Generator with ferrite magnets

[Dave B]

Hey Chris,

  I've been watching this pretty close and all I can say is be careful. Not a bad idea to get in the habbit of wearing safty glasses all the time even after shutting down the machine stuff if you don't already. Someone once told me that electricity is kind of like flying. Lots of fun but it can be very unforgiving of mistakes. I keep that in mind as I have flown some as well and respect the analogy. As usual you are leading the way and as usual you have your skeptics and protective souls but that's human nature I suppose. I for one am so glad you continue to document and post your projects here on the board. The number of views on this project alone speaks for itself. Great work, thank you.   Dave B.

[Flux]

"I looked at Glen's wiring diagram and now I'm confused already.  He's got the capacitors ahead of my 8/4 cable, then to the #2.  That's not the way I got it hooked up.  The 8/4 is my AC cable.  Why does the wiring diagram show it on the DC side?  I don't have another 8/4 cable laying around here to get power to the rectifier, and I don't want to cut that one off because I might use it for a tower drop someplace.  If I move the 8/4 cable to the DC side, ahead of the #2, then that will skew my test results I did the other day"

The only way I could see any possible benefit was with things wired as glen had shown. If the branch leads are on the ac side of the rectifier then I can't see any point in this at all.

I thought the logic was to deal with the ripple in the parts of the cable that were carrying rectified single phase where there would be a considerable ripple. The rectified two phase will have some ripple but I really cant see it bothering anything.

Flux

[ChrisOlson]

I thought the logic was to deal with the ripple in the parts of the cable that were carrying rectified single phase where there would be a considerable ripple. The rectified two phase will have some ripple but I really cant see it bothering anything.

Well, I guess as long as I got it hooked up, I'll try it and see if it does anything.  After I looked at the wiring diagram again I got confused because nobody hooks up a DC wire run that way.  At least not that I've ever seen.  The single phase branch circuits on the DC side, the way I have it hooked up, are maybe at most about 6" long to get from the individual rectifiers to the power studs where the big DC cables hook up.

I suppose I could unhook one stator and do a test run in just single phase with and without those capacitors.  Would that be useful?  The only thing I'm a little scared of is how do you unhook and discharge those capacitors without getting big sparks?  My friend at the motor shop told me how to get them hooked up without getting big sparks, but I forgot to ask about how to remove them without getting big sparks.  I'd better call him this morning and see what he says.

The more I look at this setup, the more doubts I have that it's even remotely a good idea.  I know for a fact I would never run it in a real world installation because the generator and turbine works good the way it is.
--
Chris

Edit:  I called Sparky at the motor shop.  He said to unhook the negative cable at the battery bank, then hook up my AVR to the cables coming from the rectifier with the capacitors in the line.  Slowly turn up the variable resistor on the AVR to safely discharge the capacitors without big sparks.  That AVR will handle 1,200 amps and he said it won't hurt it to just put a few tenths of an ohm across the capacitors to draw the juice out.

I'll try this in single phase a little later after I fix an oil leak on my hydraulic power unit.

[ghurd]

I was under the impression this was a test stand set up...  PMA in a lathe, wire stretched out across the yard, etc.
Figured it was just dragging cables around the yard a bit to try it.

Lots of guys rectify in the house, so they run all the wires the whole way.
In that situation, it would only involve moving the rectifiers to the tower area and adding caps.

If Jerry is following this, maybe he has a 3-ph with 14/6 running all the way to the battery?
Or a garbogen with the rectifiers at the battery?
And I bet he has a few chrome plated caps laying around over too!
G-

[ChrisOlson]

I was under the impression this was a test stand set up...  PMA in a lathe, wire stretched out across the yard, etc.
Figured it was just dragging cables around the yard a bit to try it.

Well, not exactly.  I drive the turbine with a big hydraulic motor.  These turbines are too big and they won't fit even on my big Monarch lathe.  And the lathe doesn't have infinitely variable rpm control like the setup I got.

I run a 90 foot drop down the tower from the turbine to the rectifier.  The rectifier is at the tower base.  Then I run DC the rest of the way to the bank.  I went away from running low freq, low voltage AC very far a long time ago because it takes too much wires.  You can run DC with two wires.  Three phase takes three to transmit the same power.  Two phase takes four wires transmit the same power.  Plus you get less loss with DC than with AC.  It's not much, but it's more efficient for power transmission from the tower to the bank than AC is.
--
Chris

Edit:  Just before lunch I ran the generator for a bit in single phase.  It puts out 7.3 amps @ 350 rpm, and 12.7 amps @ 400 rpm, which is less than half of what it puts out with both stators.  That's with the capacitors hooked up.  After lunch I'm going to unhook them and see what it does without them.

[ChrisOlson]

It doesn't make one bit of difference with or without those big capacitors in a 150 foot long #2 copper DC line on single phase.  Or if it does, my equipment can't measure it.

I hooked up both stators and it doesn't make any difference with that either.  I ran it at 350 and 400 rpm to test it.  The resolution of my AVR is only .1 amp.  I put the Doc Wattson on the negative wire but it's not stable enough to tell anything.  It has resolution to .01 amp but the numbers are constantly flashing different values, making it hard to use the .01 amp resolution of it with any accuracy.

--
Chris

[ghurd]

How was everything connected?

Like Flux said, if it is not hooked up with everything where it should be, it won't make any differece at all.

[12AX7]

I'd agree with the idea of adding bleeder resistors.  four 1farard caps can store an incredible amount of energy (even in series ) 250,000 uf @ 100vdc can knock you out of your shoes!    Not sure of the rating of the rectifiers,  but they need to be able to handle the in rush current that the caps might draw.

[ghurd]

On the DC side, they won't be any higher than the battery.

[ChrisOlson]

How was everything connected?

Like Flux said, if it is not hooked up with everything where it should be, it won't make any differece at all.

Starting at the generator, about 100 feet of 8/4 cable carrying AC power to the rectifer.  From the rectifer to the battery, about 150 feet of #2 copper welding cable carrying DC power to the battery.  The four capacitors are hooked in series and hooked right at the DC studs on the rectifier.  I unhooked one stator and ran it pure single phase to test it.  That didn't make any difference so I hooked up both stators for two-phase, which also didn't make any difference.

The reason I did it this way is because that for testing it uses the same wiring that a real world installation would use on the tower.  90-100 foot of drop to the rectifier at the tower base, then a 150 foot DC run to the batteries.  Again, I don't run the AC from the turbine all that way.  The rectifers on all my machines are at the tower base.  So in a real world installation, those capacitors would be at the tower base hooked to the start of the DC run.

Those capacitors are pretty dangerous.  If you try to discharge one with a piece of 14 gauge wire it will instantly burn it right off when you touch one of the studs with it.  After I got braver I tried that.  And it still didn't discharge them after burning the wire off.
--
Chris

[Purpc]

Hi Oz and Janne,

I'll attempt to reply to both questions in one post.

I'm no electrical engineer.  I just flounder along until I stumble upon something that seems to work.  With both the polyphase experiments I could not get enough copper in the stator to get the cut-in down below about 350 rpm.  That's why I tried single phase.  More coils in series helps to get the voltage without building a generator that's as big as a truck tire.  This is the second single phase one I've built - the first had 33 turns of wire in it and it ran way too fast.  This one, I put 50 turns in it and the resistance measures .81 ohm.  This is a photo of the stator coils before I casted it:

(Attachment Link)

The coils aren't ideal shape for these big square magnets because the inside is a bit narrow.  But I tried some test coils with wider pin spacing on the inside, and that didn't seem to make a lot a difference.  So I decided to try this shape and see if it works.  The rotors are 11" diameter and there's no more room for copper in this thing.  Those coils have to be scrunched in as tight as they'll go, and tied to keep them there, before casting it.

Now on to oz's questions.  I haven't really noticed any limiting - the faster you turn it the more power it makes.  I think any limiting would be more inherent in an iron core machine.  With the air core axial, in all the testing and fiddling I've done I've never seen any limiting.  I'm pretty sure you could turn it fast enough to burn it up.

I am running faster airfoils with this instead of the torquey GOE222.  They're sort of a modified GOE222, but they do like to run one full TSR faster than the GOE222's.

I had never flown a single phase generator before this.  I am finding that the voltage (AC side) runs much higher for some reason.  On the first single phase stator I built I ran test coils and stuff and decided on the 33 turns because I figured I needed 19 volts to get about 25 volts DC.  Well, it must have something to do with only one sine wave and the time the rectifier is conducting (the peak of the sine wave) because that generator was too fast and too "loose".  On this one, where I got 50 turns crammed into it I've seen that the voltage on the AC side runs much higher than what I'm used to with three phase.  I've seen it at 50 volts and I don't fully understand why it's doing that.

Even at lower outputs it'll start pushing the first tenth of an amp at 20 volts.  But by the time it gets to 10 amps the AC voltage reads 28-30 and the DC voltage is down around 25-26.  The harder you push it, the higher the AC voltage goes.  And the resistance on the line run from the stator to the battery bank is only .124 ohm.  If somebody could explain to me why it does that, that would be great.

This is only wound with 13 AWG wire, which I guess would be good for like 25-30 amps.  But I've had it over 60 for a brief period and it didn't hurt it.  I don't know what it will take for continuous power before it would overheat.  I'm thinking no more than about 25-30 amps.

As I said, it seems to deliver about the same power as the neo generator in day to day wind.  It don't make as much power as the neo generator in very high winds, but the turbine furls fine with it the way it is so I'm going to keep running it for awhile and see what it does long term.  I get the wire to wind these things for free - I get the "scraps" left over from a motor shop.  The "scraps" aren't long enough to wind a coil group for a motor, but they're plenty to wind axial coils.  So they give them to me.  And the magnets were only $40 bucks.  For that kind of money, if it burns up or I have to build another one to do some more experimenting, it's not like it's going to break the bank.

--
Chris

and
Wave winding --» http://www.fieldlines.com/index.php?topic=138287.0

Mixing the both statores, i end with this idea :


This is a stator, with 3 phases, 10 coils each vs 10 magnets.

By your experience what will be the pros and cons?

[Flux]

That will work fine but it is probably not the best way to do things. All you have done is fit 3 single phase stators in a big air gap. It gives a 3 phase output with the benefit of smooth dc with no torque pulsation but it uses a very wide air gap and will have low flux or need very big magnets.

If you look at one of your single phase stators you have a large gap in the coils and you need to devise a way to wind two other phases in there without increasing tha air gap to realise all the benefits.

That implies an overlapped coil design with all the coil legs in the same thickness air gap. You have got the windings in but you need a gap 3 times as wide. With a radial slotted core you can build a 3 phase machine easily and use the available space not used by a single phase winding. If you try it with a radial air gap machine you can still do it reasonably easily as long as the coils are not very thick. You end up with 2 types of coils, half with bent ends and the others straight. Everything goes in the same air gao as the single phase version. You do end up with a bit more resistance in the longer end windings to do the overlap but if you stick to thin coils it works well.

You can do it with an axial design but there is very limited space at the centre to do any form of overlap and you also end up with long end windings on the outside so it becomes a real challenge to wind and the resistance tends to be high. For battery charging resistance is critical. Like the radial version it works best with thin coils but unless you want a very comples construction with coils back to back you will need the coils a fair bit thinner than the radial version, it then leaves very little space for wire.

Hugh Piggott found that if you leave a lot of the coils out you can fit the whole winding in on a single layer and although it does have un wound bits, the rest is wound very effectively with short wire turns and low resistance.

The result of this is the 12 magnet 9 coil winding and its derivitives with 4 magnets per 3 coils. If you want thick coils it is the only way to go, it's easy to build and I think you will really struggle to do an an overlapped coil axial design that can beat it, I gave up, the simplicity well outweighs everything else. It will certainly work far better than the 3 overlapped stators that you propose.

I still stick to overlapped coils for radial designs where there is more space for overlap and the coils can be made thicker without being impossible to wind.

Flux

[Purpc]

Thanks for the reply.

My Goal its to pick up Hugh Piggott  axial with 12/9, with 2 strands of Ø1,5mm, 24v system and remaking more cheap, keeping the same air gap between rotors, same cut in, reduze number of magnets and wire( they coast a lot), keep the same efficiency or better. I know i will get more work.
The ideia it's to expose more wire to flux available. Using the ratio magnets/coils in 1/1.
The drawings are only to expose the ideia, i need to work on the space  inside the coils and the rotation of the phases.
I will use rotors with Ø300 (12") and magnets with 46x30x10 N42.
   

[fabricator]

That has been tried countless times with little or no gain for the extra work.

[Mary B]

Adding bleeder/balancing resistors burns power as heat in the resistors. I use them in high voltage power supplies for ham radio amplifiers and they run hot.

[charlie_ruizpr]

how many pounds of 13awg magnet wire do you use for your ferrrite magnet generator?

[JW]

This is a really good thread, from here you can explore the technical archive that of fieldlines  from this lead you can find everything.

[Bruce S]

This is a really good thread, from here you can explore the technical archive that of fieldlines  from this lead you can find everything.

Not to mention the vast amount of information from some highly knowledgeable people

Cheers
Bruce S

[JW]

Yes Bruce,
Not to mention the basics covered such as diode rectifier bridges. AC current must converted to DC to charge battery's (of all kinds).

You see a lot with new users contemplating magnet arrangements but not so much on the electrical engineering side. Windings loaded impendence etc.

9 coils by 12 magnets per rotor arrangements, single or dual rotor machines. proven designs that were pioneered by the Dans.

https://www.otherpower.com/stator.html

 

[Adriaan Kragten]

I havn't read all the posts about this subject but some of the pictures show that it concerns axial flux PM-generators. I assume that ferrite is the same as Ferroxdure as developed by Philips. The remance Br of Ferroxdure magnets is very much lower than that of neodymium magnets. So if the magnetic flux isn't concentrated, you will get only a low flux density in the air gap of an axial flux generator. So the AC voltage generated in one winding of a coil at a certain rotational speed will be rather low. This gives a rather low maximum torque level.

When I started with PM-generators about fourty ears ago, Ferroxdure was the most common magnetic material and was available in different sizes. I started with a radial flux PM-generator because for such generator, it is rather easy to concentrate the magnetic flux if the pole number of the armature is chosen large enough. My first experiments with PM-generators are described in chapter 3 of my public report KD 341. My first big 12-pole generator was using the housing of an asynchronous motor frame size 112. The 10 mm thick magnets were glued in twelfve radial grooves which make a certain angle with the axis of the armature to prevent cogging. The magnetic flux coming out of one armature pole is supplied by two rows of magnets. This makes that there is a strong flux concentration if the width of an armature pole is small with respect to the width of a magnet. The flux density in the air gap is therefore much larger than the remanence of the magnets. It appears that with enough flux concentration, it is possible to make that the flux density in the stator is close to saturation and so it is possible to make a PM-generator with a high maximum torque level. Later, neodymium magnets were developed and with neodymium it is possibele to make the armature strong enough even if it has only four poles (see chapter 4 and 5).

The price of neodymium magnets has risen strongly during the last two years and therefore use of Ferroxdure magnets might become a reasonable option again. But this also requires going back to a 12-pole armature for a radial flux PM-generator and this has as disadvantage that it is no longer possible to use the standard 4-pole, 3-phase stator winding.

[MattM]

It is possible to concentrate ferrite magnets using iron, though, so its not a total let down.  May not reach parity with rare earth magnets, but considerably cheaper to work with.  The Dans started with them.  I look at it as more of going old school.

[Bruce S]

One of the main reason ChrisO posted this as well as building these with these type of magnets was to prove that it can still be done. Seems people got lazy when NEOs came out and forgot that ceramic magnets was where most of this started out.
The added plus is that they don't burn up or have the coatings come off like NEOs do.

 Search through thru his posts, you can see how well they did. Huge in size? YUP!! Did they work? YUP.
ChrisO is still on here, but doesn't post like he used to, kinda busy enjoying the retired life these days.

Enjoy
Bruce S

[kitestrings]

I think there were two factors that influenced Chris to consider this design: the cost of neo's and corrosion, at least in some locations.  In Chris' design I believe he went to a geared alternator to increase speed and (hopefully) offset the lower flux density.  There are some compromises of course.  Perhaps he'll happen in here to explain further, but this was what I recalled of this effort.

I know Hugh has also been working with ferrites; his "F" series, and had some interesting findings:

https://scoraigwind.co.uk/2020/07/f-series-wind-turbine-designs/

[Mary B]

I think there were two factors that influenced Chris to consider this design: the cost of neo's and corrosion, at least in some locations.  In Chris' design I believe he went to a geared alternator to increase speed and (hopefully) offset the lower flux density.  There are some compromises of course.  Perhaps he'll happen in here to explain further, but this was what I recalled of this effort.

I know Hugh has also been working with ferrites; his "F" series, and had some interesting findings:

https://scoraigwind.co.uk/2020/07/f-series-wind-turbine-designs/

I remember Chris going thru transmission builds and settling on a wet case chain design like a transfer case on a pickup. Forget what oil he used but it was pretty thin.