My First Motor Conversion

[briskwinds]

After going through many posts here (zubbly's, dinges ), I was still not sure of which type of motor to buy for doing wind turbine conversion. Finally crossed my fingers and bought this second hand motor from an electric rewinding shop for the experiment.

This is a 1 HP, 3 phase, 415V motor.

The rotor is 4 inches in width and 3 inches in diameter.

From what i have guessed from looking into the stator the coil windings are like this.

First I will try to grind the rotor and put small round magnets (>= 6 mm thick ?)on it, without disturbing the coils and see what voltage I get. I will accept 12 or 24 volts. Then I will try to do serial, parallel, star, delta combination of coil and see what I get.

> Is this default winding in stator is any good? If yes how to wire the coil to get 12 or 24 volts. If no then few pointers about wire size and turns to put on stator will be appreciated.

> Should I go for 4 poles on the rotor?

I have built a 10 footer axial flux (DanB's design) before. I want to build bigger wind mill, but the magnet cost too much for 17 footer and 20 footer. From what I have learned motor conversion needs 1/4 of the magnets than axial flux. That's why I wanted to do a motor conversion.

Danb's 10 footer 1 KW type requires 24 cu. inch magnets.

Zubbly's rule 150 watt per cu. inch of magnet. That's how I arrived at 1/4.

What you guys say?

Thanks.

Fixed your typo.

[hvirtane]

If you are trying to do it cheaply, you might try making the rotor a different way.

it would probably work rather well with ceramic magnets.

- Hannu

[dinges]

Hello Briskwinds,

Crossing your fingers seems to have worked. You've got a pretty nice motor:

1. it's 3 phase
   
2. it's 1440 RPM (4 pole)
   
3. 750 W (1 HP) is a nice size with good power output. Out of the realm of 'toys'.
   
   

> Should I go for 4 poles on the rotor?

Yes, unless you want to do a full rewind. For a first conversion, I'd just build a new rotor and skip any rewinding.

But first things first, are you sure the windings are still ok and not shorted or open ? Do all phases have the same resistance when you measure them with a multimeter ? Have you tried running it as a motor ? If it runs fine, you can assume the windings are good. If it doesn't, you'd have to rewind it. And if it requires a rewind, you might as well make an 8-pole or 12-pole out of it. Same effort, better results (slower generator)...

Without a rewind, it would likely be best suited for 48V systems, otherwise it may end up with a too low cut-in.

Zubbly was the expert on re-connecting the coils; I've never done this (nor can this be done with most European motors). To me it looks complicated, but others have done it succesfully too.

Not sure if you've read these files by Zubbly yet. If not, read them before continuing:

http://www.otherpower.com/images/scimages/4/z_conversion_all.pdf

(notice that part-2 is missing in the above file)

http://www.otherpower.com/images/scimages/4/zubbly2.pdf

Lots of information packed in a few pages. Reading it just once won't be enough

Lots of pictures from Zubbly can be found here:

http://www.anotherpower.com/gallery/zubbly

As far as magnets goes, the rule of thumb of 150 W/c.inch is just that. I'd try to fit as much magnet in there as (reasonably) possible.

Don't forget to skew the rotor properly or you won't be able to turn it. I've written a short explanation on the topic, but from feedback I got I understand it's a bit too, ehm, technical. Math and stuff.

http://www.otherpower.com/images/scimages/3538/decogging_tutorial_V1.pdf

I'd also put in new bearings while you've got the thing apart. Depending on your machining capability, I'd make a complete new, all-steel, rotor with an aluminium sleeve. To me it seems the best and simplest solution, though making an aluminium sleeve is a lot of work and requires a lathe and drill press.

You have a lot of options, it's up to you to choose what best suits your situation, tools, time and experience.

Nice to see someone else starting a motorconversion project.

Peter.

[briskwinds]

Salute to you guys for such a great explanation. You are simply great.

Dinges, the motor is newly rewound by professionals (motor rewinding shop).

Forgot to tell in my post that the motor is of 24 slot type. So I will go for 360/24=15 degrees skew .

I am getting 3 types of disc magnets here. 13 X 10 mm , 10 X 10 mm and 20 X 6 mm disc magnets. Which type will be more suitable?

I have read it here that it depends on many factors. But still I wanted to know if 'X' HP of motor produce 'X' HP of power if converted nicely.

13 x10 mm magnets were purchased while preparing this comment. Is this placement of magnets and distance between the magnets good enough?

Not sure if I am approaching my limits in asking. If so please yell at me.

Regards.

[dinges]

First of all, the 15 deg. skew. Yes, the angle should be 15 deg, but in a different plane than what you have drawn in your drawing. See this image for an explanation.

http://www.anotherpower.com/gallery/decogging/figure_8

Notice that Beta should be 15 deg in your case, but, that Delta most likely isn't 15 deg.

Secondly, before continuing. I notice that your magnets are drawn pretty close together. Also notice that at the bottom, they will be closer together (and likely overlap...) because of the smaller radius of the rotor at that point.

What I'm trying to say: the way you have drawn the 'outlay' (is that proper english?) the magnets MAY not fit, depending on how you will build the rotor.

Another builder of conversions, 'Behoof', found this out the hard way...

Whether this is the case for you depends on how you will actually manufacture the rotor. Hopefully you can tell; will you make a new solid steel core, with an aluminium sleeve ? Or will you mount the magnets on the turned town old rotor (I understand that's your plan) and then cast in epoxy ? Please answer.

If you plan to place the magnets on the turned down old rotor, you should be calculating the skew angle using the TURNED DOWN diameter of the rotor, NOT using the final outside diameter of the rotor (approximately the diameter of the stator). Hope this makes some sense.

Anyway, try to read and understand page 6-13 and p.27-28 of the decogging tutorial:

http://www.otherpower.com/images/scimages/3538/decogging_tutorial_V1.pdf

If the math bothers you, give all the relevant dimensions as shown on p.28-top of the tutorial: heart-distance between the extreme magnets 'L', diameter D of the rotor-surface ON WHICH THE MARKING WILL BE MADE (important; and it depends on how you intend to manufacture the rotor) and the amount of magnets per line (7, by the looks in your above drawing). Give me that info and I can calculate your correct skewing angle for you.

"I am getting 3 types of disc magnets here.  13 X 10 mm ,  10 X 10 mm  and  20 X 6 mm  disc magnets. Which type will be more suitable?"

The most suitable magnet is the one which you can stuff the most of into the rotor. The more magnetic volume and area, the better. You'll have to make some sketches for yourself and see how much magnet you can put in. My gut-feeling is that 13x10 mm would be a nice size, though a bit high (7 mm instead of 10 mm would have been nicer); the 20x6 mm ones have a pretty large diameter so may need a lot of airgap to clear the stator. To give you an idea, my 130W genny used 12x6 mm magnets and the 3 hp used 20x10 mm ones. I think I'd choose your 13x10 mm ones but again, make some sketches to see how much magnet you can stuff in there, for the various sizes.

"But still I wanted to know if 'X' HP of motor produce 'X' HP of power if converted nicely."

I have converted a 130 W motor and did some measurements w.r.t. power output (battery charging) for various system voltages (12-24-48V). With its original winding (3 phase 400 Vac), the generator was able to put out 130 W in a 48V system; for 12V and 24V systems it was lower. You can find the measured data here:

http://www.anotherpower.com/gallery/album49/130W_genny_table_output?full=1

http://www.anotherpower.com/gallery/album49/130W_genny_graph_output_001?full=1

Notice that the maximum output was at 1000 RPM, which would be pretty fast for a windgenerator.

The above graphs are only an indication and your results will likely be different, they depend on how much magnet you put in there, the airgap, the grade of the magnets, the position of Mercury w.r.t. Venus, etc.

So, to conclude: your skew angle of 15 deg as you've drawn it is incorrect (it should be 15 deg, but in a different plane). Depending on how you intend to build the rotor (alu sleeve, cast in epoxy), the outlay you've drawn may or may not be correct as the magnets will overlap/touch eachother.

I hope the above wasn't too rambling. If some things aren't clear, let me know.

Peter.

[briskwinds]

I was too much quick on skew. The link dinges provided is too much explanatory. It will take some time to grasp things on skewing.

BTW, the nicke looking 13 x 10 mm magnets have arrived.

[briskwinds]

Thanks Dinges,

You have saved me from learning the hard way. I was wrong on drawing the magnet on the rotor plane without turning down.

I will mount the magnets on turned down old  rotor  and then cast it in epoxy.

Here I have drawn the layout. Please see if its correct or not. except the distance between the poles. I drawn 3 three rows of magnets per pole and the distance between the magnet poles came to nil.

So I have to remove 1 row from each pole. right?

[vawtman]

 Im not sure how deep the cage is on that rotor.If you happen to turn through it the lams will fall apart.

 Also,you should keep the pole width within the coil span.

 Happy holidays and good luck.

 Mark

[dinges]

Vawtman is right, the rotor may fall apart if you turn it down too much. Since your magnets are relatively high (10 mm) this could happen to you. Myself, I prefer to make a complete new rotor and loctite it on the original shaft (or machine an entire new shaft/rotor out of one piece) rather than re-use the old rotor.

I'm not sure I would remove one row per pole to create some extra space between the poles. In theory (see Zubbly's PDFs in my first reply) the pole should fit inside the coil:

http://www.anotherpower.com/gallery/zubbly/1_phase_and_skewed_mags

Notice how the magnetic pole ideally should ideally fit inside the inner coilspan. If the magnetic pole is wider, it'll distort the output waveform (see http://www.anotherpower.com/gallery/album49/130W_conversion_cancellation ) as happened in my case of the 130W conversion and it makes less good use of magnets.

In practice, for battery charging the perfect sine wave is less important.

I'm not really sure whether I would remove that extra row. I don't think it matters a lot, the last extra bit of magnet doesn't add much output (as during part of the rotation, one magnetic pole covers both legs of the coil). I think in a next one I'd make each pole just 2 rows wide (not 3 as you've drawn), but mostly to see what happens to waveform and output. I don't think it makes a lot of difference to output (it may have a big influence on waveform, though that's not very important in batterycharging). But I'm not 100% sure about it. Either way, the generator will produce useful power, as I've used both methods ('ideal' pole width vs. too large pole width) in different generators (in my 3 hp resp. 130W conversion).

Wrestling the magnets in place will be very hard, unless you use some kind of aid like a cage. I use aluminium cages, Zubbly has used both aluminium and GFRP cages in the past:

http://www.anotherpower.com/gallery/zubbly/2hp_rotor_showing_magnet_skew_1

http://www.anotherpower.com/gallery/zubbly?page=4

Making and drilling a cage is quite a bit of work but if you don't it'll be near impossible to keep all the magnets in place before the casting. Others have succesfully temporarily loctited the magnets in place, but that was using flat, milled surfaces to mount the magnets on, not using curved turned surfaces.

BTW, I noticed in the picture that it shows 75 magnets; for your 3x7x4 configuration you will need 84 magnets; if you go for 2x7x4 = 56 it should be enough with some spares in case of a mishap.

Don't use polyester ('glassfibre') resin in this application. Only epoxy will do. And make sure you properly treat the magnets (degreasing) before glueing them in place.

In your case the magnets are relatively high, which means you could have a slightly bigger airgap (nice to have a bit more mechanical clearance to the stator...) without seriously reducing flux density.

Peter.

[briskwinds]

Hey Peter, you have totally caught me unaware.

In the previous drawing not uploaded here, I made three rows with 6 magnets per pole and total 4 poles. i.e. 6 X 3 X 4 = 72 + 3 (extra) = 75 magnets. In the drawing uploaded next I drawn it with 7 magnets per row and didn't calculate the no. of magnets when buying. Sharp eye dinges.

On further inquiring I am also able to get these magnets size now.

 -> 13 X 7 mm

 -> 10 X 5 mm

 -> 8 X 5 mm

I am now planning to get 13 X 7 mm magnets for this conversion and putting 13 X 10 magnets for future bigger motor conversion.

I am planning to go with PVC water pipe for the cage.

The coil arrangement between the stator teeth is like this.

The coil is wound like 1 st to 6th slot, 3rd slot 8th slot and so on. The distance between slot to slot (or teeth to teeth) is around 10 mm. So I should aim for 10 X 5= 50mm wide magnet pole (Blue arrow in the above diagram).

The rotor circumference is 240 mm.

50 mm per pole x 4 poles = 200.

That leaves 240 - 200 = 40 mm of total non-magnetic space(?) space. That is to be divided between 4 poles.

So 40 /4 = 10 mm gap between poles. Is this OK?

(Maths and science were my strong points till 10th std. not English.)

[briskwinds]

Hvirtane, couldn't understand this.

You have placed the magnets on shaft which has poles on sides.  What I understands is that one pole should be placed on shaft and the other will come on top.

[dinges]

Hello Briskwinds,

The blue arrow seems to be correct; keep the magnetic pole width about equal to the inside of the coil. If the magnetic pole would be wider, it would 'cover' both legs of a coil at the same time. Your calculations look ok to me.

PVC for the cage would be fine if epoxy adheres to it properly. I don't know if it does, you'd have to verify for yourself.

If you already have the 13x10 mm magnets I'd use those and not order 13x7 mm ones. In case you want to do a larger conversion in the future (say, 3 hp) you will likely want to use much bigger magnets anyway (I used 20x10 mm in a 3 hp). My reasoning for using slightly lower (7 mm) magnets is that you could mount those closer together without the bottoms touching eachother. But if you plan to use 2 rows of 7 magnets per pole then you have plenty of space anyway (as you could theoretically fit 3 rows in) so the few mm of extra height of the 13x10 mm magnets is not much of a problem anyway. But, your call.

I did a quick calculation, if you use 56 pcs. of 13x10 mm magnets you end up with 4.5 cubic inch in your 1 hp conversion. For comparison, my 3 hp used 7.6 cubic inch. So you're using relatively much more magnet than I did in my first conversion, which is a good thing.

Grtz,

Peter.

[dinges]

I forgot; the lower height magnets may be better after all if it means you don't have to turn down the rotor so much. Turning down the rotor too far will make it fall apart (unless you take countermeasures). Check out this story by Behoof, for example:

http://www.fieldlines.com/story/2007/8/27/51542/1469

You can see that his rotor came apart during turning down and he had to resort to welding it back together, turn a bit more, weld again, etc. Keep that in mind when turning down. It's another reason I like to use new (solid steel) rotors, seems much simpler to me in the end. Depending on the construction of your rotor, you may/may not be able to turn it down enough without it falling apart.

[briskwinds]

Rotor turned down.

One thing to notice is that the stamping to aluminum ratio has decreased. What I meant to say that the aluminum lines have grown thicker and stamping lines became thinner.

[briskwinds]

Rotor filled with resin.



First prepared the cage for the magnet with PVC pipe and filled it with resin. But the fill area was too small (2 mm). So I removed the PVC pipe by cutting it  when resin hardened and made the filler area by wounding cello tape around the magnets. And then filled it.

Dinges told to use epoxy, but I used resin as it was at hand. Too lazy to go out and search for epoxy.