[Dr Graham]

Dr Graham asked the Naked Scientists:
   
Is there a device which can convert heat energy into electricity ?

Photo cells can do it with light, Batteries do it chemically, and there are a variety of ways to covert mechanical energy into electricity, so why not heat energy ?

What do you think?

[CliffordK]

Certainly a variety of types of steam engines and turbines use heat to produce electricity. 

Photo cells can be tuned to some extent to pick up certain wavelengths of light such as would be emitted by objects on earth.

I'm seeing a lot of notes of an ENECO semiconductor device that is supposed to convert heat directly into electricity.

http://web.mit.edu/newsoffice/2001/electricity-1205.html
http://pesn.com/Radio/Free_Energy_Now/shows/2007/01/06/9700221_Eneco_thermal_electric/

It appears to be about a decade old technology, based on older vacuum tube technology.  But, it hasn't gotten to commercial production yet.  Time will tell if it is eventually developed into a viable commercial technology, or if it just fizzles.

Keep in mind that your typical silicon solar cell is more efficient at cooler temperatures.

[MikeS]

As CliffordK said steam engines and turbines. Also petrol and diesel engines use fuel to produce heat to do work.
A thermocouple directly converts heat into electricity but not a lot.

[imatfaal]

As CliffordK said steam engines and turbines. Also petrol and diesel engines use fuel to produce heat to do work.
A thermocouple directly converts heat into electricity but not a lot.

I never really understood electricity etc - but I thought a thermocouple worked on temperature differences between two conductors.  So even when acting as a thermopile that it was the difference in temperature and not the presence of heat that caused a current to flow

[MikeS]

imatfaal

"A thermocouple is a device consisting of two different conductors (usually metal alloys) that produce a voltage proportional to a temperature difference between either end of the pair of conductors. Thermocouples are a widely used type of temperature sensor for measurement and control[1] and can also be used to convert a heat gradient into electricity."
http://en.wikipedia.org/wiki/Thermocouple

Thermocouples are obviously not very efficient at converting heat into electricity.  I mentioned them only because they directly convert heat into electricity.

[Geezer]

These guys seem to think it can

http://wattsupwiththat.com/2011/06/22/new-multiferroic-alloy-creates-electricity-from-waste-heat/

[CliffordK]

How hot does this have to be?

Think of an air conditioning unit that would actually generate electricity...   []  Or, at least use an AC heat pump to limit the energy cost.

If you could cheaply capture thermal energy down to about 0°C, you could likely have an end to global warming.  In fact, there would be the risk of it being used to excess...  and suddenly creating the next ice age.

[Geezer]

How hot does this have to be?

Think of an air conditioning unit that would actually generate electricity...   []  Or, at least use an AC heat pump to limit the energy cost.

If you could cheaply capture thermal energy down to about 0°C, you could likely have an end to global warming.  In fact, there would be the risk of it being used to excess...  and suddenly creating the next ice age.

It's still a heat engine. If we are to believe the French geezer by the name of Carnot, its efficiency is limited by the temperature difference between its hot end and its cold end.

[techmind]

Energy can be extracted from the flow of heat from a hot place to a colder place. Steam engines/turbines are one example of this.

Seebeck-effect thermoelectric devices convert the heat flow directly into electricity, and for the past 50 years have generally been based on bismuth telluride technology (They're mostly made in Russia, although China has been getting in on the act recently). As with all heat-engines, the efficiency gets better the greater the thermal gradient - but for practical Seebeck devices and realistic/compatible temperatue differences) the efficiency is really rather poor (maybe 8% at most).
It's not much use for "energy saving" devices, but can develop enough power to run remote telemetry devices off of hot pipes, or deep-space-probes containing a radioactive source (which keeps warm). I did hear of a scheme to use thermoelectric devices to harvest heat from the catalyser in a car to help power the car electrics. No idea whether it was ever going ot be commercially viable though.

For more info see http://en.wikipedia.org/wiki/Thermoelectric_effect

[Geezer]

Energy can be extracted from the flow of heat from a hot place to a colder place. Steam engines/turbines are one example of this.

Seebeck-effect thermoelectric devices convert the heat flow directly into electricity, and for the past 50 years have generally been based on bismuth telluride technology (They're mostly made in Russia, although China has been getting in on the act recently). As with all heat-engines, the efficiency gets better the greater the thermal gradient - but for practical Seebeck devices and realistic/compatible temperatue differences) the efficiency is really rather poor (maybe 8% at most).
It's not much use for "energy saving" devices, but can develop enough power to run remote telemetry devices off of hot pipes, or deep-space-probes containing a radioactive source (which keeps warm). I did hear of a scheme to use thermoelectric devices to harvest heat from the catalyser in a car to help power the car electrics. No idea whether it was ever going ot be commercially viable though.

For more info see http://en.wikipedia.org/wiki/Thermoelectric_effect

Of course, you could always use a thermoelectric device to drive a heatpump to increase the thermal gradient to increase the efficiency of the thermoelectric device. What could possibly go wrong?  []

[Escorpiuser]

(...)
Of course, you could always use a thermoelectric device to drive a heatpump to increase the thermal gradient to increase the efficiency of the thermoelectric device. What could possibly go wrong?  []

Due to inefficiencies in the processes and imperfect adiabatic isolation, eventually one of the sides will reverse cold or hot, ending the thermal gradient and the processes based on it. When you are sending the energy forth and back continuously, what you get is accelerate the end.

Perpetual movement is not viable... so far  [].

[Escorpiuser]

Going to the point, Stirling motors seems to be the most efficient mechanical devices to convert heat into movement. After that, you have to attach an efficient electric generator to convert the movement into electricity. Unfortunately, Stirling motors don't produce many HP.

[peppercorn]

Stirling motors seem to be the most efficient mechanical devices to convert heat into movement. ... Unfortunately, Stirling motors don't produce many HP.

Isn't that an oxymoron?


More correctly, Stirling engines have a relatively low power density when compared with more practical (and therefore common) heat engines.

Further, multi- or combined- cycle systems will outperform a Stirling for almost every application in terms of efficiency and kW/Kg.


And, as far as a self-powered vehicle is concerned, power density (both kW/Kg and kW/L) starts to majorly trump outright efficiency in terms of realising a real-world full-cycle efficiency.     ie. you've got to carry the engine around!

[Escorpiuser]

(...)
More correctly, Stirling engines have a relatively low power density when compared with more practical (and therefore common) heat engines.
(...)

Thank you for the correction. My explanation wasn't accurate enough.

I just wanted to contribute to the thread because I was surprised that nobody mentioned the Stirling motor.

On the other hand, a Stirling motor could be practical if you want to produce some watts out of a wood stove as in, let's say, a mountain refuge.

Also, I have read (don't remember where) that a big company is developing a Stirling motor for, I believe, more ecological vehicles. But sorry, I can't show the link to the news right now.

[peppercorn]

Escorpiuser

If memory serves me well, one of the U.S. 'big-three' played with a Stirling-powered vehicle following the '70s Oil embargo.
It was of some success in terms of efficiency gains for the time I have read, but remained woefully underpowered and was dropped.

After I replied I did suspect that you were well aware of Stirling's limitations, so I apologise if the clarification was unnecessary (hopefully useful to other readers perhaps).   IMO Stirling's will never overcome their low kW/Kg in any configuration (hybrid or whatever) to work on a road vehicle - even a large truck.

.... anyone care to prove me wrong? []

[Geezer]

.... anyone care to prove me wrong? []

If you get stuck into that tool I told you about you prove youreself right [^]

[Escorpiuser]

(...)IMO Stirling's will never overcome their low kW/Kg in any configuration (hybrid or whatever) to work on a road vehicle - even a large truck.

.... anyone care to prove me wrong? []

Probably you're right BUT...
...materials are getting better everyday. I'm thinking in materials with good strength/weigh ratio like carbon fiber, nanoparticles and so on. No one can predict what the future will bring us, except if there is some theoretical demonstration that something can not be overrun, as light speed in vacuum, etc.

When was asked about mankind ability to fly sometime, "Heavier than air flying machines are impossible" answered Lord Kelvin. Well, he was a great scientist, but it seems clear that there he put his leg on it. Also:
"Aerial flight is one of that class of problems with which man will never be able to cope". Simon Newcomb, American astronomer (1903).

(There are many of this kind...).

[Geezer]

One of the apparent advantages of the Stirling cycle is its very high thermal efficiency. However, that is based on the assumption that the expansion and compressions processes are isothermal. I was looking at the Wiki page on the subject and it seems the expansion and contraction are a lot more adiabatic (isentropic) than the typical model portrays.

If that's true, the thermal efficiency won't be much different from any other heat engine, so where's the advantage?

[peppercorn]

If you get stuck into that tool I told you about you prove youreself right [^]

I assume you're talking about that thermodynamics s/w tool - think I let the registration lapse on that sadly.

No one can predict what the future will bring us, except if there is some theoretical demonstration that something can not be overrun, as light speed in vacuum, etc.

... funny you should mention that - the speed of light thing - considering the oddities at CERN at the moment.
Seriously though, although I take onboard what you are saying, I think there is somewhat of a difference between the many and varied examples of the type of famous (or infamous) quotes of great minds that you are alluding to and the limits here - for a device like a Stirling engine - That is there are actual unavoidable upper limits to a Stirling's compactness w.r.t. HP available.  So that's not to say that it can't be improved on, but, for instance, the fact that it is (for one thing) an External combustion engine almost certainly makes it by nature, well, Big!

One of the apparent advantages of the Stirling cycle is its very high thermal efficiency. However, that is based on the assumption that the expansion and compressions processes are isothermal. I was looking at the Wiki page on the subject and it seems the expansion and contraction are a lot more adiabatic (isentropic) than the typical model portrays.

Surely there will always be a realistic limit simply as we are forced to accept a finite heat transfer index from the heat reservoirs (between hot and cold and the regenerator).
Or is there a more fundamental oversight that is also in play here?

[CliffordK]

As far as performance maximums.

There was an earlier discussion that a resistance heater is essentially 100% efficient.

According to Wikipedia,, they introduce a coefficient of performance indicating that if a resistance heater has a COP of 1, a Heat Pump has a COP of 3 to 5, or about a 300% to 500% efficiency rating which surely isn't too bad.  Hopefully nobody has said that one can't extract more than 100% thermal energy from electricity.  Obviously it is gaining thermal energy from ambient temperatures...  but....

Even with the relatively low efficiency ratings of many generators, I'm surprised that nobody is combining an air/air heat pump with a very efficient thermal-electric generator.  Or, perhaps using air as a heat source in the summer, and water/earth as a heat source in the winter.

Perhaps the future will bring a thermal-electric generator coupled with a heat-pump as the perfect method to generate power while combating global warming, perhaps even combating the urban heat islands.  Overall efficiency may be low, but if input is ZERO, it couldn't be bad.