Author Topic: 3ph conversion Magnet questions.  (Read 4934 times)

0 Members and 1 Guest are viewing this topic.

Tinkertony

  • Newbie
  • *
  • Posts: 4
  • Country: us
3ph conversion Magnet questions.
« on: April 14, 2014, 12:39:44 PM »
New to Wind PMG gennys but 25yr experienced AF acft electrician here. I am converting a GE 3ph 230/460v SF 1.15 motor to a PMG for eventual use on a wind turbine of some type. I have some questions for the experienced here about the rotor magnet arrangement and how it may affect flux strength. I am taking the advise to research EVERYTHING prior to doing anything to the motor and have read the posts by Master Zubbly and others on conversions. I will be using Neodymiums of course. I am not a Magnet expert and haven't found clear answers yet so I'm asking the experts! Here are the questions on the magnets:

1. What effect in flux will there be if the sides of the magnets touch each other instead of being separated by individual holes like in Zubblys conversion. This is to maximize the number of magnets in the pole area on rotor. The idea is to set the mags in a tighter "honeycomb" style arrangement instead of the linear style.

2. In a few conversions they used 1/2" and a combo of 1/2" and 1" dia mags. It was said to "cram as many mags in space as possible" hence the efficient Honeycomb arrangement. Would using smaller mags, say 1/4" dia, be better due to increased number of magnets in limited space and less "un-magnetized free space"?

3. Also, a benefit of the smaller diameter would be a reduced rotor/stator gap which should increase flux force, Correct?

4. How much does the height of the mag effect the strength? i.e. the taller the mag the stronger/better? Twice as tall/ twice the strength?

5. Magnet cups! Would it be good to use them to "focus" or increase strength of pole mags? Here is a page I found explaining them. Info is halfway down page... http://www.leevalley.com/en/Hardware/page.aspx?p=40077&cat=3,42363   

I have access to a mill and lathes, rotary tables, indexing plate and may have lead on CNC mill with controlled vertical rotary axis so machining the rotor is no issue. I am here to learn and this is my first post so all Jedi, please be gentle on this Padiwan! lol  ::)

I am a Disabled Vet, live in a 38ft Motor-home (nice one ;)), and want to reduce or eliminate my elect bill. Possibly grid tie and feed back into grid to earn credit for high use periods (Heating/ Air-conditioning). I live in S/W Oklahoma so wind is OVER ABUNDANT nearly every day! I live between  the Weatherford WF (147Mw), Rocky Ridge WF (148.8Mw) and Blue Ridge WF (over 325Mw).
Where the wind blows HARD 95% of the year!

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: 3ph conversion Magnet questions.
« Reply #1 on: April 14, 2014, 01:22:48 PM »
1
The more magnet you squeeze in the better but if they touch you will gain very little over narrower ones that leave a gap of half magnet width. With the cost of neo , not worth making them wide enough to touch.

2 size mainly depends on the space available in your motor, large diameter in a small rotor will make poor use of space, the more smaller ones you use the more work you make  for yourself, it is a compromise.

3
correct but you need magnets thick enough .

4 Related to last point, with close air gap you will need at least 3/8" thick magnets, if air gap is longer then 1/2" is common, Smaller will work but at reduced output, much thicker will staurate the core and give little benefit for extra cost.

5 Forget the whole silly article, it is about lifting magnets you will get in big trouble. The flux needs to be in a very different place for large lifting forces.

Hope this helps a bit.

Flux

Tinkertony

  • Newbie
  • *
  • Posts: 4
  • Country: us
Re: 3ph conversion Magnet questions.
« Reply #2 on: April 14, 2014, 03:45:24 PM »
Flux thank-you so much for the explanations. I like to know WHY something wont work, NOT just "It doesn't"! That's how I teach others when they have questions in my "fields" of expertise! lol sorry for the pun!

 I didn't think about the core saturation. Since the rotor core is metal, would it be better to strip the shaft, machine a new non ferrous plastic/resin core to mount the magnets in?

 How would that affect the field fluxes, would you still get saturation? Magnetic field theory is starting to fascinate me... Time to machine for more of smaller magnets is no issue, Lots of time to fill :)
 
 I now see where magnet cups would work for lifting and not for induction into coils. Induction requires the mag field to pass through wire as perpendicular (90deg) as possible to induce the electron flow (hence the closer to the winding the better, further away the fields are arching back toward opposite pole). The cups seem to have too short a flux path and would probably self cancel out the induction effect due to N and S being too close to each other.

When selecting a given mag, irregardless of face dia, the "N" rating is the critical measure correct? Just an example, If you had a 1/2"w x 1/8"t mag and a 1/4"w x 1/4"t mag that both were N-52's, They are essentially the same strength and could be cross substituted?
Again, please forgive if I ask a stupid question...
Where the wind blows HARD 95% of the year!

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: 3ph conversion Magnet questions.
« Reply #3 on: April 14, 2014, 05:43:30 PM »
Firstly I think I misunderstood your first question. For the magnets of one pole it is best to cram them together as close as possible, the ideal is a solid neo block of the correct shape and curvature.

What I originally thought you meant was crowding more magnets into each pole so that adjacent poles touched.

Saturation is inevitable with any iron cored machine, the point you will reach it first is in the slot teeth as they have less area than the rest of the iron circuit.

No don't make the rotor of any non magnetic material, yes it will prevent saturation, but that is not what you want, you need enough flux to bring the teeth near saturation and you need the best iron circuit in the rest of the path to do this with minimum magnet. This is why you need to keep the air gap short as that is where most of the field goes. Making the rotor of non magnetic material is basically the same as making a very large air gap, all materials other than iron behave much as air.

Finally the factor that determines the difference between neo grades is the flux density, the coercivity is similar for all grades. The higher the N number the higher the gap flux for a given thickness.

For an alternator you need the flux to pass from N pole through the stator core back to S pole and complete its path through iron in the rotor.

Lifting magnets need the flux to take the shortest path and the cup brings the N pole very close to the S pole. The flux path is very different.

Flux

Tinkertony

  • Newbie
  • *
  • Posts: 4
  • Country: us
Re: 3ph conversion Magnet questions.
« Reply #4 on: April 14, 2014, 08:43:15 PM »
HEY I got some of it right! ;D Wouldn't it be great if we could get a single magnet of correct skew and radius (curved parallelogram)? Then we could just turn down the rotor and attach mags...

After reading and re reading your explanation of flux paths, it is slowly sinking in and I'm beginning to visualize the flux and why the rotor core needs to be iron. Makes sense now...

 So simply, DO NOT just drill holes in rotor surface and epoxy mags in flush (would be the same as mag cups)...correct?
Instead turn down rotor an equal amount as the height of mags being used, then encapsulate in resin like Zubbly did?
My heads ready to explode from trying to visualize flux, saturation, coercion, density... I will be researching all this much more until completely understood, but for now I will do what you say is best! lol

I did find a pic of Square packing and Hexagonal packing, which is what question #1 was about, but not explained well by me. Where Zubbly's was "loose" square packed (not knocking it, just an observation) and I wanted to "tight" hexagonal pack, which should be better because of more mags in same pole area "parallelogram" (skewed rectangle). and yes, there will be a space between the adjoining rotor poles. BTW my rotor is 48 bars and 3.653" dia. Field is 36 slot. pretty standard...

Where the wind blows HARD 95% of the year!

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: 3ph conversion Magnet questions.
« Reply #5 on: April 14, 2014, 09:18:04 PM »
If you want to throw money at it, I would buy 8 x .75 inch wide, and 4 x .5 inch wide by .25 inch thick magnets. (for that motor, 2 inch long is probably sufficient?)

Simply machine slots in the surface of the rotor for the magnets to be press fit into the core. leave the aluminium in the core, however, cut off the disk on both sides that band the 48 bars together.

12 magnets 1 inch wide won't fit. you might be able to do 12 magnets .75 inches wide, but i doubt it.
nnn-sss-nnn-sss for a 4 pole configuration, which is likely what you have. i wouldn't bother with a two pole conversion.
the steel left between the n and s poles will short out some flux but it will reduce cogging, if the magnets are located at the correct angular position

it is easy to cut the slots deeper to make the air gap larger, adding metal isn't, so i would start with .25 inch deep slots and see what you get.

the reason i say throw money at it, .75 inch wide magnets may be hard to find.
« Last Edit: April 14, 2014, 09:24:30 PM by joestue »
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

Tinkertony

  • Newbie
  • *
  • Posts: 4
  • Country: us
Re: 3ph conversion Magnet questions.
« Reply #6 on: April 14, 2014, 11:01:49 PM »
I just measured the phase inner winding tooth spacing of the stator. It's 4 tooth and spaced @ 1.250". Seems too narrow as Zubbly's shows about 2 inches and 7 teeth between inner winding coils. Did he rewind the stator in 3 pole config and mine is still 4 pole? Ugh, Hmm will the OEM windings work for a genny or will I have to rewind stator? I'm suffering from brain fry...

Anyways, What is the purpose/effect of turning the aluminum end bars off the rotor? Some don't seem to do that...
Rectangle or cube may be a way to go. I'll keep it in design ideas...

Whoever said "do not touch the motor or order anything until you research for a month" knew what they were talking about! I wish I had someone with your guys knowledge here where I could hangout in the shop with and learn hands on... And thanks for your patience again :-)
Where the wind blows HARD 95% of the year!

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: 3ph conversion Magnet questions.
« Reply #7 on: April 15, 2014, 12:04:18 AM »
3 pole don't exist but you might mean 6 poles. car alternators are 12 poles.

i don't know which motor you're looking at but yes if there are 7 teeth between inner windings then he probably rewound it.
seems most lightweight 3 phase motors i've seen have the worst possible winding layout, and they have 3 or 4 coils per group, spread out and overlapping or they are wound concentrically.

go here:https://www.emetor.com/edit/windings/

it appears you have a 6 pole machine.
click on the box with a "2" in it, under 6 poles and 36 slots.
then scroll down a bit and click the "show winding layout" for the various options.
in order to get 4 teeth spacing you have to select the winding configuration with a fundamental winding factor of .83

8 pole machines have 48 teeth so i don't think you have one of those.
4 poles with 7 teeth between phases  would be a rewind for sure, as it has a fundamental winding factor of .96

6 poles will require nn-ss-nn-ss-nn-ss so it will be more difficult to get good magnet coverage. perhaps you can alternate half .75 inch wide half .5 inch wide.

i would suggest you pull the aluminum out completely but that is difficult, and you need the aluminum in it to support the steel laminations while milling slots. eddy currents are probably negligible, but i would just remove the outer rings, to allow you to press the magnet out if you go for a press fit.
« Last Edit: April 15, 2014, 12:12:51 AM by joestue »
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: 3ph conversion Magnet questions.
« Reply #8 on: April 15, 2014, 05:07:27 AM »
I will try to combine comments on my last post with comments from Joestue.

The hexagonal packing will get you more magnets in but I think Zubbly stuck with less close packing to have each magnet in its own pocket which helps assembly a lot, there is a lot of mutual repulsion between like magnets and you have to hold them in place until the resin sets. With the tight hexagonal packing you don't have individual holes to help you.

Yes you are right that you shouldn't bury the magnets in holes in the steel, there will be a lot of leakage flux. A small steel leakage path round large magnets will only have a small effect and Joestue does mention this. If the magnet is burried less than half its thickness the loss is not great.

You do need to establish how many poles you have, if you have the nameplate details you can get it from the speed. For 60 Hz 2 pole will be 3600 rpm, 4 pole will be 1800 and 6 pole will be 1200. These figures will be a bit less than this to include slip and will be typically 5% below nominal but obvious.

Without the name plate rating you need to understand the type of winding to decide it by looking at it.

The most common N American winding is virtual pole concentric. If 3 phase you will have 3 identical concentric coil groups . 2 pole will have one concentric group per phase. 4 pole will have 2 per phase and can be confused with a 2 pole full coil winding.

If it is overlapped coil lap winding all coils will span the same slot pitch and it will be 1/4 the full slot count for 4 pole. This sounds impossible from your dimensions.

Zubbly chose to fit the magnets to the smallest concentric coil but using magnets 2/3 pole pitch will get a similar result with gaps about half magnet pole width between them.

Depending on the magnets you can get and the cost you may get there easier with rectangular magnets, again unless you machine a shallow slot to fit them you have the issue of keeping them in place until the resin sets. Common magnets are cheaper than less common ones so you may fiond round more cost effective.

Motor conversions are hard work unless you can have the ideal curved trapezoidal arc manets that the commercial manufacturers have. The old supply of early voice coil motor magnets from computer discs dried up, these were used for many motor conversions and simplified construction.

Zubbly used many original windings but he was a master of understanding them and reconnecting them to suit the conditions. When starting from scratch he usually chose 12 pole as it had smaller magnets per pole and fitted the curvature with less air gap. More poles also mean less wire length in the coils and suits low speed operation better.

As mentioned, 2 pole motors are designed for high speed and are small diameter and don't lend themselves to good motor conversions, 4 pole are larger diameter and have more room for magnets and work better. 6 pole are less common, even larger diameter and can be an even better choice.

Whatever you do, unless you rewind it is essential you keep the poles on your rotor tha same as the stator poles otherwise you will get nothing in the worst case.

Flux


SparWeb

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 5452
  • Country: ca
    • Wind Turbine Project Field Notes
Re: 3ph conversion Magnet questions.
« Reply #9 on: April 15, 2014, 02:45:08 PM »
Hi Tinkertoy,
I can't resist throwing in my 2 cents, but Flux and Joe have you steered around the right way, I think.

I've done a few conversions, so far without re-winding any of them.  Good power from each, though one didn't have enough magnet and one had crappy wiring.  That said, I have to admit that I could improve them with re-winding, if I did so.  The magnets get you so far, which may be enough, or plenty depending on your expectations.  But if you truly want a perfect match between the prop's power curve and the generator's load curve, and match both of them to the wind available at a site, then a re-wind may be necessary.  At your stage of the game, I don't think you need it.  A motor conversion can always be re-wound after the magnet job, if it needs some extra "tweaking".

In my experience it's all about the GAP.  The more flux passing through that gap, the better.  The smaller the gap between magnet face and stator tooth the better.  If you put a flat magnet on a round rotor surface, there's a gap there too.  I've had best results from machining flats on the rotors when I convert them.  I've seen other guys just turn them in a lathe and use a file to make the flat spots.  Many people allow for some cut/fit/cut/fit as they trim down the rotor.  But watch out for pinched fingers as you keep putting the rotor into the stator with all those magnets pulling it.

Smaller magnets allow for smaller gap (also fit on a round rotor with less gap, if you must), but it does increase the cost and time required to finish the job.  I spent months on my Baldor conversion, partly because of the intricate machining (done in my spare time) and partly because I could only clamp a couple of magnets in place at t time, and there were 48 of the little bastards.

The motor you're starting with sounds similar to the GE and Baldor motors that I converted a few years ago (also 3ph / 3HP / dual-voltage).  Can yours also be wired star/delta?  Can you say how many wire leads are available?  You should have at least 6 but you may have 9 or 12.  Give you a lot of flexibility to fine-tune the speed range when the magnet rotor is done.
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, Xantrex C40, DIY 10ft (3m) diameter wind turbine, Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024
www.sparweb.ca

SparWeb

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 5452
  • Country: ca
    • Wind Turbine Project Field Notes
Re: 3ph conversion Magnet questions.
« Reply #10 on: April 15, 2014, 02:49:09 PM »
Also,
Not convinced Lee Valley sells the best magnets for this.  Theirs tend to be thin (1/8" thick) and small (1/2" diam). 
I can get any size of brick I can imagine at KJ Magnetics, for slightly greater cost, but then I'm not constrained to a small range of sizes.
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, Xantrex C40, DIY 10ft (3m) diameter wind turbine, Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024
www.sparweb.ca

Smithson

  • Jr. Member
  • **
  • Posts: 75
Re: 3ph conversion Magnet questions.
« Reply #11 on: April 15, 2014, 02:50:42 PM »
I don't understand why when the subject of motor conversion comes up neodymium magnets  are the choice.  What is wrong with ceramic magnets grade 8?  Neodymium magnets saturate the iron and cause severe cogging.  Even with skewing.  On top of that depending on how big the magnets are you would need a good size backing plate to attach the magnets to.

Those 14 pole(14 magnet) GM alternator conversions are prone to cogging.  So that would mean a gust of wind to break them free.  You are losing all that power in lower wind speeds.  Maybe not that much.

If you were going to rewind that motor count the number of slots.  Wind coil 1 from slots 1 to 4.  Then reverse direction of coil winding and wind coil 2 from slots 4 to 7, etc.   This is phase 1.  Go completely around the stator so that the last coil ends in the slot that the first coil started in. 

Note that coil 1 phase 1 finishes in the same slot that coil 2 of phase 1 starts.(slot 4)  Insulate between the different phases but not between the same phase as in slot 4 (and elsewhere) where the finish of one coil touches the start of the next coil.  This is what a motor rewinder told me.  Insulate the the different phases but not the same phase.  Now there has been some discussion about this on this site.  Zubbly always said to slide a piece of slot insulation between the two coils.  Do it the way it gives you peace of mind.  I have never had any trouble with it.  Phase 2 coil 1 starts in slot 2 and ends in slot 5.  Again go completely around the stator alternating the coils, clockwise, counterclockwise, etc.  Phase 3 would start  with slot 3.

If the output of one coil is low in place of hooking all the starts together (in star) hook the starts of phase 1 and 3 to the finish of phase 2.  Depending on the size of your stator and also the slots use 2 in hand wiring.  Maybe 15 awg.  When you get done use a blade program to match the blades to output.  Maybe windstuffnow.com.  He uses this winding.

Count the slots.  Divide it by 3.  This is your pole count.  The magnet needs to fit in the center hole of a coil.  Measure from slot 1 to 3. This would be the ideal size for magnet width.  Or less.  The length of the magnet would be the length of stack.

Another winding would be the car alternator type.  The GM had 36 slots but 14 poles with this different winding.  I had some notes on their winding but lost them.

I looked up you motor. Nice.   Archie Smith






joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: 3ph conversion Magnet questions.
« Reply #12 on: April 15, 2014, 03:48:16 PM »
I don't understand why when the subject of motor conversion comes up neodymium magnets  are the choice.  What is wrong with ceramic magnets grade 8? 

the underlying assumption is a direct driven wind turbine, and you need the highest flux density you can get.

ceramic grade 8 would work very well for a 30 pole, 36 tooth motor conversion, you would just get about one third as much voltage out of it compared to neodymium magnets, which works out to 9 times the losses for a given power output, neglecting the much reduced iron losses. 
i found a company that sells .24 by .25 by 2.25 long ceramic magnets and asked for a quote, it might just be reasonable.

my 30 pole 36 tooth conversion got me 100 volts at 1750 rpm at 445 hz with just one ohm of copper per phase, the winding stack was 1.9 inches thick.
it was previously a 1/3rd or 1/4th hp 6 pole induction motor. (maximum current output was 9 amps, mostly due to leakage inductance)
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: 3ph conversion Magnet questions.
« Reply #13 on: April 15, 2014, 05:07:24 PM »
Everything is a compromise and with wind power everything seems to work backwards. You need the highest flux to get low cut in with minimum number of turns but this comea at the price of high iron loss in low winds.

If you want the best possible low wind performance then an iron core is not the best way to go, cog can be minimised and virtually eliminated even with neo but start up is much more affected by iron loss drag.

If you look at the higher winds then it becomes a compromise of size and weight, a ferrite machine will be much hevier and larger for a given output and it will be less efficient. If weight and size are not a problem then the ferrite machine can be very effective and may not be much more costly for a given output. The low efficiency is not a big issue with direct charging as you will not run into the stall problem. The limited maximum output due to reactance is partly counterbalanced by the fact that it will be near impossible to burn out.

Ferrite air gap machines become much more practical with a speed increasing drive, but the iron loss problem makes speed increasing of iron cored machines far less attractive.

If you must use ferrite it far better suits the outer magnet drum type machine but finding suitable slotted cores becomes a big issue as wound rotor induction motors are no longer available and larger dc armature cores are also near extinct. Probably the best way to use ferrite in a motor conversion is with a stack of claw rotors with large loudspeaker ring magnets energising them ( used to good effect by Elecktro machines). It would need a rewind as the claw rotor needs 8 poles or more to be effective.

Flux

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: 3ph conversion Magnet questions.
« Reply #14 on: April 15, 2014, 06:09:29 PM »
How about curved magnets? Machine a rotor to fit http://www.magnet4less.com/product_info.php?cPath=1_15&products_id=2

Smithson

  • Jr. Member
  • **
  • Posts: 75
Re: 3ph conversion Magnet questions.
« Reply #15 on: April 15, 2014, 07:46:14 PM »
I have heard these arguments before.  I have a motor conversion.  To satisfy myself I'll rewind it to fill every slot.  What I feel is to use neodymium magnets is like paying 10$ dollars for a meal that only costs 1$.  I don't know whether to use 16. 1x1x16 magnets or keep the same rotor and wind it two 3 phases  (6 phase ?).   You are right, it sure is heavy.  37 pounds of copper in the original stator alone.

http://fieldlines.com/board/index.php/topic,128311.0.html

I'll try two in hand 15 awg.     Archie  Smith.    Ps  To really test it you would heed a neodymium rotor and a ceramic under real word condition.