Anotherpower.com Forum
Renewable Energy Questions/Discussion => Wind and Hydro => Topic started by: ksouers on February 03, 2012, 07:47:07 pm
-
Look what the Big Brown Truck brought today!
(http://i356.photobucket.com/albums/oo7/ksouers/projects/alternator/magnets.jpg)
Not knowing where to start or what to do I ordered 40 of these little magnets. They are 1x1/2x1/4 inch and intended for a small alternator. I’m not expecting much power, just getting some experience in building a working alternator. Gosh these little things are expensive! And powerful! I had one stack I was working with trying to separate one of the magnets. I set the stack down on the bench and it pulled one of the change gears for the lathe from almost a foot away! I’m going to have to make sure my work area is very clear of any iron when working with these things.
I hope you guys are up for some stupid newbie questions cause I’ll have a bunch. I don’t even know where to begin.
First I have a lousy site for wind. Most days are 3 – 15 MPH except in the dead of summer when it’s dead calm for July and August. I have lots of trees and there is a subdivision right behind me. I have a half acre but it’s covered with trees, large old maples. The only real place I have for the tower is mounting on the garage, it’s detached so I’m not worried about the noise but I know it’s going to be turbulent. As if the trees weren’t bad enough, all the buildings are going to mess up the air flow. I have to keep the tower low so the fall zone stays within my property boundries (county ordinance).
I don’t know what kind of power to expect out of these little magnets but I’ll be happy with anything that works. I hope to get about 10 watts out of it in a good breeze.
The design parameters I’ve set out to achieve are:
3 foot rotor
Cut-in about 150-200 rpm
At least 2 amps and 15 volts but I’ll settle for .5 amps.
5-6 inch magnet rotor with 8 or 12 magnets, 6 or 9 coils.
Figuring on a 3/8 inch air gap.
What do you think? Is this do-able with these little magnets?
Thanks,
Kevin
-
First off - magnets are new to our five monkey senses, there is no instinct to follow with them. GET THEM PROTECTED, back in the delivery box and up on a high metal-free shelf. If you must, find one that is already scratched/chipped/dented to play with and leave the others ALONE until the hour & minute you're going to be placing them... in a clean room w/o other metal items.
My shipment of magnets? I was flipping one around and trying to gauge the strength etc. and was disappointed... then went back to the pile of others and watched the whole stack go frictionless away from my hand enough it almost leapt to an ugly situation. If you've ever used a metal grinder, cut metal or cleaned up rusted iron in your shop the particles will find their way to the magnets. Duct-tape adhesive is about the only way to clean it off once that happens. Littler ones have given me blisters, larger ones like you have can break bones.
EDIT: I see they are smaller than I thought, not quite bone-smasher's but when they attack in packs.... Same rules apply, call it good practice for larger ones ::)
-
That's a Lot of magnet
-
Where is Sir Flux when you need him? :(
I come up with just over 8.25" for rotor diameter if you use all magnets, 20 per rotor for two rotors and have .5" spacings at the closer points. Just over 5.75" diameter with 12 magnets. I guess you are going to go three phase with this turbine as you have indicated 3 coils for 4 magnets.
I think I'd decide on the type of bearings, spindle and hub, get the rotors built and get on to a test coil. I seem to think more magnets is better so, all you have to do is decide on magnet count before you can really start the build.
Anyway, keep us posted with your progress. 8)
-
Hi,
My rule of thumb on designing alternators has been, that when you double the turbine diameter it will take 8 times the amount of magnet to make as stiff generator. That is because the power output is quadrupled, while speed is halved.
If we use that rule of thumb, and compare it to the early 10' machines, which had 24 pcs of 2*1*0.5inch magnets, your magnet lot comes in about 1/5 of that. Cube root of 1/5 = 0.58, so your magnet lot should be able to support a 5 footer, mayby even 6 if you'd be pushing it a little.
There are of course other factors that play, like how efficiently the magnets are being used, are the (turbine)rotors going to be the same performance etc.. It's important to compare it to something, that makes good use of magnets and also focus on your own design into doing the same. In the end it will only give you an estimate with all the variables in play, but these "rules of thumb" have worked quite ok for me.
The first you'll need to do is to figure out suitable size of magnet rotors. It would be best if you can size the rotors so, that there is space for one magnet in between of each magnets, but a bit smaller gap(like Watt suggested) than that is ok too. That will get you most out of your magnets.
When your magnet rotors are complete, you will then need to make the final design about the rotor size, as suitable cut-in voltage depends on the rotor size. Once you get there. This includes wounding a test coil, and then calculating the number of turns based on the required cut in voltage, and selecting a wire that you can fit into the stator with the required amount of turns.
-
I asked Flux in email - his reply follows:
I spend too much time on Otherpower to think about other boards.
In answer to the question, there is no details of the intended prop and this will have a big effect on the alternator design. There is no energy in wind at 3mph and designing for it would be pointless. For a small 3 ft machine I suspect you will struggle to get anything at 6mph but that may be a sensible cut in point.
For a fairly fast prop I would cut in about 300rpm, but 200 would be sensible to aim for and you could increase the gap if it is too slow. To get down to 100 rpm you would need a slow high solidity prop but this may be the way if he intends to use something like pvc blades.
12 magnets will be enough electrically but I see a big issue with space at the centre of the stator and no way could you get a hub in there, it would need to be a overhung design with only the shaft going through.
I think generally i would favour 16 magnets but the hub would still need careful thought. If you use 20 magnets it will stall and you will need external resistance.
Personally I would go for 16 magnets if you can solve the bearing problem. It should take a prop of tsr5 and it may stall a bit in the higher winds.
I think you will be looking at something near 80 turns for cut in at 300rpm.
Sorry I can't help more but no prop details. You should get 24w or more at 15mph if you can get it out of stall.
Flux
-
Well Kevin, it looks like you are fully addicted. ;D And us addicts love company.
You have some good advice to start, so keep us posted. My site has more or less
the same wind restrictions, and I haven't quite settled the design yet.
You mention a lathe, so some reading on Chris O's builds might be worth doing.
Best of luck!
-
3 foot rotor
Cut-in about 150-200 rpm
At least 2 amps and 15 volts but I’ll settle for .5 amps.
5-6 inch magnet rotor with 8 or 12 magnets, 6 or 9 coils.
Figuring on a 3/8 inch air gap.
Kevin, if it were me I'd use a bigger rotor. With those magnets, about 6 feet. And use all 40 magnets to build a 20 pole 15 coil generator. I did some rough calculations on this and you should be able to get about 40 watts @ 10 mph, and around 280 at 28 mph.
If you use a 3 foot diameter rotor I'm afraid you will be very disappointed on your marginal wind site. If you build at least a little size into the machine you will have a lot of fun with it because it will actually make some power.
--
Chris
-
Wow! Thanks for the replies, guys.
Dang: Yes, while fiddling with a stack my finger got pinched. Yes, it hurt. Lesson learned. I will only work with one magnet at a time. The stack will not be anywhere near the work area. I’ll also need to work on some way to anchor the rotor while working on it. I don’t need tools and parts to go flying around the shop unexpectedly. The shop vac is my friend. The magnets are back in the bubble wrap and box, they were only taken out for the photo op and inspection.
Woof: I ordered enough to make two machines. I just know I’m going to mess one up someway, somehow. It’ll happen.
Watt: Thanks a lot. This is what I really need, some rules-of-thumb to base starting points on. The parameters I started with had no basis in fact, just some assumptions I’ve made based on the descriptions of other projects. I assumed a spacing of one magnet width, maybe a little more. My plan is the first device will use 8 magnets per rotor, 16 total. Yes, it will be three phase. I’m trying to follow common practice, not blaze new trails. This first machine is a learning laboratory for me. I’m hoping my expectations are more conservative than overly optimistic.
Janne: More rules of thumb. Many thanks. That’s really what I need, critique of what I’ve set out to do, suggestions to avoid mistakes and to define reasonable expectations.
Ross: Thanks for forwarding the info on to Flux. Please pass along my appreciation for him taking the time to help me out. Of course the parameters set out above are just a starting point and will be adjusted as warranted. A 5 foot rotor is do-able, I think. 7 foot will likely start running into issues with the neighbors (not to mention the county bureaucracy). More than that and I’ll start running into structural issues and trees. Though something tells me my wife will start complaining before any of the above get a chance to :)
Bj: Thanks for dropping by. I don’t know about addicted (yet). But it is something I’ve wanted to do for a very long time. I have a smallish machine shop in a corner of the garage: 7x14 lathe, 9x20 lathe, X2 type bench mill and a larger knee mill. I also build small engines: steam, stirling and repairs around the house. The plan is to document the project here. Perhaps some other newbies will find it useful.
By the way, this is likely to be a slow build. Time, materials and motivation may not always be available at the same time. I'm on call 24/7 and of course family needs always come first. I have the motivation and would like to be flying ASAP but a more likely timeline is first wind around mid to late summer.
OK, design changes:
5-6 foot turbine, but I still want to get away with as small as possible.
Cut in ~300 rpm
Hub 8 poles, 6 coils, 6-7 inches in diameter
Bearings yet to be determined.
Wire gauge and turns yet to be determined.
Any suggestions on bearing size or style?
Coil:
I’ve read where the core should be the same size as the magnet. Is there and advantage/disadvantage to going slightly larger or smaller? Is wider better than longer?
Is there any advantage to building a pair of small test rotors first, say with four poles to help determine what coil parameters to use?
Thanks,
Kevin
-
Thanks for the leg work, Chris. Looks like I've been talked into at least 5 feet on the rotor. I'd like to stay small to be unobtrusive to the back neighbors. They are fairly close as the garage is built towards the back of the property, but still plenty of room from the fence line as long as I keep the tower low. Their properties are slightly bigger than the usual subdivision postage stamps, but not much, maybe a third the size of mine.
The only overhead clearance I have is from the garage and directly back, or above the house. And it is NOT going on the house!
-
5-6 foot turbine, but I still want to get away with as small as possible.
Cut in ~300 rpm
Hub 8 poles, 6 coils, 6-7 inches in diameter
Bearings yet to be determined.
Wire gauge and turns yet to be determined.
Assuming a 5 foot rotor cutting in at 6 mph and 8.5 TSR you have 285 rpm.
With only 8 poles using those small magnets you need 140 turns of wire to get 12 volts @ 285 rpm. You need to use more poles with those small magnets. With 20 poles and 15 coils, for instance, you can get the coil turns down to 22 @ .650 air gap, and get the resistance of the winding low enough to actually get some power from it.
If you had bought 1 x 2 x .5 mags, then yes, you could build a 8 pole. But the magnets you got will not work very well with 8 poles. I see you mentioned buying enough magnets to build two machines, so you could build a 10 pole single phase as well, which would take 20 magnets. A 10 pole single phase would require 38 turns per coil, which would also be very doable and yield a low enough resistance to get decent power from it. With the small prop the single phase configuration would also keep the prop out of stall at higher wind speeds because it is only loaded when the voltage is high enough for the diodes to conduct.
So there are several options. But I'm afraid that a 8 pole three phase is not one of them with those mags.
Coil:
I’ve read where the core should be the same size as the magnet. Is there and advantage/disadvantage to going slightly larger or smaller? Is wider better than longer?
All the homebrew books claim you should make the hole in the coil the size of the magnet. However, there are some trade-offs in performance doing that. The book designs will yield a high resistance generator that gets hot at high outputs. If you "scrunch" the poles a little closer together the flux leakage at the corners of the poles is insignificant compared to the gains in internal resistance you can make by using wedge shaped cols.
When you use a lot of turns of wire, the outer turns add resistance real fast and performance suffers accordingly. So keeping the turn length as short as possible is important, especially when dealing with small units like this where you'll wind with small wire and internal resistance ends up being the killer in getting any amps out of it.
So I would tend towards a design using wedge shaped coils with probably about 1" top to bottom, about 5/8" wide at the outer part of the coil and about 1/4" wide at the inner part. Using this wedge layout will require about two more turns per coil, but the gains in less resistance will be significant compared to using a winder configuration that has a rectangular hole.
--
Chris
-
OK, 20 poles with 1/2 inch spacing puts me at a ~8.5 inch hub. Still not bad. Funny how this thing keeps getting bigger :)
Flux also mentioned more poles to make it work. Guess I better defer to experience :) 20 poles it is.
I'd rather stay with 3 phase to get power from the lower wind speeds since that's what we have most of the time. Wouldn't a single phase need more torque to overcome cogging?
8.5 TSR seems rather optimistic with all the obstructions I have, but I really have no basis for that opinion. I just assumed I'd have lousy efficiency no matter what I did.
You built a multi-stator machine at one time, didn't you? How did that work out?
Thanks,
Kevin
-
For 5' you're going to need all that magnetic material you have - so 20 pole is a good way to go forth.
To add to Chris's post, the only time your design would suffer from smaller than a magnet sized hole in the coil, is when the magnet is coming to the middle of the coil.. At the time when the voltage is at the lowest. With battery charging devices, no current is flowing at that time, and no loss will be made. So that coil geometry would probably be a good way to go, in fitting the most copper wire possible into stator. A good thing to remember, is the more copper you put into the stator, the more less you're going to lose power internally in it.. Meaning less chance of burnout and better electrical efficiency.
I'd avoid single phase. You will get better electrical efficiency if you go 3 phase. This is because in 3 phase the maxium current is less - hence smaller I^2*R losses. If stall becames a problem I'd rather deal with it externally(longer wire, or resistors in series), than heat up the stator. Not saying single phase cannot also be succesful - my first machine was a single phase 5 footer :).
-
You built a multi-stator machine at one time, didn't you? How did that work out?
Beautiful. They're two-phase 10 pole geared units, driven with 3.8 meter rotors. I"m flying three of them as we speak. They use ferrite magnets and are considerably more powerful than the three phase single stator neo generators they replaced on my 12G turbines. They will develop 1.8 kW continuous without overheat and I've pushed them as high as 4.3 kW.
The rear stator is skewed 90 electrical degrees from the front one and they use IRP (Individually Rectified Phase) configuration with four wires down the tower
[attachimg=1]
They're also heavy. The generator assembly weighs about 80 lbs.
Here's another photo of a complete 12G turbine head with a dual stator generator laying on its side in the wheelbarrow - this is the one I took down a couple week ago when I put up my new 3.2 meter MPPT turbine:
[attachimg=2]
That turbine head weighs about 340 lbs. Those generators will take anything a set of 3.8 meter GOE222 blades can put to the shaft. It is one heavy duty sucker.
--
Chris
-
I would use 2mags for each pole shaped like a t. And shape the coil's in a wedge shape.
And use 12 poles and 9 coils.
That would be a lot simpler to build than with 20 poles and 15 coils.
You can push the mags on each pole within about 1/4 " apart as long as they are seated on the steel rotors.
Or use for 8 poles and 6 coils
-
That's damned impressive, Chris! I sure do like the way you farm guys build stuff.
jlt: Thanks for stopping by. Interesting suggestion, I must say. I may try that when I start experimenting.
Just got off the phone with a friend that has access to a water jet where he works. They keep 1/4 inch plate stock on hand, so arrangements will be made to have the rotor plates cut within the next couple weeks. All dependent on the jet operators schedule.
My plan is to face cut a wide groove in the plates maybe 1/8 inch deep to hold the magnets, result will be something similar to what Chris did with the cleats. Then all potted in resin.
Kevin
-
I've read some comments that mention a "stall" condition.
Could someone describe this, please?
When I see that term I think of an aerodynamic condition where an airfoil loses lift, usually at low speed and high angle of attack, but it can occur at any point in flight under the right conditions. Essentially, it's cavitation on a large scale.
-
I've read some comments that mention a "stall" condition.
Could someone describe this, please?
When I see that term I think of an aerodynamic condition where an airfoil loses lift, usually at low speed and high angle of attack, but it can occur at any point in flight under the right conditions. Essentially, it's cavitation on a large scale.
That's right.
In a wind turbine, the blade RPM slows when the PMA is loaded more.
If the PMA is loaded too much, the blades slow down too much.
When the blades slow too much, the angle of attack becomes huge, and the blades can't make much lift (or power).
In a wind turbine, typically when it is 'stalled', the power output kind of hits a plateau, even when the wind speed increases.
Say it makes 2A at 8MPH, 2.5A at 9MPH, and 2.5A at 15MPH? That's an indication it is stalled pretty hard.
Make sense?
G-
-
An easy way to keep mags from sliding off is to make a band around the outside of the rotors.
Just make a band out of mild steel or stainless.Make the band 1/4 inch smaller than the rotors.
Heat the band to a cherry red and it will expand enough to slip over the rotor. when it cools it will
fit very tight.
If you use mild steel only leave 1/8 sticking above the rotor. If you leave the band to tall there will be some cancellation.
-
Thanks ghurd. I've been having trouble trying to visualize the aerodynamics of the blade, now I have more to think about. I can't quite get my head wrapped around the relative wind hitting the airfoil on the bottom (very high AoA) then the transition to rotational motion with a frontal relative wind. Sometimes I think I have a grasp of it then it slips away. It's the transition that gets me. It'll click eventually. At least now I know what people are talking about.
jlt: I saw that on the "other" site. Seems like a pretty good trick. Any drawbacks to cutting the groove? I don't have any welding equipment though I can silver solder using propane, and I have a broken band saw blade laying around that I was going to repair, though that's a very high carbon content. I think I've read of problems with high carbon steel laminates.
-
Hey Kevin, Kevin here.
I'm watching you start up with interest. You are asking some good questions, especially on this new forum.
You stated earlier in the thread that you weren't sure about the addiction thing. As they say, Denial is the first sign.
I am curious about this:
My plan is to face cut a wide groove in the plates maybe 1/8 inch deep to hold the magnets, result will be something similar to what Chris did with the cleats. Then all potted in resin.
Are you planning to recess the magnets in the rotor? IIRC there was a discussion over at the other place about this. (haven't looked it up 'cause I'm not sure about linking to over there) Being 50+ with occasional CRS, maybe one of the experts will chime in. I am under the impression that recessing the magnets will disrupt the flux pattern. To my uneducated logic, recessing a 1/4" magnet 1/8" into the rotor will cause problems. Disruption of flux and leakage out the back.
Hopefully one of the experts will correct me. Just didn't want you to find this out too late. Keep it up.
-
Hi Kevin. Thanks for stopping by.
I'm watching you start up with interest. You are asking some good questions, especially on this new forum.
Thanks. I'm trying. I have a lot of curiosity and little knowledge on the subject.
You stated earlier in the thread that you weren't sure about the addiction thing. As they say, Denial is the first sign.
I will admit addictions to curiosity and experimentation :)
Are you planning to recess the magnets in the rotor? IIRC there was a discussion over at the other place about this. (haven't looked it up 'cause I'm not sure about linking to over there) Being 50+ with occasional CRS, maybe one of the experts will chime in. I am under the impression that recessing the magnets will disrupt the flux pattern. To my uneducated logic, recessing a 1/4" magnet 1/8" into the rotor will cause problems. Disruption of flux and leakage out the back.
Welcome to the +50 crowd. I blame a lot of stuff on that :)
The plan was any groove, just deep enough to keep the magnets from slinging off the rotor. Maybe not 1/8 inch, but a few thousandths, if it's not detrimental. I've also thought maybe a ring of non-ferrous metal (brass?) or a ring of plexiglass or fiberglass would work. Though I suspect getting the plastics to bond, and stay bonded, to the steel rotor may be more trouble than it's worth. The metal band mentioned by jlt is also intriguing. Simple and elegant. I thought a pocket would mess up the flux but maybe a groove would not. I just don't know. I, too, am hoping someone will have an opinion about it.
Thanks,
Kevin
-
Kevin,
Use a quality epoxy adhesive with a very thin bond at the magnet. Clean both surfaces very clean with maybe acetone or similar. Use the same epoxy to run fillets around the magnets after the initial bonding has set up. The key is proper preparation of the surfaces and quality epoxy. Completely potting the magnets looks pretty but I have never found it necessary. Do not use Bondo, it is not an adhesive, it will crack, hold water, rust and destroy your magnets. Yes, flying magnets could be an issue using Bondo, it's how this all got started about needing to fully pot the magnets. It's a problem that does not exist when properly done. Dave B
-
I am under the impression that recessing the magnets will disrupt the flux pattern. To my uneducated logic, recessing a 1/4" magnet 1/8" into the rotor will cause problems. Disruption of flux and leakage out the back.
Some folks claim that, but it's not true. I made a video once of testing that with the "claws" on the outside of my ferrite magnet rotors. I thought I had uploaded the video to YouTube but I can't find it right now. At any rate, putting 1/8" thick x 3/4" wide "claws" on the outside of even those very weak magnets did not make one bit of difference in the magnetic performance. Not even .01 volts on a spin test with a generator that doesn't have them vs one that does. All it does is redirect flux that leaks out the perimeter of the rotor anyway.
I know I had sent that video to Hugh (Piggot) at one point, and I thought I had uploaded it to YouTube. If I find it I'll post it.
--
Chris
-
Dave B: Thanks. I never even considered Bondo. I usually run from the stuff. Hideous! I had planned to use West Systems as it's readily available. I've also had good experience with Devcon under adverse conditions.
-
Hey Chris
Some folks claim that, but it's not true.
I don't know that I'm ready to agree with this blanket statement. Your big honkin ferrites are sitting on a full thickness rotor with "claws" on one outer edge only. I'm talking about thin neos recessed into the rotor. I don't know for sure but, to me it seems to be apples and oranges. It's probably a moot point as it sounds as if he (ksouers) will be using another method. I would be interested to see the modeling of neos sitting in pockets on a rotor.
Kevin
BTW I enjoy seeing and reading about your projects. Gives us wanna-bes something to shoot for.
-
I don't know that I'm ready to agree with this blanket statement. Your big honkin ferrites are sitting on a full thickness rotor with "claws" on one outer edge only. I'm talking about thin neos recessed into the rotor. I don't know for sure but, to me it seems to be apples and oranges.
Actually, the best modeling is to actually do it and spin the two varieties on a real world stator and see what you get. I recessed 1/4" thick wedges that I bought from Ed Lenz that didn't have any holes for pins years ago and never had a single problem with it. Got the same voltage from those generators as the books and spreadsheet calculators said I should get. I used to leave a .150" rim on the edge of the rotor when I machined them.
--
Chris
-
I would be interested to see the modeling of neos sitting in pockets on a rotor.
Kevin
I had already discounted using a pocket very early in the decision. Mostly because I didn't really want to stand in front of the mill for days on end carving out 40 little pockets in hot rolled steel plate!
My intention was to machine an => inch wide groove in the rotor face, maybe .030 to .125 deep leaving a rim around the outer edge ~.020 - .050 inch. Still holding out deciding what to do seeing how this argument plays out. For those keeping score I'm siding with Chris so far. Hard to argue when someone can prove by example it doesn't make a difference.
Kevin
-
My intention was to machine an => inch wide groove in the rotor face, maybe .030 to .125 deep leaving a rim around the outer edge ~.020 - .050 inch. Still holding out deciding what to do seeing how this argument plays out. For those keeping score I'm siding with Chris so far. Hard to argue when someone can prove by example it doesn't make a difference.
You could set up the mill and mill out pockets. But it's a lot easier and quicker on the lathe. I've always machined my rotors by facing them so they run perfectly true. I always took .150" off to get all the humps out and get the face of the rotor nice and flat, and left a rim about .050" around the outside for the magnets to butt up against.
On later turbines I went to using bar mags with holes for pins so I didn't do that.
On the ferrite turbines, each one of those magnets weighs over 3/4 lb. I didn't trust the little lip like I've done in the past. The centrifugal forces on those are astronomical when you got a generator running at 1,200 rpm like it does in my latest turbine. So there I welded 1/8" x 3/4" pieces of flat stock to the outside of the rotor to hold the mags against the force. Somebody told me when I posted pictures of the first one that it "steals flux". And that's why I made that video on it. I still haven't found that confounded video. I think it's in my computer with a name like 100-3823.wmv or something and I'm still trying to locate it.
Hah! I found that frickin' video. I'm uploading to YouTube.
OK, here's the video. Later, after I made this video I bought a gauss meter. What I found is that these axial rotors saturate around the edge of the disc and you get flux that leaks because even a 1/2" thick disc will saturate right near the edge. Leaving a "lip" or weld on "claws" only stops the flux that leaks normally anyway, and it doesn't make one iota of difference in airgap flux between the two rotors.
--
Chris
-
Chris, thank you for sharing that video. 8)
-
Chris, thank you for sharing that video. 8)
You're welcome. I was beginning to think I had deleted it or something. I know that when I made it I sent it to Hugh and Flux to watch. Hugh commented that he had always wanted to try that but that all the "science" says it's "bad". But what the "science" failed to predict was that you get flux that "leaks" out the perimeter of those rotors anyway. This is evident if you build a turbine and use steel stator support bolts that are very close to the rotors. The generator will actually "cog" because of the flux that leaks and attracts the steel stator supports.
Putting those on there, or machining rotors with a lip merely stops some of the normal leakage that you get at the edge of the disc. So I hope this helps clear up some of the "old wive's tales" about how "bad" this is. Sometimes these things get propagated as "fact" and somebody that realizes that it's just not right has to challenge it.
--
Chris
-
Thanks ghurd. I've been having trouble trying to visualize the aerodynamics of the blade, now I have more to think about. I can't quite get my head wrapped around the relative wind hitting the airfoil on the bottom (very high AoA) then the transition to rotational motion with a frontal relative wind. Sometimes I think I have a grasp of it then it slips away. It's the transition that gets me. It'll click eventually. At least now I know what people are talking about.
jlt: I saw that on the "other" site. Seems like a pretty good trick. Any drawbacks to cutting the groove? I don't have any welding equipment though I can silver solder using propane, and I have a broken band saw blade laying around that I was going to repair, though that's a very high carbon content. I think I've read of problems with high carbon steel laminates.
You know a plane must be moving forward (relative to the air) to lift off, right?
Lets say it takes 50mph to create enough lift to lift/take off.
If the plane is on an aircraft carrier moving 25MPH, then only need 25MPH more speed?
But if the plane has 55MPH wind straight up from the bottom, then it is a different story.
It is not the transition, it is the AoA.
Yes, I know I didn't explain it worth a hoot.
"paper clip video"
Lets see if I understand what you are saying...
Without the claw, the flux is NOT shorted (can't argue with that).
And the paper clip hangs.
WITH the claw, you can no longer hange the paper clip.
Looks like WITH the claw has less flux in the disk, and the only place that flux has to go around is through the claw.
Since it only has a path through the claw to go around (in the video), then the flux IS shorted through the claw.
Since it IS shorted through the 'claw', then there is less flux in the coil
Since there IS less flux in the coil, then there IS less voltage, and less power.
The claws are helping to direct the magnatism to a shorted condition, which makes no power.
It is not going through the back side of the plate, it is simply wasted in the "claws".
If the paper clip sticks, then the disks are not thick enough.
Making the magents less effective (shorting their flux) so a paper clip will not stick is like cutting the legs off a sprinter so he will not run as fast. Sure, he will not run as fast, but that doesn't make him more efficient.
I had a guy tell me his solar panel was not working as well as it should. The problem, as he explained it, was the wire was too fat. It took too much power to push electric through that fat wire, and skinney wire was easier to push power through, he said, because the power didn't need pushed through so many pounds of wire.
His explaination made sense to him. It was totally incorrect, but it made sense to him.
He did not understand what he was doing.
-
The claws are helping to direct the magnatism to a shorted condition, which makes no power.
It is not going through the back side of the plate, it is simply wasted in the "claws".
That's not how it works. All the plate does is provide a medium, that's better than air to complete the magnetic circuit between poles. Except the pipe that carries the flux has a leak in it. All the claws do is plug the leak. I will quote Flux himself here, as I don't think he'd mind, and he and I and Hugh discussed this at length when I did it:
"There is significant leakage flux with the wide air gaps of these dual rotor machines which you won't avoid and some tinkering with the path of these leakages will have virtually no effect on the real gap flux."
He did not understand what he was doing.
If that's a hint that I don't know what I'm doing, at least I'm doing it so other people can see how it's done.
--
Chris
-
Chris,
The claws you have are not "plugging a leak". In fact, as Ghurd stated they're making it worse. In addition to the magnetic circuit between poles you mention, there is a local magnetic circuit for every magnet, between it's own N and S pole. In the paper clips-video the claws are not enhancing the inter-pole magnetic circuit, it just increases the local leakage magnetic circuit. As a result, the magnetic flux in the plate is reduced enough, so that the paper clip is not sticking anymore. The real solution is to increase the plate thickness, so that there is no leakage behind the plates, as Ghurd stated.
-
I have conclusive proof that it does not alter the magnetic performance one bit in the air gap. When you place those magnets at the edge of a rotor the flux lines do not go directly from pole to pole - they bend. There are some lines that bend outside of the active air gap. I invite you to assemble a generator and model it with a gauss meter yourself like I have done.
All those claws do it stop some of the lines that bend outside of the active air gap from doing so.
I have built seven of these ferrite magnet generators so far - they all have "claws" on the rotors. The latest one develops 3 kW continuous, without overheat, @ 100 volts and is only 12" diameter. You can't make too many stupid mistakes, like "shorting flux", and get that kind of performance from a 12" axial with ferrites.
Edit for correction: The three phase 16 pole generator shown in the below photo is a 14". The dual stator two-phase units I built with ferrite magnets are 12".
The rotors in that latest gen also use another concept that is supposed to be "bad", and that is crowding poles like this:
[attachimg=1]
Except that crowding poles so there's minor leakage at the corners also has negligible effect on real gap flux and allows dropping the internal resistance of the winding by using a different coil geometry.
When it comes to this type of stuff there's always going to be "armchair quarterbacks" that can come up with all sorts of "wisdom" and tell how it "should be", and maybe some of them have even built a "book design" generator once. Then there's the guys that actually play the game and got hands on experience with what works and what doesn't. I don't follow the book. I design and build my own stuff simply because I enjoy exploring the outer limits of what can be done and what can't.
And that's exactly where this comes from - experience and hands on testing.
--
Chris
-
Well, I have heard it said that years in the library can often be replaced by an hour or two in the shop...
Sadly, this stuff gets regurgitated over and over and eventually becomes misrepresented as a fact simply because it has been restated so much.
Not sure what your paper clip test shows but any with:without comparisons in use should show the reality.
I also think the game really changes with the lower power magnets.
So much flux in a NEO that a bit of diverted flux is perhaps a non issue as well as strong enough to go where it cannot in a weaker magnet?
If I had a dollar for every "fact" I got from a book / theoretical education that was later proven to have fallen from the south end of a north bound bull I could easily buy a dual rotor alternators worth of of big NEOs.
Interesting discussion. Glad to see the old bulwarks of [perhaps]misapplied theory being chiselled away!
Chris; Should you ever decide to produce these to sell I would love to give one a try. Really curious how they would work if tailored to heating water with common high voltage water heater elements. Always seem to need heat in some form or other. Bathing, planned hot tub, heating spaces. Our windy months are also heat demand heavy months.
I am no builder but find this stuff very interesting. Keep it coming. K.I.S.S has its place but can lead us into stagnated pools of design.
Thanks for posting this great stuff!
Tom
-
So much flux in a NEO that a bit of diverted flux is perhaps a non issue as well as strong enough to go where it cannot in a weaker magnet?
When I got my gauss meter I tested some neo rotors that have 2 x 1 x .5 N42 bars and the leakage on those is pretty severe. It's way less with the big ferrite blocks. Once you put two rotors face to face, with the ferrite blocks I had .120 Tesla @ 1/2" from the perimeter of the rotors and .167 in the air gap. With the neos I had .350 @ 1/2" from the perimeter - over double what the ferrites got in the air gap.
With ferrites you really cannot make ANY mistakes in construction or design or it won't make any power. To put it in perspective, take a "book design" neo generator with 1/2" thick x 2" round mags and open the air gap up to 2-7/8". Now, try to get that generator to make some power.
Chris; Should you ever decide to produce these to sell I would love to give one a try. Really curious how they would work if tailored to heating water with common high voltage water heater elements.
I do build these machines for other people who can't build their own and want one. My clipper is a three-phase delta configuration of standard off-the-shelf water heater elements and I got a pretty good match on that to the available power. With a Classic it's necessary to have that because when the bank reaches absorb (and especially float) the controller simply unloads the turbine and lets it speed up to the point where you have the clipper drive set for, then it engages the clipper to keep the turbine out of an over-speed condition.
So I think it could be done on pure water heating. It would take a careful match of wire run resistance with the elements to get it so it matches available power. I don't think you'd ever get the same accuracy in power tracking that you get with a Classic controller, however, and I think it would be only be suitable on an excellent wind site. What happens is that the clipper tends to work good, and provide a good match, to shaft power at really decent wind speeds (where it's needed most to control over-speed). But when the wind dies down the clipper grossly over-powers the turbine and stalls it.
That would be an interesting one to try to work out the details on because I think it would require a "staged" loading system to let the turbine get started. :)
--
Chris
-
To get this back on-topic for Kevin (because he's got a turbine to build), I did some figuring on the magnet configuration, just for giggles.
Somebody had mentioned putting them in a "T" fashion on the rotors to increase the surface area of each pole. This works really good and I've seen other generators (like one Russel built in Ireland) where this has been done with great success.
If you used 32 magnets (two magnets per pole, per rotor) and built a 8 pole 6 coil, you will not get as much from the magnets as using the same 32 to build a 16 pole 12 coil. The doubled surface area of each pole does not have the same power yield as 16 poles with half the area.
Part of the problem is that when placing them in a "T" configuration the width of the top pins in the coil winder has to be increased to take advantage of it. You have double the number of turns per coil that you have with a 16 pole at half the area per pole. Since the outer turns in the coils add turn length faster than the inner turns, it ends up losing out to using one mag per pole, per rotor.
The most powerful configuration I can come up with, using what you have, that should be a fairly close match for a 5-6 foot rotor is using all 40 in a 20 pole 15 coil generator.
I think that had already been decided earlier. Just wanted to point out that doubling up magnets on poles to get double the surface area usually can't match efficiency using the same magnets to build with twice the poles at half the surface area.
--
Chris
-
To get this back on-topic for Kevin (because he's got a turbine to build), I did some figuring on the magnet configuration, just for giggles.
Thanks, Chris, but I was enjoying the arguments. It was pleasant to see everyone defending their positions while remaining civil.
I need all the information I can absorb at this point, so these kinds of debates really give me a lot to think about. With several successful generators flying, it would seem whatever distortion in the flux field the cleats or claws may create is really insignificant. Therefore I have no worries about mounting the magnets in a shallow groove.
OK, the plan is a 1/4 inch thick rotor 8.500 diameter with a .030 deep groove to mount the magnets in. That should leave about a .134 wide rim to retain the magnets, and an absolute maximum of 6.388 inches for a hub. I may go 3/8 thick on the rotors. I'll have to call my friend and see if they have it in stock otherwise I'd have to order it. I can get 1/4 inch rotors water jetted for time and materials plus a six pack as they don't normally do custom work.
I'm thinking about a 3 inch diameter chunk of carbon steel for the hub, 3/4 inch diameter for the spindle. Any suggestions for bearings? I'm thinking radial load would be better, I won't have to deal with incorrect pre-load. I can stack them up say 3 per rotor, 2 inside and 1 outside, total of 6 bearings on the hub. Or would a pair of double-row bearings per rotor be better?
Thanks,
Kevin
-
You know a plane must be moving forward (relative to the air) to lift off, right?
Lets say it takes 50mph to create enough lift to lift/take off.
If the plane is on an aircraft carrier moving 25MPH, then only need 25MPH more speed?
But if the plane has 55MPH wind straight up from the bottom, then it is a different story.
It is not the transition, it is the AoA.
Yes, I know I didn't explain it worth a hoot.
ghurd,
I think I got a handle on it now. The trouble I was having was visualizing a stationary blade with wind hitting directly on the bottom, then having the blade start to move and that wind change to a relative wind with a forward component over the leading edge. I ended up thinking about an airplane in stall (something I'm familiar with) where you have a high angle of attack and the airplane is falling downward, basically the same configuration as a stationary blade. As you lower the nose (lower AoA) you start to get a relative wind component coming over the leading edge, just as you would when the turbine blade begins to move. The airplane also picks up speed and the relative wind becomes even more horizontal, or a more forward component, again reducing AoA further.
Anyway, that's the way it finally made sense to me.
Thanks for all the help.
Kevin
-
Hi Kevin,
With my 4kw motor conversion I counterbored in 84 16mm diameter bores 1mm deep so I could mount the N50 grade 16x13mm thick round Neo's. I found by seating each one in the counterbore held it snug and one the first pole I was pretty scared all 22 neo's would go flying out but they stayed firm. I do that now with with all my wind generators and will do in the future. By having a pocket machined in the right place does make it easy for both placing the magnets and most of all purfect alignment of the magnets.
Regards Bryan
-
Thanks Bryan. It's great to have other confirmation that recessing the magnets isn't detrimental to performance.
Kevin