Motor shutdown - help

[trietje]

Hello Fellow e-bikers,

For the moment I am driving my Dahon Cadenza with a Crystalite HS2540 in the front wheel and a battery pack of 36V 15Ah.
With full pack I can drive 35 km at 40 km/h and I pedal moderate

Yesterday, in rainy weather, I had strange motor behaviour.
With strange I mean that the motor does not always respond to the throttle and then after trying a couple of times it does again.
Also the engine just keeps going after releasing the throttle.
After 1 hour of such behaviour the engine stopped and reacted at nothing.
Battery is ok and feeding the controller. All cables checked.

Symptoms are:

when the cable is connect to the engine, the engine gives resistance in driving on human power only. This is even so with battery disconnected. (seems very strange to me).
when cable is disconnected the engine runs freely, without resistance in driving on human power.
e-DR display is working normal, no notification of bad connections or failure.
I can see some vapor/fog of water in display the day after it rained a lot.
Controller LED active and normal function.

What could it be and what could i do to make the engine/setup work again?

Thank you in advance,

Dimitri

 

[neptronix]

I would check the hall sensor wiring and throttle wiring, if you experience this intermittance.
With the bike upside down, i'd hold the throttle steady and jiggle the hall sensor and throttle wires to rule this out.

Another thing is that you could very well have water in your motor. Hub motors are poorly waterproofed. Open it up and do a check.. take pics if you do so.

 

[dogman dan]

I think the resistance points to a short in the phases in the controller, or perhaps just the plug. You could have tiny shorts from wet connections in other places too, like the halls in addition to the problem causing the high resistance to pedaling.

Hopefully if you dry everything out, it will be ok.

 

[neptronix]

You are probably right dogman, but intermittent power means there are probably 2 issues at play.

My bet: unspeakable things will be discovered when the motor is opened

 

[boppinbob]

I'm a gamer and in the old days when computers weren't very fast I used to water cool the cpu, gpu and southbridge chip with an aquarium pump. After 3 yrs it sprang a leak and the screen went blank. I thought it was toast. I let it dry out for 2 or 3 days and voila, it worked again. If you take the cover off your components to expedite drying, all might be good again.

 

[docnjoj]

Unfortunately water is the Achilles (sp?) heel of our electric systems. Ther are some ways to "waterproof" them but none are really perfect. Yep! Dry everything out by opening it up and perhaps a hair dryer blowing on the boards, connections and wires. I'm afraid opening the motor may be necessary but try all other parts and connectors first! Rain and electrics don't seem to mix yet! some folks drill holes in the motors to allow the water to get out!
otherDoc

 

[trietje]

Thanks for all the replies.
Today I opened the controller and dried it with a hot air blower.
I could not see any moisture inside.
Also took apart the connections, seems to be dry all.

Only now what is left is the engine ...
Just take apart the engine through the screws? And then dry everything with a hot air blower?
an I install anything to prevent that in the future?

WHat can I expect to see when opening the engine? It is a new one since oktober 2011 ...

 

[dogman dan]

At worst, rusty water pours out. Likely you have moisture, rust, and some dew in there. One helpfull thing for riding hubmotors in wet places is a drip loop. Leaving the motor, the wire goes down before turning and going back up the forks or frame. Water will collect, run down the wire, but drip off before running down the wire and into the hub.

Once you have the cover off, you can drill two holes in it, perhaps a half inch size. Cover it again with good tape. On a dry day once in a while, pull the tape off, and run the motor long enough to heat it up. Any humidity inside will then steam out through the holes. Then tape it back up again.

It could be completely unrelated to water, if you have nicks in the wires anywhere. Cut wires where they enter the axle are common.

Did that resistance we talked about go away when you plugged in the dried out controller? If it didn't, you have a shorted fet in the controller. No resistance unplugged indicates the phase short is in the plug or the controller.

 

[trietje]

Yes, after I dried the controller out (it was not even wet in my opinion), and connected to the motor, the resistance in the wheel was still there (like before).
Only with an unplugged engine, the resistance is gone.
So then the problem is in the controller? Can I fix that? WHat can I do about a phase short in the plug?
No need to take apart the engine then?
The idea of a drip loop sounds smart

 

[trietje]

If I connect another controller to the engine, the wheel runs without any friction.
SO it is probably a problem in the controller.
On the side of the print where the switchcover is on the controller, there seem to be a little oxidation. From water?
How can I clean this up or repair if this might be the problem?

http://desmond.imageshack.us/Himg849/scaled.php?server=849&filename=dsci0298.jpg&res=landing


http://desmond.imageshack.us/Himg51/scaled.php?server=51&filename=dsci0297.jpg&res=landing

 

[trietje]

Cleaned the oxidation with alcohol ... most of it seem to be removed.
But still resistance on the wheel when the controller is connected to the motor.
So now i take motor apart ...

 

[trietje]

Motor seem to be ok ... cleaned anyway ...
When reconnecting I still have resistance on the motor when connecting the controller ...
So broken controller like dogman predicted ...

http://img404.imageshack.us/img404/1079/dsci0305o.jpg
http://img24.imageshack.us/img24/1914/dsci0303s.jpg
http://img21.imageshack.us/img21/3009/dsci0302p.jpg

 

[Ykick]

Motor seem to be ok ... cleaned anyway ...
When reconnecting I still have resistance on the motor when connecting the controller ...
So broken controller like dogman predicted ...

http://img404.imageshack.us/img404/1079/dsci0305o.jpg
http://img24.imageshack.us/img24/1914/dsci0303s.jpg
http://img21.imageshack.us/img21/3009/dsci0302p.jpg

Correct, output FET's likely shorted.