My first AC conversion

[opo]

Hello,

I'm new to this forum. This is my first post so go easy on me.

I have been collecting hard drive magnets for a while to build a wind generator. But a couple of days ago a came across a siemens 1 hp ac motor, 120/220 v running at 3535/3510 RPM 16.5/6.6 A. This is what Icould readfrom the plate (or what it is left of it).

The stator has 36 slots, and about 18 coils (it is pretty tight in there) which I leaved untouched  

After reading for a while the post in this forum, I got a little disapointed about it. Today I decided to go ahead anyway and to the lathe I sent the rotor, making it 10mm thinner, because my magnets are a little less than 5mm thick , and decided to put two poles on it N S , (because the RPM are too high): each pole consisting of 12 hard drive magnets (i.e. 6 HDD magnets per pole) arranged in a 4X3 skewed array to avoid cogging (actually there is "little" cogging). The converted motor can be turned with the fingers.

The motor has 6 leads coming out: black, red, yellow,green, white and blue. By pairs I got open voltage readdings of

black-red :  5.5 V

blue-white : 3.9 V

yellow-green: 3.9 V.

These readdings where obtained by turning the motor by hand ussing a wrench bolted to the motor (yes very scientific!).

I started playing around with the connections, and I found that connecting

red and green

yellow and blue

and taking measurements at black and white I got 10 V with the same method. This is what I cannot understand (the three phases are in series here making it single phase?).

Please tell me what to do next. I'm thinking in opening up the coils to see if this can be improved or at least find out what is going on but I'm afraid I might destroy the motor or spend too much time on it. It looks pretty tight inthere.

Do you guys think it is worth the effort to continue with this conversion?

What tests should I perform?

Thanks guys,

Octavio

[hiker]

black-red :  5.5 V

blue-white : 3.9 V

yellow-green: 3.9 V.

use three rectifiers[you know the block type]

then parelle the outputs...

voltage might be a little high for a 12v battery bank..

but would still work..24 or 48v batt. bank would be even better..

[ghurd]

Connecting the leads in series increases the resistance.

It can increase the voltage but it Greatly reduces the output because of the resistance.

Try to get it spinning at a few known RPMs and check the voltages.

Check the resistance of each set of coils.

I expect the motor was multiple speed.  You may find 1 set of coils works OK, one works marginally, and the other is nearly useless.

Adding more magnets will bring up the voltage and output.

Reducing the gap between the magnet and lamination will bring up voltage and output.  If you can add a piece Half as wide (for mechanical clearance) on top of the existing magnet, to take up some of that space, it really helps.

A 1HP will have low resistance.  Get past cut in voltage, and the amps shoot up fast.

Yes.  Continue.  

It may not be great, but the more you do to it, the more you will know before starting the next one.

G-

[FishbonzWV]

Opo,

Did you cut the hd mags in half?

A picture of the rotor would be helpful.

Bonz

[opo]

Yes I did cut them in half in oreder to have the same polarity. I'll try to post pictures latter on.

[wdyasq]

"This is what I cannot understand (the three phases are in series here making it single phase?)."

I don't think you have a 3PH motor. I think you have a single phase 2 pole motor with start windings.

Ron

[opo]

I think you are correct. Actually I forgot to mention there was a capacitor connected to the red lead which I set aside. So probably the red-black corresponds to the running coils and thats why I'm getting higher readings from them.

Thanks for your comments.

Octavio

[FishbonzWV]

The caps are on the starting coils.

There should also be a centrifical switch attached to the rotor to disengage the start windings once it is up to speed. You should remove it.

The start windings will hit cut in voltage first.

Three bridge rectifiers paralleled will still work.

[opo]

Thanks for clarifying this to me. I bypassed both the capacitor and the switch. I see no reason to leave the capacitor there. Is there?

I'll get the rectifiers soon and see what DC voltage can be produced when paralleled.

Thanks again for all your info.

Octavio

[opo]

Thanks for your advice. I think there is enough room to put a second layer of smaller/thinner magnets.

The resistances are as follows:

black-red: 3 ohms

green-yellow: 2 ohms

blue-white: 2 ohms

And when all in series it goes up to 5 ohms.

Thanks again for your suggestions.

Octavio

[ghurd]

If there is space to add more magnets, do it.  Pretty surprising jump in numbers for a little more magnet.

"arranged in a 4X3 skewed array" has me wondering.  Not going to be well phrased...

Is there space between NNN and NNN to add more N magnets?  If so, add them.

I believe it is better to have the Ns close together, the Ss close together, and the space between the Ns and Ss.

NNN, NNN, space, SSS, SSS, space.

"Regular" meters will not accurately read resistance that low.  Accept you don't really know the resistance and don't get a headache over it.  Close enough is close enough.

Can test it more accurately with other methods if you Need to.

Verify the coil groups (you called phases) are not connected.

Will be interesting to see numbers for a 1HP with HD neos.

And pictures are nice too.  

G-

[opo]

Hi ghurd,

Instead of 4X3 I should have said 3X4 array which means 3 rows by 4 columns of half HD magnets and "skewed" meaning that I put them in a "close to" 10° angle with respect to the rotor axis: a planar version of it would be

N N N N      S S S S

 N N N N      S S S S

  N N N N      S S S S

There is enough space, I think, to add one, maybe two, extra column(s) of Ns and Ss and there is enough clearance between rotor and stator to add a second layer on top of them. I can also make more room by placing the magnets closer to each other (fixing them may be a little more difficult thogh).

How close can the Ns and Ss be? And I mean at the boundary between the Ns and Ss poles.

I checked for continuity (?) between coil groups, they seem to be isolated from each other: so the motor has three coil groups, one starting group, two running groups.

Thanks again,

Octavio

[ghurd]

The More the Merrier!

I suppose I wouldn't worry much about "the last column".

I don't think I would be at all concerned with the distance between N and S groups on a  2 pole machine. (I pretty much just pack in as many as fit on something like this one, and hope for the best)

The last column in the boundary between N and S is a bugger,

and it is worse with a round rotor using flat magnets.

The like magnets will try to push it away, the opposite magnets will try to flip it.

I have a feeling the last column is not worth the effort on something like this (2 pole with HD neos), though no 2 of mine are similar enough to make any solid comparisons.

More magnet surface area and less distance from the magnet to the stator laminations will make a dramatic difference.

I think you are doing pretty good.  I would not have expected that kind of voltage with a 1HP and a few HD neos.

Not sure how fast you had it spinning and speed is a very important variable.  I know I can't get them moving fast with that test method.

G-

[opo]

Yes, and it is a painful method. My pointer finger is all scratched now. I'm getting better now at spinning. I can easily see 6.5 V now, sometimes can see 8V and even 9V. I guess the speed is between 120-150 RPM.

I'll take the motor down, together with rectifiers, to the lathe shop and see if they'll let me do the testing to some known RPMs rectified and not rectified.

I started asking the plumbers around and they all agree the motor must be from a jacuzzi/hydromassage type bath tube. I first though it was from a water pump, but I was told the ones in water pumps are mostly 1/2 hp. I live in Mexico City by the way.

Pics to follow latter.

Octavio

[FishbonzWV]

Here's how I manage to squeeze the mags together.

http://www.fieldlines.com/story/2009/3/13/233643/638

[opo]

Here are the results from after spinning the motor in the lathe:

RPM   VDC (open)

180.   12.7
     
240.   14.4
     
300.   25.0
     
     

This is with each set of coils rectified and connected in parallel. I only had a chance to perform one test with an UPS battery (12V 9Ah) connected to the gizmo at 300RPM the voltage, measured at the batery, was 19.4VDC. The charge of the battery went from 5.1VDC to 8.7VDC in a couple of minutes.

The guys in the shop where very busy so I stoped making more tests. I wanted to know at wich point the running coils reached the ouput of the starting coils. I guess this happens somewhere inbetween 240 and 300 RPM, an thus the jump in voltage. I forgot to measure amps :-(

Cheers,

Octavio

[ghurd]

The photos are too large for the board rules.

But not bad results!  I may have to try something that size.  I usually stay smaller than that.

The magnets are super glued on?  I do not believe it will hold well.

Be a good idea to get a good layer of epoxy all over (and under) them.

I paint the entire magnet with epoxy to stop corrosion of the neo.

Making power with junk is fun.  

G-

[DanG]

Octavio: The four photos at 3.4 megabyte file size couldn't stay.

Graphics at 640x480 pixels and/or 150kb are the largest allowed.

Hope you reduce the sizes and repost!

[opo]

I apologyze for the previous enourmous pics. Here they are in smaller version:

Original rotor:

Rotor turned and with magnets:

The magnets in the picture are superglued and then I used a couple of wraps of brown tape to help keep them  from moving around and also help in the inserting of the rotor.

I disasembled the motor last night and found out that I just removed 12mm but from the diameter, not the radius as I originally planned (thinking on a second layer of mags). So I'll not be able to put a second layer of magnets on top of the ones I have alredy in place. That is I have about 1mm of clearance between the flat magnets and the stator.

I did put an extra column of magnets, so now I have two poles each consisting of 3x5 halfs of HD magnets (between 4mm and 5mm thick). Apparently there is "less" cogging now, or probably it was too late at night and I was dreaming, who knows.

Today I got some 30 useless seagate HDs for me to tore appart and keep collecting mags for a bigger/better project (axial flux?)

Ok.

What would be the next step with this motor?

2m diam 3 blade prop? I'd like to see how fast the wind can turn this motor. Where I live there isn't too much wind. We see mostly 5-15 miles per hour winds with some gusts in the afternoon.

Cheers,

Octavio

[ghurd]

Next step is test it for open voltage, and amps into a battery, at a few different known RPMs.

Then hope somebody good with blades can tell you something.

G-

[opo]

Thanks ghurd!

I happen to have a cheap digital multimeter. But I think I'll do.

Now, by open voltage test with battery connected I guess you mean testing the voltage/Amps at the battery poles while battery is connected to the paralelled rectifiers which in turn are connected to each set of coils and the motor is beeing turned by the lathe at a known RPM. Is this correct?

Thanks in advance for all your help.

Octavio

[ghurd]

My fault.

Connect the bridge rectifiers to their coils.  One set of coils gets one bridge rectifier.

Parallel the bridge rectifier's + and -.

Set the lathe at a known RPM.  Test the battery voltage.

Test the voltage without a battery.

Then test the amps into a battery.

Set the lathe at a different known RPM.  Re-Test the battery voltage.

Test the voltage without a battery.

Then test the amps into a battery.

Do it maybe 4 or 5 times total?

Set the meter on 10A or 20A.

Do not run each test more than a few (10 or 15?) seconds.

Some cheap meters do not appreciate carrying more than a few amps for very long.

The battery voltage will stay closer to the voltage it started.

I would be very careful about spinning that too fast with only super glue holding the magnets on.

If it was mine, I would get them epoxied down with 24 hour cure (Not 5 minute) epoxy,

Before a test.

Mine tend to have less cogging with more HD magnet pieces.

I think it is because with more pieces of different sizes and shapes, it sort of randomly "skews" itself.  

Not well explained, but I expect it does have less cogging.

This one has almost no cogging,

3 phase, 1/3 HP, 3550 RPM.  64 pieces of magnet (2 deep).

http://www.otherpower.com/images/scimages/2050/3phHD.jpg

And that photo is also too large to post.  I will fix it someday.

G-

[ghurd]

"Parallel the bridge rectifier's + and -."

Connect all 3 + together.

Connect all 3 - together.

The + goes to the battery +, the - goes to the battery -.

To use the meter for reading Amps,

Disconnect the rectifiers + from the battery's +.

Connect one meter wire to the rectifiers +, connect the other wire to the battery +.

G-

[opo]

Thanks ghurd. I haven't been able to take it to the lathe. But by wrapping a chord 5 times to the motor I was able to read 2.4 A dumped into the UPS battery having 6.4 v of charge. The open volt average with that same method was 27vdc. But even with the chord I don't think I was getting even close to the same rpm as in the open case.

Cheers

Octavio

[ghurd]

Any progress?

G-

[opo]

Well,

My work and stuff to do have prevented me from taking the motor to the lathe shop. Instead I have been working, mostly at night, on the tower (done), a mounting plate (done), tail (done) and a hub to mount blades which is giving me troubles: I went ahead and fabricated 3 PVC blades out of 6" pipe thick wall, they spin ok until cut in is reached and the hub fails.

I'm using the motor conv. described in the post.

First I used an 12v 9a/h UPS battery only to realize it was dead with only 6v. I have been using an old 12v car battery since wich has about 12v of charge. With both batteries I have had the same problem with the weak hubs I made.

Also the winds are not cooperating (exept, of course, right now and I have a broken hub).

As soon as I get a chance to take the motor to the lathe I will to turn it another 10mm to double the amount of magnets and run more reliable tests.

As it is, it seems to be "working" (exept the hub detail), I'd like to lower the cut-in RPMs a little bit by using more magnets and build a proper blade set (wood) and call it. I have a 1/2 HP motor 24 slot core, that I'd like to rewind and convert, have been collecting HDD, degaussing coils and keeping an eye open for a 3ph motor to convert.

Thanks for your interest in this first project. I have learned a lot with my first mistakes and from reading this forum. It rules!

Cheers,

Octavio