I'd like to recreate Tom's experiment

[Jack]

As per title, I'm curious and have some time available.

I don't think I fully understand it though, so this thread is for me to ask questions about the setup Tom had and also what data I need to look for to make this believable for everyone on here.

Let's discuss this experiment. Try to be civil and only talk about the facts or data you'd need to be satisfied and I'll try to set it up that way.
I'm trying to be Switzerland here. Curious Switzerland.

I have easy access to this Stirling engine from a local ebay variant.

Screenshot_20240722_170123_Lazada.jpg (149.29 KiB) Viewed 6391 times

Screenshot_20240722_170132_Lazada.jpg (91.56 KiB) Viewed 6391 times

Would this be sufficient?

[MikeB]

As per title, I'm curious and have some time available.

Excellent. Though be prepared for a large quantity of "healthy discussion" sooner or later.

I'll let Tom have his say on how to set up the experiment (could anyone stop him?? )
But for me, what I want to see for pretty much any experiment, is:
1) An accurate measure of the energy/heat going into the engine.
2) An accurate measure of the heat coming out of the engine - by ANY route - not just via the cold-end.
3) An accurate measure of the _useful_ work done by the engine - most engines like the one in your picture don't have ANY load, and purely fight against internal friction etc.

It would also be nice to know:
4) The average temp or the working fluid
5) The average (or max/min) pressure of the working fluid
6) Physical dimensions

Good luck!

[Fool]

Tom allegedly made modifications to the engine he tested, such as replacing the metal through bolts with plastic bolts, and increasing the piston stroke. I don't know what else. Typically he broke or melted parts, so there is the random element as well.

To duplicate his experiment you will need to "insulate" the cold side, which was often the top and piston side. He used several different insulating materials for several tests. A verification that you've succeeded might be the engine running faster immediately after adding the insulation, as reported by Tom. Also he reported the cold staying at or below room temperature. Looking at all his YouTube videos might be helpful. Do not fear lack of success as failure. All good lab practices produce valid results. We want to know either way.

Try it in a cold room, hot room. Cold, hot, or room temperature insulation. Hot on top, cold on top. Ambient on top, hot or cold on the bottom. Experiment.

Once a verification can be duplicated as above, I recommend an improvement in testing as below.

Tom had a bit of trouble with the piston condensing moisture from the air and stopping the engine. I think that can be eliminated by building a stand and turning the engine upside down and heating the piston side. That should tend to burn off any moisture from the moving parts. It also lets the cold side be on top for better running, cold on top heat on bottom. The cold side will be flat and far away from movement. It will be easy to insulte the cold side with a flat piece of fiberglass, blanket, or Styrofoam, or dished to fit in the flat plate side, now the top.

Heat or cold could just be ambient room temperature.

I am looking for a potential difference between running the experiment right side up, and upside down.

Tom started using an infrared camera thermometer, then a thermal camera, then thermal couples. He had several different models, and types. Even one that had an entire cold side of acrylic.

I recommend that temperature of both inside and outside of both the hot and cold plates be measured using a calibrated set of thermocouples. They could all be calibrated in an ice, and then in a boiling, bath even when attached to the plates.

I recommend a small electric heat source be used, a few watts 10 W or so, and voltages and currents be measured.

I recommend a homemade dynamometer be built and calibrated. Could be some rubber bands for friction production and force measurement, or springs, a small scale and a torque arm. Keep it simple. Keep it inexpensive. Calibrate it with known forces and torques. Cheap optical tachometers are available for RPM measurements.

From torque and RPM, power out can be calculated. For a small LTD engine it might be in the milliwatts range. Note: 10 W input might over drive engine, or over heat it. Experiment with that too.

Running on ambient heat and a calibrated ice pack on top will give some idea how much energy and time it takes to absorb enough on the cold side to melt and stop running. Try it insulated and not insulated.

Just some memories and thoughts of mine, nothing definitive. And thanks for trying.

Thanks MikeB.

[Jack]

Sounds reasonable.

I'm thinking temperature sensors on both the hot and cold plate.
Pressure sensor or gauge.
Total machine well insulated.
Freeze the whole apparatus including a cold reservoir.
Heating element around the hot plate with a lab power supply.
A way to measure work output at the shaft.

Edit, this was my answer to MikeB

[Jack]

To fool,

I agree with keeping it easy and simple as well.
Should be doable.

I mean to play around with it a lot to try and get a feel for what different solutions do to the results.

[Fool]

Excellent.

Also:

The temperature difference between inside and outside the hot or cold plates can be used to calculate the heat moving through them. It may be too low a difference to measure. 1/10 a degree or less.

An indicator diagram would be good too. Measured pressure verses crank angle, correlated with volume at crank angle.

[MikeB]

Sounds reasonable.

I'm thinking temperature sensors on both the hot and cold plate.

That's a good start, but merely measuring the temperature of either/both plate doesn't directly tell you how much energy is passing through - a measure of both sides of each plate would be better.

[VincentG]

This is the only feasible way I see to conduct this test.

Use a small electric heating element on the bottom plate and insulate it well enough that the majority of heat is forced through the engine. Put the whole engine in a quality cooler and measure the temperature rise over time inside the cooler with an engine under load, no load, and not running.

The heater should only just supply enough btu and temperature to run the engine at its best power.

All in all it's going to be a very difficult test to get accurate and repeatable results.

[matt brown]

This is the only feasible way I see to conduct this test.

Use a small electric heating element on the bottom plate

inside engine

and insulate it well enough that the majority of heat is forced through the engine.

make the whole case out thick, but low density wood, like balsa

Put the whole engine in a quality cooler and measure the temperature rise over time inside the cooler with an engine under load, no load, and not running.

I don't think this will be required

The heater should only just supply enough btu and temperature to run the engine at its best power.

yep, make sure element has variable output

All in all it's going to be a very difficult test to get accurate and repeatable results.

Jimmy the Greek would have loved covering bets on this one...

[Jack]

If I have a way of controlling very detailed how much heat I'm putting into the engine, I don't think freezing it would be necessary.
I'd be looking for the right delta T, where I start on that scale shouldn't matter. Right?

[MikeB]

If I have a way of controlling very detailed how much heat I'm putting into the engine, I don't think freezing it would be necessary.
I'd be looking for the right delta T, where I start on that scale shouldn't matter. Right?

From my point of view, Yes/Agreed.

No need to freeze anything, unless you are very specifically wanting to test an ambient-heat engine.

[Jack]

I think the ambient heat idea could be deduced from the results.
But since we want to measure heat in and work out it doesn't seem to matter.

[Jack]

I'm reading old posts by Tom.
Maybe I should also try to find a way to track the pressure in the engine compared to timing.

[Tom Booth]

As per title, I'm curious and have some time available.
(...)

I have easy access to this Stirling engine from a local ebay variant.
Screenshot_20240722_170123_Lazada.jpg

Would this be sufficient?

Not to be critical of your apparent interest but "Tom's experiment" is too vague, considering I've done dozens of different experiments.

"Would this be sufficient?" Is not answerable. Sufficient for what?

What experiment specifically are you interested in replicating?

I've done many different experiments using different setups for different purposes to test different things over the course of several years of research.

Presumably you are talking about a video recorded experiment posted to YouTube. If you are genuinely interested, that would be the logical and sensible place to start.

Post the video of the experiment you intend to recreate so everyone, including me, can have some actual idea what you're talking about.

[Fool]

I'm reading old posts by Tom.
Maybe I should also try to find a way to track the pressure in the engine compared to timing.

Good idea. I recommend copying exactly whatever experiment you wish. That will give a repeat of Tom's results. Then add the modifications. That way you will have a control group. So to speak.

What you are describing is an indicator diagram. Do a search on "indicator diagram". In the beginning the diagram was made by paper and pen. Pen attached to the pressure gauge. Now electronic instruments are used. An Arduino can also be used, or any computer, and sensor, system data logger.

.