My contribution to the ECE

[Tom Booth]

The importance, and/or influence of expansive cooling in relation to work output to a load (load = what the engine is powering, such as a generator, Matt seems to have some odd definition of "load") is seen in free piston Stirling engines (NASA type)

A problem with these free piston engines is that the piston depends on expansion cooling to effect the return of the piston. Part of that cooling is the result of work output, or conversion of heat to work. That is, the load on the engine is partly what keeps it running cool and limits the stroke.

The heavier the load on a free piston engine, the shorter the oscillation. If the load is removed, not only can the engine overheat, but the oscillation can increase to a degree that can become destructive to the engine, as there is nothing to restrain the piston from flying outward but the cooling effect of converting heat into work output, so with no load on the engine, the planar springs can be stretched beyond their design limitations and the piston can bang into the engine housing.

How all this translates over to a crank type Stirling is something to consider, but the idea of ports to relieve the excess pressure seems to address the issue.

Under load, the ports would likely not be necessary, but under no-load conditions can act as a pressure relief valve.

A free piston engine runs at a steady frequency, so under no-load conditions the amplitude of the oscillations increases. A crank engine generally will respond to the "excess" energy by increasing in speed as well.

[VincentG]

Unfortunately, or fortunately, I'm still not sure, this little engine has become an obsession. I have gone through half a dozen wildly varying cam designs, as well as different displacers. For the displacer, I've found the stock foam piece is best, with the addition of my top foam piece with the cylinder cut out. It's clear to me that one thing is crucial, the displacer must make full contact with the hot and cold ends to "shut off" their influence. Also what is clear, at least with the current compression ratio, the displacer must have a full 180 degrees of motion with the crank shaft.


The influence of the displacer is MUCH faster than I thought. That is to say, the speed of the displacers movement seems directly correlated with the speed of ignition, and there does not seem to be any real thermal lag(expansion latency). For instance, when operating the components by hand, the movement of the free piston is proportional to the movement of the displacer. Stop moving the displacer, and the piston stops moving. It does not continue to rise as I would expect if there was any thermal lag. At least not much. The "not much" part means that the piston does travel a bit further than 1:1 compared to the displacer when the displacer rest fully on the ends. This I'm sure is due to nearly stopping any temperature mixing.

So it now seems clear to me that to provide a rapid, "diesel" like impact on the piston, the displacer must move much more rapidly than with my cam, again at least with stock compression. Further testing with solenoids is in order. But for now the cam does the job.

This leads to my next observation, the working cylinder at this point, is miniscule compared to energy available right now with the system. This video shows the power the piston is exerting. It absolutely blew me away. The steel bit driver in the video is quite heavy, and seems to be of little setback to the piston. With no weight, the pistons movement is the full extent of the cylinder and then some. Note: the last thing I did was compress the cold air from the top of the cylinder to the bottom. The results speak for themselves. I NEED to make a top end with a much higher compression ratio and more swept volume.

Here is the steel weight I used

free piston power test.jpg (249.15 KiB) Viewed 5376 times

The video
https://photos.app.goo.gl/ic6xaKQeTd46T8Gv7

[VincentG]

And just for reference, this is the same weight on a completely cold engine to show the baseline air spring effect.
https://photos.app.goo.gl/yLskVKkxjec8h2Qz9

[Tom Booth]

.... It's clear to me that one thing is crucial, the displacer must make full contact with the hot and cold ends to "shut off" their influence. Also what is clear, at least with the current compression ratio, the displacer must have a full 180 degrees of motion with the crank shaft.

This is not true for the acrylic top engine I have. The acrylic is basically an insulator so holds in the heat, so the displacer does not need to cover the cold side to prevent heat loss and at times the displacer only makes a very slight lift off the hot bottom plate.

Your metal top with ice is just the opposite of that and will draw away heat unless covered.

... So it now seems clear to me that to provide a rapid, "diesel" like impact on the piston, the displacer must move much more rapidly than with my cam, ...

A Diesel produces heat for ignition by adiabatic compression. There are two ways to produce compression adiabatically, (without heat loss).

1. Carry out the compression very rapidly so there is not enough time for heat to be carried away by a metal cylinder, or other cold surfaces in contact with the working fluid

2. Have an engine constructed of non-heat conducting material. Or insulated. In that case the compression can be slower without loosing heat, as the cold heat conducting surfaces have been eliminated. (Like using an acrylic top).

The steel bit driver in the video is quite heavy, and seems to be of little setback to the piston. With no weight, the pistons movement is the full extent of the cylinder and then some. Note: the last thing I did was compress the cold air from the top of the cylinder to the bottom. The results speak for themselves. I NEED to make a top end with a much higher compression ratio and more swept volume.

Here is the steel weight I used
free piston power test.jpg

The video
https://photos.app.goo.gl/ic6xaKQeTd46T8Gv7

The engine certainly seems to have a lot of potential bang for the buck. I haven't had any of my little LTD engines blow the piston out of the cylinder like that when doing a similar test.

[VincentG]

This is not true for the acrylic top engine I have. The acrylic is basically an insulator so holds in the heat, so the displacer does not need to cover the cold side to prevent heat loss and at times the displacer only makes a very slight lift off the hot bottom plate.

Indeed, I think this may be the case after I enlarge the working cylinder. We will see. But in my tests, a small lift of the displacer equals a small movement of the piston.

Tom, what is your take on the fact that a regenerator seems entirely unnecessary for this engine. I always felt that the regenerator was just a band-aid for engine types(especially beta and alpha) that cannot effectively block off the heat source.

I was thinking about trying a displacer made of rockwool, for its greater(potentially) insulating properties and high heat tolerance. I think part of the reason the aluminum clad displacer fared poorly is that the entire displacer became heat soaked after an extended run. So I now think the displacer should be insulative, and also have as low a thermal capacity(density) as possible, but not necessarily permeable. An additional benefit would be less weight.

[Tom Booth]

This is not true for the acrylic top engine I have. The acrylic is basically an insulator so holds in the heat, so the displacer does not need to cover the cold side to prevent heat loss and at times the displacer only makes a very slight lift off the hot bottom plate.

Indeed, I think this may be the case after I enlarge the working cylinder. We will see. But in my tests, a small lift of the displacer equals a small movement of the piston.

Tom, what is your take on the fact that a regenerator seems entirely unnecessary for this engine. I always felt that the regenerator was just a band-aid for engine types(especially beta and alpha) that cannot effectively block off the heat source.

I was thinking about trying a displacer made of rockwool, for its greater(potentially) insulating properties and high heat tolerance. I think part of the reason the aluminum clad displacer fared poorly is that the entire displacer became heat soaked after an extended run. So I now think the displacer should be insulative, and also have as low a thermal capacity(density) as possible, but not necessarily permeable. An additional benefit would be less weight.

The displacer on the acrylic body/magnetic engine displays this behavior only when running at very high speed. That is, the faster the engine runs, the less the displacer moves, but the engine has a tendency to run at a steady speed. If you push it to run very fast with a lot of heat, running over a steady source of steam from boiling water for example, the displacer hardly moves, just making a slight "skip" at/near TDC.

An advantage I've seen in a regenerator in an LTD type engine is (I think, or my impression is) it allows air to flow through more easily without the hot and cold air on each side mixing. In a side by side test, the engine with the regenerator built into the displacer ran at a higher RPM, but this did not seem to conserve a lot of heat as I had anticipated. The regenerator equiped engine stopped running sooner, the other slower engine without a regenerator kept chugging along for several minutes. One test however doesn't say much, but the engine on the right has the regenerator:

https://youtu.be/QqN80ZqJLoQ

I've also had regenerators that do not seem to allow any measurable heat to pass across them at all. That is, the cold side (top) of the engine does not rise in temperature above the ambient air temperature, but I'm also finding this to be often true, regenerator or no regenerator.

Generally, if the engine can run, heat getting to the cold side can be traced to heat being conducted through the engine BODY or around the engine by air convection (outside of the engine) rather than through the actual internal working fluid., in my experience, and even just logically. For the piston to be able to return, the internal pressure must drop below atmospheric pressure. It seems to follow that if the pressure drops that low, the temperature must also be dropping very low. Apparently that is what actually happens according to what I take to be computer modeling (?).

From the Wikipedia article: "Temperature versus crank angle"

Temperature_vs_angle.png (5.1 KiB) Viewed 5351 times

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

[Tom Booth]

Apparently the regenerator or "economizer" was originally based on the idea that heat was a fluid (caloric) and could be stored and reused, so if heat is energy, that throws the whole idea into question.

A heat engine though, does not, or apparently, or theoretically, does not necessarily run on energy but rather on contraction and expansion of the elements.

Some gases, at certain temperatures and pressures actually contract when heated and expand when cooled. I would theorize that such a gas could conceivably power an engine by taking away energy, cooling it so that it could expand and do work. Then we would have to heat it up to get it to contract I suppose.

So, I guess a heat engine doesn't run on heat directly, it runs on the response of molecular attraction and repulsion to temperature changes.

[VincentG]

A heat engine though, does not, or apparently, or theoretically, does not necessarily run on energy but rather on contraction and expansion of the elements.

I think this sums it up well. We are driving a piston after all, and all the piston cares about is PSI. To make heat "energy" do work, may be akin to making a solar sail. Not theoretically impossible, but beyond the scope of this project to say the least.

I am now on the 3rd revision of my 3d printed housing, the first two being plagued by air leaks. On this print I used 0% infill so it is solid nylon. Small prints are easy to make air tight but it gets more difficult as the size increases. As an added precaution I will seal the outside of the housing with superglue or something of the like. To make that easier I have abandoned the water cooled jacket for now, and will add that back as a separate print, as I should have done in the first place. After a short while of running the previous housings let in water.

I have done further testing with the power of the piston to lift weight over simmering water and ice with the second(failed) housing. Even with a significant air leak, the piston was easily able to lift a nearly one pound weight after a full compression cycle. I have calculated the compression ratio of the engine(with the exaggerated stroke of 1.2") to be just over 1.1:1. I think this is a much more useful number that the swept volume ratio, which is 1:9 in this case. The stock engine being over 1:60.

With a compression ratio of 1.1:1, and a delta t of no more than 170 degrees, the piston with an area of .34sqin was able to lift one pound of weight. That means a working pressure on the piston of over 40psi!? The force of the vacuum drawing the piston in after the displacer is pushed back down is equally impressive, though I will need to attach a scale to measure this. Again, this all happens the moment I move the displacer its full travel.

I have made a small pressure port to check this with a gauge when the 3rd housing is dried. The little engine is looking quite used at this point but still functions well.

3-11-23 3d printed housing.jpg (231.06 KiB) Viewed 5335 times

[VincentG]

Here is some theory on my new housing design. I now believe that the ideal displacer should neither be hot or cold. It should essentially be highly insulative and inert and function simply to shuttle the gas back and forth while blocking the hot end while at rest.

I was thinking about thermal conductivity and it seems to me that the best way to impart temperature change, is for the object to be one with the heat source. So maybe the fastest way to heat air, is to rapidly introduce preheated air.

This sketch shows how the displacer rests on the bottom of the cylinder, blocking the heat from entering the system, but at the same time traps a small volume of air that has time to become super heated. The added benefit is the displacer is not in direct contact with the hot plate and stays cooler this way. When the displacer lifts, the super heated air instantly imparts its energy into the cold air above. This is the theory anyway... but the engine that lifted the 1 pound weight incorporated this design.

air gap displacer.jpg (128.21 KiB) Viewed 5333 times

To further this idea, maybe with a more precisely machined engine, the displacer can seal this super heated air so effectively, that it must be actively held down. Then, when the displacer is called to lift, it is mechanically released and explodes upward with the speed we need to create ICE like ignition, all while requiring less work from the crankshaft.

[VincentG]

EDIT: I meant potentially over 4psi.. as the can shot up with speed. I know, I need to get more scientific with experiments at this point.

With a compression ratio of 1.1:1, and a delta t of no more than 170 degrees, the piston with an area of .34sqin was able to lift one pound of weight. That means a working pressure on the piston of over 40psi!? The force of the vacuum drawing the piston in after the displacer is pushed back down is equally impressive, though I will need to attach a scale to measure this. Again, this all happens the moment I move the displacer its full travel.

[Tom Booth]

EDIT: I meant potentially over 4psi.. as the can shot up with speed. I know, I need to get more scientific with experiments at this point.

With a compression ratio of 1.1:1, and a delta t of no more than 170 degrees, the piston with an area of .34sqin was able to lift one pound of weight. That means a working pressure on the piston of over 40psi!? The force of the vacuum drawing the piston in after the displacer is pushed back down is equally impressive, though I will need to attach a scale to measure this. Again, this all happens the moment I move the displacer its full travel.

I'm glad you made that clarification. 40 psi would be insane, but even 4 psi, I think, is amazing. I think, in general, pressure changes in an LTD type engine are on the order of 1/10ths of a PSI (or maybe that was ATM, I'll have to check)

I also wanted to say regarding my last statement, that a heat engine does not run on "energy". I think it would be more accurate to say - does not run on heat. Perhaps.

At any rate I was thinking of Joule-thomson inversion curves and coefficients. A compressed gas released through a throttle can either cool of heat up depending on which forces dominate, attractive or repulsive. There are also several different types of forces and it all gets rather mind bogglingly complicated. Add in the fact that Air is a mixture of gases as well as water vapor

What I mean to say is that an attractive force is still "energy" presumably, though causing a gas to contract, and an expansive force is more obviously "energy"

Joule-thomson cooling is what makes refrigerators and heat pumps etc. function, and I'd say, to one degree or another also Stirling engines. So JT expansion resulting in a temperature increase is unusual, but is true of helium at normal temperatures.

From my conversations with people in the business of building NASA type helium filled engines, it seems these work more reliably as cryocoolers and they've all but given up on building engines.

I can't say I know what's going on on a molecular level but I think it is important to take into consideration. Mostly I rely on observation/experiment. That this seems to deviate from mathematical "ideals" does not particularly surprise me, as for an "ideal" gas, the JT effect does not even exist.

Anyway back to your project.

[Tom Booth]

This graph, actual readings from an LTD type Stirling, show a pressure difference of barely more than 1 kpa (kilopascal) which converts to about 0.15 psi.

If you are getting a 4 psi pressure differential, then it looks like that would be what?

25 times more than "normal" or what was recorded here

Resize_20230311_131144_4431.jpg (118.08 KiB) Viewed 5322 times

Heat source "a cup of hot water".

That graph BTW was generated by these experiments under load I think. Then the no load pressure differential would be even less

https://youtu.be/dvomod6SsA0

[VincentG]

To be clear, the high mep was with a simulated high compression by hand. With the engine running with the mechanical Guage attached by hose(unavoidable increase in volume) I am still seeing a 1psi pressure swing. And about 1.5 psi when I slowly turn the motor over by hand, giving it more time to react.

[Tom Booth]

The idea of preheating a small volume of air to a very high temperature keeping it trapped until needed makes sense. I think certainly air itself would transfer heat to more air faster than any solid surface ever could.

I've had some model engines where the displacer did not reach the bottom but left some air gap. I'd usually try to adjust the displacer down to eliminate this gap, but I can't say that this change actually made for any noticeable improvement in performance. Those engines seem to run pretty well with the displacer just fluttering in the middle. Maybe your theory explains how engines like that run as well as they do. The air, more or less pooling in the bottom has time to heat up before being mixed.

That gets me to thinking, what about some very heavy gas that would naturally tend to pool on the bottom as a heat carrier.

I must say you seem to have a genius for coming up with great ideas.

[Tom Booth]

Now you've got me thinking about using something like that to improve my Ringbom conversion idea:

Resize_20221206_055806_6714.jpg (192.22 KiB) Viewed 5297 times

Maybe just put a filter in the air line and something like silicon carbide sand in the bottom of the displacer canister.

When the compressed air enters the canister it would blow the red hot sand around.