC but they are rarely past about 30w Qmax.

Any tips on the practicalities of running them in reverse as thermal to electric generators other than dont exceed the thermal limits of the internal solder junctions on the hot side?

There is a bit of ambiguity here...

You mean power them in reverse - which would just swap over the cold and hotsides. Although I think perhaps you understand this...

If you keep one side cold and the other hot the unit will produce a low voltage with a very small current from the attached wires. The polarity will depend on which side you keep hot etc - the usual wire colourings of red and black are pretty meaningless on peltiers except as a method to identify one wire from the other.
Whilst they are essentially the same units as those for cooling peltiers for generation are tweaked in manufacture to help them work and they are often refered to as TEG - ThermoElectricGenerators. this distinguishes them from those used mainly to cool - TEC ThermoElectricCoolers. This is what I normally call them but I stuck to "peltiers" for this thread to avoid confusing anybody.
The thermal limits of the solder in peltiers is usually around 160ºC unless you have high temp modules. TEGs are usually high temp spec. The key to generation is maximising delta t as opposed to having a high temp -not surprisingly it is also the key for efficient TEC operation.
 
Last edited:
I played around with one of my 168 watt Peltier coolers today and heres what I got while running it for about 2 hours at 11 volts 7 amps input using a big aluminum heat sink on the hot side and an old CPU heat sink on the cold side. The other Peltier cooler and the quarter are for size reference. These coolers are approximately the same size as a standard CPU.

My IR thermometer says the hot side is about 83 F and the cold side is about 22 F. No heat sink compound or clamping was used.

DCP01889.JPG

I dont know about you but that just seems to say mini beer cooler to me! :D
 
I played around with one of my 168 watt Peltier coolers today and heres what I got while running it for about 2 hours at 11 volts 7 amps input using a big aluminum heat sink on the hot side and an old CPU heat sink on the cold side. The other Peltier cooler and the quarter are for size reference. These coolers are approximately the same size as a standard CPU.

My IR thermometer says the hot side is about 83 F and the cold side is about 22 F. No heat sink compound or clamping was used.

View attachment 36130

I dont know about you but that just seems to say mini beer cooler to me! :D

To start I think whoever sold you the TEC appear to of quoted the specs at hotside temp of 50ºC.
This increases the voltage and current rating and also the Qmax, i.e. makes it look better... Then they went and quoted the input power instead of the Qmax. It is the Qmax you really want.
I tend to work with hotside 27ºC specs. At 27ºC Bismuth telleride couples have a voltage of 0.1227v so a 127 couple unit like yours, at hotside temp of 27ºC has a voltage of 127 x 0.1227v = 15.6v this is the true voltage of the unit (people quote variously 15.2 - 15.8 ) and your current will be 10amps (it's a standard value.) so your unit is a 12710. 12710 have a standard Qmax of around 90w (it varies a bit depends who makes it.)
So to me you a 12710 90w unit.

The reason for clamping is that in normal use TECs vibrate. This vibration is sufficient to affect the heat transfer also eventually after long continued use it can break down the solder joins. TECs will work happily without clamping but if you do clamp you will get a better performance from better heat transfer and a longer life for the TEC. Also the use of a quality TIM will make quite a difference.

Your difference of 60ºF while it might seem impressive is nothing for 11v in for a 15.2v unit, it is only 15ºC, in terms of performance it is piss poor. Perhaps your heatsinks don't help but with clamping, TIM and voltage tweaking you should be on at least 25-30ºC difference (77 - 85ºF) or even more, after all you very little load (it is under load even in free air.)

I got 28ºF from an 80w Qmax. (15.2v) unit running at 4v !! I had a vapochill micro on the hotside, silver TIM, a 10mm hotplate and a 5mm coldplate and it was running in free air same as yours with an ambient of 20ºC (68ºF it was outside.) And I thought that was not very good !!!
 
Last edited:
well like I said. I didnt clamp it or use heatsink compound it. Its just sitting there as is.
 
So what can be expected from a peltier like this being used as a TEG with a 75 C temperature differential between the sides?
 
So what can be expected from a peltier like this being used as a TEG with a 75 C temperature differential between the sides?

Generally you wont get a peltier with a DTmax that high...in the main it is 69-71ºC.
I know some Kryotherm peltiers go higher their highest rated is 74ºC.

Secondly I can't answer your question because I have never had any interest in peltiers for generation partly because they can't generate very much but mainly because I am only interested in cooling. I have never attempted to use a peltier for generation, I don't consider it worthwhile for any application I may have.

You have a peltier try it and see.....but I am not interested.

I know some people used to think you could use a big TEG to power a small TEC but you can't it is way off, no where near enough current. you could parallel them up I suppose but I really don't think it's worth it.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top