6 pole motor vs 4 pole for conversion

[bigrockcandymountain]

So I already have a 4 pole 200v 5hp motor but it seems not quite right for conversion.  It's only 3 wires, has grooved stampings, kind of a small rotor, not very beefy mount, etc. 

I found a listing for a 6 pole 254u frame 5 hp 208/220 440v so likely 9 wires.  It is $100.  It is Canadian general electric and loks more beefy.  U

My question is has anyone converted a 6 pole and what were the results? 

[SparWeb]

It will definitely be more beefy - 3 phase motors seem to go up in weight by at least 50% for every pair of poles that are added.  Just needs that much more iron to get more copper inside.  I too have a 6-pole 5 HP motor in my garage which I picked up (from scrap) intending to convert it.  I soon realized that I would require a huge tower to support this machine because it weighs 200 pounds!  Probably a similar frame # so you will probably have the same difficulty.  I'm keeping mine in case I ever move someplace where a hydro generator would be useful - it would be perfect for a medium-size hydro turbine.

You didn't mention if the listing you saw was for an old motor or a new one, so I'd like to make a second point, though not a strong one: I prefer to use old motors rather than new ones.  There is a lot of "embodied energy" and materials in electric motors and they serve a valuable purpose as motors.  When I take an old one for conversion, it was going to the scrap heap, but I give it a second life as a generator. 

I did once convert a 2-pole motor and it was very disappointing.  Since I've always been satisfied with 4-pole motors, by extrapolation a 6-pole should be fantastic.
If you do find a 6-pole motor, but of a 2 or 3HP size, and you can deal with the mass of the machine, then you could have a prize in your hands.

[bigrockcandymountain]

Awesome, I was hoping you would chime in sparweb. 

I think I might go for it.  It's an old motor and looks fairly well used.  Probably needs bearings but I would do that anyway.

I'm not terribly scared of the weight since my planned tower is only 30' high and I'll likely use some 7" or similar well casing pipe.   I want the tower to be capable of supporting a larger machine if I ever need one so it will be over build for this purpose. 

I have 48v and I get the feeling with 6 poles it will hit that voltage at a very low rpm so maybe I can hook it parallel and have the higher amperage potential.

Thanks for the great advice again.  I'm hoping to get to my inlaws machine shop next weekend and actually get this project going.

[joestue]

if you can dig into the winding and separate each coil, potentially you can rewire it for a 6th of its nameplate voltage, but its just as likely you will be limited to wiring it for a third of the nameplate voltage.

and if you wire it in delta that will be a further reduction by a factor of 1.73.

at a minimum it should be trivial to wire it in delta, which would be a nameplate voltage of 137 rather than 240/480. it is usually very easy to find the hidden star point and bring out those three wires.

[SparWeb]

7" or similar well casing pipe[\quote]

Now you're talking!

Concur with Joe, too: Getting at the star point and soldering a few more wires is usually trivial.  Tie everything back up carefully is the only part that seems to be forgotten sometimes.

[bigrockcandymountain]

joestue that is an idea I wouldn't have come up with myself.  I'm guessing separating each coil wouldn't be easy, but it makes a world of possibilities.   A third the nameplate voltage would work good I think though, since I need 48v. 

[Adriaan Kragten]

The procedure how to transform an asynchronous motor into a PM-generator is described in my report KD 341 which you find at the menu KD-reports on my website: www.kdwindturbines.nl. The maximum torque level of a generator or a motor for which a certain magnetic flux is realised in the air gap, is about proportional to the armature volume lying within the stator stamping. The mechanical power is proportional to the product of the torque times the rotational speed. So a 6-pole asynchronous motor which can supply a certain power at a rotational speed of about 964 rpm (at f = 50 Hz) needs an armature volume which is about a factor 1.5 larger than for a 4-pole motor which can supply the same power at a rotational speed of about 1450 rpm.

However, a direct drive PM-generator made from an asynchronous motor, is normally used at a much lower rotational speed than the original nominal motor speed. The windmill rotor supplies a certain maximum torque depending on the wind speed and the safety system which limits the power above a certain wind speed. So the armature volume of the PM-generator must be at least that large that it can take the maximum torque level of the rotor. It doesn't matter if the motor has originally 4, 6 or 8 poles, it is the armature volume which counts.

However, if the original winding winding is used, the open voltage which is generated at a certain rpm, is proportional to the pole number. So the open voltage of a 6-pole PM-generator is a factor 1.5 higher than for a 4-pole PM-generator. If the PM-generator is used for 24 V battery charging, it might be that this is possible to use the original winding of a 4-pole motor but that the voltage of the original winding of a 6-pole motor is too high or that a 6-pole PM-generator with the original winding can only be used for 48 V battery charging. It is possible to connect the first and the second layer of the winding in parallel in stead of in series and this halves the voltage and doubles the current. The voltagre can also be reduced by a factor square root of three, if the winding is rectified in delta in stead of in star. However, rectification in delta has the disadvantage that the sticking torque is rising faster than for rectification in star because higher harmonics can circulate in the winding and this may give starting problems (see report KD 78 for measurements of different onloaded torque curves).

Theoretically it is possible to rewind the generator completely with a number of turns per coil and a wire thickness such that the matching in between rotor and generator is optimal for a certain nominal battery voltage. However, a special winding may have extra cost more than that of a complete new motor with a standard winding.

[Mary B]

Well casing is some pretty flexible stuff... and 7" is going to be HEAVY!

[bigrockcandymountain]

Mary the 2 7/8" tubing and oilfield steel like that is very flexible.  The 7" I'm talking about is fairly rigid.  Actually I think it was 6 3/4 or something.  We used it for surface casing on core holes.  You could lift a 10m joint in the middle with a pallet fork and not see hardly and deflection.  It had about a 1/4" wall.  We'd run a drift through to check the ID and if it stuck, it got thrown out.  I would like to get one length of the thrown out stuff.  It sells for scrap price around here.

You may be thinking of deeper well stuff with a heavier wall.  I don't have any experience with that but it would be way heavier. 

I plan to use a tractor to tilt it up rather that a winch.  it lifts about 3000 pounds and pulls harder yet. 

Adriaan, I looked over that report.  It definitely told me what I wanted to hear about the armature volume.  I like your way of embedding the magnets on edge deep in the rotor but it looks quite difficult to machine and would require building a shaft.  I think I'll do thinner magnets mounted on the surface of the rotor more like your 34 pole. 

[bigrockcandymountain]

I got the six pole motor.  It has a rotor diameter of 7.5"  54 tooth stator.  9 wires.  3" rotor length.  My plan is 84pcs  1 1/2" x 1/2" x 1/4" n42 neodymium magnets.  They will be thin but hopefully make up for it with a really small air gap.  They each have two countersunk holes and work out to $150 usd plus shipping. 

Any other ideas will be appreciated. Otherwise I'll get the magnets in the mail.   I'm thinking just turn it down on a lathe and not worry about milling flats.  I think I'll skew the second row of mags 3.33 degrees. 

[SparWeb]

That sounds like a very good start.

If you haven't turned down a motor before, expect it to come apart if you turn it down too far or too aggressively. 
How far is too far?  Is too far less than or more than the amount you need to turn it down?  I don't know.
That's a tough question to answer for certain without examining the rotor . 
The size of your rotor hints that the aluminum bars embedded inside the steel laminations could be thick and deep.
I succeeded in modifying the rotor of my Toshiba conversion this way.  The other motors required new solid rotors.
When you start penetrating into the aluminum bars expect the torque on the cutting tool to cause the steel laminating rings to shift, so turn medium fast and feed slow.

Have you considered an adhesive you want to use?  Screws are great, but moisture can condense in the gaps between rotor and magnet unless you seal it up.
Might as well use a strong structural adhesive.

[bigrockcandymountain]

I am planning on using t88 epoxy for the glue.  I have quite a bit left over from an airplane flap project. 

That's good advice on the machining.  Hopefully it doesn't come apart. 

[bigrockcandymountain]

Here is my rotor turned down.  I have a paper template on and I'm center punching holes for the magnet screws.  168 holes!

This is mostly an attempt to upload pictures on here for the first time.  If they are too large or something, let me know. 

[SparWeb]

Good photos, they look fine to me.
That's a mighty big rotor - you'll get lots of magnets on there   

[bigrockcandymountain]

Yes, it was 7.5" diameter.  I turned it down just under 7" and will put on 1/4" magnets.  1/2" wide by 1 1/2" long 84 pcs.

 If someone else thinks of doing it this way, I would suggest finding magnets with only one hole.  There is no need for two.  It's not going to come apart though. 

[hiker]

Nice..I used 98 small mags double stacked on my 4 pole conversion..super glued the mags down..then used epoxy stick...in all the gaps...all around the rotor..that was about 4 years ago...still holding ..just fine...good luck on yours..

[Adriaan Kragten]

An important disadvantage of using the original armature is that it contains alumium bars which are cast in the steel lamination. If you remove the outer shell of the armature you will see that a large part of the surface is aluminium. Aluminium is guiding a magnetic field very poorly so your armature will become less strong than an armature made out of massive iron. But may be it is strong enough for the torque level which you need for your wind turbine.

[bigrockcandymountain]

Ok so I finally got this beast together. I broke two taps but otherwise it went fine.  6-32 is apparently the worst tap for breaking so if anyone wants to go this route, use a metric or fine thread standard.

 The voltage is a little lower than I expected. 

I only did a really poor hand crank test but it showed about 30vdc at 100rpm.  That is in 2Y

It doesn't cog even a little bit.  I'm very happy with that.

The air gap is very small.  Probably about 1/32" but I didn't measure.   

That means my 48v cut in is 160rpm.  That might work very nicely.  I need to think on it I guess.  I never did extract the star point and it looks like I may not need to. 

Thanks to everyone who helped so far with this project.  I would be nowhere without this community.  Any further advice would be great too.  I'm hoping to set up a hydraulic motor to test input and output properly, but I don't have a tach yet (one coming from ebay $10) so that will be awhile. 

Hiker, do you have any pictures or specs on yours?  There aren't actually a lot of motor conversions up and making power to compare to. 

[SparWeb]

Great results!
You could try a lathe to test it (such as the lathe you used to turn the rotor).
Chuck the motor shaft into the lathe, bolt a 2x4 to the feet of the case, restrain the case from turning with the other end of the 2x4 on a bench.
I was able to estimate the torque this way, and multiplied by RPM equals the input power. 
Output power (charging batteries) divided by input power equals the efficiency.

[Adriaan Kragten]

Yes, a strong lathe can be used to test such a PM-generator, especially if it has a variable rotational speed. Only clamping of the generator shaft in the head stock of the lathe won't be enough. You need a rotating centre at the other side. To prevent the housing from rotating, a lever has to be connected to it. If the lever is supported by a balance, you can measure the reaction force and by multipying this force by the length of the arm, you measure the torque. The mechanical power in W is the torque in Nm times the angular velocity in rad/s. The generator should be loaded in the same way as it is also used in the windmill. Most people and even most manufactures only measure the electrical output but the Pmech-n curve is needed if you want to check the matching of the generator with a certain windmill rotor (see for matching report KD 35 chapter eight). Both the Pmech-n and the Pel-n curves are needed to determine the efficiency curve for a certain load and to determine the Pel-V curve.

[bigrockcandymountain]

Just an update.

I never did get a proper test done on the motor conversion. It makes power when you crank it.  Takes exactly 2 ft/lb to start at the worst sticking spot.  Other than that, the cogging is very weak.  Seems to have 6 sticking spots per rev. 

I couldn't find time to gather the stuff and head up to my friend's place with the lathe.  I took a slightly different approach to blade design. 

I was reading Oztules chainsaw blades posts and he made what he called real mongrel blades with improper twist and too much chord and said they performed great, so that's what I've done.  I'm hoping unloaded they are about a TSR 8.

They are 4.5" at the tips and 7.5" at the root.  They have a straight taper.  It is a 3.8m diameter.  12.5'  Maybe 1 degree at the tip and 10 degrees at the root.

I still need to build a hub and chassis and tower.  A little progress though.  I'll try to get some pictures at some point. 

PS I fully expect to have to build a few sets of blades, or test the generator eventually.  I just got excited about the chainsaw blade thing and had to try it out.  They aren't even close to perfect, but I think will work to some extent.

[joestue]

Just an update.

I never did get a proper test done on the motor conversion. It makes power when you crank it.  Takes exactly 2 ft/lb to start at the worst sticking spot.  Other than that, the cogging is very weak.  Seems to have 6 sticking spots per rev. 

They are 4.5" at the tips and 7.5" at the root.  They have a straight taper.  It is a 3.8m diameter.  12.5'  Maybe 1 degree at the tip and 10 degrees at the root.

can you tilt the blades 3 degrees? 4 degrees at the tip is about right, and the outer half of the radius is where all the power is.

your repeating cogging torque 6 times a revolution may be fixable. can you remove one of those magnets (trial and error to find which one) to fix it?

[bigrockcandymountain]

Hello joestue,

I could carve a bit and tilt the blades in their hubs.  The face is more like a 3 degree at the tip I think. The 1 degree I talked about was the the chord line of the airfoil, not the windward face.  I'll check exactly what they are and let you know. The boards twisted after carving slightly with the trailing edge curling downwind. That will help a bit too. I have a prop protractor angle checker that I'll try today and see exactly what they are.

The 1 degree I talked about was the chord line of the airfoil, not the windward face.  I'll check exactly what they are and let you know.

The magnets are glued and screwed so likely no.  Could you explain what you think is going on so I don't do it next time?

Somebody posted that slight cogging could be thought of as a bearing saver in that it doesn't let the blades spin until there is almost enough wind to make power.  I'm hoping this works out that way. 

[Adriaan Kragten]

If a rotor blade is calculated using the aerodynamic theory (like you find in my free public report KD 35), you calculate the blade angle for a certain blade radius and a certain local speed ratio. The blade angle is the angle in between the rotor plane and the zero line of the airfoil. It depends on the airfoil how the zero line is defined. For most NACA airfoils, the zero line is the line in between the nose and the tailing edge but for most Götingen airfoils with a flat lower side, the zero line is the flat lower side. The geometry of the NACA 4412 is almost identical to that of the Gö 623 (both have a maximum thickness of 12 % of the chord) but the difference in between the zero lines is about 2 degrees. So if you have calculated the blade angle for a NACA 4412 airfoil and if you measure the blade angle from the flat lower side you make a mistake of about 2 degrees (see figure 5.7 of report KD 35).

[SparWeb]

No, you still have to change bearings every few years. 
Consider that it's spinning about 1/2 the time on a good wind site, which is about 4,000 hours, or 52 million revolutions per year.
I wouldn't worry about cogging for now.  I've done a few conversions and they all need a bit of torque to turn. 
Your 2 ft-Lb, or 24 inch-Lb is a little high but not much higher than what I measured on my own.

Depending on how you mount the blades, shimming with wedges can be practical to get the angles of the blades to agree.
Make sure you balance your blades.  Get their weights roughly equal first, then mount them on the hub and balance the hub.
Also try to get the tips to follow the same path.  Depending on what you do to set the tip angles, you could cause the tip paths to wander.
Not the end of the world, but it will run smoother and quieter.

[Adriaan Kragten]

The lifetime of bearings depends on the load, the rotational speed and the condition of the grease. My experience with direct drive PM-generators made from asynchronous motors is as follows.
Both the front and the back bearing (which are identical) have a certain radial load, mainly because of the gyroscopic moment, but the back bearing also has an axial load because it takes the rotor thrust. So one would expect that the back bearing is weared first but this isn't the case. I use rubber sealed bearings for both places. The back bearing cover is closed so no water can penetrate the back bearing. The front bearing has an extra spring loaded oil seal and the space in between this oil seal and the front bearing is filled with grease. After in between four and eight years, the oil seal starts leaking and this finally results in penetration of water in the front bearing. This bearing starts rusting inside and therefore gets much more friction. This results in increase of the starting wind speed. Because of the rusting, the front bearing also wears and starts to make noise. After replacing the front bearing, the oil seal and the grease in between them, the generator can run for about another five yeas. So to my opinion load and rotational speed are not critical for the lifetime of bearings of generators made from asynchronous motors. It is the penetration of water which finally ruins the bearing. Only the seal of a rubber sealed bearing isn't enough. On needs at least one extra oil seal at the front bearing.

[joestue]

So to my opinion load and rotational speed are not critical for the lifetime of bearings of generators made from asynchronous motors. It is the penetration of water which finally ruins the bearing. Only the seal of a rubber sealed bearing isn't enough. On needs at least one extra oil seal at the front bearing.

could also find the space to put a pound of silica gel in the motor to absorb moisture.

[bigrockcandymountain]

The bearing discussion is interesting.  I used open bearings since this motor is set up for greasing the bearings.  It has a grease fitting and a drain plug on each bearing.  I plan to flush out the grease every year and replace with new grease.  If it doesn't work or moisture proves a problem, I'll use sealed bearings next time. 

Here is a picture of the blade.  I never did get a picture of the magnet rotor when it was done.  It looks pretty cool with all those magnets.  I get kind of excited and in a rush sometimes.

The blades are about 3 degrees at the tips so I think I'll try them they way they are.  They are pretty rough but should make some power.  I'm thinking cedar shingles would work for wedges to adjust them if need be.

The second pic is the house so far.  It's been a long process, but we're getting close to moving in.  Obviously it won't be done when we move.   

[Adriaan Kragten]

could also find the space to put a pound of silica gel in the motor to absorb moisture.

That is certainly a good idea. Although the housing of an asynchronous motor is very well closed, some water is still created, probably because of condensation and not of leakage. One of my generators had a strong inside oxidation of the cast iron housing at the lowest point after about twelve years. May be one should drill a little hole there so that the water can escape and so that no under pressure is created when the housing cools down at night. The undere pressure might suck air with water vapour from outside along the seals and so water comes within even if the housing is completely closed.

[joestue]

need a diaphram to keep the air pressure equalized. negative pressure will suck in moisture and air eventually.

[bigrockcandymountain]

This motor actually has a 1/8" npt plug at the bottom.  It says on the side that when operating in very wet conditions you should remove it for a drain.  I was thinking maybe put short pipe nipple on it.  Should it be open or closed do you guys think?  I also thought I could just check it when I grease and close it up again.