Changing an induction motor voltage

Originally posted by woody Originally posted by coulomb Originally posted by woody If you're feeling keen you can pull apart these insulated wires and maybe convert to 2 pole 30kW by reversing A2,B2,C2 connections. Can someone help me imagine this correctly?    .... Back to crazy ideas: Rewind for 2/3 voltage: Cut the jumpers between the concentric windings and re-pair them as Outer + Inner, Inner + outer, Middle + Middle for approximately the same resistance. ..... cheers, Woody

Dear all,

just bought a 7.5 kW four pole. Windings in 400 V Delta, 690 Star (or thereabout). Originally thought that this might be a good motor to test (it was dirty and cheap)inverter etc., but now also starting to consider to use this as a drive motor for a lighter car.

I know in Germany drives a Opel Corsa with a 5.5 kW 4-pole, and that is working.

However, getting something better for free remains very appealing, so I was thinking about rewiring it into a lower voltage. My inverter does 55 kW continous, and I would keep the clutch + gearbox for a start.

My question is, how to recognize the several windings. Should I post a picture here, or is there some general knowhow that I can immediately use to determine the type of winding?

Pictures is difficult as I do not know how to do that :-0, and also as the motor is in storage a bit farther away.

Many thanks for your thoughts about this, regards,


Huub

Just in case it's useful I tracked down that contact-free shaft encoder I mentioned a few posts up. Unfortunately it's only a 256 CPR.
It's an AVAGO. Around $25.
http://www.avagotech.com/pages/en/motion_control_encoder_products/magnetic_encoders/aeat-601bf06/

Edited by Johny - 24 February 2010 at 7:25am

Originally posted by a4x4kiwi More progress...

Good to see.

Just one comment... with an 800 pulse per revolution encoder (so 0.45 degrees between pulses), maybe the slight movement from that coat hanger wire might be enough to cause jitter in the least significant few bits? Perhaps more so when changing from reverse to forward, and vice versa.

0.45 degrees is about 0.2 mm at 25 mm radius. Would it move that much/little?

I don't have any idea whether it might move that much or not, but it seems possible that a totally rigid connection from resolver/encoder to the motor case may be needed. For simplicity's sake, I hope not!

Originally posted by weber Mal, thanks for the info re the encoder, and the photo on the lathe. We need to put a taper-lock hub on our fan shaft to couple to our aircon/power-steering pulley, so Ross Pink suggests adding a shaft extension for the encoder by precisely drilling and tapping into the end of the shaft and threading a 12 mm dia post in.

Be careful that it does not unscrew itself when you reverse the motor.

I'm not sure about these but sometime in the last 2 weeks I got an email from some company (regular advertising stuff) about a contact free shaft encoder. I following it up enough to find that farnell stocked them but lost interest because I figured I already had an 8mm shaft.
They claimed a wide tolerance for mis-mounting and use a magnetic coupler which is glued to the end of a shaft. I'm sorry I can't make it easier but if you think that the motor magnetic field won't interfere it may be worth looking into.

Woody, that 2/3 rewind idea is clever. You've solved the different resistance problem, but unfortunately not the different waveform problem, so there would be circulating harmonic currents as coulomb mentioned. But how bad they would be is another question. But based on coulomb's and my experience of trying to find the crossover wires on a motor with _clear_ varnish, the chances of finding all 12 of them on Mal's motor are slim to nonexistent.

Thanks for a neat notation too, Woody.

Mal, thanks for the info re the encoder, and the photo on the lathe. We need to put a taper-lock hub on our fan shaft to couple to our aircon/power-steering pulley, so Ross Pink suggests adding a shaft extension for the encoder by precisely drilling and tapping into the end of the shaft and threading a 12 mm dia post in.

I'm wondering if that can be done without a lathe and without removing the shaft, by powering the motor and mounting the drill in a vise or something?

Hi all,

I selected the encoder based on the Danfoss input max frequency with a push pull driver is 65khz. This equates to 886 pulses per revolution at 4400rpm.

Encoders are usually available in 100 increment steps. They did not have it x stock but it arrived from Germany within 2 weeks. Give them a call. Jill was very helpful.

The price was $491 +GST. The price had just came down by $50 due to exchange rate changes.

Originally posted by coulomb Originally posted by woody If you're feeling keen you can pull apart these insulated wires and maybe convert to 2 pole 30kW by reversing A2,B2,C2 connections. Can someone help me imagine this correctly?

The 2 pole would probably be less efficient that a dedicated two pole, more like a dahlander 2 speed.

That aside:

Each of A1 and A2 with the same current would now be making a field in the same direction (e.g. left to right) instead of against each other (e.g. outer to inner) + a consequent/phantom pole (inner to outer).

So that means only half the flux lines for the rotor to cut, i.e a quarter of the torque.

We can't increase the current as we're all out of voltage.

I'm not helping, am I.

going back to dahlander:

Originally posted by Motor Book Constant torque Dahlander is For constant torque, the motor is usually connected two-parallel star for high speed and series-delta for low speed. For constant horsepower the connection is two-parallel star for low speed and series delta for high speed.

Both of dahlander windings require series delta.

Back to crazy ideas:

Rewind for 2/3 voltage:

Cut the jumpers between the concentric windings and re-pair them as

Outer + Inner, Inner + outer, Middle + Middle for approximately the same resistance.

I.E. A is currently made up of ('=' means coils '-' means series connection)

1=12-2=11-3=10 in parallel with 19=30-20=29-21=28

Reconnecting the coils as:
1=12-21=28 and 2=11-20=29 and 3=10-19=30 in parallel

This gives Mal 480V star (720V * 2/3) and 275V delta.

Worth a bash when you have the money for a rewind when it all goes pear-shaped.

275V should give 230Nm @ 2200rpm (56kph) i.e. 53kW
0-90 in 17 seconds instead of 31.
top speed 120 instead of 95

cheers,
Woody

Originally posted by a4x4kiwi I am also fitting an 800ppr encoder. See my blog for details and full resolution photos.

Hi Mal, coulomb and I are in the market for an encoder now too.

In another thread acmotor suggested one for about $200 from RS Components, but the only one I could find that matched that description has a 4 mm diameter solid shaft, which means it needs a flexible coupling and precise alignment. It seems designed for much smaller motors as I couldn't find a flexible coupling that could tolerate the typical end-float of a 132-frame induction motor. And Ross Pink warned me against using solid shaft encoders anyway and swears by hollow shaft encoders because flexible couplings don't last and mounting is allowed to be much less precise with a hollow shaft.

So can you tell us a price for that INGH 12HS 33AM 00800 encoder? Why did you go for 800 ppr? Ross Pink suggests the version with differential outputs for noise immunity. And we'd only need a 2 m cable so that would be an INGH 12HS 27AM 00800. I guess it depends what they keep in stock?

Just having a look at the piccies and in reference to Chapter 4 of the "Electric Motor Repair":

Looks like a "Consequent Pole" 4 pole motor with concentric windings

(starting at the top going clockwise)

Phase A is
A1: 1-12 2-11 3-10
A2: 19-30 20-29 21-28

Phase B is
B1: 13-24 14-23 15-22
B2: 31-6 32-5 33-4

Phase C is
C1: 25-36 26-35 27-34
C2: 7-18 8-17 9-16

I can see fat wires coming out of the outer of slots 6, 24, 30, and the inner of slot 33.

There will be 6 wires coming from the terminal box.

It looks like A1 & A2 etc. are in parallel already, looking at the connections on top around slot 27-36.

If you're feeling keen you can pull apart these insulated wires and maybe convert to 2 pole 30kW by reversing A2,B2,C2 connections.

Probably the best thing to do is take lots of photos, work out which wires go where as best you can, and put it all back together without breaking anything.

Then you can do some homework, figure out if anything good is possible, and pull it out again later and do it.

cheers,
Woody

I just pulled my motor out with a view to placing the phases in parallel in order to convert it to 220v star. Unfortunately the wiring is all gooped up with a hard brittle substance preventing getting into it.

I am also fitting an 800ppr encoder. See my blog for details and full resolution photos.

Here are some photos.

Originally posted by antiscab free energy? was that post meant for this thread? i think you've lost me Nevilleh   *confused* Matt

Hi Matt,

Neville is referring back to the Loo Tech Electricity Amplifier whose invention was indeed first announced in this very thread back in July 2009.