Temperature Amplifier: Ideas & Input Requested

[zytrahus]

Hello guys,

Introduction: I would like to get some input about some kind of temperature amplifier.

What I have, is:
- an electrical heater (basically a resistor that is heating a square piece (20x20 mm, 50-100 W) of copper). It's insulated on 5 sides of the copper cube. Only one side is actually free to be connected to ... anything.
- I am measuring the temperature with a thermocouple in the center of this copper side: T0.

What I want:
- I want to keep T0 lower than a certain temperature, let's say: T0max = 60 C
- I want to put a device on top of this heater. The input of the device is mechanically attached to the heater. The output has its own copper surface with a thermocouple in the center, T1.
- I want to have T1 > T0. Basically I would like to reach something like 200 C.
- on top of the device, I will put a cooling device to reject heat to the ambient air.

the device should not use any other energy input than the heat that is coming out of the heater.I thought of different things that will not work (I was thinking of something like this: http://www.imeko.org/publications/tc12-2004/PTC12-2004-PL-003.pdf )...

anyways I would like some input about this. :)

 

[Integral]

Sorry, if I am reading your description correctly, what you are asking is impossible. You will never be able to get T₁ > T₀.

You may want to read up on the laws of Thermodynamics.

 

[cesiumfrog]

What you describe is impossible: heat will not flow from cold to hot.

Something similar should be possible. Thermocouples and such can be used to generate power from the temperature gradient between the heater and the (near-ambient) heatsink, then the (low entropy, electrical) power can be used to heat a sufficiently small and well insulated copper surface to arbitrarily high temperatures.

The trick is that the device must let heat flow around (not through nor from) the "amplified temperature" surface.

 

[zytrahus]

right, but what is actually happening in the experiment with coupled heat pipes?

 

[cesiumfrog]

Which experiment, and what "coupled heat pipes"?

 

[zytrahus]

second part of the abstract here: http://www.imeko.org/publications/tc12-2004/PTC12-2004-PL-003.pdf
I think I just don't get it, maybe you can help me

 

[cesiumfrog]

I think it's completely unrelated to your original question.

They've just noted that boiling points of any fluid will vary with pressure. Basically, it's an apparatus to maintain separate boiling flasks of different fluids, all at the same pressure, because the temperature of anyone flask can then be determined from the temperature of any other flask. (It's neat because it effectively let's them amplify variations in temperature so as to produce a more precise thermometer, but irrelevant because each flask individually requires heating and cooling elements.)

 

[zytrahus]

Yes I understood that but I am basically wondering what would happen if you are cooling the first condenser with the second flask evaporator. then you cool down the second condenser with a heat sink.

the way I understood this paper was that the second heat pipe fluid would boil at a higher temperature; now if you have both heat pipes apart, and at one moment you put them together the way I described above: what happen is it's just going to stop working. no?

 

[zytrahus]

if you take this backward:
you use a fluid with a higher boiling point in HP#1 than in HP#2.

then you always have T1>T2, so heat can keep moving. so you can connect Condenser#1 to Evaporator#2 and still have the system working.

 

[cesiumfrog]

Yes I understood that but I am basically wondering what would happen if you are cooling the first condenser with the second flask evaporator. then [..]

So, for that to happen, the second flask evaporator must be cooler than the first condenser (and hence cooler than the first evaporator also). You've deliberately made the second flask much colder just now, contrary to the design (that specifies externally supplying far more heat). Consequently, the second flask will not be boiling.

Sure, we can use a hot boiling oil to make warm boiling water, but it isn't so easy for warm boiling water to make hot boiling oil. Take Integral's advice regarding Thermodynamics; I've given a solution to your puzzle above, and come to understand a new device, but let's not spend too long individually debunking such similar variations of perpetuum mobile.

 

[MrXow]

What you describe is impossible: heat will not flow from cold to hot.

Not impossible. Heat is the energy of particles and basically their speed. They move randomly. There is no law of thermodynamics that says heat flows from hot to cold. It is just assumed that this should happen and for large systems has not been observed not to.

 

[cesiumfrog]

It is statistically impossible.

Not impossible. [..] There is no law of thermodynamics that says [..]

I'm sorry that you haven't yet learned the 2nd law of thermodynamics. (Specifically, the heat formulation due to Clausius says exactly that.)

 

[MrXow]

sorry if you don't like things quoted from wikipedia, but
The second law of thermodynamics is an expression of the universal law of increasing entropy, stating that the entropy of an isolated system which is not in equilibrium will tend to increase over time, approaching a maximum value at equilibrium.

will tend to increase over time

 

[MrXow]

Oh you and I know that the chances of it happening are like 10^-54 but that is far different from 0

 

[cesiumfrog]

sorry if you don't like things quoted from wikipedia, but [..]

Perhaps you could manage to read a whole four paragraphs down, instead of presuming you knew everything just because you found a word you like ("tend") in the first sentence of the article.

How old are you? You've made a scattering of recent posts to different threads, in which you briefly state something completely irrelevant and misleading while including a grain of truth to protect from quick refutation. Are you trolling deliberately or obliviously?

 

[MrXow]

Even if it does say something else a whole four paragraphs down I am still right. Temperature is just the movement of particles. There is no law that dictates that the particles have to diffuse to lower concentrations. It just happens practically all the time.

 

[cesiumfrog]

There is a law that says that, and that law is called "the second law of thermodynamics".

It is also true that the law is only valid for macroscopic systems, such as the systems which are the intended topic of this thread. For microscopic systems the more general "fluctuation theorem" should be used instead, but it's a mistake to assume we don't already know that. Note the OP wasn't asking about a machine that has less than a one-in-a-billion chance of amplifying temperature for one second out of the age of the universe; the OP was asking for a machine that can reasonably be expected to amplify temperature for every second of a day or year.

Hence the exception which you are emphasising is impertinent.

 

[MrXow]

much much less than one in a billion first of all
but i was responding to what you said about it being impossible
I am emphasizing your fault in saying impossible

http://en.wikipedia.org/wiki/Maxwell's_demon

 

[russ_watters]

Let's back up a little...

Sorry, if I am reading your description correctly, what you are asking is impossible. You will never be able to get T₁ > T₀.

You may want to read up on the laws of Thermodynamics.

What you describe is impossible: heat will not flow from cold to hot.

Heat will not spontaneously flow from cold to hot, but the OP said he wants to use a "device" to cause heat to flow from cold to hot, and you most certainly can force heat to go from a cold area to a hot area with a "device". Most people own such devices. Our houses and cars are equipped with them...

What the OP did not specify is if he wants an active or passive "device" or how much energy should flow from cold to hot (he's asking for a temperature, not a heat flow rate and he doesn't specify if there is anywhere else energy is being provided or rejected). An air conditioner is a "device" that uses input electrical power to run a pump that pushes heat (in a working fluid) from a cold area to a hot area. A heat pipe uses the temperature difference to help move the heat energy around. It is passive (it requires no energy input other than the heat it is moving), and harnesses heat energy that is already moving from warm to cold and just makes it move where it wants to go, but in a different way. Ie, a heat pipe is a passive heat pump that pushes heat where it already wants to go.

Now, it appears that the device referenced in the OP is designed to create a third temperature (not described in the OP, but shown in the reference), one different from the source and sink and, in fact, lower than the temperature of the source. Now I know the first instinct would be that such a thing is impossible, but it isn't. The net energy is still moving from warm to cold, but part of the energy gets sent somwhere else. Call it a "thermodynamic advantage" - it works the same as a mechanical advantage. A Large weight can be used to lift a smaller weight above the large weight's original height. A hydroelectric plant can be used to pump a smaller amount of water above the level of the reservoir. Heck, a sterling engine could be used to power an air conditioner. Now there is no net gain in power this way, but it should be clear that there are cases where a mechanical or thermodynamic advantage could be useful.

It appears that someone has found a way to passively generate a "thermodynamic advantage" using heat pipes.

So. All that said, what exactly is your question, zytrahus? It seems you have found a device that can do what you are asking. Does it not meet your needs? What are your needs (constraints)? From the description in your opening post, it sounds like a normal powered liquid chiller should be acceptable. Are you requiring that it be passive? Or that it use only thermodynamic energy (no mechanical or electrical)?

 

[cesiumfrog]

Russ,
I stand by my statement that it is impossible to make a device that satisfies all of the constraints described in the OP. Note that external power sources (like for a normal fridge and for the heat pipe device referenced by the OP) are ruled out by the OP's emphasized criterion (and Xow's complaint is muted by the implicit criterion that the device typically works how the OP desired).

As I described earlier (but you seem to have missed), it is possible to construct a device that functions in spirit as the OP wants, but close reading of the strict criteria eliminates even that (by the implication that the insulation continue up along the sides and be completely spanned by the second copper surface, such that the only heat sink be the one connected to the top of the hottest surface).

 

[russ_watters]

Heh - not sure how I missed the part about no external energy, but now I'm wondering what is wrong with your thermocouple idea? Or a Sterling engine powering a heater? The whole device has two outputs, one where T1 is measured and another where heat is rejected to the atmosphere.

The stuff about insulation is pointless because nowhere are any heat transfer rates mentioned - only temperatures. It doesn't say the surface of T1 is insulated, but it doesn't even have to be if heat is being pulled from the T0 surface fast enough through conduction.

And since there is a provision for heat rejection, the 'no external power' provision is really irrelevant - we can generate our own power to run a heat pump.

 

[zytrahus]

Let's back up a little...
Heat will not spontaneously flow from cold to hot, but the OP said he wants to use a "device" to cause heat to flow from cold to hot, and you most certainly can force heat to go from a cold area to a hot area with a "device". Most people own such devices. Our houses and cars are equipped with them...

What the OP did not specify is if he wants an active or passive "device" or how much energy should flow from cold to hot (he's asking for a temperature, not a heat flow rate and he doesn't specify if there is anywhere else energy is being provided or rejected). An air conditioner is a "device" that uses input electrical power to run a pump that pushes heat (in a working fluid) from a cold area to a hot area. A heat pipe uses the temperature difference to help move the heat energy around. It is passive (it requires no energy input other than the heat it is moving), and harnesses heat energy that is already moving from warm to cold and just makes it move where it wants to go, but in a different way. Ie, a heat pipe is a passive heat pump that pushes heat where it already wants to go.

Now, it appears that the device referenced in the OP is designed to create a third temperature (not described in the OP, but shown in the reference), one different from the source and sink and, in fact, lower than the temperature of the source. Now I know the first instinct would be that such a thing is impossible, but it isn't. The net energy is still moving from warm to cold, but part of the energy gets sent somwhere else. Call it a "thermodynamic advantage" - it works the same as a mechanical advantage. A Large weight can be used to lift a smaller weight above the large weight's original height. A hydroelectric plant can be used to pump a smaller amount of water above the level of the reservoir. Heck, a sterling engine could be used to power an air conditioner. Now there is no net gain in power this way, but it should be clear that there are cases where a mechanical or thermodynamic advantage could be useful.

It appears that someone has found a way to passively generate a "thermodynamic advantage" using heat pipes.

So. All that said, what exactly is your question, zytrahus? It seems you have found a device that can do what you are asking. Does it not meet your needs? What are your needs (constraints)? From the description in your opening post, it sounds like a normal powered liquid chiller should be acceptable. Are you requiring that it be passive? Or that it use only thermodynamic energy (no mechanical or electrical)?

I have not found anything :D - originally experimenting stirling engines (basically making pressure, temperature measurements from a stirling engine installed on an electrical heater). Not that I have worked on this for a long time but a good way I found to improve efficiency of this stirling engine was to increase the heat load. The heat load just gets a higher temperature on the hot side of the stirling engine. the cold side being liquid cooled with a chiller at 25C, the delta of temperature is then higher and so is the efficiency.

increasing the temperature of the hot side of the stirling engine is not hard, but I have constraints:
- max temperature in my heater of 60C

- and because the main application would be to eventually convert this work into electricity, I would like to use a passive (or a low energy consumption) device to do the job...
today, I used a TEC between my heater and the stirling engine: it does what I wanted: keep the heater in my range and increases the hot side of the stirling engine...
but a TEC is not very efficient and is sucking more energy that I could harvest. I'd be satisfied if there was a TEC with a capacity of 200W with a 60 DT with only 20-50 W consumed... but there is not; it's even faaaaaaaaaaaaaaaaaaaar from here :D