Micro LTD Advice

[Paulie1982]

Hi everyone,

I was just wondering if anyone has any input. I am trying to build a really small LTD Stirling Engine. Its approx 50mm tall and 25mm in diameter.
The displacer cylinder is also 25mm and 14mm high. The problem is that i absolutely suck at maths. So far i've been going by the trial and error method, i think i may have gotten away with that with a larger stirling but not with such a small one. The problem is that i don't know what the ratios of the volumes of the displacer cylinder and power piston should be etc.

Any help at all would be appreciated.
Thanks

[Ian S C]

Paulie, I read some where on the net a ratio of 20:1for LTD motors, for yours I'd try something like 6 mm to 8 mm, the volume ratio can be adjusted by altering the stroke, as I did on my 6" dia LTD. It would be great to see a little motor like yours running, you don't happen to make watches in your spare time? I have heard of a smaller high temp motor, a V type with a piston of maybe a bit less than 1/4" dia, and a displacer to match. Ian S C

[jimlarsen]

Jan Ridders has made some micro Stirling engines. You can see one here. http://www.machinistblog.com/a-couple-o ... stirlings/

[Ian S C]

Hi Jim, thats small, but did you look at the smallest, I wonder which has the most power, the motor, or the house fly? That one is now the smallest I'v seen a photo of. I believe that there have been ones built at microscopic size. Ian S C

[jesterthought]

Hi Paulie, I like tiny engines, too. Have you seen these? At the end, a tiddler is going fairly fast:
http://www.youtube.com/watch?v=dICFVknQ7_Q
Here’s some more: http://www.stirlingparadise.de/coll_sch ... hager.html. One is a “thumb nail sketch”: http://www.stirlingparadise.de/coll_sch ... aum_99.jpg.

Those are LTD engines, as developed by the late Ivo Kolin and James Senft. Senft has written several practical books on hot air engines including “An Introduction to the LTD Stirling Engine”. They are listed on http://www.amazon.com/Miniature-Ringbom ... 0965245535 and I found them all really helpful.

But LTD engines have almost no power because they draw very little heat, so making them small is tricky. Maybe start with a higher dT, like this one (which also shows that you don’t have to use a conventional layout.): http://www.youtube.com/watch?v=5t5qb0GWnsg.

You ask about displacement ratio. The 150mm LTD model in Senft’s book uses 15, which should be a good starting point. (Note: this is a swept volume ratio not a diameter ratio.) The ratio is independent of size, although it is higher for low dT.

For interest the principle is that air expands by 1% for each 3degC temperature rise from 27degC – eg, by 10% between 27 and 57degC. But that expansion has to do more than just fill the displacement of the power piston. It has also to produce enough pressure to provide the torque – only “half” the expansion is used to move the piston, the rest to give the force. Only the displaced air (swept volume of the displacer) gets heated and, sadly, most of that doesn’t get right up to temperature. Worse, some expansion is wasted through leakage and some through compressing the inevitable dead air. So don’t worry about maths, it doesn’t work anyway - the displacement ratio needs to be much more than calculated. I’ve used far higher than 15:1.

Miniature engines are a challenge! Because the power is so small, friction must be minimal. The bore is best lapped and the piston lapped to it. I find friction is worse than slight leakage, so ease off the piston. Graphite is best for pistons (try old motor brushes). Also, I use light flywheels and thin wire crankshafts to minimize the friction torque in bearings.

An advantage of wire crankshafts is that an extra bend in the crank arm can be re-set to vary the stroke (use a round bend or it will eventually snap).

Another problem with small engines is the limited cooling area. That allows the “cold end” to warm up and stop the engine. So I make the displacement cylinder from wrapped thin film to minimize conduction. I use longer cylinders and thicker displacers than Senft, too, to reduce the temperature gradient (and conduction) and to improve regeneration (in the gap around the displacer). Balsa displacers are less compressible than foam (equivalent to dead air) and take higher temperatures.

It is good for the displacer to rest against the end plates (as in a Ringbohm engine – see Senft), to minimize dead volume, as well as heat up its faces which then help heat the air.

To micro-ize, can you re-arrange the mechanism so that the flywheel lies parallel with the top plate – like closing a powder compact?

Here’s an engine with no flywheel at all: http://www.youtube.com/watch?v=kkpiq8wwRrc&NR=1. (See it slow and stop as the cold end warms!) Its works: http://www.youtube.com/watch?v=lljxMZTBIyU .

Good luck with your fascinating project. Keep us up to date.

Jester.

[Paulie1982]

Hello Gents,

Thanks for all the great advice, i'll definitely look into all that has been mentioned above.
Ian, i've never built watches, ha, i wish i did, maybe then i wouldn't be having so much trouble, i also wouldn't mind a watchmakers lathe, that'd come in handy.

Jimlarsen, Jan ridders disgusts me :), his engines are always masterpieces, he's a great man to learn things from though.

Jesterthought, not only have you given some great info, you've also provided me with quote of the week "So don’t worry about maths, it doesn’t work anyway" :).

Once again fellas, thanks again, i'm just about to start the production stage, i have the 3d model designed and tomorrow i will start re-making parts - i've tried already and was not happy with it, so i'm going to start from scratch.
I've set up a blog to chart my progress, i don't know whether this is a good idea or not as the chances of me failing to create such a small LTD are high and there will be proof of my failure on the net for a hundred years, ha, oh well, its worth a bash. If any of you have blogs or youtube add me, i'm sure i'll be picking your brains for quite some time ;).

Blog-
http://microstirlingengine.blogspot.com/
Youtube-
http://www.youtube.com/user/paulie1982?feature=mhee

[jesterthought]

Oops! When I said the displacement ratio of Senft's LTD engine was 15, I quoted the wrong engine. The ratio he uses in N-92 hand held engine (designed for 6degC dT) described in his LTD Stirling Engines, is over 150 - nearer the ratios I've used for really low dT. Sorry to be so misleading when I meant to help.

When I said that maths didn't work I didn't mean by getting the maths wrong! I was referring to the academic approach to thermodynamics, for example. Senft describes the difference between the theoretical Carnot efficiency of an engine operating between a 37*C hand and a 20*C room = (37-20)/(37+273) = 5.5% (only!) and the actual practical thermal efficiency of nearer 1%, after allowing for the air not reaching the temperature limits, skin being an insulator, heat exchange losses, flow losses, conduction, etc. This is further reduced by friction and leakage. Most coffee cup engines only just go, producing no power at all. Thus, by definition, their efficiency is zero. So much for maths and 5.5%.

Not only is the efficiency very low, so is the heat input from a hand, for example. No wonder that it takes longer to make an engine go than does to make it. The upshot for us is that low dT requires very careful design and build - especially for micro engines. That's the fascination!

Jester.

[jesterthought]

My user personal album won't accept a brief video so I've just uploaded Poppette on You Tube: http://www.youtube.com/watch?v=4l_ss9OcLdU. It is 4mm dia x 10mm high.
It is running on a 70*C hotplate, but only just, so it is not really LTD although it uses similar constructional techniques. Its frequency is around 600Hz (the variations must be due to strobing with the frame rate). Despite no cooling fins it seems to run indefinitely.
The simplicity of the basic design allows miniaturization. However I prefer to see engines working, so I hef plenze. It's a bit tricky, though, so don't hold your breath.
Jester.

[celinaaniston86]

Oops! When I said the displacement ratio of Senft's LTD engine was 15, I quoted the wrong engine. The ratio he uses in N-92 hand held engine (designed for 6degC dT) described in his LTD Stirling Engines, is over 150 - nearer the ratios I've used for really low dT. Sorry to be so misleading when I meant to help.

When I said that maths didn't work I didn't mean by getting the maths wrong! I was referring to the academic approach to thermodynamics, for example. Senft describes the difference between the theoretical Carnot efficiency of an engine operating between a 37*C hand and a 20*C room = (37-20)/(37+273) = 5.5% (only!) and the actual practical thermal efficiency of nearer 1%, after allowing for the air not reaching the temperature limits, skin being an insulator, heat exchange losses, flow losses, conduction, etc. This is further reduced by friction and leakage. Most coffee cup engines only just go, producing no power at all. Thus, by definition, their efficiency is zero. So much for maths and 5.5%.

Not only is the efficiency very low, so is the heat input from a hand, for example. No wonder that it takes longer to make an engine go than does to make it. The upshot for us is that low dT requires very careful design and build - especially for micro engines. That's the fascination!

Jester.

Hi jesterthought! I have read the other replies but I think your reply is very convenient for this problem. Really you have great knowledge.
Nice explanation!