linear alternator

[rotornuts]

Last year I think Monte350c did an exellent conversion using metalic poles with magnets sandwiched in between with the poles facing each other.

I was thinking this would also work nicely to produce a linear alt as circular magnets polarized around the rim are a hard find.

Here's an image of my thoughts

Aside from why I would want to do this and how will I drill the magnets(diamond bit/low speed). Does anyone see any big issues with this. I realized the spacing of the coil/magnet relationship probably isn't the best but experimentation should help.

I have a couple dozen 1/8 x 1" Neo's so I'm into playing around with them.

Mike

[maker of toys]

Rotornuts-

what's the central shaft gonna be?  will it be strong enough to take the force reversals without fatigue?  how about the potential for a magnetic short circuit?

(and why not just get some flat-pole torus magnets to experiment with?  for a first cut, a bunch of ceramic toroids would do dandy. . .?)

other than that, it looks remarkably like an energy-recovery shock absorber.  I noodled around with a similar idea, but never got to the energy recovery stage; I was looking for a fade-free cockpit adjustable shock for a college project.  (formula SAE)  turns out to be too much unsprung wieght using ceramics.  Plus, of course, the heating and wear problems.  (even on a track, there's a remarkable amount of dirt gets into the underside of a car.  I'm sure I could have licked most of the troubles, but mountain bike shocks were cheaper, lighter and easier.)

just being nosy,

-Dan

[rotornuts]

The Neo's are 1/8' so if I double them between ferrous spacers of 1/4 then I'm thinking I'll use about a 3/8's stroke. I was considering brass for a center shaft to preclude the need for a   roller bearing. A highly polished brass on polished steel bearing surface shouldn't present alot of friction and avoids the magnetic short. Brass should hand the load just fine. The actuation will be via a lobed cam converting rotary motion to linear. The slope of the lobes will be slow(3 lobes around the circumference).

I'm bouncing this idea around my head to get around the low rotational speeds of a Vawt. I suppose it would also work in a dynamic feedback kind of shock absorber too but I can imagine the controls being complex to provide an appropriate dampening response. Energy recovery or using it as an alternator connected to a vehicles suspension would be a fun experiment.

Nasa has been playing with linear alternators power by stirlings for a little while now(built on contract from the patent holder)

Sandia Labs is  playing with a two stroke hydrogen engine the runs a linear alt directly off the piston firing in both directions.

The applications for a linear alt are numerous once you start seeing all the linear motions kicking around. We have all trained ourselves to covert linear motion into rotary because of the internal combustion engine and the need to turn wheels.

Mike

[maker of toys]

Analog electronics to the rescue; a user operated switch inteface chooses the jounce and rebound rates via resistors, simplistic FET hardware with resistor-capacitor-opamp/comparitor frontends handle the damping by regulating the volts across what would otherwise be a short-circuit once a threshold dV/dt is reached.  simple. but not perfect; the diode drop required to keep the FETs happy was a stumbler.  took me about 10 minutes to wrap my head around the idea that a short offered more damping than a 10 megohm resistor; after that i was hooked.

dumping that power into a battery is a bit more complex, but doable. (mainly because there's a lot of 'fretting' that goes on that needs damped but doesn't generate enough volts to 'cut in'.)

Now if you wanted to eliminate the springs and go to a fully active suspension, THAT gets a little more involved.  . . .linear encoders, uControllers for each corner, a serial bus for them to talk to each other. . .  it also eats power. . .  Not really cost effective; that much copper costs and weighs more than the steel it replaces.

actually, i'd say the linear-to-rotary training started with Watt, et al. . . as soon as they figured that there were plenty of things to do besides pumping water. <G> or maybe even earlier, when some un-named dreamer started thinking of a way to get the wind or a river to turn the grindstone in his grain mill or pump his water so he wouldn't get dizzy himself.  funny how energy mostly wants to go straight, and humans mostly want it to go in circles. <G>

and now here we are, with all this nice rotary motion, and we want linear motion from it.  (geez. . .humans! <BG>)

there's been a little undercurrent of free-piston engine research for the past 30 or 40 years; and I have to say that there's an attraction there. . . steam-powered air-compressors and waterpumps are one very practical application.  (provided that you have steam available. . . )

Back to topical matters:

'Bearing' bronze on steel might be a little better from a wear standpoint. but if you already have the brass, run with it.  Lubrication is going to be the make-or-break on this, not material choice.

What are you using for restoring force? I'm thinking weights; gravity is free and it doesn't leak or rust. <G>

Occurs to me that setting this thing up to operate trip-hammer style might be the mechanical equivalent of a boost converter-  push the alt slow one way, then a fast return the other to generate power at a predictable, useful voltage even in very low winds.  It'd probably have a lower average current (even in good wind) than a purer version of your vision, and cogging (and noise) could be more of a problem, too. tanstaafl. sigh.  Still, if the grid's a long way away, some power 70% of the time is better than a lot of power 5% of the time. . . .

-Dan

[electronbaby]

in your posting (picture) you show like poles facing each other with an iron spacer in between. The only way you can do this (i think) is if the iron had enough mass to absorb the flux field. Otherwise they would repel each other. If the flux was absorbed, it would then not radiate any flux (or a weaker pattern) into the coils. Not sure this would work the way it is drawn. Maybe im missing something. :-/

Roy R

KB2UHF

[monte350c]

Hi Mike,

An interesting thought!

As you probably gathered from that conversion I did a while back, one of the trickiest parts is in assembly of the magnet stack. But all that repelling does result in a pretty good field ejected from the iron 'pole' pieces.

I wonder if you could use a very thin 1" ID tube to assemble this thing into. Something non-magnetic - like some stainless, brass, or aluminum tubing. The thinner the better since it would be in the airgap. Then you could create a way to hold the tube at the ends. This could be something as simple as an endcap with appropriate fittings on it. If the caps and the ends of the tube were threaded, you could put the squeeze on the stack of pole pieces and magnets by screwing the caps on.

The pole pieces could just be cut from plain round steel bar stock.

I am getting a headache thinking about the coils....

This sounds interesting - keep us posted!

Ted.

[finnsawyer]

I wouldn't do it this way.  I'd use alternating bar magnets arranged vertically with non magnetic strips between them.  I'd then make the coil assembly the movable part.  As a final touch to increase performance, I'd place an iron plate above and below the entire structure to concentrate the magnetic flux and increase performance.  I assume you are aware that if you drive this with a crank off of a wheel your output will be a sine wave.

[finnsawyer]

They will repel each other in any case.  The iron doesn't "absorb" the flux.  It only channels it.  Flux can not be created or destroyed by unmagnetized iron.  The flux must flow from the north pole of the magnet back around to its south pole.  In doing so it is squeezed due to the presence of the same polarity pole opposite it and forced to go up or down.  This squeezing results in the repelling force.  As the intervening iron becomes wider the force rapidly becomes negligible as the squeezing effect drops off.  

[wphfla]

Check out this link.  Gives plenty of formulas and results using the type alternator your proposing.  Am currently building one with my son for his science fair project.

http://www.lancs.ac.uk/staff/bakern/papers/PEMD2004.pdf

Kevin

[monte350c]

Thanks for the link.

Wow. Those guys got a grant to play around with this! I'm envious.

Ted.

[windstuffnow]

  I've been toying with the idea of linear alterntors myself.   Although, not for the wind turbines... more toward a regeneration system for an electric vehicle using the suspension movement.

  I've gone over the mass of movement and either moving the magnets or coils there is a tremendous loss on each end of the swing.  More weight means more input power needed to make the transition.   You can, however, use a dashpot-spring-etc. to help store the energy on one swing to reuse.

  I've been leaning in another direction... which isn't set in stone as yet but the numbers are far better.   All the details haven't been worked out either.   In any case, a geared assembly that rotates the magnets.   This cures the biggest problem, the mass of moving a row of steel and magnets.   You can use long square magnets over the coils which also reduces rotary mass.  All we need to do is change polarity based on an input force.   As well, you don't have to have speedy input just brute force, the spin of the magnets make up the difference between the two.

 Just some thoughts to ponder...

.

[Gary D]

Ed and others, there was an article earlier this year of someone developing a backpack charging system for the military. The article claimed 4 to 7 watts continous charge with a hinged 30 to 60 lb. pack (while hiking). Not sure how you'd google for it (or do a patent search),or even if it would give you any relevant info. but a car weighs a bit more...  Gary D.

[rotornuts]

Thanks everyone for the comments and thank you very much Kevin for the link, that's exactly what I'm talking about. It's far simpler to get round Neo's polarized axially.

Ted, I was going to part off 1" round 1/4" thick discs on the lathe after boring a 1/4" hole through the center. Using a 1/2" brass rod I was going to turn a section to 1/4" long enough for the stacked assembly plus generous extra to accommodate the repulsion. If I thread the 1/4" section to just beyond the thickness of the stack then I could use magnets across the stack to short thier flux(like Ed suggested) and bolt the stack together against the 1/8" shoulder of the 1/2" section. The Neos would need to be drilled. Cut the extra off after assembly and surface the end of the 1/2" side with a radius.

As for the coils. I think if you just wound them in a round coil winder of about 1 1/8" diameter then stack them in a form with 1/4" spacers between then fill with epoxy leaving the outside of the form oversize the stators should be a cinch. I guess a frame would have to constructed for running the ends of the stack through, mounting the stator and providing mounts.

An image of how envision using them

 



That would be a 2 phase orientation. so that they counteract the return springs and balance out. I think with a 5 lobed cam you could use three for three phase and still have the forces balanced.

Mike

[kitno455]

you will have pretty extreme vibration with that design, as you have both mag pistons moving at the same rate, in the same direction.

suggestions:

1. sinusoidal cam engine. do a google search
   
2. why does it need to be linear? wrap the thing into a big hoop, and spin it.
   
3. why do the mags need to be on the inside? i could see a toroid spool of soft iron wire as a core, with the copper coils around that, and then a pigot-style dual rotor around that. the rotors would need to be shaped like a bundt-cake pan, and the mags would have to be the same pole across the gap, but you should be able to get the entire coil to be under a mag...
   
4. linear alternators dont make quite sinusoidal power. with the constant acceleration,  you tend to get periods of no output as the velocity approaches 0.
   
   

allan

[MountainMan]

Nuts,

I'm just curious, when you came up with the idea were you trying to solve a particular problem, or just letting your creative juices run all over the floor?  Either answer is laudable in my view; just curious which it was.

jp

[rotornuts]

A bit of each actually. I just like to think of diferent ways of doing the same thing(generate power).

The first time I thought of a linear alternator was when I had an idea for a tall slender wind generator that has a single blade facing directly into the wind standing vertically.

The idea was to fix the blade at the top with an elastometric type fitting of some sort to afford some vertical(up and down) movement. The base of the blade would be fixed on a bearing point running from front to back to allow some sideways movement(although a fixed base may work better). If the blade is made to the correct elasticity then with a quick stalling profile you should be able to get the blade to vibrate slowly in the wind. A realy intelligent design would have it occilating back and forth as it lift/stalls oneway then lift/stalls the other way. Fix a linear alternator in the middle and presto you have power. You could maybe do it by fixing the trailing edge of the blade to a tether.

Have you ever taken a blade of grass and put it between you thumbs so you could blow in between your thumbs an make a squeaking sound? It's the same idea.

Mike

[windstuffnow]

  Mike,

   I made an osolating wing turbine with some similar thoughts on running a linear alternator.  It was based on an experience I had shortly after completing an Avid flyer experimental.  It incorporated the drooped "flaperon" system.  It worked for both flaps and ailerons.   Any way first flight out at about 300 ft there was a vibration, started small and became more prominant.  I decided not to take a chance and lined up on final when it got real bad and BANG!  Luckily I was on final with no real wind problems and all went well.   found the flaperon had sheared 4 3/16 bolts from their mounts.   Aileron flutter, is dangerous and even deadly.   Like a venition blind in the window with the wind catching it just right it goes wild!   Very powerfull.  Anyway the turbine was based on a controlled version of aileron flutter.

   It only worked at one wind speed and wasn't very predictable.   It may have worked alright with an adjustable balance but I never made it that far... another dust collector.   After that I had made the "flappy" turbine... shown below... it was slow but quite powerful, converting linear movement to rotary...using lift on both up and down movement.  From the little unit I calculated the output of a wing I have in the hanger ( 4.5ft chord x 14ft long ) and the lift in a 20 mph wind against the machine would be fairly overwhelming.   Not to mention the fatigue created in moving it in an up and down motion over extended periods of time.

  Lots of crazy stupid ideas... but I sure do have alot of fun !

[rotornuts]

I remember you posting the "walking" turbine as I like to call it. I like the "crazy" ideas. Although they may not always work they're an invaluable tool for advancing knowledge and something learned from every crazy idea can usually be used to help out a sane one.

Somehow I'm not surprised you've tried this. It's something I'd like to try someday for kicks but I have too many other crazy ideas to try first.

The other reason I want to try my hand at a linear alternator(and the lastest motivation) is winter is on it's way. I'm sure many others have thought about it but I can't help but wonder what the potential is to utilize the tempature differential between the air and the ground 6 feet under when it's 30 below to run a stirling engine?

Mike

[windstuffnow]

  I agree about the stirling in winter.  I often thought a large low temp unit through the wall might do quite well.   I've actually been collecting parts just for that purpose.  I worked out a design some time ago using easy to find cheap parts.  So here's the plan...

Displacer cylinder - 55 gal drum

Displacer piston - layered 1" urethane foam

Power piston - 10" diameter inner tube

The drum needs to be the type with a removable lid - the seal will be silicone rubber when completed.  All the components ( flywheel, crank, displacer rod etc ) will be built into the lid.

Now when the unit is assembled it can be stuffed through a wall for winter use. Actually I would most likely use a smaller hole to duct the cold in through an insulated dryer vent tube and use a low watt fan.  In the summer it can be placed in a water base with a solar batch heater.  I'm estimating about 40 to 50 watts from it.... or it won't work at all... It will be fun to build!  I have just about everything I need to build it except time.

I haven't figured out how much heat (BTU) loss from the house to run it in winter though.  It might be better to run it off the radiant chimney heat from the wood burner.  The foam should handle up to 200 degrees which might make it a bit more powerful.   Anyway, another wacky idea that I have to try...  I'll bet I never finish my "todo" list in my lifetime.  

.  

[rotornuts]

Once again Ed i'm not surprised you had similar thoughts. I don't think I'll ever finish my to do list either. Too many ideas get spawned by other ideas that came about from an idea to solve a problem associated with another idea that I don't even remember anymore!

Mike

I might just sink an old oven thermostat probe in the ground before it freezes to see what the tempature stays like down there this winter. I'm thinking between 40 to 60 degrees differential when it's 0 to  -20 outside.

[maker of toys]

 I agree that doing this sort of thing with a single-throw crank would be bad.

but, while there will be an unbalanced inertial force on the bearings, I dunno if the vibration will be all that worrysome, and (in any case) could easily be countered by putting on a second cam arrangement with another pair of in the alts 180' out of phase.  

my reasoning:

a) this is for a low-RPM rotor, with inherent damping on the oscillating force.

b) the rotor and structure will mass much more than the reciprocating parts.

c) there doesn't seem to be a significant torsional vibration or unbalanced rotating mass problem.

d) the motion will be simple harmonic in one plane- easy to counteract.

e) the stroke proposed is miniscule compared with other dimensions of the machine.

so as long as the bearings can withstand the inertial loading, i don't see as there's a problem.

the point about there being significant lost motion vs. power generation is valid, and IMHO, unavoidable with reciprocating motion.

Rotornuts has taken some steps to mitigate that by making the stroke 50% longer than the magnet (coil?) spacing. In any case, the design goal, (as i understand it) is to enable some power generation below the RPM threshold that a rotating field alt would cut in.  And, of course, to have fun. <G>

-Dan

[rotornuts]

Thank you Dan and of course the goal is to make power at low rpms and definitely have fun at the same time. A linear alternator would be a great fun project and I can see myself running around looking for linear motions to connect it to.

I doubt I would build the three lobed cam(I was getting tired of generating the image so I skipped a couple steps) I would likely built a five lobed cam and use three linear alts so the vibration potential, which I think is negligable anyway, wouldn't be a problem at all.

As far as waisted motion is concerned I don't really see it. As you noticed I've made the stroke longer than the coils are wide so really when the magnet stack changes direction it's not much different than when a magnet passes the center of a coil on a regular alternator.

It would be nice to build a set of cams using different stokes to see if passing a couple coils each stroke would work better but I'm thinking a single coil a stroke will be best as the additional travel would require more return spring travel and more force in general. Larger diameter cams would move the work farther from the fulcrum of the lever(cam lobe) reducing the mechanical advantage.

Still some details to work out but I think the idea is prototype ready.

Thank you for actually looking at the design.

Mike

[elvin1949]

Mike

 I had a thought.

put your linear alternator's in a circle

around a single throw crankshaft.

Think radial aircraft engine.

later

elvin

ps Ed will know what i am talking about.

[asheets]

You might want to check with Bose Electronics... seems that they are trying to come up with an all-electronic shock absorber/suspension system along the same lines you are.  Their intent, though, is to make the ride as smooth as possible, not generate electricity.

[hvirtane]

Hi,

this idea is a very good one in principle.

Just some thoughts.

1) I've discussed ideas of

linear alternators with a friend

some time ago. His opinion was that

it would be best to start with the

same arrangement of the magnet(s)

as it is with loudspeakers.

Jerry might know

it really well...?

2) I've been interested in linear

alternator to be used with simple

stirling engines. In addition to

the good ideas by you, Mike and by

Ed, there is one really interesting

development as being done here:

http://members.aol.com/hstierhof/

Please see also this:

http://www.webcom.com/sknkwrks/stirling.htm

- Hannu

[Palentier]

Any Updates?  I found that i'm in need of a Linear Alternator like the one proposed here.  Although I do not need the stack to have a hole through the center as I will not need it connected to anything (think No-Batteries Flashlight).

I think monte350c had perhaps a good solution for what I am looking for in his responce:

"I wonder if you could use a very thin 1" ID tube to assemble this thing into. Something non-magnetic - like some stainless, brass, or aluminum tubing. The thinner the better since it would be in the airgap. Then you could create a way to hold the tube at the ends. This could be something as simple as an endcap with appropriate fittings on it. If the caps and the ends of the tube were threaded, you could put the squeeze on the stack of pole pieces and magnets by screwing the caps on."

Would it be possible for someone to help me design something like this? I would need it to slide back and forth inside a tube.  Maybe some drawings or some detailed explaination.

Unfortunately I have the Vision, but not the know how.  Please email me if you can help me with this project.  

Also an update on this one would be great.

Thanks!

[Newton980665]

Hi all,

I hope I'm not digressing too far from the main subject of this particular forum, if so apologies in advance......

I am in the process of making my own Linear Alternator, but have no idea how to regualte the power that it will output. For example, what is that little box of tricks on a car alternator that always regulates the voltage to around 12 to 14 volts?

Could such a device be bought or manufactured to regulate the output of a linear alternator?

Cheers..............

[waternwaves]

Also not to vary to far from the topic.....But the title of this thread seems to be slightly off....

But since you are discussing using permanent magnets.....  This would be called a permanent magnet linear alternator not a regular linear alternator.

Linear alternators utilize the same principles as rotary alternators with winding voltages controlled by power supplies/regulators instead of magnets.

But I enjoyed all of the discussion so far.