Ever had an ebike breakdown?

Joined
Dec 2, 2016
Messages
216
Location
Land of the Florida Man
This has happened to me twice in a week, now! Riding 20~30 miles, then it stops working. Both times at night, and in the same area. The first time was the on/off switch plug had come undone. Easily fixable had I known more about what I was looking at.

But, this time (tonight, 30 min ago) it was the geared front hub motor screeching and grinding to a halt. Felt like someone slammed on the brakes. I was cruising along at 22~23 mph, and with no warning it seized up. A little smoke with electrical smell was coming out of the frame bag, so I opened it up. No wires torched or anything.

I think the controller blew up. Battery pack still has good charge at 3.8v, and all 14 parallel groups are good. So I think the expensive part is okay. The wire that comes out of the hollow axle got spun and yanked really tight, from whatever happened inside. Must have been pretty awesome carnage from what it sounded like. Something also happened with the bearing on the opposite side of the wire.

It was really hard to push if you walked it, and felt like a constant uphill in the granny gear if you pedaled it. So, the remaining 20 miles home was not feasible. A $16.80 taxi fare it is! Always carry cash, debit card, DL, and bungee cords as breakdown supplies. Not including your tubes, CO2, tire levers, multitool, wrenches, cellphone, snacks, water, etc.

And, I'll tell you what. Any day spent riding or solving a breakdown issue is far more entertaining than playing games on my computer. Reality is unpredictable and amazing, and it's about time I got back in to it.

I'm thinking a leaf 1500 DD front hub motor on a 700c rim, plus a 6 fet sinewave controller would be on its way shortly.
 
My bike has failed me plenty of times. Sometimes ride able, sometime not. Pedaling or pushing a 90lb bike sucks.

What kind of motor were you using? Sounds like the gears destructed, causing it to spin the axle, shorting the controller and smoking it. I hate it when that happens. At least the battery sounds like it survived.

Time to consider an upgrade.

I'd love to see pictures of the insides of the motor. I bet its real ugly. You may find a clue to preventing this in the future.
 
Put the 1500W leaf in the back, you'll thank yourself. It will power wheelie from a stop like none other. Also you'll need torque arms with that motor wherever you end up putting it. The sine wave is a good choice but for the love of god, don't get a Leaf and then power it with a 6 fet! I have a 18 fet 4110 and wouldn't dream of using anything less powerful to drive this motor. I put 5KW peak thru the thing and it goes 56 with me in a tuck.

Have had good luck with breaking down, personally. It hasn't happened anywhere remote but I always carry a full tool bag pump and tire stuff.
 
hypertoric_amplituhedron said:
The wire that comes out of the hollow axle got spun and yanked really tight, from whatever happened inside. Must have been pretty awesome carnage from what it sounded like. Something also happened with the bearing on the opposite side of the wire.
If the wires are twisted up, then that usually means the axle spun in the dropout; this usually means whatever torque arms you have were insufficient to hold the motor in place.

Since the geared motors usually have a freewheeling clutch, which would make it the same to ride with or without power, then something else probably went wrong, too.

Most likely either the clutch failed in a way that locked it up, or the gears disintegrated, and when this happened, the rolling torque of the bike applied directly to the axle of the motor, and turned it in the dropouts, twisting the cable up until it cut into the phase wires.

When the phase wires then shorted together (or to the axle/frame), the controller blew up (FETs shorted by the twisted wiring).

Probably the battery is ok, but most likely the controller will need repair or replacement. If the motor actually has internal damage, and didn't just have an axle spinout, then it may need repair or replacement. Otherwise just the wiring needs to be repaired, which can be done by pulling it thru from the inside and cutting out the damaged area, then splice to the existing interior motor wiring. Then replace the torque arms with better ones. :)
 
I have fried motors and controllers to fill a garbage can, before I achieved reliability at the power that I like to ride. Since then, I fried a motor last fall by stupid mistake after 4 years of rough riding. As for small electric problems, I have had none that I was not able to diagnose and fix on the spot in a few minutes.
 
Were you already pushing 72V? As indicated in your thread here - https://endless-sphere.com/forums/viewtopic.php?f=3&t=86999&p=1271894#p1271894

Breakdowns blow chunks man. Sometimes you can walk it up hills and coast down them and maybe even pedal slightly on flat? Other times like this situation, it’s taxi, subway, Uber, friend, hitchhiking for a pickup truck, etc.

Whatever it takes to get home and then what? Often out of commission for a while. That’s why I’m a firm believer in more than one eBike. Share the expensive battery pack if you must but it’s not that much money to assemble a 2nd eBike of some sort.

That way, you have the luxury of comparison along with a spare to ride while you get the other eBike back up and running.
 
Pics, because it definitely happened:

9fX08ja.jpg


The axle spun several times, which unthreaded the right nut, but kept the left one tight. Thank god, since my wheel could have come off at 22mph. Not good, man.

The motor doesn't have that much torque to even remotely spin it. But, if the motor were to lock all of a sudden at 22mph, well there's a lot of momentum to dissipate, straight to the weakness of no torque arms.



FfUbI4j.jpg


Spread apart the dropouts a bit. I'll try to close them with a vice tomorrow. Maybe just replace it, might have a spare one laying around the shop.



wI9jQSg.jpg


The phase and hall wires got wound up pretty good.



Avr1M1B.jpg


On closer inspection, only one of them was physically damaged, a simple signal wire.



I4q1aV6.jpg


Aha, the red phase wire and some others had burned up into a metallic foam. Controller couldn't take the power, which I think is 52v 20a. So, this may be what caused the motor to lock.

And, the real question is:

If 80% throttle was being applied, and one of the phase wires burns up, would the motor actually lock up?



GEABWei.jpg




wRuW3nh.jpg


You can see the chernobyl-style Elephant's Foot that formed, when the solder was melting from the current.



6rTxYoP.jpg


Time to extract the cover, and see what's inside, expecting shredded internals ....



iuvv9O2.jpg


Actually ends up being a direct drive motor. This may be the one of the few DD motors that AmpedBikes sold. I believe they sold mostly geared hub motors, but I could be wrong. This is good news, since the motor is okay. It also means I don't need to worry about the issues with riding a geared motor too fast for too long. I really can put 72v into it, and run it at higher speeds. No issues with nylon gears getting mushy.

It has decent torque from a dead stop. You don't need to get it going first. Is this indicative of a high turn winding, like 8T or 10T? It also has a funny power band, with good responsive torque at low and high rpm, but a lull zone in the middle.



VS4vWoj.jpg


Hall sensor wires



OuIBrqn.jpg


Phase wires



A9ugLBJ.jpg


My friend Talon performing his magic. Since the wires got damaged from the twisting axle, we had to re-solder all 8 of the wires. After the job got done, the motor was sitting on the chair, and slipped off.

The axle banged on the ground, which cut through the newly fixed wires. Had to do the whole process all over again. Seriously, WTF man. Just a freak incident, when the universe doesn't want you to get off that easily.



LrKhuf5.jpg


All fixed up, ready to go. Apply that sealant ....



9yb4HVt.jpg


I'm gonna ride out like Predator, with the guts of my new kill hanging from my handlebar. It's a warning to all other lesser-capable controllers to watch out, here comes the GT e-Windstream.

Since the controller took a crap, I'll need a new one. So, that's when talon pulls out a dusty, old 12fet spare. Ole' Reliable is now back in commission, packing a 30 amp continuous, which is an upgrade to what I had. Now waiting on a 3-wire throttle, that will arrive on the 6th.
 
Torque arm this time ?
Doing the mistake once is part of the learning, doing it twice is plain stupid.
 
One blown controller in approx 15 000 miles. That"s it. Oh yes, and a one broken thumb throttle.
36V controller that i run with 54V. Velomobile with DD and burnt out controller is one hell of a thing to pedal. I managed around 10mph on flats.
First ride of the year yesterday! "19 wheel works better than "21. It"s still pretty cold here, around +3-4C during the day.
 
On the subject of torque arms....

I thought the same, so here's my breakdown story: My motor is a 500W Q128c (geared, 260RPM@48V) rear hub on a 700c wheel with exactly 540W@48V controller input. Although I felt it didn't produce enough torque to spin out in the dropouts, I had a BMSB torque arm gear clamped to one side just in case.

I was just leaving work when I tried to out-accelerate a pickup truck from a stop sign, and my rear wheel came free of the dropouts while WOT and pedaling madly. I was lucky the single gear clamped torque arm prevented the motor from spinning freely and shearing the wires. The wheel came out of the rear dropouts nonetheless and caused me to do a stoppie at 15-20 kph and nearly crash - I even heard the driver of the pickup truck gasp wtf as my rear wheel raised 2 feet off the ground!!. I was able to reattach the wheel with my ever present toolkit and I very carefully rode home. Later that evening, I cut myself some custom fit 3/8" thick hardened steel torque arms and carefully DP420 epoxied them to the rear dropouts. It's well worth the peace of mind.
 
I suppose if it was from a high enough speed and weight was far enough forward, then during the stoppie the rear of the bike could lift up quite high, taking the rear wheel with it since it was still attached to the torque arm.
 
With those pics and it turning out to be a DD rather than geared, it makes a lot more sense for the failure. Also, I did not catch that the controller was in a bag, which probably caused it to overheat repeatedly during use, eventually causing a failure. Probably has cooked capacitors from heat, too, which would make any voltage spikes from the motor potentially pass right into the FETs and can cause direct failure of them.


I'd guess the controller itself blew first, from the caps or other heat-induced failure(s), and then locked up the wheel by shorting phases together...but it could have been the motor coming loose that blew the controller.



Two scenarios I've seen:

--during acceleration the axle spins out, twisting the wires and shorting two of them together (even if the insulation appears undamaged, during twisting the insulation can be pressed hard enough to split on two or more wires, two phases or even one phase and a hall, and cause a short--but when the wire is manually exposed and not under twisting/tension/etc., it then rebounds and appears undamaged until probed with a fingernail, knife, pin, etc.). Then the shorted wires blow a FET in the controller, which then continues to add drag to teh motor even if the wires don't remain twisted, by continuing the short started by the wires.

--during braking with regen, or coasting downhill faster than the motor's unloaded speed, if the controller fails to do what it should then it could cause a FET failure, which would then lockup the wheel due to the short, and high currents caused by the plug braking could melt wires inside an already-overheated controller.

Some other notes:
hypertoric_amplituhedron said:
The axle spun several times, which unthreaded the right nut, but kept the left one tight
SInce the nuts are threaded in opposite directions, that can be a normal occurence. But it can also happen that the rocking bakc and forth of acceleration/coasting (or regen) can actually loosen the nut(s), and then vibration can even completely undo it and cause it to fall off. Iv'e seen that happen even on a non-powered bike wheel.


The motor doesn't have that much torque to even remotely spin it.
It actually does--just a little on each acceleration, bending the dropouts open, loosening the nuts, so that it just comes right off and spins out once it reaches a critical point. All the motor's torque is put on the dropouts in the opposite direction as the normal rotation, and that's a very tiny area to put all that pressure.


If 80% throttle was being applied, and one of the phase wires burns up, would the motor actually lock up?
If the phase wire burned that badly, it is from a short either to battery positive or negative, either thru a FET on the board that blew, or from a motor wire that shorted (usually to another phase), or from another type of short that then (probably) blew a FET.

You can see the chernobyl-style Elephant's Foot that formed, when the solder was melting from the current.
It is possible, depending on what else was also touching the case, that the solder itself is the root cause. If the solder melted from controller heat (like if the controller was kept in a bag, instead of in open air, which will cause the controller to overheat, usually by a lot, repeatedly during use), and then the solder touched the case, if the case was already touching a trace or pad or wire on the board inside it (or the plastic FET screw insulators melted from the heat and allowed a FET to touch the case), then that short could blow a FET which would then lock the motor up and cause the physical damage, etc.




Actually ends up being a direct drive motor. This may be the one of the few DD motors that AmpedBikes sold.
I have one of the rear DD's from them, and it is by MXUS, essentially the same as the 9C 2807 (see ebikes.ca simulator for capabilities of that motor).


It also has a funny power band, with good responsive torque at low and high rpm, but a lull zone in the middle.
That probably has more to do with your controller and/or battery voltage than the motor itself. Check out the ebikes.ca simulator and try various configurations, see how well they do or don't match your system.



The axle banged on the ground, which cut through the newly fixed wires. Had to do the whole process all over again. Seriously, WTF man. Just a freak incident, when the universe doesn't want you to get off that easily.
Yeah, it works like that sometimes. I'd just finished rewiring all the lights and controls and whatnot on SB Cruiser, when a dumb mistake caused a fire in that new wiring, so I had to spend more hours redoing it all (and I was already cold and exhausted before I finished the *first* rewiring...but had to redo it anyway so I could work in the morning, IIRC).
 
flat tire said:

As near as I can tell..

When the rear wheel became disengaged, it partially locked up when it wedged against the seatstay and it pushed my weight forward since I was standing on the pedals. I had no idea what had happened and automatically hit the brakes hard. Since the rear caliper was no longer engaging the rear disk, it left only front brake to engage and the forces I applied resulted in a very nice stoppie. I think I'm used to feeling how fast I'm stopping and I adjust the brake pressure accordingly, and with only the front brake engaging plus my weight shifted forward, it's lucky I compensated fast enough by releasing the brake to avoid a flight over the handlebars.
 
Okay, just learned something. When I got the bike home, it started to have that locked up, hard to roll issue again. Now, this morning it doesn't. I found it happens when the phase wires are touching, which explains the failure. It was the controller that locked up the wheel with faulty phase wire signals.

1. The controller overheated from a combo of being in a frame bag, and being run for +4 hours continuously each night for 7 days in a row. I went through plenty of cold spots next to water, which could have helped, but not as well when insulated from the open air.

2. Red phase wire + others got burned up, from melting solder over time. It's hard to tell how long it took for the elephant's foot to form, or if it was the cause of the short. The controller was sitting at an angle, and looking at where it's located, maybe too much solder drained off the phase wire, and accumulated on the opposite side.

3. Bad/wrong phase signal caused the wheel to become a brake at 22mph, with 190 lbs of mass. The wheel locked with the axle, which made it spin CW looking from the right hand side. Torque washers were terribly insufficient, axle spun, which loosened the right, tightened the left.

Here's a better look at the root cause :

xNLXXWm.jpg


Had I known about the phase wires, I could have unplugged them, and pedaled the bike home. Now I know what to do next time :D
 
amberwolf said:
I have one of the rear DD's from them, and it is by MXUS, essentially the same as the 9C 2807 (see ebikes.ca simulator for capabilities of that motor).

Whoa, awesome. I wondered which motor it was, since I wanted to program it in the sim. Maybe you mean 9C 2707?

Here's my exact bike in the sim, true to empirical results as tested:

- 9C+ 2707
- 52v 21ah useable battery
- 22a controller
- 27.9" wheel height
- 0.7 CdA + 0.004 Cr frame drag
- 190lb vehicle weight
 
Boy, I bet that smells bad. Looks like maybe the shunt overheated and desoldered itself, then started arcing. At some point the FETs shorted, causing the motor to lock up and spin.

Thanks for sharing the pictures.
 
hypertoric_amplituhedron said:
Maybe you mean 9C 2707?
No, the 2807; you have to go to the bottom of the list and pick "show all" to see it, cuz it's ancient tech. ;)

I don't know what hte 2707 is like; havent' seen one, but it's probably very similar, just a 1mm narrower stator, at a guess.
 
Breakdowns happen to all of us. Don't panic. Always have a plan to get to point B. Worst case scenario you walk a few miles with the bike. Almost all of my breakdowns have been due to a flat tire, but I did have 1 controller box go out.

For me, I always have a backpack with a can of tire sealer, a poncho in case it rains, and a knife in case I get stranded in a ghetto area.
 
fechter said:
Boy, I bet that smells bad. Looks like maybe the shunt overheated and desoldered itself, then started arcing. At some point the FETs shorted, causing the motor to lock up and spin.

Thanks for sharing the pictures.

Yeah, I had it in my room, taking pictures of it, and set it aside. Not too long after, I realized it reeked pretty bad, and should be in the garage, or sealed in a plastic bag, or something.

Yep, people love pics.

amberwolf said:
No, the 2807; you have to go to the bottom of the list and pick "show all" to see it, cuz it's ancient tech.

I don't know what hte 2707 is like; havent' seen one, but it's probably very similar, just a 1mm narrower stator, at a guess.

Ah, found it. The '9C' is spelled out. It shows a little lower speed than the 2707, but pretty identical.

lester12483 said:
For me, I always have a backpack with a can of tire sealer, a poncho in case it rains, and a knife in case I get stranded in a ghetto area.

Lol, on that note, after the breakdown, I pushed it really hard about a quarter mile to the parkinglot, which was pitch black, but 500ft from a main roadway. Five minutes later, a car full of people pull up and park about 100ft away from me. It's 10pm. I have my three bright lights pointed towards them, but not at. It's a Stella Dual 150, and a Bontrager Ion 700.

They stayed there for about 15 minutes, getting out of the car, opening the trunk, searching around the car, doing weird shit. I felt very vulnerable with my broken down bike, in the dark, with a car full of people I don't know just hanging out. This is Orlando, after all.

I've had bad experiences in other places, too. It was in Redmond, WA, at the bottom of the NE 80th street hill by the elementary school, with a car full of people at night. Car pulls up, 5 dudes telling me to come over to them. I stayed put, 100ft away. Dudes jumped out of the car, and started yelling and running after me, on my pedal bike. I got away.

Then, one of them starts walking up to me. I'm holding my 700 lumen light in my hands at this point. If he tries something, he'll be completely blinded. Turns out to be a highschool kid, or barely out of HS, asking for a lighter. I would have been a lot more concerned if he was 25-ish. They were more intimidated with me, as some dude hanging out in the darkness with bright lights, and they're trying to find a spot to smoke weed. Lol.

I need to start a journal or something. It's an adventure every time I go out.
 
amberwolf said:
No, the 2807; you have to go to the bottom of the list and pick "show all" to see it, cuz it's ancient tech. ;)

This is the fifth year i run my MXUS 2807:) Stator says 9x7.
When my controller fried there was no increased wheel friction. It just died smokeless death without begging any attention.
 
Back
Top