Author Topic: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo  (Read 5984 times)

0 Members and 1 Guest are viewing this topic.

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
I have a Samsung DC31-00049A washer motor. 36 pole stator, 24 magnet rotor.  It came equipped with nice curved ceramic magnets, I presume to minimize motor cogging/noise. It appears to be rated at 0.94HP as a motor. (see attachments)

I'll obviously be using it as a 3-phase alternator, but my application needs to maximize power output at very low RPMs... so I'm contemplating upgrading the magnets to Neo's.  This is NOT a wind turbine, my application has lots of torque, but runs at a very low speed.

My question is about stator finger size to magnet width ratio... and how to optimize this.

The 36 slot/finger stator has an OD of 9.862".  Each pole-piece (stator finger) width is 0.700", with approx. 0.160" gap between fingers.  (height is 1.270").

With the original magnets, the ID of the magnet 'tips' is 9.940", giving real close to 0.040" magnet-to-stator clearance (1mm). Each magnet overall width is 1.340" at its widest point.

Would my goal be to match the stator finger width of 0.700" with the magnet... or go oversize to maybe 1" wide (common)... or go undersize to 0.625", or  ?

The calculated "tip-to-tip" distance of the OEM magnets is 1.297" (the "chord" length of a 15 slice at 4.970" radius). If I assume I want 50% of that as my magnet width, I end up with 0.6485" magnet width & gap... close to 5/8" or 0.625".  This seems logical...

I'm not concerned with the sine-ness of the output waveform, I'm concerned with maximizing power and efficiency at low RPM.   

I'll be using a power conversion circuit that properly loads the copper, so we can ignore Voltage/Current issues here.

I know... there's millions of variables... but I'm hoping someone has real magnetic machine design engineering experience here... not just more guesses like I have.   ;)

Thanks in advance for any help!
« Last Edit: April 17, 2016, 02:49:09 PM by ronbot »

joestue

  • Hero Member
  • *****
  • Posts: 1365
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #1 on: April 17, 2016, 03:47:10 PM »
Increase the number of magnets, rewire it. also looks like you can fit a lot more copper in there.

You have many options, all of them are better than 24 magnets, I don't know why they used that number, but they might have been running into iron loss limitations at the intended rpm. alternative connections here: https://www.emetor.com/edit/windings/

Anyhow since you want low rpm torque iron loss doesn't matter, limit is the magnetic flux and saturating the core.

you don't need much for neodymium magnets, you could even use 15 magnets, and 15 blank pieces of metal, once the core is saturated there is no further gain with more magnets.

so i would choose a number of magnets that fit the motor . 1 inch wide magnets would fit the space you have with 30 or maybe 32 poles.
if you use 1/8th inch thick magnets you will have to fit steel backing pieces to get the magnet up close to the pole.

1/4th inch magnets are cheaper per cubic inch though but probably overkill. maybe 1/4th inch thick magnets, half of them, the other half metal blanks.

oztules

  • Hero Member
  • *****
  • Posts: 1473
  • Country: aq
  • Village idiot
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #2 on: April 17, 2016, 06:25:16 PM »
I have to disagree with Joestue... which is dangerous, as he knows what he is talking about.

But in this instance, the long salient poles mean that current limiting will occur very early, and as such your current may be as low as 5-8 amps max. It will not increase beyond there regardless of the load impedance..... so if you stick to normal magnetic theory, you will do little better than that.

However, empirical testing shows, that the only way you will do any better is to ignore the steel altogether ( almost). As soon as you saturate, it well be seen by the magnets as air core not iron core .... and so at the expense of magnetic efficiency.... you stack as much neo as you can... It won't take much to saturate it anyway, so forget that, and drive as much flux as humanly possible into the coils.... Thick as you can, as tight as you can... and ignore inter magnet leakage arguments, they are all true, but it is the cost of driving flux down into the coils.... there is no other way.
Even then it is probably only a doubling plus a bit in current that will be attained... air core is not efficient, but works to make the thing more powerful.

Beware, in it's original ferrite guise, you will not burn it out shorted all day, with the neo's stacked in there it will be possible to over heat the thing I suspect

If you rewind then that will be a different matter, but the same logic applies.

A rewind that gives you  a lower volt, higher current winding by using more poles for the same voltage, will allow higher currents before the current limit kicks in.

In these motors/alternators, it is the back MMF that will dictate the performance at the top end. When the coil back MMF can get to equilibrium with  the magnet MMF, then current limit sets in, and cannot be beaten by rpm or anything else.

If you can let the voltage rise as the current stalls, then your electronic gadgetry will save the day, and you can use Joestue's answer instead, as the current stall is not the end of your power output in the case of a buck converter... let the rpm rise, voltage rises, so power increases.

......oztules
« Last Edit: April 17, 2016, 07:18:33 PM by oztules »
Flinders Island Australia

joestue

  • Hero Member
  • *****
  • Posts: 1365
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #3 on: April 17, 2016, 09:44:54 PM »
Increasing the number of magnets gets you proportionally that many more volts.. which means less amps. however, the magnets are closer together which means more leakage inductance.

It might be a wash, in the end with regard to current limits caused by leakage inductance.

Biggest difference between 24 poles and 30 poles is the winding factor of .866 compared to .966, quite a bit more voltage for free, so to speak, efficiency will be improved when it comes to copper loss, but iron loss will be 25% higher.

you will want to measure the volts per hz and calculate the flux density as it is now, with ceramic magnets. its possible there will be little improvement with neodymium.

oztules

  • Hero Member
  • *****
  • Posts: 1473
  • Country: aq
  • Village idiot
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #4 on: April 17, 2016, 11:23:31 PM »
It has been done a few times on the backshed forum... there are increases to be had.

You will do much better using capacitors, and leaving the stator with the ferrites... they are close to saturation already I think.

.............oztules

Flinders Island Australia

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #5 on: April 30, 2016, 03:30:48 AM »
It has been done a few times on the backshed forum... there are increases to be had.

You will do much better using capacitors, and leaving the stator with the ferrites... they are close to saturation already I think.

.............oztules

-------------------

OK... I'll bite... capacitors to do what???
Please expound or link?

There's a lot of core here... (maybe 6-7 pounds) I'd really be surprised if it's anywhere near saturation with the original radiused ceramic magnets.

More/narrower magnets... if I run 32 magnets, that seems like it would give me more of what I'm wanting... more flux transitions per revolution.  HOWEVER... the magnets will be bumping right up against each other. The ID of the "can" is 10.75", which gives a 33.77" circumference, divided into 32 pieces is 1.055"..... but since the 33.77" is at the very outer edge of the INSIDE of the rotor can, the angle between adjacent magnets will cause their faces to be up against each other... and may not even fit depending on thickness.

If I use 1/4" thick magnets, 1" wide, glued right up against the "can" wall, there will be 0.005" gap between the front edges of the magnet faces.  This increases the magnet-to-stator gap of 0.194", up from the original gap of 0.040".  (it seems like a 0.750" wide, 0.375" thick would work better... less inter-magnet leakage, and better stator 'coupling'.


I ran some tests today, re-wired to Delta to get reasonable voltages I that can convert with my eventual MPPT inverter.
(first round of testing gave voltages in excess of 200V DC, but it was very low current)


electrondady1

  • SuperHero Member
  • ******
  • Posts: 2836
  • Country: ca
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #6 on: April 30, 2016, 06:42:16 AM »
you should check out the back shed site . they have many years of experimenting with those washing machine motors as alternators. the capacitors mentioned are part of voltage doublers and triplers

oztules

  • Hero Member
  • *****
  • Posts: 1473
  • Country: aq
  • Village idiot
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #7 on: April 30, 2016, 03:46:17 PM »
Yes Gordon ( Gwatpe) did a lot of work with caps on F&P over on the back shed, and managed to get 1kw out of a single stator... it had doublers and tripplers.

However, you can get a very substantial increase just with series capacitors, as attested by Glen here http://www.thebackshed.com/forum/forum_posts.asp?TID=3411&KW=neo+conversion+&PN=0&TPN=2

There has been a lot of contentious claims  by some very smart people about how the caps actually do it, but the net result is they do, and no-one really nailed it as to why.

The backshed is full of material on this, as was this site years ago, me, flux, and a few others all had a go at it..... would not know where to find it here.

You can get if your very  lucky X2 increase with neo, more usually 1.5x..... you will do as well with caps in series/parallel. It changes the way the core MMF works against the magnet field. Your figures would indicate that the current limits around the 3+ amp mark... add caps and see what happens is the best way to do it.


.............oztules
Flinders Island Australia

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #8 on: April 30, 2016, 05:29:24 PM »
Increase the number of magnets, rewire it. also looks like you can fit a lot more copper in there.

You have many options, all of them are better than 24 magnets, I don't know why they used that number, but they might have been running into iron loss limitations at the intended rpm. alternative connections here: https://www.emetor.com/edit/windings/

Anyhow since you want low rpm torque iron loss doesn't matter, limit is the magnetic flux and saturating the core.

you don't need much for neodymium magnets, you could even use 15 magnets, and 15 blank pieces of metal, once the core is saturated there is no further gain with more magnets.

so i would choose a number of magnets that fit the motor . 1 inch wide magnets would fit the space you have with 30 or maybe 32 poles.
if you use 1/8th inch thick magnets you will have to fit steel backing pieces to get the magnet up close to the pole.

1/4th inch magnets are cheaper per cubic inch though but probably overkill. maybe 1/4th inch thick magnets, half of them, the other half metal blanks.

= = = = =

Educate me a bit here...  I want to understand the math behind this all.

If I have 36 stator slots/teeth, 24 magnets gives me a clean 3-phase output with a little cogging.  It's easy to visualize how the 4:3 ratio works out.

Moving to more magnets/poles would obviously give me a higher frequency output... this is what I need.  Do I lose anything by going this route? It 'looks' like it would be lower in power (flux paths seem odd)... but I'm 'looking' at it from an uneducated perspective.   (I won't have the tight flux path I do now, if/when I switch to neo's)

Rewiring (reconnecting) all of the stator teeth is a possibility, but rewinding is not... for now.

Again... I don't require any specific output Voltage/Current... just Power. My regulator circuit wants the rectified DC voltage from the alternator to be above 15V, less than 300V, and my need is for at least 400Watts, and my "desire" is under 300RPM to minimize the drive ratio. I'll be stepping up from about 30 RPM constant at around 90 pound-feet of torque. If I run a 10:1 drive ratio, I have 300RPM and about 9 pound-feet torque.

I'm unable to find the Back Shed posts about capacitors... but multiplying voltage doesn't really help me, and capacitors won't multiply power.

oztules

  • Hero Member
  • *****
  • Posts: 1473
  • Country: aq
  • Village idiot
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #9 on: April 30, 2016, 08:06:29 PM »
"I'm unable to find the Back Shed posts about capacitors... but multiplying voltage doesn't really help me, and capacitors won't multiply power."

Then suggest you learn to search better... It will save you a lot of wasted effort. That site has done it all before in every way and combination.... you need caps well before you need neo has been the consensus...... or to quote the man who owns the site... Glen....

" I know from the comments posted o this forum over the years that neo magnets can easily saturate a laminated iron core. Fitting rare earth magents to a F&P hub will give a gain in power, but not 5 to 10 times that you would expect, closer to 1.5 to 2 times the power from the factory ceramics magnets. The neo magnets saturate the iron core, and a lot of the flux is wasted. Once a core is saturated, adding more flux wont increase power.

Its proven easier to increase the F&P output with series caps and a few other tricks, leaving the original ceramic magnets in place. "
============

So I suggest you run your tests again with a few hundred uf of series caps, and check your results again, parallel caps will make a big difference too.
Then reconfigure the windings for lower voltage....still more current ( see backshed for configuration and how to...)

have fun with it.... and it is no crime to re-invent the wheel, just not sure it is useful.

.............oztules
« Last Edit: April 30, 2016, 08:12:29 PM by oztules »
Flinders Island Australia

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3190
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Direct Drive Washing Machine motor as Alternator: upgrade PM's to Neo
« Reply #10 on: April 30, 2016, 11:30:10 PM »
Hi Ron,

I experimented with these on my turbine, a few years ago.

http://www.fieldlines.com/index.php/topic,140926.0.html

Don't worry about the missing images from the first few posts - I replaced them in an update recently.
Hope it fills you in a little.  Everything I tried was inspired by the Backshed discussions at the time.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

oztules

  • Hero Member
  • *****
  • Posts: 1473
  • Country: aq
  • Village idiot
Good discussion there Sparweb..... but in this case I think Flux hit it on the head... this fellow is not trying to increase voltage from the looks of it, just to up the grunt... for the F&P the only way is really to use capacitive coupling ( not doublers persay ) to get rid of / mitigate the leakage of the long poles.
It also is suitable for this as the pole count is high, so Cx is more effective per farad.

The F&P and the induction conversions all suffer from  reactance limiting, and so a current limit wall will be hit... but the F&P do it too soon for practical purposes and need a fix.

Flux had this to say.... and I think he is mostly correct too.

"At the high end series capacitors can neutralise the leakage reactance and make the output dependent only on the winding resistance as it is in an air gap machine. This is where you can gain real top end power.

If you are using the  F & P or similar salient pole cored machine there is a very strong case for using the capacitors. For any other machine there is not a similar argument "



It will be noted, that the winding resistance is the least of your limitations with the F&P.... if it were you would be able to burn it out, but it is not possible, as the reactance will kill off the power well before that occurs.... only much more magnet... or simply more capacitance will partially cure this.

Experience tells us, caps are the best answer, and a few more tricks in the wiring will be the best.

Doubling and trippling were for wind matching purposes, we still need the series caps to get useful output, or stack stators as well.


..................oztules
Flinders Island Australia

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3190
  • Country: ca
    • Wind Turbine Project Field Notes
You're right (of course) Oz.  It's been a while since I worked on these so that distinction wasn't clear.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
Much more testing later... and I've come back to bother you some more...

I'm not going to be able to get the output Power (Watts) I need at the low RPM's that I'll be able to turn this at... with it being "stock".

I did some research on F&P models, and they had one with 36 slots (like mine), and 48 magnets... twice of what mine has.  Their stator wiring was exactly the same as what I have now... simply ABCABC.  The main difference is that, due to the narrower magnets used, the stator fingers need to be narrowed a bit. 

This actually looks fairly straightforward... other than these stator fingers/poles are grooved. (I assume it's to further minimize cogging noise).  I can trim the tips of the stator poles to decrease the width from 0.700" to about 0.625". Then I can use 0.625" (5/8") wide magnets (by 0.375" thick). I'll still have a 0.060" gap between the stator fingers and magnets, which is about 0.004" closer than stock.  Gap between magnets (at the tips) will be 0.030", which should be easy to manage using alum. shim stock.

Here's my question...
I referenced the EMETOR Electric motor winding calculator, and for both the 35/24 and 36/48 is shows different winding configurations... it shows AaBbCc for both.
https://www.emetor.com/edit/windings/

Why is that?  Why would it not use ABC like I've seen the F&P motors?

The "end result" would be a 2x output frequency multiplication, and the effective increase in Voltage. I'll be using N45 magnets (cheap), so Current will also increase - apparently until it saturates the core.  For my application, this drastically reduces the complexity of the drive system... spinning it at 300 RPM will be like 600 RPM as it stands now... and maybe a bit better with the stronger magnets.

As I said before... I'm pretty much a noob in this "motor winding" field, but I don't see anything glaringly obviously wrong with this idea... but I try to be teachable!

Thanks for any/all help!!  :-)

joestue

  • Hero Member
  • *****
  • Posts: 1365
The emetor chart looks the same as the winding convention posted,  AaBbCc. you have two choices: wind every other tooth, or wind all the teeth. (the winding factor is the same for both btw, 0.866) click the button for the 2 layer winding layout..(in this case, not actually two layers)

For 30 poles, winding every tooth gets you a winding factor of 0.933, winding every other tooth gets you 0.966.. (but winding every tooth gets you nearly twice the surface area for cooling if you can blow air through the coils. )

for the moment i am convinced that higher winding factor means higher amps, (when operating at the reactance limit). but you're only going to get a 10% increase from .866 to .966

Anyhow, yes you get twice the volts out with twice as many magnets. but the frequency is twice as high which means the leakage inductance makes twice the inductive reactance. which means you've not gained anything except double the iron loss.

Also because the magnets are closer together, there is more leakage flux. but I'm not sure it matters once you get beyond about 8 poles.
there is a huge increase in leakage flux when increasing the pole count from 2 to 8. this is why high pole count induction motors (6 or more) have significantly lower power factor than 2 and 4 pole motors.

if you spend the time to do it, you should be able to simulate the motor on that website.
« Last Edit: May 07, 2016, 03:31:55 PM by joestue »

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
The emetor chart looks the same as the winding convention posted,  AaBbCc. you have two choices: wind every other tooth, or wind all the teeth. (the winding factor is the same for both btw, 0.866) click the button for the 2 layer winding layout..(in this case, not actually two layers)

I'm a bit confused by the above... both the Samsung motor that I have (36 slot, 24 mag), as well as the F&P motor mentioned (36 slot, 48 mag) are wound ABCABCABC, no stator poles have reverse polarity windings. What's the difference  / benefit of side-by-side Aa Bb Cc windings with adjacent reversals?  It seems it would defeat my goal of higher frequency at lower RPM's.


Quote
... but the frequency is twice as high which means the leakage inductance makes twice the inductive reactance. which means you've not gained anything except double the iron loss.

The frequency related losses would be the same for me, since I would have to spin the rotor twice as fast (with 24 magnets) in order to get the loaded voltage I need... 600RPM for 24 magnets, 300RPM for 48 magnets, so it seems like a wash.  At low speeds, the 24 magnet version just is too inefficient.

I looked into trying a simulation, but I have far too many unknowns... lamination material, magnet material, etc.

I've drawn the motor in Autocad (actually, the free clone "DoubleCAD"), with both 24/48 magnets, and with 48 magnets, 180 electrical shift only requires 7.5 of physical rotation... it all lines up perfectly "as wound" (ABCABCABC), and gives 24Hz/revolution.


Again... I'm trying to learn, hoping not to seem stubborn, I just need it to make sense.

joestue

  • Hero Member
  • *****
  • Posts: 1365
The ABCABC format you mention is the same as emitor's UuVvWw when describing the motor we're talking about.

V means the wire comes up out of the motor towards you, then goes back down the motor through "v".

Vv can be replaced with "A" as you describe if the "A" refers to a complete coil, not half of a coil.

It gets confusing to refer to a complete coil as "A" (and a coil wound in the reverse direction as "a") when you need coils that overlap, such as a 36 slot, 4 pole winding which would look like this:
AAAbbbCCCaaaBBBcccAAAbbbCCCaaaBBBccc. (coil 1 goes up through the first "A" and down through the first "a" etc, for 18 complete coils)
Quote
my need is for at least 400Watts, and my "desire" is under 300RPM to minimize the drive ratio

Honestly I don't know if you can get that much torque out of it. seems doable, your motor is far larger than my conversion, which could deliver 300 watts at 850 rpm no problem with 1.875 cubic inches of magnets.
« Last Edit: May 07, 2016, 10:05:16 PM by joestue »

ronbot

  • Newbie
  • *
  • Posts: 10
  • Country: us
The ABCABC format you mention is the same as emitor's UuVvWw when describing the motor we're talking about.

V means the wire comes up out of the motor towards you, then goes back down the motor through "v".

Vv can be replaced with "A" as you describe if the "A" refers to a complete coil, not half of a coil.

It gets confusing to refer to a complete coil as "A" (and a coil wound in the reverse direction as "a") when you need coils that overlap, such as a 36 slot, 4 pole winding which would look like this:
AAAbbbCCCaaaBBBcccAAAbbbCCCaaaBBBccc. (coil 1 goes up through the first "A" and down through the first "a" etc, for 18 complete coils)
Quote
my need is for at least 400Watts, and my "desire" is under 300RPM to minimize the drive ratio

Honestly I don't know if you can get that much torque out of it. seems doable, your motor is far larger than my conversion, which could deliver 300 watts at 850 rpm no problem with 1.875 cubic inches of magnets.


Confusing indeed... I had no idea ABC was not equal to UVW (or UWV). 

I always thought the lower-case letters (like AaBbCc) meant the same as a letter with the "-" (like U+U- V+V- W+W-).... meaning coils with the "-" would be connected in opposite polarity, such that two physically adjacent coils "U+ U-" were two stator teeth connected in opposite-polarity-series.

On the Emetor page, I see this for the 36slot-24mag-1layer:



...and this for the 36slot-48mag-1layer:




This is how these stators I'm discussing are wired, except I have mine connected in Delta, not Wye:


I see this described as ABCABC...


Is this the same as:   U+ U- W+ W- V+ V-   ??


SparWeb

  • SuperHero Member
  • ******
  • Posts: 3190
  • Country: ca
    • Wind Turbine Project Field Notes
Ronbot,
I wish I was much more familiar with motor winding that I could help you directly.
If you're willing to invest in a book that would help you, I recommend this one:
Rosenberg, Electric Motor Repair
It's an old book, more like a shop manual, and very thorough.
Part of the problem I'm reading in your last post is the naming convention that NEMA now uses versus the older one, which you can see in this old book.  Rosenberg is much more interested in showing how it's done than how the wire coils are named.  The book is literally 50% illustrations.

I just found it on Amazon and Alibris, but you had better keep looking (maybe e-bay).  It's a bit spendy... 
I think they classified it as "university textbook" but this thing seriously ain't going to be seen at any university. 
I'm really disappointed to see the ridiculous pricing, now that I've checked.  I wouldn't recommend a book that I thought was too expensive to buy.
I only paid 40 or 50 bucks for my copy.  With a little hunting you should turn up a better price.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

joestue

  • Hero Member
  • *****
  • Posts: 1365
Confusing indeed... I had no idea ABC was not equal to UVW (or UWV). 

I always thought the lower-case letters (like AaBbCc) meant the same as a letter with the "-" (like U+U- V+V- W+W-).... meaning coils with the "-" would be connected in opposite polarity, such that two physically adjacent coils "U+ U-" were two stator teeth connected in opposite-polarity-series
.....
I see this described as ABCABC...
Is this the same as:   U+ U- W+ W- V+ V-   ??

when you refer to the letter A, you are referring to a complete coil spanning one tooth, wound in say, a clock wise direction. yes, this is equivalent to Aa, or U+, U-
It gets confusing to use the letter "A" to refer to a complete coil when you need a coil to span more than one tooth and overlap another. (how would you know which slot it goes in?)

The diagram you see on emetor shows you the teeth, wire goes up through the page towards you and back down through the page away from you.
the diagram you posted is for an 18 coil machine. you need to click the button for show winding layout for 2 layers. it will say:
 "The above winding pattern is repeated 12 times in order to represent the entire winding!"

so yes, the wiring is the same for 24 and 48 poles, except the motor will spin in opposite directions. this is why one is UVW, the other is UWV

the terminology people are using on the other websites etc, to show how to wire concentrated pole motors all assume that you are winding every tooth, "A" then refers to a clockwise coil, a refers to a counter clockwise coil. this doesn't work for overlapping coils spanning more than one tooth.
this is what you want:
http://i.imgur.com/BasuvvS.png
in that image you see three coils, all wound clockwise.. repeated 12 times..
« Last Edit: May 08, 2016, 11:19:26 AM by joestue »