Peltier Effect

No you are thinking of the "Seebeck effect" where two disimilar metals are held at different temperatures the junction produces a small electric current. This led to the discovery of the "Thermocouple" Thomas Seebeck born 1770 - 1831.

Thermocouple an electric temperature measuring device made by welding two wires made from different metals together at their ends. A current flows in the circuit when the two junctions are maintained at different temperatures. An EMF is produced this is measured by a millivoltmeter and is proportional to the temperature difference.

Yes Guest...

The peltier effect can be used both ways, i.e. supply a current through the device to have a cold side and a hot side (depending on the direction of current flow).

also you can heat one side / cool other side to get a current from the device.

Incidentaly, thermocouples can and do give very high current outputs albiet at a low voltage, they are used as safety cutouts for gas operated equipment to cut off the gas if the pilot light is extinguished.

The pilot light heats the thermocouple and the output is taken through a coil to hold a spring loaded valve open, in effect an electro magnet, the current produced is in excess of 5 amps!!

John.

I used to repair boilers but never got any thing like that.

It was only 10 years or so ago when I was demonstrating a gas flowmeter I had designed that a gas appliance designer showed me how the thermocouples worked in this way!

Quite surprised me as well - I had always thought they operated on some sort of expansion principle, but no the copper sheath is needed so as not to allow much voltage drop - take apart any gas tap and look at how the knob, which you have to push in actuates a safety valve which will only stay in place if there is enough force in the solenoid coil to hold it open.

John.

Electroman,

I believe that your 5A experience was probably with a thermopile, which is made up of multiple thermocouples in series and/or parallel.

Is there a way to measure the electric output of the peltier?

For some experiments You can find at www.conrad.de [international.sales@conrad.de], Peltier elements;for example :Peltier Elament TEC1-3105

nominal voltage 3.7Volt, max current 4.6 A ,dimens. 20x20x3.9 mm

Applying heat to one end of a thermocouple will produce DC current at the other end that can be measured with a standard multimeter.

Guest,

Brainwave and Electroman are both right. Producing a cooling/heating effect from a Peltier device is called the Peltier Effect. Producing electricity from the same device is the Seebeck Effect. Maybe they should have called the thing a Peltier-Seebeck device.

With regard to calculating the output the answer is yes. There are quite a few websites that address your question. Just Google Peltier Effect, Seebeck Effect, Thermoelectric Effect and/or the related Thomson Effect. You will find information ranging from simple to very complex explanations. Here is a good place to start: The Thermoelectric Effect Wikipedia.

Here is the bad news. The efficiency of a Peltier device is very low when producing electricity. I believe it ranges anywhere from 10% to maybe 20% (if your lucky) depending on how well you are able to manage the temperature differential between the heated and cooled surfaces. It is equally inefficient in producing a cooling effect compared to other forms of refrigeration. It takes a lot of DC to remove a little heat. Don't dispair...Things are looking up! There is ongoing R&D that is improving the efficiency of these devices.

Yes. The a deep space probe (Galileo, I think) is powered by the thermoelectric effect- a hot piece of plutonium at one end of the stick and and the other end is just allowed to get real cold.

Thanks every one useful information exchange. I have been working with various TC for last sevaral years as a part of my job. Thinking of making an electrical energy generator using TCs; many in series and many in parallel. Though efficincy is low if it is close to Solar panel in power generation and much cheaper than a solar panel; it will be a great thing; is not?

Having a good source of heat for a Peltier array is only half the battle. You also must have an efficient way to remove the conducted heat. Peltier device efficiency is dependant on a temperature differential between the hot and cold substrates. The more heat you can remove before it migrates back through the junctions the more efficient your thermoelectric generator will be.

What is the difference between parallel and series?

Is one more effective for the peltiers than the other, or is it the same?

We would need an expert here, but I would bet for that for stronger effect it should be on series, if it would not fry them chips. Parallel would probably operate them normally, only draw more juice.

I'm not sure here. You need someone who tried this for fact.

These are not cheap, mind you.

I've played with Peltiers some years back, they are now very cheap, after all most of those small table top / portable fridges and warming 'ovens' use them, they sell for £10 complete!!

Using them in parallel increases the power of heat transfer - similar to having two gas flames heating a vessel as opposed to one.

Using them in series increases the maximum difference in temperature from one side to the other for a given heat power transfer. You can buy them stacked in series operation for cooling CCD devices or other small chips and mirrors for measurement of dew point etc...

They are really useful devices - Buy one and play with it!

John.

Interestingly enough, Peltier devices are both series and parallel devices. If you examine one, you will see that thermal-wise the junctions are in parallel. Since the voltage drop is very small across each, they are wired in series electrically. The heat is transported across the junction by electrons, so the amount of heat moved is proportional to the current. Once there is a thermal difference established, a back-emf is generated (Seebeck effect) that opposes the applied voltage. The biggest inefficiency is due to the conduction of heat back through the device.

Please see my posting
#26

Got it.

Well, I'm more of a reader or a spectator on this thread, than as a writer, so I only offered my common sense, not an expert's advice.

One thing to note here, a Peltier device is not really considered a thermocouple. It is an array of discreet NP junctions sandwiched between two conductive substrates. It is actually a solid state device.

Back in the day (1985 +/-), I was a tech in the USMC. I was involved in testing a thermo-pile based portable power source. It burned kerosene, was about the size of a standard 2000 watt portable generator. The output was 48VDC at around 500 watts. It was pretty quite, blowing hot air out one side. It was well built, like a commercial generator, so it may have been a commercial product. I don't know much more about it or whether or not it was ever used. DISCLAIMER: Allow some fudge factor around the specs, as it was a long time ago.

Your quite correct thermo pile generators have been around for a while.I have an propane gas powered one sitting in my workshop. I think it is a unit that was used in the armed forces for an emergency back up supply

There is a company in Gibraltar that claims to have a very efficient Peltier device. Does anybody know anything about this? Do you suppose these will be cheap enough for hobbiests to play with?

http://www.coolchips.gi/

No I don't think that they will be cheap enough or hobbies, but I have a good idea for my future and that may takes some money. (I am only 17 that doesn't make it easier for me with the money) so, if someone can help me to find a good cheap efficient peltier, then write it :)

Look on ebay you will find peltier devices selling for £5 ($10) each and as they are low weight the postage shouldn't be much.

That's where I bought most of my peltiers from 5 years ago, but then they were about £25 ($50) each!!!

John.

Thanks for very interesting coolchips !

http://www.coolchips.gi/

These devices are to cool hardly can be cooled by conventional heat-sinks & fans.

these are flat, extra-thin devices with 2 wires to be connected to an external Sink [which draws current in opposite to SOURCE which produces Electric-Energy] and energy dissipated.

These devices are not connected in parallel or series; but dealt individually.

Application areas may be CPUs of Laptops/ Notebooks or such tightly packed equipment.

But what is if I want to generate energy with them and there are more then one?

How would I connect them then?

It depends on what you want at the output: if you need a high voltage use them in series.

For high current you need to use them in paralel.

Bigger systems of coarse need to have a combination of both.

Typical elements are a combination of NP and PN elements in series, one element is only capable of generating little voltage differences. In one type (NP) it is the electron that moves with the heat and in another type (PN) it is the hole that moves against the heat flux.

There are many attempts to use this as power supplier, as small thermal differences do exist everywhere and these elements can do something with small differences. giving only little power but in some cases sufficient to power the application.

can you explain again with the NP and PN I didn't understand that well.

Semiconductor junction, NP (Negative to Positive) and PN (Positive to Negative) to indicate electrons flow.

You cannot 'generate energy' with them. You cannot 'generate energy', period.

You can only convert existing energy, and this way, use some part of it.

We know that but it is easier to say it like that!

Actually the device was to dissipate HEAT in a confined-area where you cannot use conventional Sinks & coolers etc. It converts heat to DC potential which is conveyed to an external location & dissipated in a power-resistor [load].

Try if you can generate energy "with them and there are more then one".

Is there an other thing than the peltier, but similar?

(for getting energy)