E-bike power cuts out under load

[Glen Hendry]

I am having trouble with my e-bike. The power completely cuts out under load. The LCD goes blank and a key reset is required to get going again. The cut out happens very fast, after only about 15 seconds. The bike has been going well for about a year, and after reading these forums I made sure that I bought higher quality specced equipment. It is a 26" MTB based e-bike with a 1000W rear wheel kit bought from Dillenger electric bike in Australia. I have recently replaced the controller and the LCD screen.

Specs: 30Amp - 48V - 12 MOSFET Upgraded Wiring Controller, Brushless 1000W motor, Li-ion 10Ah 48V (upgraded wiring and LCD).

I looked inside the motor, and all looked fine. Smelled normal. When it cuts out there is no heat on the motor or controller or battery. I plugged in my spare controller and the same behaviour. Motor spins fine indefinitely off the ground. I examined all the connectors and plugs and connectors and all seem strong and secure and of good quality workmanship. All the posts about cutting out pointed to the battery being out of balance, and so I charged it for 4 days and the same thing happens. Photos of charger attached. The "charged" green light comes on and then stays on, I am not sure about how good the charger (and BMS) is at balancing the pack?

I pulled the battery apart last night hoping to find something obvious amiss, but all the connections and solder looked good. Smelled good. The key mechanism looked good, the wires were solidly connected. I looked into the BMS and it looked fine too. Images attached. I measured the voltage of the whole pack after charging for 4 days, 54.2v - is this too high for a 48v battery?

I couldn't get to the individual cells, but I measured across all the pins on the BMS (must be the cell groups?): 0v, 4.16, 8.35, 12.52, 16.70, 20.8, 25.0, 29.2, 33.3, 37.5, 41.6, 45.8, 50.0, 54.2. This means the groups? are (I think): 4.16, 4.19, 4.17, 4.18, 4.1, 4.2, 4.2, 4.1, 4.2, 4.1, 4.2, 4.2, 4.2. Charger says output +48v 2A.

<please note that I have posted this same Q in another forum, but I was not able to post images, thanks for any input>

 

[Woosh]

check your Hall sensors.

 

[JuicyBike]

You're right to suspect the battery but there's little value to checking individual cell voltages - they will balance up naturally in very little time when there's no drain.

I think you might want to double check the battery switch. Perhaps bypass that temporarily.

 

[Nealh]

Battery max charge is 54.6v, 48v is the nominal battery of approx. 3.7v per cell.
Voltage though of cells 5, 8 &10 are out of balance so bms is either not balancing correctly or the cells have lost charge.
Most likely cells can't deliver the amp draw for the 28/30a rated controller and some cells are seeing a voltage crash as the amp demand is too high, battery is only 10ah and under spec for the controller now it is a year old.

For 30a controller draw you will need very good 10/15/20a rate dcells and not cheap mediocre cells.
A pic of the cells and ceil make used.

 

[Nealh]

Cell voltages only tell us the state of charge for the parallel groups, unequal voltages can be under performing bms or weak cells groups.

If bike has a throttle check each cell group voltage drop with the wheel off the ground, if necessary apply the brake gently to induce more load/amp draw but not to stall the wheel.
If not you will have to rotate the cranks to simulate the drive, you may need a hand unless you have a bike stand to help you.

 

[Nealh]

Under load if any one cell group collapses down to below 3.3v/3.4v then you will have your answer of a knackered battery or one that isn't able to sustain high amp draw.

 

[Nealh]

Replacing parts willy nilly can work out expensive as ideally you need to know the reason & cause. A controller and lcd not needed is a waste of £50 +.
Having ruled out those two, battery is the next step for fault finding if you are able to carry out the basic voltage/load checks to confirm what might be happening.

Fault finding /checking can be tedious and like finding a needle in a haystack but is the only way to find a solution without haplessly fumbling around in the dark.

 

[Nealh]

Also no use trying charge the battery any more as the charge won't switch on because one or more of the cell groups are at max voltage.
Batt will only charge and balance once all voltages have been discharged some, then you can charge up and leave it on a for a hr or two after the green light has come on to see if it balances better.

 

[Ajax]

check your Hall sensors.

Also speaking from experiance, I found a combination of dodgy power connections and faulty brake connections will also give the appearance of cutting out randomly under load.

So yeah, i would try the set up with the break connects removed from the
controller (don't ride the bike with no breaks though, just use the throttle to test the bike under load).

 

[Woosh]

random cutouts are difficult to diagnose, especially when you don't have another similar e-bike or kit at hand to borrow stuff that you can swap out, made worse by the fact that the LCD went off.
I would have swapped out the LCD to rule it out first if I could, then the battery, then the controller and last, the motor.
Without another kit, the Hall sensors are easiest to test and rule out, after that, rig up a discharge test to check the battery.

 

[Nealh]

Faulty brake lever/signal won't cause power to shut the lcd off, all they do is prevent the motor working until remedied.
Any issue with lcd turning off is a power supply issue either the controller or battery.
A continuity check on the wiring from controller to lcd would rule out possibilty of a intermittent wire break.

 

[KirstinS]

Just my tuppence but “most” of the similar issues I have diagnosed and repaired have been bad connection somewhere. Usually at

1. point battery connects to cradle (or the wires that attach to the cradle connector)

2. where a weld comes undo inside the battery itself. The pack works when the weld is pressed into contact , hit a bump and power goes. Hit another comes back. Also weak connection can causes switch off under load. So works on the bike stand and not in real use

3. other one has been phase wire bad connection giving same result (either where the bullet style connectors had wiggled nearly apart or where bending had frayed most of the wire at point of solder to phase connector




Sent from my iPhone using Tapatalk

 

[KirstinS]

Faulty brake lever/signal won't cause power to shut the lcd off, all they do is prevent the motor working until remedied.
Any issue with lcd turning off is a power supply issue either the controller or battery.
A continuity check on the wiring from controller to lcd would rule out possibilty of a intermittent wire break.

I would only add that on two cheapo dd conversions I sorted , one had lost a magnet on the hydro brake sensor. What happened was the system started and you could pedal off. Within 15 seconds an error message showed . About 30s later the system shut down. Crappy programming in my view but it it is possible to have a brake sensor issue shut the system down (albeit unlikely and with an advance error warning )


Sent from my iPhone using Tapatalk

 

[Glen Hendry]

This video shows the behaviour. It cuts out at 1m20s and 2m20s into the video. Not that the voltage doesnt drop noticably before the cutout. Then I notice something funny, it still shows voltage at the battery terminals after key is turned off. Must be voltage coming back from capacitors in the controller??? Or key mechanism problem?

https://photos.app.goo.gl/Amw3Tx8BuyUjEJkGA

 

[Nealh]

The voltage does collapse dramatically eventually for the whole battery, did you try increasing the battery draw by gently applying the brakes ?

If it is a cell group then voltage will bounce back when it has recovered, you need to do the same test but for each of the 13 cell groups to see which/if any single group collapses. You will only see 4 - 4.2v reading, any poor cell group/s that collapse will indicate weak cells and the issue of your problem.

 

[KirstinS]

Having seen the video I think Nealh is going to be right. Your battery is the likely issue (weak cell/s)

If you can locate another 48v battery on loan you could prove this fairly quickly .


Sent from my iPhone using Tapatalk

 

[Nealh]

As we often see posters never include their location so it's difficult for any one to even offer help locally.
It's not that hard to put your county/town or first four digits of a post code in the settings.