Stirlingengineforum.com

Jack wrote: ↑Sun Dec 17, 2023 2:36 am I don't really see how regenerators in the general sense do anything else than basically change the timing of heating and cooling. So the most work would be done at a different crank angle.

For my idea I'm trying to store some heat in the fluid so I can use it to extend the heating cycle. Haven't fully figured it out yet though.

I won't say anything is "impossible" though in this case I'm tempted.

If you add heat to the working fluid it wants to expand, and the expanding is the means of extracting power. So it sounds like you are endeavoring to "store" heat in the working fluid but somehow inhibit expansion.

To a degree I guess that happens at or near TDC when the piston is holding the gas back under pressure throughout the heating process. So, I have to say it's not only possible, but happens to some extent in all engines during "isochoric" heating

https://en.m.wikipedia.org/wiki/Isochoric_process

To some extent I also think a "thermal lag" or the popular little so-called "thermoacoustic" engines with the long relatively voluminous chamber partially plugged with a stopper, partially isolated from the power cylinder, at TDC is effectively an isochoric heating chamber, especially with the piston held firmly down over the opening by a connecting rod attached to a crank, like a lid on a pressure cooker getting ready to explode. A boiler in a steam engine stores the hot working fluid under pressure until the valve opens allowing the high pressure steam to rush into the power cylinder to expand and drive the piston. Similar isochoric heating process.

I'm not sure there is any other way to store heat in the working fluid except as described here - under PRESSURE.

With steam, high pressure tends to raise the boiling point, inhibiting phase change, so I think when released the "wet" steam flashes to a "dryer" more expansive gaseous state.

Storing heat in metal is a way of putting the heat into a latent form using the high heat capacity and low expansion coefficient. The heat is "stored" in the metal with little expansion, but transfered into the gas the gas expands a lot.

An ammonia absorption refrigerator is getting pretty far afield but the ammonia gas is "stored" in water. It greatly expands when released from the water by a relatively low grade application of heat.

So my initial knee jerk response, that it would be "impossible" to store heat in the working fluid itself as it would expand and become useless, I have to think again, not impossible, but tricky. There may be other possibilities I haven't thought of, of course.

I'm approaching it from the rotary vane angle. It allows me to create chambers where I hold the fluid and move it to where I need it.

Right now I'm at the point where I use the fluid at its hottest to power my cold rotor. There will be excess fluid that, in stead of pushing it with the hot rotor through the cold rotor, I hold and bring together with a new "load" of fresh fluid in my "regenerator" rotor.

It seems like a nice idea on paper, but I'm not sure it works the same way in real life. I might have to start learning how to work with CFD software.

For a long time I've been wanting to give the gravity pump idea posted early on in this thread a fair try.

I couldn't imagine it could actually work as presented, but, I couldn't really figure out exactly why not.

As water runs out, it creates a vacuum that draws more water in.

Well I finally decided to set up a little test in the kitchen.


As can be seen from the collapsed bottle, the setup certainly does result in a vacuum.

Actually I did, with a little poking at the inlet tube, get water to start running, up and out of a bowl in the sink into the milk bottle then out a tube in the bottom of the.bottle back into the sink drain... Continuously...

But, cheating a little, so far.

The siphon outlet is still lower than the water level in the bowl. So, just a siphon with this contraption in the middle. No pumping water up hill so far

But the water is still running and the vacuum increasing, presumably, but the milk bottle is also being crushed by the vacuum/pressure.

I'm going to call it inconclusive at this point. It isn't really a fair test as A LOT of potential vacuum is being lost due to the flimsy container. Also the sizing and length of the tubing, and those upright tubes that look like some kind of vapor traps are not included in the setup

Would making some "improvements" get it to work to pump water up hill? I very very much doubt it.

Unfortunately just now I hear air being sucked into the milk bottle from a leak around the hole where the tubing is inserted on the bottom

All I can say is a surprisingly high vacuum has been generated without air bubbles getting drawn in backwards from the end of the bottom drain tube

A bit of video:

https://youtu.be/83wjLTWKXDk?si=Umo82cShViQOqXc1

This (the following, not mine) video does a good job of explaining all the necessary elements, many of which my test experiment is obviously lacking.

https://youtu.be/xXJAkQscS2A?si=nMNDVQWzehQnFGYG

Some videos include a foot valve on the bottom of the inlet pipe.

My test shows at least that a container with air, in line in a siphon does not break the siphon.

The upright tubes are apparently to trap air bubbles that presumably, will eventually get drawn in through the end of the outlet pipe to rejuvenate the flow. Presumably this could only work until the upright pipes become full of air, if it works at all.

At best I think it can only prolong the inevitable and the "siphon" will eventually be broken.

On the other hand... As water flows out, more flows in.

Also, the inlet pipe is supposed to be of a narrower diameter than the outlet, until the end, My test uses only one size tubing throughout.

Does the weight of all this additional water in the larger outlet pipe help in some way?

Pressure in a column of water is calculated by the following equation :

Pressure = density of water times g gravity times height

P = d•g•h

The equation doesn't have diameter in it, only height and density. A bigger diameter will have a higher flow at the same pressure.

In other words, whether it works or not depends only on height, not total mass.

After watching this video:

https://youtu.be/vsTIWhZ-6HM

I've been trying to think how to utilize both the hot and cold side of a Peltier device without having to use two engines.

Heat pipes?

That might work, but kind of difficult to implement and bound to lose a lot of heat in transport.

What if, ...
Use the Peltier device as the displacer. Or embed the Peltier device within the displacer. Or multiple possibly layered Peltier devices.

After reading several articles about Peltier device efficiency and COP, when used as a heat pump, efficiency can be greater than 100%

This is because the device is not only moving heat but the applied power is also converted to heat as well.

When using it as a heating device, the efficiency will be above '100%'. You will always have the applied power turned into heat, therefor you will never generate less heat than that. How much heat you move, largely depends on the temperature difference between the cold and hot sides. The smaller this difference is, the more heat you can move. In the case of your example device used as a heater, you would apply 150 Watts and get 150 to 250 Watts worth of heating

https://www.reddit.com/r/AskEngineers/c ... nt_how_is/

For a low temperature differential situation, a COP > 1 is not impossible. (For refrigeration)

If utilizing both heating and cooling, does that mean the "efficiency" could be combined?

Say you have a 6 volt 2 amp Peltier device.

It produces 12 watts of heat plus MOVES an additional 12 watts.

If you have an LTD Stirling that can generate 12 Watts... Hmmm...

Might be an interesting experiment.

Somehow the above scenario seems highly unlikely to me.

For one thing, the efficiency of Peltier devices generally is notoriously rather poor.

For the above to be true, it would suggest that ALL the heat generated by the input current would somehow only appear on the hot side of the chip.

At any rate, an anonymous Reddit post is hardly a credible source of information.

In my limited experiments so far however, trying to use the chip as a cooling device only to run the engine on ambient heat, the enormous amount of "waste heat" being generated to be removed by the heat sink and fan nearly overwhelms the scant cooling effect which disappears almost immediately when the power is interrupted.

As seen in that previous video: https://youtu.be/vsTIWhZ-6HM the engine on the hot (top) side is running much more vigorously than the one on tbe cold side.

So these Peltier devices, whatever the actual efficiency, DO generate a great deal more heat than cold.

It seems rather remarkable then, they could ever generate a refrigerating COP > 1

Quite obviously, it seems to me, a Stirling engine relying on a Peltier device displacer to provide a ∆T / heat could not function with a 0°K cold side alone.

The Peltier unit is transferring the heat to run the engine from the cold side to supply heat to the hot side.

That would make some heat input into the engine from the cold side essential I think.

Maybe transfer the electrical current through via some contact strips ?

Or maybe the wires could be routed up through a hollow connecting rod:

Or maybe generate a current within the displacer?

I think your first pic, of the contact strips, is possibly the best, but I would put them on opposite sides of the displacer, and use a metal wheel as the other side of the connection.

I'm worried about the weight that all of this would add to the displacer though.

MikeB wrote: ↑Thu May 30, 2024 4:32 am I think your first pic, of the contact strips, is possibly the best, but I would put them on opposite sides of the displacer, and use a metal wheel as the other side of the connection.

I'm worried about the weight that all of this would add to the displacer though.

The most interesting thing about using a Peltier device for experimenting is that Peltier devices are only about 1/4 the efficiency of a vapor-compression refrigerant based system.

Logically, if something could work using a Peltier device, it could only work better using a conventional heat pump.

Experimenting with a Peltier device, though, is obviously MUCH much easier than building an entire vapor-compression refrigeration system.

Also, in a conventional heat pump, heat separation is not an issue at all. The condenser and evaporator are completely separate units.

But if a Peltier device could work, it would be a simple and easy proof of concept demonstration that would justify putting the time and effort into a more robust heat pump / heat engine experimental project. Which is so often dismissed as "impossible" and a waste of time to even think about.

And a vapor-compression version would likely not involve any additional weight on the displacer at all.

I was just browsing the internet for small HVAC compressor pumps.

There are a couple companies making mini refrigerator compressors.

One is Rigid:
12 volt DC Compressor, smaller than a can of soda pop.

Not really a lot different from some of the little butane refrigerators/ac/heat pumps IMO.

Anyway, while thinking about the possibility of scaling up a Stirling/heat pump combination engine, in my browsing a link to an automotive ac compressor appeared.

Not really what I was looking for, but it got me thinking.

I wonder just how much torque, horse power it takes to turn one of those things.

Anyway, car ac compressors are not something I've ever gotten into, so they are kind of a mystery. Anyway I found this interesting tear down video.

Lots of interesting parts inside:


https://youtu.be/urFVLKlpf1I

Since they are mass produced they are also pretty readily available and fairly inexpensive.

Probably, I'm guessing, a lot of the power draw from these things comes from the electromagnetic clutch. Possibly not necessary if driven directly by a Stirling engine to heat and cool itself.

I think I have a new project!

Now I'm wondering why I didn't think of this before.

Probably an automotive ac compressor just seemed out of the question since my general assumption/impression has been you need a 300 horsepower engine to turn the thing, and, well, they don't seem very reliable. At least the ones I've had in used cars I've owned. The car engine slows down when the ac is on, seemed like it must be a tremendous load.

But, after seeing how they are built, Now I wonder just how much of the load is actually just the electrical draw from the clutch, putting a load in the alternator.

Also, how high pressure is the system startup? Likely that could be reduced using a capillary rather than an expansion valve.

Tom, the Peltier displacer should work just fine. It will produce a temperature difference, hotter and colder inside the engine. The entire displacer chamber could be insulated and it will still work fine, maybe even better.

I liked the hollow displacer rod and the wires running up through it. Put a big 'S' turn in the wires, or make a horizontal spiral out of the wires like a clock spring with the rod in the center and the outside attached to the frame. Excellent.

Just a word on COP's from the following web page:

https://www.meerstetter.ch/customer-cen ... r-elements

Had the following graph for the amount of cooling.
The up shot is that the COP drops off very fast even at a minimal temperature difference. At a 15° C/K delta, about all you will get is a COP of 2, at 20% max I for 12V•2A•0.2% = 4.8 watts. n=(315-300)/300 =0.048 or 4.8%. For 0.23 watts maximum output from an ideal Stirling. If you get more work out than that it will be a Carnot anomaly. Don't feel bad if only getting half of that.

Carnot maximum follows:
Engine efficiency:
n=(Th-Tc)/Tc = 1/COP

Heat pump COP:
COP=Th/(Th-Tc) = 1/n = RefCOP+1

Refrigerator COP:
RefCOP=Tc/(Th-Tc) = COP-1 = 1/n-1

The wattage separation for a Peltier is Qc+Qh or COP+RefCOP. Pretty excellent. But at very low power and Delta T.