increasing displacer 'dwell' time

[tmk]

Hi folks, new to this forum, nice to meet you all

I haven't built an engine yet, so please bear with me :)

I've been thinking about how to increase the displacer dwell time, and i came up with a few ideas:




left: default engine

mid: 2 gears, with teeth on only a small portion of the outside. note the side view, they are 'double' gears, and rotate opposite directions. Perhaps a magnet at the top to keep the displacer at the top position if it's too heavy to stay (or can't be used horizontally)

right: oval shaped gears.. should change the speed of rotation.

Yes, i know my connection points/timing are wrong in the drawings.

Would these work? is it worth it? i know that the compression/expansion stages of the piston with the default flywheel configuration 'recover' some engery lost by the fixed timing by compressing/expanding the volume, thereby heating/cooling it slightly.

cheers
-tmk

[tmk]

Also, here's a clever electronic circuit that could be used to move the displacer at the correct time.



Explanation:
the switches are momentary, normally open switches. They will be 'connected' when the power piston reaches the end of its travel (or as close to it as you put them). The lefthand circle is a motor, and the righthand one is a (fly)wheel attached to a dynamo of some kind, to convert the up/down into rotation and into electricity. This could also be a linear generator instead of a flywheel/dynamo, but those are slightly more tricky to use, and less flexible (eg can't add gears to change speed)

The motor and dynamo are both DC, and the capacitor should be sized so it has JUST enough power to move the displacer motor/piston.

The 'M' in the circuit diagram is the lefthand motor, attached to the displacer piston.

OK, assume the engine in the drawing has the heat-source at bottom.

When the power piston reaches the top, it will close switch-a, which will cause the capacitor to charge, and that charge will have to go through the motor, so it will turn and move the displacer. Once the capacitor is full, it will no longer pass current, the motor will stop, and the displacer will be at the bottom.

Now that the displacer is 'blocking' the hot side, the power piston will start to move as the working gas cools. it will release switch-a, and travel downwards. Once switch-a is released, there are no viable paths for the electricity to flow down, so the capacitor will keep its charge.

When the power piston reaches the bottom, it will close switch-b. Now the fully charged capacitor will discharge in the opposite direction, and again it has to go via the motor, but it will turn in the reverse direction from before (since the current is reversed) and will move the displacer back to where we started. When the capacitor is fully-discharged, the motor will stop and we'll be back at the starting position, ready for the gas to expand.

The + and - are assumed to be the capacitor/battery that your generator is charging.

Some efficiency notes:

DC motors probably aren't very efficient, so this may not be worth the trouble vs a normal flywheel design, or some of the geared designs mentioned above.. This is because we are converting motion to electricity, then electricity back to motion. we'll lose efficiency both ways.

Also, you will need to use diodes to keep the generator motor from spinning backwards if the power-piston stops, and that eats up a bit of voltage. This means that this may not work on a small scale, and to test it, it's probably best to use a battery as the power-source, and just forgo the generator on the power piston. Gearing may also be required to get the motor turning fast enough to generate the amount of voltage you need.

Hope this is useful or at least interesting.. it's a very elegant circuit at the very least :)

-tmk

[alpha stirling]

Hi!
The electrical system is out of the question because of it's low efficiency. You easily understand it by considering the difficulty our small engines have with friction losses.
Furthermore, there's no point in large scale gamma stirling, since it has very low power-to-weight and power-to-volume ratios. Large, powerfull stirlings are often either beta or alpha stirlings of which the free-piston alpha stirling is the most efficient up to date I quess.

There's also no point (other than showcase model) to make complicated mechanical systems for small LTD stirlings. Their power output is so low that it aint' worth the efford. For something larger... why not. But with reasonably large stirlings we again hit the efficiency issue. The basic aim of your idea is to give the working fluid more time to heat and cool. This can also be arranged by facilitating the heating/cooling processes with heat exhangers, although in a small and simple engine your idea would probably work better...

On the sidenote, the piston movement of the free-piston alpha stirling tends to be purely sinusoidal, thus it prompts thinking about the utility of your idea. Don't misunderstand me, I would like to see how well your engine would run in practice.

[tmk]

Thanks for the comments.

My plan is to try and build one of these simple LTD type engines and swap out the flywheel etc with the gearing type mechanical linkages and compare how it works.

I would *expect* that the gearing models would rotate more slowly but have higher torque, but i'm no M.E. nor stirling expert so it's just a guess :)

I think most of the designs above are simplest in the LTD type, but can be adapted to other designs without too much trouble.

For my own information, i thought the LTD type were beta stirlings.. is that not the case?

-tmk

[Cartech]

tmk,

From what I've seen, all the LTD's in the small model variety are gamma's. It's always fun to make something new or different even if it doesn't work any better only that it in fact just works. The problem I see with adding electical actuators to a Stirling would be that I expect the engine to then generate the needed electricity, see my point? Then add in the energy convertion loss and your bound to lose any possible gains. Now then if your talking a mechanical system you have something. I've seen it done with cams running the displacer and I've heard of magnets being messed with as well. You mention your wanting to build a LTD model with this system and it sounds like a good project. I'd be a bit worried about friction using gears or anything with many moving parts. I'm not an expert but I've seen my engines come to a rapid halt with just the smallest extra friction from simple things, like a connecting rod twisting slightly in it's bearing. I hope it works out.

[jhead]

I am currently building an engine with a cam on the crankshaft that works the displacer. It is similar to the way a cam works a valve in a internal combustion engine. The way I plan to raise the displacer to stay in contact with the cam is with a counterweight. The cam I have designed resembles a triangle (with rounded edges) with the pivot point at one point so the maximum length of time is at the top and bottom. I am in the early construction phase now and have not tested it at this point.
joe

[tmk]

Since for the displacer, all you need to do is move it up and down, and it's not a power-producing part, you could just make a 'guide' on the flywheel, shaped such so that it would push the displacer up and down as it rotated.

picture, thousand words, etc:



The whole thing is the flywheel, and the blue part would be raised - a rail of some kind that a guide for the displacer would rest on (or be pushed into).. so as the wheel rotates, the displacer rests on the edge of the blue part and is rasied and lowered as the blue part moves up and down.

of course, this may be exactly what you meant :)

-tmk

[jhead]

that is exactly what I am talking about. I have toyed with the idea of a roller on the bottom with a counterweight to keep the displacer in contact with the cam OR a sort of hooK going around the cam and rolling on top of it and let the weight of the displacer work the downstroke and the cam would pick it up. . . any ideas which is better?
joe

[tmk]

I was thinking that a weight to move the displacer would probably be counter-productive, as you want the displacer to move as freely as possible with as little effort as possible. every bit of energy required reduces the efficiency of your engine.

At the same time though, the weight of the displacer alone may not move it fast enough; if your engine does 500 rpm, could the weight of the displacer move it down fast enough?

A spring might do the trick, and should only put pressure on the flywheel bearing. if mounted on a bolt or something, the tension could be easily adjusted. But the spring would need to be compressed each cycle, which takes work.

You could make the cam shape a 'track' that the displacer arm would ride in (with a wheel or something).. You'd just have to make sure the displacer arm was restricted so it could only move linearly.

I think the track method is probably the most efficient (no need to compress a spring), and if you properly counterweight the displacer, there should be very little force necessary to move it.

The cam idea is great in that the rotation of the flywheel produces a very predictable,controllable motion, and it's perfectly linear.

-tmk

[jhead]

the track idea sounds great but how do you make it work without friction... router in a piece of wood? Would a roller (bearing) at the top and bottom of the cam work? I was planning on a slower but more powerful engine geared to a higher speed fan. I have also toyed with the idea of a water pump that pumps water through a coil of tubing attached inside the top of the compression cylinder for maximized cooling... is the energy made going to overcompensate the energy used?
joe

[tmk]

yeah like a router in a piece of wood was exactly how i would make it.

Then just a small wheel with a good bearing that fits inside the track with just a tiny bit of room all around.. it will sometimes roll on the 'inside' edge and sometimes on the 'outside' one.

Perhaps a small lego wheel or something..

Or if the track is small enough, just a smooth metal rod riding in the track might be ok.. probably more friction that way, and the whole track would need lubrication, vs the axle of the wheel.

-tmk

[tmk]

For reference, went to the maker faire and met a guy who used to run a company who designed stirling engines for power generation.

He cleared up a few things for me, but in particular he said that it 'seemed like' changing the dwell time of the displacer should make it more efficient, and that people occasionally talked about it, but no-one had ever proven it one way or another. He said that might work, but it would increase the complexity of the engine.

He also said that the weight of the power piston shouldn't make much of a difference (he compared it to a pendulum), but the displacer should be as light as possible.

I asked about building regenerators into the displacer, and he said that some people do that (mostly in cryocooler stirlings) but that it makes the displacer heavier.

-tmk

[jesterthought]

A simple way to get displacer dwell would be to use excess stroke and a springy conrod. The compliance needs thinking about, maybe.

Energy used in compressing the spring (at either end) is returned past t or bdc. There might be some loss due to any increased loading of the big end and crank bearings causing greater friction.

I have yet to try this but plan to see if the swings beat the roundabouts.

Jester.

[spinningmagnets]

I recall a patent about 100 years old using oval gears to increase dwell time of a double acting steam piston. I Googled "oval gears' and one link looked useful:

http://decidedlyodd.com/cw/cs285/log.html

As for camming, I would start the first prototype with a vertical displacer (for visualization purposes) and the displacer rod tip would have an "F" shaped end. One of the horizontal arms of the F will be attached to one side of the vertical rod, and the other arm would be other side.

The horizontal camshaft would have two cams side by side. When a given cam wasn't lifting/depressing its F-arm, there should be an air-space equal to the lift of the other cam. The displacer should be centered by light springs.

You could also do something similar with one horizontal arm, using a cam above, and a cam below. I'm partial to the single camshaft.

A third option might be to use a crank or scotch yoke (T with a hollow horizontal bar) that has a longer stroke than the displacer cylinder can fit.

The displacer piston rod can be two-piece, a hollow pipe on the bottom 3/4ths, and a solid rod on the upper part. The bottom of the upper rod would have a circular plate attached that just fits the ID of the lower pipe-rod.

This flat circular plate could have a spring on the top, and another on the bottom, with appropriate ring stops above and below that are attached to the pipe ID.

When the displacer piston bottoms out against the stops inside the cylinder end, the rod end will continue to cycle.

All three ideas would work, but whether any of them are worth the extra trouble, cost, effort, complexity, and efficiency loss, I don't know.

Just a few ideas to kick around...

[spinningmagnets]

Sorry, jesterthought, I hadn't read your post yet. Two separate guys independently invented the telephone at the same time. Bell patented it first, can't remember the name of the other guy.

He probably was headed out to the patent office before Bell, but first he had to drive his wife to the store in a horsecart (to look at shoes and dresses) while she complained "why don't you ever invent something useful?"