Identifying Winding (leads) on 12 lead Generator.

[stair]

I have a 12-lead generator which has been cut out by monster workers that has no labels or color code. I can take an ohm meter and determine the 6 pairs of windings; however does anyone have a method to determine what each pair is in relation to one another; that is, is a particular winding t1-t4, t2-t5, t3-t6, t7-t10, t8-t11, or t9-t12. I want to connect this generator in low voltage, parallel, wye; which would require parallel of t1-t4 with t7-t10, t2-t5 with t8-t11, and t3-t6 with t9-t12. Also, how does one know which end of the winding it is; that is, is this particular lead of the pair t1 or t4. Would it be out of phase if one paralleled t1 to t10 and t4 to t7? Maybe there is a way with a tone or RF signal? I suppose one could generate with a very low field and try to juggle the 12 leads and meter or scope it out, but that would be quite cumbersome.

Any ideas?

 

[waross]

You could try exciting the machine with AC. Then turn it slowly. A voltmeter will tell you when the field is lined up with each winding. That is each winding will show a voltage that varies from zero peak as the rotor is turned slowly. The voltage on each winding will peak as the field passes the center of the winding.
Step two:
When you have identified all three groups, connect each pair in series and continue checking. If the connection is correct, you will see double voltage. If one of the coils is reversed the voltage will be zero.
Step three: Connect the windings in star. Continue turning as you measure the phase to phase voltages. The voltages should be equal. If one voltage shows about 55% or 60% it is reversed.
It may be well to remove the injectors or spark plugs before starting. A compression release is a boon. With the compression relieved you should be able to turn the set with a wrench on the front of the crank shaft.
Good luck.
Let us know how it turns out.
Good luck.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

PS: You must bypass the diodes on a brushless exciter and energize the field directly. You may have to use a long cable and let it wind up on the shaft as you turn the shaft. Then unwind and go again.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

If you want to run the machine, connect any two leads together and measure the voltage. If the voltage is double the voltage across one coil or zero, you have identified a phase group. If the voltage is zero reverse one coil.
You should be able to identify all three phases and connect them in parallel. Then connect the groups in star and see my last post for the expected Volt readings.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

It often helps to look at the voltages induced in each winding. And the remanent rotor field can be used. So no need for excitation during this work.

1. Rotate the rotor, find one winding and trig on that one.
2. Then use next channel on the scope/recorder and find the winding that is in phase with the reference winding. Mark the two winding's phase and polarity.
3. Still using first winding as a reference, find winding that is 120 degrees after first.
4. Find second winding with same phase. Mark the two windings.
5. Same thing with the two remaining windings. It is quite easy, actually.

You can use anything to rotate the rotor. You don't even have to rotate it at nominal speed but it helps if rotation is fairly constant in speed.




Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[EmergencyPowered]

I've used Skogsgurra's method in the past and it was simple and obvious once you see the scope waveforms. I am no engineer, just a power systems technician.

 

[PHovnanian]

Skogsgurra's method works with a voltmeter and a bit of trig. You'll still have to identify phase rotation once its all hooked up.

 

[dicer]

So when using Skogsgurra's method, do you ground one of the windings wires and then trigger the scope on the loose end?
And ground one end of the winding that is being checked and on and on?

 

[Skogsgurra]

Yes something like that. Not necessarily ground. But ground it if that makes it clearer.

PH, you get phase rotation from the sine wave sequence. No need for extra identification. But easy to do if you want to.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[PHovnanian]

PH, you get phase rotation from the sine wave sequence. No need for extra identification. But easy to do if you want to.

Yeah, but voltmeters are cheaper (an important factor if you've ever seen me poking around in a live junction box). ;-)

 

[Skogsgurra]

Not cheaper than a RIGOL scope. It costs less than a Fluke DMM. Two channels, good quality and fantastic specs for the price.
I bought one together with the Scopix 4 channel 200 MHz scope. And I use the RIGOL more on the bench than the Scopix. Scopix is great for field work.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

If it is possible to energize the field with AC, the voltage induced in the windings will be mostly independent of the speed of rotation. the rotor may be positioned so that one winding is at the maximum voltage and the rotor left in that position until the companion winding is located and connected. The testing may be done with only one revolution or less. Same principle as a selsyn.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[dicer]

Do you have alink for scopix, I googled didn't find it.

 

[PHovnanian]

It costs less than a Fluke DMM.

You'd trust me with a Fluke? I was thinking Harbor Freight.

 

[Skogsgurra]

One link here. There are several.

http://www.test-meter.co.uk/index.php?_a=viewProd&productId=806

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.