Motor growl problem

E-HP

10 GW
Joined
Nov 1, 2018
Messages
6,201
Location
USA
Problem: I hopped on my bike this afternoon, pulled out of my driveway, and after about 20 feet, the motor made a loud growl and shuttered. A little further and it seemed to subside. I went up the hill and things seemed fine then as I got to a level spot in the road, it started again. Fortunately, it was downhill all the way to my house. I was bummed, since I got off work early today, and was eager to go on a ride.

Background: I rode my bike pretty hard on the weekend, but everything was working perfectly after the ride. I set aside Sunday afternoon to mount my new knobby tires and convert the wheels to go tubeless. Of course I had to uncouple the motor connections, but I also took the opportunity to lengthen some of the wires from the controller (throttle, PAS) since they were a little too short to work with. I also added Andersen connectors to the motor phase wires to make tire changes easier (instead of the insulated block that uses screws/nuts). Afterwards, with everything back together, I test rode the bike for about a 10 block loop (uphill for the first half, down the second), and everything worked fine (tubeless conversion held). Oh, and for this short loop, I also tested out increasing regen (C13=2), increasing it one level.

Observations/diagnosis: First thing I did was set C13 back to 1 and 0, with no effect. Tried running with PAS level 0 (throttle only), but with the same results. Lifting the rear wheel and apply full throttle, no movement; but doing the same and barely pushing the wheel, maybe a quarter inch, and the wheel spins up to full speed, although with a moderately loud hum (usually it's near silent with the sine wave controller). Any load, and the motor shutters and growls. I unplugged the hall sensor wires and plugged them back in, in case there was a loose connection, but same result. I also tried a few of the motor phase settings (C2) which had no improvement. The growl is there whether the motor is activated by throttle or PAS.

I need some hints on where to start on this. The odd part is that everything was working great a few days ago. Any help would be greatly appreciated. Seems like something related to the hall sensors, but I'm hoping it's not the sensors themselves.

Tech Specs:

Motor:
26” Rear Hub - MXUS model: XF 40 1000W 1610 0020
48V, 1000W

Controller:
Sine Wave - RisunMotor model: KT36/48SVPRK-SLSO2G
Maximum current: 45±1A
Rated voltage: DC36V/48V
Rated current: 22A
18 MOSFET

Display:
KT-LCD3

Battery:
Li-Ion – Unit Power Pack - 52V, 20Ah, 30A BMS
 
The most likely problem is a connection or wire failure in phase or halls, especially since the wires and connectors were handled and moved and modified.

It's unlikley to be actual sensors, because it's intermittent.

The first most likely connection would be halls, but if you have already verified they are good connections, with no contacts backing out when plugged together (you won't always see that when they're apart), and no bad crimps or solder joints, or wires breaking off at the back of the contact, etc., then phase is next.

Since you are using andersons, make sure that:

--Wires have clearance within the housing (insulation is not in contact with the plastic housings. (if there is no clearance, contacts cannot float, and may not fully seat against each other, causing poor contact, high resistance, poor performance, etc.).

--Wires are routed straight and loosely into the backs of the housings, not twisted or curved against them (will also prevent contact float). And also not tied down close to the housings, for the same reason. (the housings can be tied down, but not the wires as they exit from them)

--Contacts are properly crimped to the wires.


Then check to be certain no wires are damaged in any way at the motor wire exit point. Even a tiny nick in insulation in the wrong spot can allow a short to occur to a wire next to it with a similar nick, or to the motor axle, or housing, or the bike frame, etc.


Then check for continuity on all wires (hall and phase) from the connector contacts all the way to the inside of the motor itself, and from teh connector contacts to the controller PCB, while moving the wires all along their lengths. It is always possible for a conductor to break inside the insulation in a way you won't normally see visually, and it can make intermittent contact, or high resistance contact, causing exactly the results you see.

I recommend it as the last step of the above because it's a PITA and a lot of work to take things apart, and has the risk of actually causing as much of a problem as it can reveal. :/
 
amberwolf said:
The most likely problem is a connection or wire failure in phase or halls, especially since the wires and connectors were handled and moved and modified.

It's unlikley to be actual sensors, because it's intermittent.

Thanks a LOT. When I checked the hall connector, I just disconnected it, wiggled things around and reconnected. I didn't check for pins backing out and stuff. I didn't change the hall wiring, but I did split them off from the phase wires to make it easier to connect and disconnect. I'm wonder if I nicked one of the wires when removing that portion of the housing. I'll check the connectors first, since I'll need to remove the heat shrink tubing to check the insulation.

Also, I didn't know that about the Andersen connectors, about making sure the wires are floating. I didn't crimp the wires, but soldered them in place. I recall on a couple of them, I had built up too much solder and it was a tight squeeze into the connector housing, so not floating for sure. This should be easy to check and fix, if they are the cause.

I'm crossing my fingers that one or both of these things are the problem, rather than having the pull the wheel cover off. I was also worried that it could be controller related, but even when I pushed it hard, it never heated up, so I ruled that out for now.

Thanks! I'll report back.
 
I reckon you've blown some fets. Open the end of the controller and have a smell/look. A friends (kt cont) did exactly this after turning up his regen. Very suspicious! I don't think kt have figured out how to control the Regen currents precisely-ive heard of a few dying after Regen related events. Other than that I've found them to be great cheap controllers.
 
kdog said:
I reckon you've blown some fets. Open the end of the controller and have a smell/look. A friends (kt cont) did exactly this after turning up his regen. Very suspicious! I don't think kt have figured out how to control the Regen currents precisely-ive heard of a few dying after Regen related events. Other than that I've found them to be great cheap controllers.

Ugh, I worried about that. I posted on another topic about about a motor braking issue. And, it did occur after a long ride, and felt like the motor was braking for a while before the PAS would kick in. Regen was enabled at the lowest setting on that ride. Maybe increasing the setting fried something.

I guess if the wiring all checks out, then opening the controller would be the next easiest thing, before cracking open the motor.

Thanks!
 
amberwolf said:
The most likely problem is a connection or wire failure in phase or halls, especially since the wires and connectors were handled and moved and modified.

:bigthumb: You were correct sir!! It was the Andersen connectors that I used to connect the phase wires from the controller to the motor. I ganged them together, three across, so it felt like everything was snapping together correctly, but the yellow phase wire had an intermittent connection, because it wasn't floating.

I thought it was the halls, because I got the motor to start when I jiggled the cable, so I redid them, but same problem. So, I jiggled the phase wires individually, and got the motor to start when I pushed the yellow wire to the side a little. So it looks like a little soldering tomorrow morning, and I'll be back in business. :bolt:

Thanks again for the tip. I was fearing a lot worse. :mrgreen:
 
Glad it was that easy. :)

Andersons are great and I like them for higher current connections, but they do have their limitations.
 
All back together and running great. I decided to just go with crimp on barrel connectors since that makes it easy to pull through the opening of the bag I'm using for all of the connections. Makes it easier to diagnose too. I took the opportunity to lengthen the phase wires coming out the controller, since those were ridiculously short. Should be easy to disconnect when I need to service the wheel/tire. Thanks a bunch!
 
For others, intermittent was the key here. That is almost always wires or plugs.

halls fail, they stay failed, same with controller fets.


Andersons are great, but fussy about wire size, and alignment of the wire on the contact. Best to crimp andersons. 4 mm bullets next time you have a problem, or rebuild.
 
Back
Top