Hot motor = crook windings ?

[Hyena]

Hey guys,
I've got a bit of a weird problem with a kit I put together for a guy recently.
It's a crystalyte HT35, running on 18S lipo.
I've done a fair bit of testing with this configuration and they normally run quite cool and efficiently with the lower speed wind.

So the guy assembled the kit on his bike and after the first ride he said the motor was boiling to touch and the battery quickly went flat. When I checked the CA the top speed was only around 40km/hr but the whr/km was up around 45 (should be sub 20 on this config). The current limit was only around 40 amps so it certainly shouldn't have been running that hot.
He bought it back with a stuttering problem which I assumed was halls. At first I thought it was a wire cut issues but the wiring looked fine. Each hall voltage changed with the wheel spun but not quiet as it should - only slight voltage changes. I figured maybe the heat cooked them and took the easy way out and fitted a sensorless controller. I took it for a spin down the driveway and back and it ran smooth and fine so handed it back to the guy.

A few rides later he says it's scortching hot again when he tries ride it now it just shudders. Cooked windings.

So, I'm kinda stumped as to what the problem is. Given I replaced the controller with a sensorless one it really has to be the motor.
I've never seen it before (contrary to popular opinion I try *not* to blow up motors ) but my guess would be a short in the windings somewhere from new. It ran smooth on the bench and under load on the short test I gave it but apparently was just dumping all the current the controller allowed as heat.

Thoughts ?

 

[nechaus]

was the side cover drilled out at any stage?

Maybe need some pictures of the insides to see if their is any physical damage.
possible have the wrong phase combo and it ran but eventually caused issues ?

its seems like a strange problem

 

[Jeremy Harris]

Sounds like a shorted turn in the motor, most probably a manufacturing defect, perhaps from the wire insulation having been damaged during winding. The effect would be for the affected phase to draw a lot of current, making the motor heat up. I bet if you take the cover off you'll find one phase winding cooked if this is the fault. Often the sides of the windings on these motors are hammered flat during manufacture, in order to fit in the case with enough clearance, a process that doesn't look to carefully done in some of the motor winding videos that are around.

 

[John in CR]

A short in the laminations leads to extra heat too. That's how Farfle melted his double Pi. I don't know about the shorted windings, since I think that would cause obvious problems and not run at all, though I guess that could depend on the nature of the short. I definitely sounds like a manufacturing defect...maybe simply not wound correctly.

 

[Jeremy Harris]

A motor will usually run OK with a shorted turn, it'll just get hot and have a high torque ripple under load. Often they'll run reasonably well at no load, depending on the severity of the short. All the shorted turn does is act as transformer, effectively loading one winding down as current increases. It's a fairly common problem with industrial motors, such that there's a special bit of kit for detecting it, a Growler (http://en.wikipedia.org/wiki/Growler_%28electrical_device%29). We had one at a place I used to work, a remarkably crude but effective way of testing for faulty windings.

 

[Lebowski]

What Jeremy describes I suspect is also the failure mode for my cooked motor. I burned out my motor
but it still turn when I power up, just no torque anymore. If you have an inductance measurment thingy
you should be able to detect this as the affected winding will have a much lower inductance...

 

[John in CR]

What usually are "shorted turns" ? Scraped enamel on the magnet wire shorted to the stator iron, or shorted to itself on a different turn?

 

[Hyena]

I suspect as Jeremy said, that's it's shorted to it's own turn (rather than another phase) hence why it still runs smoothly. Well, did... :|

I like the sound of the growler Jeremy, if for no other reason than it's a cool sounding name :lol:

 

[Jeremy Harris]

A shorted turn is usually within one winding, often where two wires have crossed, twisted or got kinked during winding, damaging the enamel and leading to a breakdown of the insulation. It's quite rare to get a short to the stator unless the stator is really poorly insulated, although it can happen at the edges.

The Growler got it's name because that's the sound you hear when the strip of steel you use as a pointer finds the shorted turn, it growls at mains frequency, often quite loudly.