3.2 meter ferrite magnet MPPT wind turbine

[ChrisOlson]

..Do you think I should use another bearing assy. behind my generator head

Truthfully I don't think you need a pilot bearing on there.  An 1 1/2" bearing will handle as much radial load as the axle bearings in the rear axle of a 1/2 ton pickup truck.
--
Chris

[ChrisOlson]

It was a miserable failure. The magnet pulled most of the JB-weld up and around the magnet. Very little actually stuck to the aluminium. I worked - sorta. But nowhere near as well as I had expected!

Ross, I could see where that would be a problem on aluminum.  But I've built better than three dozen turbines using JB on the mags and never had a single problem with it - unless I wanted to get some mags off some plates later.  With both JB Weld and pins in the center of the mags it's about impossible to get them off without breaking them.
--
Chris

[Wolvenar]

Question Chris.

Does JB weld work differently on ceramics?
I've also tried it on neo mags, and had the same experience as Ross.
It migrates most of it to the sides of the mags and corners rather quickly, WAY to fast to set up.
Like I said it looks like a very thick ferrofluid experiment.
Are the ceramics just weak enough that it does not do that?

[ChrisOlson]

Does JB weld work differently on ceramics?
I've also tried it on neo mags, and had the same experience as Ross.
It migrates most of it to the sides of the mags and corners rather quickly, WAY to fast to set up.
Are the ceramics just weak enough that it does not do that?

No, it'll do the same thing on ferrites.  You're putting too much on there.  All it takes is a very thin film of the stuff applied evenly to the surface of the magnet that goes against the rotor.  If you put too much on the magnet will "swim around" on it and the shear force of attraction between the magnet and steel forces it out the edges.  You want to put enough on so it does squeeze out and seals the magnet from getting water under it.  But it's a case where too much ends up being a mess.

Also, I machine my rotors so they're really rough on the magnet side so the JB Weld has something to "bite" on to.  Putting the magnets on a perfectly smooth surface might work, but then you're going to have to use a lot less of the epoxy because virtually all of it will squeeze out the sides.

If nothing else, rough the discs up severely with a disc grinder.  I cut them on the lathe with a fast rate on the cross feed so they almost got "threads" in the surface of the disc.

Neo rotors with pins are easy because once you put the magnet over the pin it stays put while the JB Weld sets up.  The ferrite rotors with those big blocks are a pain.  Just because they're ferrites doesn't mean you can't get your fingers pinched and suddenly have a magnet get sucked off the plate by one that's adjacent to it, suddenly flip upside down in mid-air and go "SMACK" on top of the other one.  I got sick of that and cut out a template made of UHMW poly to hold those ferrites in place while the JB Weld sets up.

Those 16 pole rotors with the 2 x 2 x 1" thick ferrite blocks are more dangerous than the 12 pole neo rotors with 2 x 1 x .5" thick neos.  The total attraction force between those 16 pole rotors is more than it is with the 12 pole neo rotors.  So you have to be careful with that during assembly.
--
Chris

[ghurd]

JBWeld give me fits.
It crawls all over everywhere.
If it is a situation like a motor conversion that requires magnets to be set over several days or a week, then it requires 5 hours of babysitting to be safe.  A glop of JB in the wrong place, at the wrong time, is a serious mess.
I do use it again, mostly, but only place magnets when I have 5 hours to check it every 10 minutes for 5 to 8 hours.
Just when you think it won't crawl any more, and you go to bed, the stuff makes a pool in the worst possible place.
I have thrown things away after the glob showed up where the thing was not salvagable.

It sticks well if the surface is clean.

If the neo's nickel plate is damaged (IE: HD neos), then the plate pulls off the neo.

"Pins and Claws" weaken the magnetic field. You yourself proved it with the paper clip test.
(might try looking at the physics of a old-school fridge magnet hook, because we are correct about it, even if you don't get it... and it follows the exact same concept as your comment as Ross' magnets on AL).

[Dale S]

Boy! Somebody got his corn flakes peed in, are veiled personal attacks by moderators allowed on this board?

[ChrisOlson]

"Pins and Claws" weaken the magnetic field. You yourself proved it with the paper clip test.

That's only in theory.  My gauss meter shows no difference in the air gap, and a spin test (rpm/volt) shows no difference with pins or claws, or without.  So in practical application, if somebody asks me if they can do that I'm going to tell them yes because I have tested it and it works fine.

I have built a lot of turbines and tried a lot of different things.  Theory is fine.  But there's a difference between what theory says you SHOULD do vs what you CAN do.  Only experience tells you what you CAN do.  Experience tells me that using JUST JB Weld is not going to work if there's any significant shear or tensile load on the adhesion point.  Adding pins or claws allows me to spin very heavy axial generator assemblies at speeds never before attempted in homebrew turbines without affecting rpm/volt.
--
Chris

[Wolvenar]

Boy! Somebody got his corn flakes peed in, are veiled personal attacks by moderators allowed on this board?

@Dave  or anyone that might have actaully seen it this way

Hoping to diffuse this early just in case there was something serious about that comment

Just because ghurd has a "title" doesn't mean he wont have an opinion.
Heck not even sure if he remembers that he has that "title", everyone has been so civil so far here moderation has not been needed for anything but improving the forum look and moving a couple topics here and there.

These titles can become a hampering thankless job, as has been seen elsewhere.
Many wish not to have a title at all, just so they can contribute something that may not be well taken, without feeling they are in some way going to cause the forum problems if someone has a problem with it.

Lets not have a  "them against us" thing happen here..
Lots people have had different experiences for who knows what reasons.
It can give us all different views based on different experience or training.

Were all people with a passion for RE, and long as we can all keep it civil,
agreeing to disagree should happen time to time.

Please continue with our regularly scheduled programming

[tomw]

Boy! Somebody got his corn flakes peed in, are veiled personal attacks by moderators allowed on this board?

Dale;

People will be people but, no, attacks are not acceptable. "Veiled" is a judgement call. I often make comments to folks I know well that may appear to be attacks to an outsider.

I guess I am thick today? I don't see that. PM me with the details and I will look into it closer.

Thanks.

Tom

[ChrisOlson]

Guys, the comment about
"might try looking at the physics of a old-school fridge magnet hook, because we are correct about it, even if you don't get it..."

may be interpreted several ways.  I've built enough turbines to have a general idea of "how it works".  The concerns with using pins in magnets, or claws on the outside of the rotor, to retain them against centrifugal forces are only founded in theory.  I have done enough testing, and built enough turbines, to know that it does not make any difference in gap flux or rpm/volt performance of the generator whether you have them on there, or not.

What will make a huge difference in gap flux is if you DON'T use a method to retain those magnets to the rotors and they decide to leave the general vicinity of the generator when it is spinning at 1,000 rpm.  That causes a huge loss in power, and subsequently (usually) a runaway turbine.

JB Weld has a fairly low tensile strength and when you're using it to install magnets on rotors it pays to calculate the stresses on the magnets on the rotors before you rely on it to hold them in place.  With these big heavy ferrites at .77 lbs each on 355 mm rotors the tensile strength of the epoxy is exceeded at only 233 rpm.  On this particular turbine it would not even reach cut-in before the magnets fly off.

Take a look at this generator on a 12G turbine with all those "claws" on there to retain the mags:
[ Guests cannot view attachments ]

The guys that are embedded in theory are going to see that, choke on their beer, cough and tip over in a coma because it "won't work".  Those of us who hands-on build it know that it does work.

So I might not "get it".  But I DO know how to build wind turbines that work.
--
Chris

[Wolvenar]

Just want to point out.. Zubbly put it to me once, "If it gets the job done, why care if it's not theoretically perfect."
Take a look at his conversions, he has some of these not so perfect things going, and they work.

[rossw]

Take a look at this generator on a 12G turbine with all those "claws" on there to retain the mags:

Chris - I'm often a stickler for details, but as Tom says, sometime "just good enough is all it needs to be".
I think this MAY be one of those. Where you may be losing some theoretical power, but in the greater scheme of things it makes no difference.

I'm curious if you ever considered using something non-ferrous for those claws? Like brass?
Sure, you probably can't just grab the MIG and whack 'em on, but the blue spanner will do it

Or, is the extra effort simply not warranted in your experience?

[ChrisOlson]

Or, is the extra effort simply not warranted in your experience?

It's just that the extra effort isn't warranted.  Some guys put a non magnetic stainless band around the outside.  When I started fiddling with these ferrite generators I had one fly apart during testing.  I tried to drill the mags with a cobalt drill for pins and couldn't even make a scratch in the surface of the magnet.  Neos can be drilled.  Ferrites require diamond cutting tools.

So I contacted Hugh and Flux and asked them if they'd ever tried "claws" on one.  Neither of them had, and both said it would cause problems with low voltage.  Just because somebody says you can't do it never stopped me before.  So I started experimenting with it.  I found not even one tenth of a volt difference with or without, and my gauss meter showed .17 mT in the airgap no matter if I had them on there or not.

Hugh and Flux and I discussed it at length and the conclusion put out by Flux was (after watching my movie I made on it):
Certainly steel screws and roll pins in the holes of neos will have no ill effect, using stainless may be convenient but is not necessary.  In the second case the disc should act as a conductor of the flux from one magnet to the other and in theory there will be no flux on the back of the disc. I think it unlikely that 2 such size ferrite magnets will saturate 1/4" thick mild steel with the other poles open circuit. Probably away from the edge you would detect little. It is really a case of how much here, I suspect the leakage needed to support a paperclip is not measurable on the gap flux by fluxmeter of by volts induced in a coil so it is safe to ignore 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.

Been doing it ever since.  We arrived at the conclusion that all the "claws" do is redirect flux that was otherwise leaking anyway.  It don't change anything in the air gap where the work gets done.
--
Chris

[Dale S]

So Flux doesn't get it either.

[ChrisOlson]

So Flux doesn't get it either.

Well, I don't know about that.  It didn't do what everybody expected.  The thing with flux is that you can't see it.  You can attempt to model it in different ways.  But the one thing that nobody considered is how much of it leaks, and by leaking it's not useful.

Hugh's comments on it were (I don't think these guys will mind me quoting them in the interest of education on these machines):
In the first case with one disk and one magnet, the flux is obviously spewing out all over the place and coming around the back in force.  The second magnet traps a lot of that flux and reduces the leakage around the back.  A second disk with magnets would further reduce the leakage and focus flux in the right place, through the coils.

Adding a 'claw' seems to trap flux that would otherwise leak out and around the back.

The voltage is unaffected by the claws, which suggests that either the amount of leakage is not significantly increased, or the edge of the steel disk is saturating, so the extra flux you get in the claw would not otherwise have been available.

So we all learn as we proceed.  In order to learn and make advancements you have to question things.  The greatest advancements in technology and knowledge were not made by people who blindly accepted that the earth is the center of the universe and everything revolves around it.
--
Chris