Model LTD max power effort

[VincentG]

I finally got around to extending the stroke on my testbed LTD engine and thought it would be fun to see what kind of power we can really get out of these things. I'm still up in the air whether to use a pony brake or generator to measure power.

The stroke has increased from .32" to about 1" with the same 5/8" glass piston. This brings the displacement to about 5cc from the factory 1.6cc. The crank shaft and big end of the connecting rod are now on ball bearings, with the conrod in single shear to make stroke adjustments easier. I'd like to order a larger diameter piston and cylinder to max out the displacer chamber after this round of modifications. For now I can make the displacer larger to increase the compression ratio.

The only factory part left to the rotating assembly is the crank throw for the displacer as there is no need to change that. I'll have to make a larger flywheel but this is the largest round stock I had on hand for now.

long stroke ltd 2.jpg (187.29 KiB) Viewed 11570 times

long stroke ltd 1.jpg (185.71 KiB) Viewed 11570 times

[VincentG]

I was able to complete the first test run of the long stroke engine today. The results seem extremely promising.

long stroke ltd v1.00.jpg (133.65 KiB) Viewed 11526 times

The swept volume ratio is now 9:1. When the engine is cold, the expansion and contraction air spring effect is quite high, enough that I was doubtful the engine would run. But run it does, and with quite alot of power from what I can gather. I have added a heavy brass nut for added flywheel effect, though I will explain below why it seems unnecessary.

In this first video I show how the engine starts rapidly from both TDC and BDC.
https://photos.app.goo.gl/7vMFMB8aDGFSqxBt9

The second video shows the engine fighting through a wet power piston, which would have stopped the stock displacement engine dead in its tracks! Just a bit of moisture on these glass pistons causes an incredible amount of friction.

https://photos.app.goo.gl/UPE8hYzMZhrbM1Th9

Now for the really promising part. Although the cold engine displays a large air spring effect, when at operating temperature there is next to zero compression or expansion work required at all. The power piston falls right down to TDC when the displacer is covering the hot plate and vice versa. So it seems even at this temperature ratio we can increase the compression much further. Also, the flat plate heat exchanger is likely ineffective at transferring heat, so this should get even better with proper exchangers and thermal isolation from the housing.

There are also air leaks from the o-rings seals and various places. At this scale these leaks are significant. Hopefully my new concrete housing will be more air tight.

[matt brown]

The second video shows the engine fighting through a wet power piston, which would have stopped the stock displacement engine dead in its tracks! Just a bit of moisture on these glass pistons causes an incredible amount of friction.

This mod appears more powerful, but if the surface tension of water stalls PP...

Is that a moat for ice cubes on the cold plate and what led to wet PP ?

I'm all for increasing PP volume, how about 2x further increase and maybe via 1"/+ diameter bellows with same stroke. Another crude possibility is a "rocking piston" scheme where a loose fitting ping pong ball is rigidly attached to the conrod with a "rolling sock" seal under ping pong ball (where rolling sock is a weak balloon diaphragm).

[VincentG]

The water on the piston is from boiling water over the stove. So this issue should be resolved with the internal heat sink I have planned.

I would like to try a diaphragm type piston but it won't allow a piston port exhaust without extra valves.

After some testing I see there is a huge(relative) pressure gain available from a full compression cycle. By that I mean a full charge of air with the piston at BDC and the hot plate covered. I saw a guage spike of over 3psi when simulating this by hand. And there is an equal vacuum potential when reversing the cycle.

Even at this displacement, there is more expansion available than the power piston can utilize. I think I have a source for a 1" glass piston and cylinder.

At that point with a 1x1 bore and stroke we would have 13cc of displacement giving us a 3.5:1 volume ratio.

[VincentG]

Well I tried running the engine over a tea candle and saw much better results with no water contamination. I think most of the water vapor was getting in around the worn out o-rings that seal against the plate.

Here's a test run where I let the full weight of a standard #2 screw driver down on the flywheel. The tapping noise under no load is actually a slotted wrist pin in the connecting rod that allows a bit of power piston dwell and lets the piston slap into the rod for an extra shove. I have a hunch that it gives the engine a bit more power at TDC and BDC and makes timing a bit less critical. Look close and you can see how much the plates are flexing now with the added compression. This might actually be helpful until the power piston is large enough to utilize all of the expanding air volume.
https://www.youtube.com/watch?v=fMrivcQ-tWg

After that I removed the large brass nut to see how it would run. Turns out quite well. I think the video speaks for itself. The knife is not even sharp!
https://www.youtube.com/watch?v=5AF_fhhdbJs

I noticed with the stock engine but it's even more pronounced with this one, the engine really makes the best power around stall speeds. Toward the end of the video, I nearly stall the engine. When it is allowed to accelerate again, it does so with more force than any other time. Tom seems to be right here in that more load means more of the heat is used up.

[VincentG]

This is just to show the resistance from compression and expansion with a cold engine and light flywheel. The cylinder pressure was equalized at MDC for this test.

https://www.youtube.com/watch?v=82Ztn5UMWqs

Notice in the brass cutting video from the last post when the engine nearly stalls at BDC but then the cold stroke powers through. This to me is the single biggest advantage to the Stirling cycle for industrial use. As a true one-stroke engine it has an incredible resistance to stalling.

[matt brown]

The water on the piston is from boiling water over the stove. So this issue should be resolved with the internal heat sink I have planned.

This sounds interesting.

I would like to try a diaphragm type piston but it won't allow a piston port exhaust without extra valves.

I've always thought that a bellows might work on small scale (especially 'model') but most videos I've seen tend to look a tad jiggy and rather limited to low speed. Years ago, I saw a scheme where a relatively large bellows was loosely fitted over a solid cylinder that (1) reduced inner volume to practical value (2) served as a guide for bellows.

At your scale, I favor a large RC model engine PP (remove head, invert on cold plate, connect conrod) with 'hot' exhaust thru lower ports and cold plate snifter valve a la Otto.

At that point with a 1x1 bore and stroke we would have 13cc of displacement giving us a 3.5:1 volume ratio.

During my deep dive on gammas, I stumbled upon an "ideal" cross value equation that unifies various cycle values: (Thigh-Tlow)/Thigh=Vpp/Vdisp or in simple terms (for your project) when Vdisp:Vpp ratio is 4 then the ideal thermal ratio is 300-400k and when Vdisp:Vpp ratio is 3 then the ideal thermal ratio is 300-450k. Your would be 3.5 is 300-420k (not 300-425k) but this 'ideal' is merely a uniform guide across a wide range of cycles values (aka comparative analysis) otherwise, unrelated to ideal anything.

[matt brown]

As a true one-stroke engine it has an incredible resistance to stalling.

Or do you mean: one cylinder engine...

[VincentG]

Matt I am wondering what the real limit will be at lower temperature ratios. I suspect if I can keep compression isothermal it will be higher than we think.

And no I meant one-stroke. As in there are 2 power strokes per 360 degree rotation.

[matt brown]

The swept volume ratio is a good thing to track since it's fairly easy to measure. Meanwhile, the compression (and expansion) ratio is more of a guess due to phasing. However, considering that your aim is displacer dwell, then your idealized compression ratio is directly related to your swept volume ratio. Assuming extreme dwell with hot plate obscured, then your relatively massive PP still yields only a meager compression ratio...if PP=10cc and disp=40cc, then swept vol ratio=4. But if we consider this similar an ICE, then the compression ratio (and expansion ratio) is only 50/40=1.2 !!! and not much heat is generated by any compression at this paltry ratio.

As I see it, the whole LTD design flaw is not cooling DURING compression, but cooling PRIOR compression. IOW if the entire "displacer" gas mass is heated prior expansion, then any heat that isn't converted to work will have to go somewhere just to lower the PRESSURE so that PP can compress back. In a real gamma, this somewhere is a regenerator, otherwise this somewhere is a heat sink (or engine will stall). All the typical regen mumbo-jumbo is second fiddle to the basic reality that regen only works within a specific temperature range where its value is most effective (and efficient) over a wide temperature range which just happens to favor 'impossible' isothermal input. Once you rule out isothermal input, regen quickly follows, then the typical massive LTD displacer volume (relative typical tiny PP volume) will have to sink massive heat..."just to lower the PRESSURE so that PP can compress back". This issue decreases as relative PP size increases, but you still can't heat up 40cc to whatever, send 10cc to PP, then recover the heat from the other original 30cc without a strict PVT scheme.

So, regen is closely associated with isothermal input, but nixing both is not a bad thing. It simply means that typical LTD scheme is a no-go...and you can't heat more gas than you send to PP. But nixing typical LTD scheme also releases inherent low compression ratio constraint which is a major ECE bugaboo.

[VincentG]

I think in operation a Stirling engine equalizes its internal air mass to balance compression and expansion. The original engine seemed to gas out after a while but this one sustains operation much better. I attribute this to the higher pressure swing that pulls as much air into the leaks of the system as it pulls out.

So in reality the compression ratio is half of what the volume ratio would imply.

Sure the power piston should be large enough to use all the expanding air available from a given displacer, thus using as much heat energy as possible. But in practice I think maybe the heat input needs to be limited by the exchanger to keep efficiency up.

If compression could stay near an isothermal ideal, there may be unlimited expansion available for the power piston(otto), unless the mass of gas in the engine is balanced(stirling). But given enough cooling for the compression and heat input for expansion, I think a significantly higher compression ratio is possible for even the stirling cycle.

[Tom Booth]

(...)
As I see it, the whole LTD design flaw is not cooling DURING compression, but cooling PRIOR compression. IOW if the entire "displacer" gas mass is heated prior expansion, then any heat that isn't converted to work will have to go somewhere just to lower the PRESSURE so that PP can compress back.
(...)

More often the problem is the heat input doesn't keep up with the work output, so the return stroke is actually much stronger than the expansion stroke.

This becomes clearer in a "free piston" type engine when the piston starts smashing into the engine body with very forceful "contractions".

When drawing on "internal energy" the engine doesn't always know when to stop.

Or you could think of it like a slingshot.

During expansion is like slowly pulling back on a slingshot. Your working against atmospheric pressure, but the "heat" gets used up at the same time, then the return stroke is like letting go, letting the slingshot fly. On the return stroke you are NOT working against atmospheric pressure, atmospheric pressure is helping, and the internal pressure is effectively gone, so the piston comes back like a bullet, with a lot of force

Or think of expansion as pushing a stalled car up a hill against gravity. If you then let it roll back down what happens? It can accelerate rapidly and get out of control. It comes back down the hill with gravity a LOT faster and harder than it went up against gravity.

Same thing in an engine. During expansion your working against atmospheric pressure. The return stroke is WITH atmospheric pressure. It's like a free fall.

Then you need additional heat input to soften the blow. As the piston returns, you need a cushion of expanding hot air to slow it down

[Tom Booth]

(...) given enough cooling for the compression and heat input for expansion, I think a significantly higher compression ratio is possible...

I think you are right about higher compression being possible, though, maybe for different reasons

. I attribute this to the higher pressure swing that pulls as much air into the leaks of the system as it pulls out.

I don't think it has anything to do with "leaks", it's just that a Stirling engines capacity for converting heat into work has gone largely unrecognized, even denied entirely.

Your engine obviously is not having any problem pulling back on the return "power stroke".. Increasing the compression ratio has resulted in the engine running stronger, you're obviously doing something right.

When I eliminated "leaks" in this engine by switching to an oil lubricant, at first I couldn't get it running. I had to modify the piston. Make it shorter and slightly looser to accommodate for the high viscosity of the oil (relative to dry graphite). I think it also needed a very slight timing adjustment.

But once I got it going it had a LOT more torque.

Previously, out of the box, this engine would not even light the LED and would stop from a feather falling on the flywheel.


https://youtu.be/D6F_cDjrEEU?si=NkYggQmTe-zzSaxP

Some might think a "high temperature" engine is apples to oranges compared with an LTD but I don't. If anything, IMO, an LTD design is just generally much better at heat utilization.

Your doing some fantastic work with that little engine.

[VincentG]

Thanks Tom. You had me laughing with dragging that knife on the flywheel, something just makes you want to try that to test torque lol. I should get one of those engines to compare power output with. For reference, my LTD was just running on the little tea candle like the unlit one in your video. The flame was about 2 inches from the bottom plate. Those tea candles seem to be a pretty steady 30 watts of heat output so I think I will use them for further testing. Could you possibly recreate that test with just the heat from a tea candle?
I agree, there is more going on than we give these things credit for. Not only does the cold return stroke have significant power, but once at operating temperature the compression and expansion work appears to be very little. Hopefully this trend continues with a larger power piston.

[Tom Booth]

... Could you possibly recreate that test with just the heat from a tea candle?
...

I think that engine needs rebuilding. It ran even better with a pumice stone displacer, until the pumice stone broke.

There was this similar high temperature engine. A different subject, but this engine required a lot of very high heat to run.

After replacing the solid aluminum displacer with a wooden displacer it could run easily on much much MUCH less heat. Like a tea candle with the wick trimmed down as much as possible and the flame just glancing off the tip of the engine. It ran very quiet, smooth and steady with a wooden displacer, but I didn't try the brake test.

The wood displacer got charred earlier while running it with a propane torch, but I soon realized it no longer required so much heat with the non-heat-conductive displacer.

https://youtu.be/bQ44Rm40unA?si=mdurAkcIsFcmcWPi

I'd almost bet with oil lube it would run on "heat of the hand", but the small surface area really limits heat input.

If I find the time, I may see what one of these can do, but I really don't think a "high temperature" engine can outperform an LTD, due to the very limited surface area.