Conversion of a Bultaco Sherpa

[Elektrosherpa]

I think I posted some pictures in my ktm thread when I changed bearings.

Found the pics :thumb:
If I get that right, there are only 2 bearings, one in each side cover of the motor?

 

[Elektrosherpa]

This kind of problem is often linked to a hall sensor problem.

And this can happen from one day to the other just like that, without having changed anything in the settings?
How can I find out?

 

[Elektrosherpa]

What does the Kelly say, fault codes?

No error message/fault code at all...

 

[Elektrosherpa]

... a new problem appeared- the motor made very bad noise, then was like it wanted to move but couldn't.

Wrote to the guy from QS who sold me the motor/controller combination-
of course they hide behind the fact that I have received the motor more than one year ago...

So thanks again to the battery seller a***holes, who caused the huge delay in my build
by their enormously delayed delivery and the garbage BMS/charging port...

 

[larsb]

I see you’re in france. Any chance you can find someone local to borrow a controller from for a test? My gut says you’re into a hole with the Kelly, a test with a different controller would give a quick answer.

 

[PITMIX]

Tu es en France ?

 

[Elektrosherpa]

Tu es en France ?

Oui, en Alsace...

 

[PITMIX]

Ok I am near of Paris. It's so far !
If you don't have fault in Kelly is not a hall failure.
But I don't think it's a rolling problem either.
You can very easily disassemble the motor flange on the pinion side, you will already see the bearing on top.

 

[j bjork]

I think I posted some pictures in my ktm thread when I changed bearings.

Found the pics :thumb:
If I get that right, there are only 2 bearings, one in each side cover of the motor?

Yes

 

[Elektrosherpa]

Just to keep you updated:
Until now, I havent had time to do anything - too much work.
Anyway I am not really in the mood right now to invest more time into this ...

In the meantime, QS has offered me to repair the motor for free - but:
"Shipping cost back by air express USD160
Optional shipping by train door to door(45days): USD90"

So either I wait an eternity, or I pay a lot of money, to finally get a working motor...
Both alternatives are not really tempting...

 

[larsb]

I’m not sure it’s the motor so i think it’s generous of them. But if it isn’t then problems will remain..

 

[PITMIX]

I agree with Larsb. The likelihood of this being the engine is extremely low. Save your time and money and look for another reason for the problem you are having.

 

[Elektrosherpa]

The problem is that the motor doesn't move anymore.
If I turn the throttle, there is a humming sound (not really loud), but the rear wheel/sprocket/shaft does not move at all.

I can move the wheel/chain/sprocket/shaft by hand, so the reason for that is probably not a motor bearing which is completely blocked.

I have no idea where to start looking for the problem.

 

[PITMIX]

The first thing to do is disconnect the 3 motor phases.
If the wheel rotates freely while driving the motor without forcing then the motor is ok.
Be careful, the motor phases must not touch each other when you do the test by rotating the wheel. Otherwise it's as if there was a short circuit
If the motor is blocked only when you connect the 3 motor Phases with controller it means that the controller is dead on the mosfets or that the phases are shorted.
If the motor blocked even when the phases are disconnected from the controller then there is a short circuit on the phases or the winding of the motor. In this case you will be the first that I know who has burned out his new engine QS ! :wink:

 

[Elektrosherpa]

Thanks, Peter.
This sounds like something I can do (disconnecting phase wires).
I will report.

 

[PITMIX]

 

[Elektrosherpa]

The motorbike still refuses to spin the wheel when I switch everything on and turn the throttle.

So first I tried to move the wheel (and with it the motor, too, via the chain), and I was able to move it without applying a lot of force.

Then I removed the chain to the rear wheel and disconnected the phase wires from the controller.
No difference when I tried to move the sprocket, with phase wires connected or not.
It moves without applying too much force (but not as freely as the rear wheel moves alone, without the chain).

I also tried to do the Kelly Controller's "Identification Angle", and it failed.
While it was trying to do the "Identification Angle" procedure, I observed that it did do something (because the BMS display showed various currents, between 0.2 and 4.5 A), but the sprocket didn't move at all.
I had been doing this a few times in the build process, and each time it worked, so I know how it should go if everything is normal: The sprocket should rotate and stop a few times...

The Controllers "monitor" function then showed the error message "Identify Err",
and gave an absolutely impossible reading for the controller temperature (246°C):

(the controller really had ~25°C , what was the ambient temperature this afternoon)

This means what now???

 

[larsb]

Do you see the halls working in the kelly program, if you rotate the wheel?
Did you ever identify what (if any) temp sensor you had?

A wild guess just by looking at a kty-84 characteristic is that resistance would have tripled to show a reading at 246 deg C and the halls and temp sensor share ground line. Could indicate a failing ground wire/circuit.

 

[Elektrosherpa]

Do you see the halls working in the kelly program, if you rotate the wheel?

Yes, the "1"s and "0"s alternate when I move the sprocket/wheel

 

[PITMIX]

The good news is that your motor does not seem to have any electrical problem. So he's not dead. The fact that it presents resistance to rotation is normal because Neodymium magnets are very powerful. Disable the engine temperature safety device, your sensor has been saying nonsense from the start. Your controller is probably in motor overheat protection.

 

[Elektrosherpa]

I re-connected the phase wires and when I wanted to switch on the main power, there was a huge spark under my controller cover.
Opened it, and it looks now like this:



The fuse is still unchanged, but the controller seems to be dead now...
WTF is happening?

I didn't change anything except diconnecting the phase wires yesterday, and connect them again today.

Just for comparison-this is what it looked like before:

 

[PITMIX]

Yes indeed before burning it was not burned...Now the motor is stuck right ?
There should be a mass on the blue phase.
It looks like a ground or a wire with another voltage has shorted here.
You didn't forget to remove the 13 key to tighten the phase before powering up?

 

[Elektrosherpa]

It looks like a ground or a wire with another voltage has shorted here.
You didn't forget to remove the 13 key to tighten the phase before powering up?

No way- I even had closed the cover before switching on !
There is/was no lose cable inside the controller box, and no tool either.

I dont understand from where those small balls of melted metal come from.

And the motor(sprocket/shaft) can still be moved by hand, it feels the same as before.

 

[PITMIX]

It's really very strange because to make a short circuit at this level it is absolutely a piece of metal between 2 phases or between the phase and the mass of the frame of the motorcycle for example. If the motor is not blocked it is surely that the controller is not in short circuit, but it would be necessary to open it to see inside, if there is damage. A friend of mine had a problem with a brand new 600A Kelly controller that cost over $1200. The controller burned out on first power-up. There was an internal short between +5v and +VCC.

 

[Elektrosherpa]

Do you see the halls working in the kelly program, if you rotate the wheel?

Yes, the "O" and "1" alternated.

Did you ever identify what (if any) temp sensor you had?

Not yet, because it seemed evident that this sensor shows nonsense,
and then the motor stopped moving at all,
so I decided to deal with that first...

A wild guess just by looking at a kty-84 characteristic is that resistance would have tripled to show a reading at 246 deg C and the halls and temp sensor share ground line. Could indicate a failing ground wire/circuit.

This (ground wire problem) might be proven by the most recent incident ?
(see above-a flash at the controller's blue wire inside the closed controller box when switching it on)

But where?
Inside the motor?