Motor inductance discrepancy between phases

[bunya]

Hi All,

I borrowed an LCR to measure the inductance of my two cyclone motors, I was surprised to find that both motors were the same but the inductance across the three phases was different.
1kHZ LCR measurements:
A-B 188.9 uH
B-C 190.2 uH
C-A 121.7 uH

Both motors were within 1uH of each other but the inductance red and black phase wires was only 121uH compared to the 188-190 of the other two.
The motor seems to work just fine. I imagine this descrepancy wouldn't help with torque ripple but it must be there for another purpose or a fault in both motors?...

 

[Miles]

I think this just demonstrates the affect that rotor position has on inductance.

 

[mateusleo]

According to what Miles stated above, spin the motor a bit and measure again so we can have more info

 

[Doctorbass]

To measure the proper inductance and see if there is any variation over the 3 phases you must remove the rotor :wink:

If the inductance is still too different between phases then you might meed the same problem i had with my first opened never used MXUS 3000, witch mean two of the 16 strands was broken ! that was the inductance variation i measured !

A motor MUST/Should have all 3 phases inductance and resistance nearly identical. I would say the tolerance of the inductance would be proportional to the 1/(parallel strands number).. ex 16 strands mean a tolerance that should be less than 1/16 of the inductance value...

Doc

 

[cycborg]

As an aside, this is how some sensorless field-oriented controllers can determine the rotor position at a standstill - the inductance of the three phases is "measured" by applying a high-frequency excitation (just like the LCR meter does), and the discrepancies between the phases tell you the alignment of the rotor with respect to the stator poles.

If you have broken strands, this will show up in the resistance as well. Are your resistances equal?

Strange that you get the same result with both motors, but maybe it's just coincidence that the same phases were aligned in both when you happened to measure them. Spin to different positions and see what happens. Or turn the rotor slowly - if you can feel the individual cogging points, as you step through them you should see the inductances cycle accordingly.

 

[bunya]

Thanks for the prompt replies, I'll take another measurement today with the rotor removed. I did spin the motor by turning the wheel of my bike and take another measurement which was also 120 uH but it could have gone around by full revolution etc. I had only measured resistance via a MM not via the constant current and voltage drop method, I can also do this today to confirm.

 

[bunya]

Rotor position it was, I removed the rotor and took some measurements, resistance was measured with a CC power supply at 5.00A.
Phase A-B: 154.2uH and 59.9 mOhms
Phase B-C: 154.5uH and 59.7 mOhms
Phase C-A: 154.5uH and 60.1 mOhms

If anyone wanted some reasonably precise measurements of the 1200W cyclone motor here they are:

Also I was surprised by the resistance of the A-B phase wire which was 13.2 mOhms, I'll probably replace this with some AWG10.

 

[Miles]

If you could possibly get accurate measurements for the velocity constant (Kv) and the no load amps at two different full open throttle speeds, we'd have all the info needed to enter the Cyclone motor into the comparison spreadsheet http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=65757

 

[Doctorbass]

Rotor position it was, I removed the rotor and took some measurements, resistance was measured with a CC power supply at 5.00A.
Phase A-B: 154.2uH and 59.9 mOhms
Phase B-C: 154.5uH and 59.7 mOhms
Phase C-A: 154.5uH and 60.1 mOhms

If anyone wanted some reasonably precise measurements of the 1200W cyclone motor here they are:

Also I was surprised by the resistance of the A-B phase wire which was 13.2 mOhms, I'll probably replace this with some AWG10.

Does your 59.9mOhm measurement include the wire 13.2mOhm?

Personally when i post motor data i always get rid of the wire measurement so everybody can compare with an "absolute winding only" value whatever the wires they installed. In fact it is easyer to apply current to the external wires and to measure voltage at the winding-wire junction... witch in fact is a true Kelvin 4 wires measurement :wink:

Doc

 

[bunya]

Yes the 60 mOhm measurement was "absolute winding resistance". Voltage reading was taken directly from point where the motor windings get soldered onto the phase leads, current was flowing through leads so overall voltage drop at the terminals of the CC power supply would have been higher.
So Doc*, it makes sense that you provide a measurement from just the windings as 2 lots 13 mOhm would have huge influence on that reading: 60 mOhm vs 86 mOhm...

@ES, I am happy to try and contribute, though I have an LVC that I don't know how to disable, I can take a no load current measurement at 52V but to get another I could try running the motor at a slightly high voltage, or disable the LVC and run it at 26V. Is the measurement an AC phase current measurement as I know my controller will be more lossy at lower voltages and this will affect the accuracy of the results if I measure DC in from the battery.

Is Kv measure via a scope? Full throttle and measure the frequency? (feel free to refer me to a post about this, I am lucky enough to have access to a range of equipment that would be needed for these measurements)

 

[crossbreak]

yep you can measure Kv with a scope. or a multimeter that can measure frequency. or a laser tacho. or a cycle analyst you connect to one hall sensor.
and of course measure battery voltage and current at the same time. Just measure DC, controller loss is negligible at no load, even for a weak controller. AC would be hard to compare as many True-RMS multimeters differ somewhat. For no-load measurements i always use the same 12fet sensorless controller have in spare

A simple way to measure 2 no load speeds is to do one test with full battery and one with an empty battery. or connect a 12V battery in series to your lithium pack, make sure you dont exceed your controller max voltage rating.

Would be very valuable to have these 2 no load figures and Kv. Thanks!