Naked Science Forum
On the Lighter Side => That CAN'T be true! => Topic started by: alancalverd on 11/02/2023 19:58:43
-
I have converted an old barn into a music room. Some of the instruments being almost as old and fragile as myself, I have installed an air-source heat pump to maintain a reasonably constant indoor temperature of 18°C.
The heat pump uses electricity to run a refrigerator system that extracts heat from the outside air and uses it to circulate warm water under the floor. So most of the time there is a temperature difference between indoors and outdoors - unless the outside temperature happens to be exactly 18°C, in which case it is unnecessary.
Suppose I have a huge thermocouple, with one junction inside the barn and the other outside. Let us suppose that Tin>Tout, which it will be as soon as I walk into the room (Tout is just above freezing tonight). So a current will flow through the thermocouple pair. OK, the voltage will be small but I could make a lot of junctions in series (a thermopile) to get more volts, and/or use a very thick wire and fat junctions to get a lot of current. Enough, in principle, to start the heat pump. Which will increase the temperature differential, and thus the power available to drive the heat pump.....
Never mind the engineering practicalities: I'm in physicist mode and assume that my alter ego can build the necessary inverters etc. so I can heat my barn to any temperature I like without expending any additional energy - it all comes from the fact that it starts off slightly hotter than the air outside.
And why not?
-
"Can I get infinite "green" heat from my barn?"
No.
-
I know that, but it's difficult to explain why.
-
I know that, but it's difficult to explain why.
Not very.
Do you imagine a thermocouple (or thermopile) would be good thermal insulation?
Also, do you realise that one form of heat pump which you could have used is a Peltier cooler.
At that point you have essentially connected a bunch of Peltier elements together and expect one of them to spontaneously generate a temperature difference
-
But the heat pump moves more energy than it consumes ;).
-
But the heat pump moves more energy than it consumes ;).
And the thermopile moves more heat than it produces.
-
Theoretically, if the heat pump and thermopile were perfectly reversible processes you might well be able to get back all the energy you used to pump the heat that performed the round trip. But they're not perfectly reversible devices. And further, a lot of the heat you pumped in leaks out through the walls, ceiling and floor, and so cannot be used to generate electricity anyway. So there's inevitably a gap.
-
Interestingly, a semiconductor thermocouple pair can have up to 30% efficiency, and a heat pump COP can exceed 4.That looks like a net gain!
What is interesting about this scenario is that it treats the external environment as both an infinite sink (which it effectively is) and an infinite source (ditto)!
Never mind insulation and stuff - I'm just looking at the physics (i.e. the world of weightless strings and spherical cows) not engineering (the world of wobbly girders and hungry humans). If you want to go in the direction of sordid practicality, remember that a good passive house or even a "passive school" in UK latitudes can be heated entirely by the people inside it, and if you keep the door closed, an igloo is adequately habitable until it melts.
-
remember that a good passive house or even a "passive school" in UK latitudes can be heated entirely by the people inside it, a
Yes.
Remember that maintaining a temperature differential requires an energy input (which may be in the form of food).
-
Admittedly I have taken the essentials of life for granted, but the object of a heat pump is to add comfort - the barn already has a kettle and a toilet! When I stop eating food I probably won't care too much about keeping the piano warm, and eventually the other guys will realise that the bass player hasn't turned the page because he is actually dead.
Speaking of which, would you like to read my favorite viola joke?
-
Speaking of which, would you like to read my favorite viola joke?
Only if it's a lot more polite than the cello joke.
-
The orchestra was on tour. On the second day the conductor fell ill. They held an emergency meeting.
Second viola said "I played this book a couple of years ago under John, and I have his notes, so I'm happy to conduct until he recovers."
All agreed, and the tour continued.
The conductor recovered in time for the final concert.
Second viola took his usual seat and first viola said "Where the hell have you been for the last two weeks?"
-
Interestingly, a semiconductor thermocouple pair can have up to 30% efficiency, and a heat pump COP can exceed 4.That looks like a net gain!
Nope. Show your working.
-
Any thermoelectric generators I've seen are very low on efficiency and a lot of the heat passes directly from the hot side to the low side so they are only used on waste heat. I'm sure I've heard your joke before, Alancalverd.
-
Many thermoelectric systems are fairly small, with a short distance between hot and cold junctions.
But there is no theoretical limit to the dimensions of the wires joining them - I'm looking right now at the temperature probe for a multimeter which uses a tiny sensing junction at the end of 2m of 16SWG wire to a thermistor-compensated reference junction.
So the feasibility of my absurd suggestion depends only on the ratio of the thermal and electrical conductivities of the connecting wires.
The viola joke is a classic and I may have even repeated it in these forums. It was told to me by a conductor, who happily admitted that his profession is for wannabe percussionists who can't be trusted with two sticks.
A true bass story: the MD was introducing our brass band to a local journalist. "John plays the tune on lead cornet, Sally plays the counterpoint on the tenor horn, Ruth takes the trombone solos....and those blokes who look like bass players, well....they play bass."
-
I'm very rusty on this stuff so I pulled out my battered copy of "Rogers&Mayhew". They state that the max ideal efficiency is T1-T2/T1, same as the Carnot heat engine. So you would need a decent thermal differential to achieve any reasonable output. They also state that in the case of thermocouples one only gets a few μV/°K, way too low to be useful.
-
But as much current as you like because the current output depends on the junction area.
-
So the feasibility of my absurd suggestion depends only on the ratio of the thermal and electrical conductivities of the connecting wires.
Yes.
Specifically, your plan requires a thermal conductivity of zero and an electrical resistivity of zero.
You can stop now.
-
They also state that in the case of thermocouples one only gets a few μV/°K, way too low to be useful.
It's a small voltage but that doesn't stop us using it.
https://support.celestialfireglass.com/faqs/thermocouples-everything-you-need-to-know-and-then-some/
-
Yes indeed, BC, they are extremely useful for temperature measurement. What I should have said is that thermocouple output will not be of use for Alancalverd's eco-barn.
-
Yes indeed, BC, they are extremely useful for temperature measurement. What I should have said is that thermocouple output will not be of use for Alancalverd's eco-barn.
That's not just a measurement. The current from the thermocouple actually does some work. It opens the solenoid valve.
But it's not a lot of power.
-
Thermocouple current is used in domestic gas boilers to hold the main valve open. The hot junction is heated by the pilot light flame so if the pilot light blows out the main jets will not be activated. Not a lot of volts but tens of amps.
-
Thermocouple current is used in domestic gas boilers to hold the main valve open. The hot junction is heated by the pilot light flame so if the pilot light blows out the main jets will not be activated. Not a lot of volts but tens of amps.
If only someone had mentioned that earlier.
Oh.. hang on. It seems I did
-
Not quite. Your link refers to maintaining the main flow with a manually-triggered constant-flow system like a patio heater (why? if it's cold, go indoors...) or barbecue, as long as the main burner is alight. A domestic boiler opens the main valve automatically if the pilot light is active and water is flowing through the heating coil. Nitpicking, but accurate.
-
Making the distinction between a gas boiler powered by mains gas and a gas barbecue powered by bottled gas is, as you say, nitpicking.
Here's what the thermocouple does.
[ Invalid Attachment ]
-
Tens of amps from a pilot light thermocouple? I don't think so. More like millivolts o/c and milliamps s/c.
-
Tens of amps from a pilot light thermocouple? I don't think so. More like millivolts o/c and milliamps s/c.
Nope, Unexpectedly, Alan is correct.
The voltage is small, but the resistance is very small so the current is large.
-
Hi BC, I have looked in vain to find data on the s/c output of pilot light thermocouples. Any chance of a reference?, I would be much obliged.
-
Say 50μV/K, 1600K between flame and ambient: 80 millivolt. Resistance of a 2 mm diameter iron wire is about 0.03 ohm/meter so you could get a couple of amps flowing through a fairly small coil from a practical thermocouple in a gas flame.
-
That gives ~2.7A. The expression "tens of amps" is what I queried. I am going to do a test on this, a thermocouple in a flame, short circuited and current measured with a dc clamp meter( hall sensor ). Also can we assume that the source of emf has no resistance, neglecting the leads involved?
-
That gives ~2.7A.
Depending on how long the wire is. Not that it makes much difference since the magnetic field is measured in ampere turns per meter so a longer wire = less amps x more turns. Given the size of my old boiler I would have thought 10T was entirely achievable.
-
Given the size of my old boiler I would have thought 10T was entirely achievable.
I doubt that.
I think 10T is somewhere between 400MHz NMR and levitating frogs.
10 A is just about possible.
-
That gives ~2.7A. The expression "tens of amps" is what I queried. I am going to do a test on this, a thermocouple in a flame, short circuited and current measured with a dc clamp meter( hall sensor ). Also can we assume that the source of emf has no resistance, neglecting the leads involved?
You probably can't neglect the leads here.
But you might be able to measure the hold-in current of the electromagnet when the valve is open.
Tens of amps might be optimistic. But it's a much bigger current than people usually consider.