Stirlingengineforum.com

In searching the forums I haven't found much detail on linear alternator design, and I wanted to open a discussion on the subject.

Alternators / generators work by forcing wire coils to cut through magnetic fields. There are lots of ways to move magnets near coils that are not productive, and linear alternators are different enough to be non-intuitive.

I believe the commercial folks clearly discussed radially facing magnets. To me this implies lots of small round magnets in rings that alternate North or South facing out. This would seem to project lines of force into the surrounding coils.

I've also seen the standard piston shaped magnet sliding back and forth inside coils. Yes this will work, but is it optimal? I have doubts about this one.

I've also read about mounting magnets on edge such that they slide between narrow slots, this seems to me to be a very optimal solution but one which is more difficult to fabricate.

Mr Barumman's Thermomechanical Generator moves a magnet a few millimeters such that it alternately completes and breaks a magnetic loop that includes the magnet and laminated metal with coils wrapped around the lamination's. Magnetic lines of force love to go through loops and this would seem to be a fairly effective method, especially considering the power he seems to get out of his modest device.

The mention of power brings an issue to mind. Everyone loves measuring voltage. Voltage means nothing. 1000 volts at 1/1000th of an amp is still just 1 watt. In any meaningful discussion of alternators Watts are what are important. I would gladly take 1 volt at 1,000 amps (1kw) over 12 v at 1 amp (12 watt).

I would bet linear actuator design would be a good source of design material, if it makes a good linear motor then it should make a good linear alternator, unless of course they are simply brute forcing the pistion-insde-coils by slamming it with power.

What are your thoughts on great designs for linear alternators? What are the best and the worst that you have seen?

Dave

It didn't occur to me at first that this post might be a bit off topic for the forum, but since I've started it I'm going to run with it for a bit.

At its simplest, a generator is wires cutting through magnetic lines of force. The more lines of force that can be cut, the more wires that can cut them, and the faster the wires move the stronger the generator. Generators are generally configured such that continuous lines of force are available to saturate across the area the wires cut through, either by using multiple magnets, or a horseshoe shaped magnet (for classroom use), or by ducting the magnetic lines of force alternately through a core that the wires are wrapped around.

One common theme is that multiple magnets or multiple poles of one magnet create a continuous field which contains the wires. Either the field or the wires might move, but the field through the wires is created by pairing magnets:


The other common theme is to wrap the coils around a core and drive an alternating magnetic field back and forth through the core. Each time the field flips all of the lines of force must cut through the coils, creating an efficient generator.

This second approach is similar to the concept implemented by Barumman.

After seeing these, I get a clearer picture of what the design of a real linear alternator would look like. One obvious method would be to stack numerous C shaped stationary laminations wound with wire and slide alternating magnets through the gap thusly:

Another alternative would be two rows of magnets with a wound lamination between the rows thusly:

which is an image directly from the wikipedia article on linear motors.

While reading the book "Stirling Engines - the next big thing" I kept screaming inside as they twisted themselves into engineering knots trying to make a 60hz cycle. I wanted to smack them with the book and tell them to rectify whatever power they got into DC and feed it into power shaping circuitry exactly like a solar or wind inverter. It was a nice hardcover book, smacking them with it would have been very therapeutic for me. It also seems apparent from reading that book that the design they described was the first linear design above, though now the literature on their web sites describes radially facing magnets. I can't get my head around how radial magnets would be beneficial. The goal would be to create a magnetic field all the way across the coils and then cut the lines of force with coils. Hey, that would work if the coils wrap in a full circle around the linear axis, but honestly it seems more complex than it's worth:

I dismiss the idea that this could be anywhere close to what they are really doing, but I can't think of another radial design that would offer a saturated magnetic field for the coils to cut.

Here is an interesting configuration that will not work, a saturation field created by two rows of magnets with a coil of wire as pictured:

What direction does the current flow? This is the Right Hand Rule for generators.

The first finger points in the direction of the magnetic field (first - field), which goes from the North pole to the South pole.
The thumb points in the direction of the motion of the wire in the magnetic field (thumb - torque or thrust).
The second finger then points in the direction of the induced current in the wire (second - current).
So, when a wire moves perpendicular to a magnetic field, cutting through the magnetic field lines, a voltage is induced across the wire. The faster the wire moves, the larger the amount of induced voltage and current. If the wire moves parallel with the magnetic field such that it does not cut through magnetic lines of force then there will be no induced voltage or current.
If the coil pictured above is moving upward then the right hand rule tells us to point the first finger left, the thumb up, and the second finger points toward us. If both sides of the coil are pumping electrons toward us then it isn't going to circulate in the coil and you don't have a generator. Seems almost the same as the radial design just described, except that the radial design would actually work whereas this one will not.

How about the much beloved magnetic rod oscillating inside a coil of wire? Yes it creates some electricity as it will inevitably cut some magnetic lines of force with the wires. It's easy and cheap to make, but without a continuous magnetic field to cut through I doubt if subsequent outer layers of wire are actually doing anything. The inner few layers are probably cutting fields and giving you some power, but I don't think you are going to get real energy from it. Just my opinion, feel free to argue a different point.

Dave

Linear alternators are at the heart of those "battery-less" flashlights that you shake back and forth to light the LED. Their problem, as I see it, has little to do with the alternator design, rather the energy losses when reversing the mass of the moving magnet (or coil assembly) at the ends of each cycle. Add to that frictional losses on the moving magnet (or coil assembly) which are likely to be much greater than losses on a well made bearing for a rotating mass. Those two factors I think make an inexpensive efficient linear alternator impractical.

My interpritation of the linear alternator; for the magnet take a magnet from a microwave oven, or a large speaker, this is mounted on the end of the pison rod. Now for the coils, I use modified transformers. Remove the laminations, and reassemble with all the Es facing one direction, you don't need the I s, I use four of these on my free piston engine the coils are placed around the magnet, with a gap of a few thou, there are three faces of the cores that the magnet passes, the magnet is in the form of a disc with a hole in it one side north, other south, the coils can be connected in series, or parallel, to give the required voltage/current. Ian S C

Hey DavesPlanet,
Here is a video of a Stirling, Free piston cryo-cooler..

[youtube]http://www.youtube.com/watch?v=Ig3PrQBdf2c[/youtube]

There is a cut-away section starting at about 55 secs and if you pause it at about 1.05 you can see the linear motor winding setup. Depending on the magnet configuration in the power piston, maybe the linear motor can be used as a generator as well?? like in some conventional electric motors.

Coleman also made a 12v Stirling cooled camp fridge some years ago. I have often wondered if they could be used as a free piston stirling engine without any real conversion trouble, even with a generator already built in. (Buy a second hand fridge for $300 and 'bang' you have a free piston stirling engine). Its pressurised with helium as the working gas. The linear motor has a consumption rate of 48 watts.
http://r1.coleman.com/Manuals/5726-750.pdf





The cooler in the video and cooler in the Coleman could originally be from the same company.. maybe. At least they look like a very similar configuration to me.
I think the Coleman 'compressor' is Manufactured by 'Stirling Ultracold'..

http://stirlingultracold.com/global_cooling

vamoose

some examples of linear alternators/motors for stirling engines..



It could be possible to use the linear 'motor' of a speaker as a generator also,
and use the speaker diaphragm In Situ as the spring of the power piston...



and one could even consider modifying another speaker (ie, remove windings and/or magnets) and using this floating diaphragm as the displacer spring. It wouldn't be a tin can Free piston Stirling as such, but could reduce requirements for machining and sourcing parts etc for hobbyists..

It might also be worth considering the use of speaker diaphragms in conventional Alpha Beta and Gamma Stirling piston designs with flywheels etc. But that's probably another thread..

vamoose

vamoose, the speaker idea is ok, but the voltage output, and current would be very low, it would require a bit of modification, like a whole lot more wire, but that is a basuic linear alternator. Ian S C

Great examples, thanks!

What's with the odd triangles indicated in the displacer piston? I've seem them a lot but never understood what they were representing.

Dave

I think as an experiment, I'll try using the stator/ field coils of an electric motor to see if that is effective or better than the modified transformers that I'v been using up to now, just got to find one with the right sized hole to take the magnet. Ian S C

DavesPlanet wrote:What's with the odd triangles indicated in the displacer piston? I've seem them a lot but never understood what they were representing.

Dave

I've also wondered, then recently came across this image with domed Caps/Ribs in the displacer.
It's a cut away of a 300w CHP engine sold by Viessmann, that gives a good view inside, and another better quality pic that gives a very nice glimpse at many of the engine's internals.
(i'm not sure weather they make this generator or on-sell it...?)
(Ad Edit: I think its made by Microgen)
http://www.viessmann.com/com/content/da ... boiler.pdf



I have not read what they're for, but i'll have a crack. I imagine they give the displacer piston some internal rigidity for coping with the oscillating pressures on the walls (without increasing weight too much), and the domed shape helps with reducing the effect of the heat expansion of the metal which could change the piston diameter significantly as the displacer and ribs start cold and progress up to operational temperature.

vamoose

I'm not sure either, but my displacers have a light weight disc about half way along, I attach the rod to this, and pass it through the cold end cap, some say that the disc helps to restrict the transfer of heat via the heated air in the displacer, it also makes the rod more stable than it would be if just attached to the cold cap end. Ian S C

Yep Ian, that makes sense, thermal (convection) baffles.. (plus with some other added pay-offs?).

Found this image with a layout of the components from (i think) an 'Infinia' free piston engine..
(its low resolution, but is the best quality I could track down). It helps to see the magnets of the power piston and also the generator stator lamination's and winding's.



vamoose

For our size engines the donut shaped magnets ex microwave oven, or speakers, the letter take a bit of work separating them from the speaker. These are easy to attach to the motor. Ian S C

RonT wrote:Linear alternators are at the heart of those "battery-less" flashlights that you shake back and forth to light the LED. Their problem, as I see it, has little to do with the alternator design, rather the energy losses when reversing the mass of the moving magnet (or coil assembly) at the ends of each cycle. Add to that frictional losses on the moving magnet (or coil assembly) which are likely to be much greater than losses on a well made bearing for a rotating mass. Those two factors I think make an inexpensive efficient linear alternator impractical.

Though it's technical problem little bit critical.But i think Linear alternators have greater efficiency. also more compact because of it's higher output for a minimum amount of power required to turn it.

RonT wrote:Linear alternators are at the heart of those "battery-less" flashlights that you shake back and forth to light the LED. Their problem, as I see it, has little to do with the alternator design, rather the energy losses when reversing the mass of the moving magnet (or coil assembly) at the ends of each cycle. Add to that frictional losses on the moving magnet (or coil assembly) which are likely to be much greater than losses on a well made bearing for a rotating mass. Those two factors I think make an inexpensive efficient linear alternator impractical.

I prefer rechargeable battery flashlights rather then battery less flashlights. It's my personal opinion. Because I'm using flashlights for a long time but I've not got any single problem while I shacked back it or doing other funny things.