troubleshooting an intermittent throttle/pedelec issue

[winkinatcha]

bunch of potential causes,

current symptoms:
Throttle works intermittently, and seems to cut out and stay cut out when WOT is applied

Pedelec input intermittent, but better than throttle... that is, when engaged, consistent cadence produces on-off-on-off power sensation, maybe half a second on, then off for unpredictable time.

(when working correctly pedelec produces similar sensation to WOT... consistant uninterupted power.)



History:

Existing Vintage road racer build, running mini geared hubmotor rear.

bog standard Cohnis kit 9 FET controller 20-25 amp, 62 volt caps.. presumed 75v FETS, modified to 10-15 amp by removing one of two shunts

Controller has been run and is capable of being run at voltages from 24 thru to 60

thumb type throttle shell physically modified but electronics untouched (hall sensor and magnet)

pedelec ring sensor on Bottom bracket, with plastic 6 magnet ring on axle inside crank.

(mmm think I just figured where I need to check... Pedelec sensors have been a "challenge" in the past on other builds)

.......

Build was initially working absolutely fine and dandy, running 12s LiPo, nominal 500w, through a 36v QT120 equivalent style min-geared-hubmoter
Motor is Nominal 250w but seemingly more than capable at 750w for a road oriented build.

I broke the first motor, literally by accidentally jamming a huffy drift trike into the back wheel at 25km/h

I just got some new stock, different manufacturer, mini geared hubmotors, nominal 24v... meaning fast wind...

I installed one of these wheels into the existing build, swapping out the busted 36v wheel, Saturday night, preparatory to a group ride here in Melbourne AUS on Sunday.

Got lost getting to the ride Sunday AM and ended up riding the bike hard, a fair amount of road work, lots of WOT or full pedelec, CA saying system is drawing around 600 watts maxed out.

GOOD SPEED, hit 50km/h with the motor still pulling on the flat.

Found the group ride, just as I killed my 12s LiPo battery... (Next minute... 12s pack is reading 30v Holy Carp.. I just killed my battteries.)


Borrowed a 58v pack...

Wass following a mad down-hill local (Full-Throttle) and inadvertantly found myself launching my roadie and belting into a few things, including the ground, though managing to stay upright.

got onto som clean bitumen cycle track and managed to wind the bike out to 55

caught up with group, lots of full throttle and hardcore abuse, including attempting a hill-climb and stalling out...

Began to have "throttle issues" as above

Got a lift back with bike in mates trailer

swapped battery for a 36v monster and went for a 50 km ride this AM

throttle issues initially not bad but worse and worse til eventually rode final 15 odd KMs on pedal power.. (Advantage of lightish road build... not a total shit-fight to pedal back home )

..............

After writing this... the first thing I am gonna do is swap out the pedelec sensor, and see if that makes a difference... Had issues with them in the past... may have got bashed on the trails...
If swapping out sensor makes no diff, will take off all electrics and put in a box marked "WTF?" and install new stuff... I did give the poor thing a hell of a hard time and hard to track issues like water ingress (went through a few puddles n a little mud) as well as a bunch of other stuff could mean stuff well beyond my current knowledge of testing n repair.

But if anyone's got another idea about what might be causing these throttle/pedelec issues... (maybe not even the controlls but other stuff going on, like internal shorting in the hub or summink...)
It would be awesome to get your input

mmm wonder if I have smoked the motor... ohhh and it's not halls... when the motor runs it is smooth.

Cheers

Joe



..........

 

[RoadWrinkle]

OK Geared hub motor, system is cutting out, could be a bad motor and not the throttle or PAS at all. Can you disassemble the motor and confirm good gears, especially the ring gear? Hill climb stalling out means your motor went from a motion device to a space heater. Most hub motors do not transmit heat efficiently to the outer dish assembly, so if the external is hot to the touch, its like the third ring of hell on the inside.

 

[NeilP]

As last poster, since same thing happens with throttle and PAS, to me that rules both out.

First thought was LVC cutting in, maybe due iffy batteries, but seems you have swapped battery pack so that is ruled out.

How long since removing one shunt wire has this started to happen?

I was thinking controller / CA issues

Need to read your post again when on big screen and not iphone

 

[winkinatcha]

No stress guys and thanks heaps for your input...

I've been hammering all shades of shizzle out of a bunch of mini geared motors recently
I got a bunch in new unexpectedly and accumulated a couple of second hand ones at the same time... managed to sell some good legal (in AUS) builds and kits based on these motors, well within tolerance (15ah, 10-12s). 12s I run through old stock 9FET kit controllers (originally with DD kits) that are more robust than the 6FET ones that come with the mini hubs.

the controller on this build is one of my original 9FET shunt mods, about 18 months of pretty constant use, been on a variety of builds from a dd 20" kick scooter running 58v 10amp, good service time on a dd running 24 to 60v and maybe 2 months around 500km on the current build... All post the shunt mod (removing one of two shunts to REDUCE the amperage from nominal 20 to nominal 10)
Note originally and still set with LVC of 21v for 24v nominal operation

so controller has had a long life but also well tested and not really stressed...

BUT on this run I stressed it... particularly full stall up an old bituman hill climb thing... and that (in retrospect) was the last time the throttle worked well.

Max amps during the stall could have had a variety of ill effects from controller through to motor.

Stoopid me not thinking the controller itself could have had a brain fart...

Good call on the "both pedelec and throttle misbehaving suggests problems at controller end" or some such similar words NeilP

RoadWrinkle... deffo motor was hot to touch several times and during the Hell-for-leather road ride it would have been hammering for a good 20 mins straight... That hill climb though, didn't check temp... third ring of hell a possibility.......


Much obliged for your considerations and deliberations... I am off for the night but first thing tomorrow I am gonna try to spin the wheel on a different controller (check hardware first), and work backwards from there (if wheel spins... original controller into the "WTF" box! )

Joe

EDIT/PS: At this stage, f I haven't fried the wheel, I am happy to continue using it without opening it up... mechanically it seems there is nothing obviously wrong with the wheel, when there was power to it the wheel engaged and did its thing... hadn't entirely died by the time I opted for pedal only... though I have zero clue what a failure of a geared motor feels like..Still... I wanna try spinning it up before I disassemble.

 

[teklektik]

BUT on this run I stressed it... particularly full stall up an old bituman hill climb thing... and that (in retrospect) was the last time the throttle worked well.

Max amps during the stall could have had a variety of ill effects from controller through to motor.

You might check the halls to see if the BBQ episode has compromised one of them. I had a hall failure last summer where the output had been marginalized and operation got spotty before finally giving out. It seemed that the output level was not consistently being seen by the controller.

• Unplug the motor phases, power up the controller, and using your meter check the voltage from the black hall lead to each of the three hall wires (B,G,Y) individually while slowly turning the wheel backward (so the clutch engages). For each hall the voltage should switch up and down about every 2 degrees of wheel rotation. The voltage should switch from close to 0v to 4.5v or so.
  
  You have to turn the wheel extremely slowly and carefully when testing a gear motor because of the speedup of the rotor caused by the internal gearing - the hall transition is very easy to miss with a meter. If you are using an auto-ranging meter - lock it to 'volts' range so it won't waste time doing ranging.

Anyhow - just a thought....

 

[winkinatcha]

Wheel spins up fine on a different controller (with different throttle and no pedelec sensor... why does each manufacturer use different connectors on their conrtrollers?? Grr!)
and continues to work under load (brakes applied to wheel to near stall, motor still pushing at WOT, done a bunch of times, no throttle issues post load test.)

So at least the wheel is good, making me happy that these buggers can take a bit of mild punishment.

Still unknown if pedelec/throttle issue or controller issue, though suspect from others' input here that controller is likel;y to be cuplrit now.

Not gonna get a chance to do much more on this build today, but intentions are to test existing throttle & pedelec sensor on new controller... if all spins up and operates fine, then controller will be isolated as issue, if not, sensors will be isolated as issue.

Joe

 

[winkinatcha]

Problem resolved...

Somewhere on the ride, I managed to mash the e-brake signal lines causing an ongoing and intermittent short... whic would of course, cut the controller power. Found this problem as I began stripping of the wiring and original controller.

Easy fix, bike is now fine.

Joe

 

[cosmicman]

Had the exact same problem, was about to start stripping the whole bike looking for loose connections but I read this first and disconnected the brake switch wires - straight away the problem was resolved. After putting a meter on the brake switch wires I soon found the problem, loose switches in the brake levers - the plastic cap wasn't tight enough to hold the brake switch away from the brake lever, so even the slightest touch on the brake lever (up or down, not in or out) was causing the power to cut off.