Problem with Powabyke Commuter Controller

[Subspace]

Hi,

I am having trouble with a Powabyke Commuter which I think is a controller problem. A week ago it started cutting out randomly with no red LEDs. Sometimes turning the key on and off a few times would fix it but not so much any more. I put a multimeter across the red and black wires to the controller and it is around 3v when "dead" and 38v when "alive". However if I disconnect the controller from the power connector when "dead" the voltage at the connector goes back up to 38v. So something has too high a resistance and is causing a large voltage drop in the failure state. Any thoughts as to what it could be? Can't see any obvious signs of damage on the board. I am tempted to start replacing the transistors and diodes as there are not many of them.

James.

 

[Deleted member 4366]

Sounds like a faulty key-switch. Take the top off the battery and short out the two wires to the key-switch to see if the problem goes away. The problem is often the solder joints on the switch, so, if faulty, try resoldering them with a powerful soldering iron. You can buy the switches from ebay or Aliexpress.

 

[Danidl]

Hi,

I am having trouble with a Powabyke Commuter which I think is a controller problem. A week ago it started cutting out randomly with no red LEDs. Sometimes turning the key on and off a few times would fix it but not so much any more. I put a multimeter across the red and black wires to the controller and it is around 3v when "dead" and 38v when "alive". However if I disconnect the controller from the power connector when "dead" the voltage at the connector goes back up to 38v. So something has too high a resistance and is causing a large voltage drop in the failure state. Any thoughts as to what it could be? Can't see any obvious signs of damage on the board. I am tempted to start replacing the transistors and diodes as there are not many of them.

James.

Hi I do not know this controller , but I would certainly consider replacing components at random as unwise.
Are you sure you are actually dealing with the controller itself .?
The display and switches are not necessarily the controller . The electric sequence for an ebike is. Battery, keyswitch , if fitted , controller , motor.

The display , throttle if fitted and other switches are inputs and outputs to the controller.
From looking at older websites is this a 2007 vintage brushed DC motor mounted in the front hub.? If so it will have a more rudimentary controller than more recent models. Is it a model with lead acid battery and if so how old. There could be a value in checking the current capability of the battery. 10 meter s of thin hookup wire connected momentary across the battery It should get hot quickly do not use mains cable....
I would suspect that rather then the 3v being due to a a failure mode it is a signal level voltage perhaps for the leds or a relay and that a low power switch is not turning on the main power . I would look at any switches or relays in the power path which is not letting the current through .

 

[Danidl]

Sounds like a faulty key-switch. Take the top off the battery and short out the two wires to the key-switch to see if the problem goes away. The problem is often the solder joints on the switch, so, if faulty, try resoldering them with a powerful soldering iron. You can buy the switches from ebay or Aliexpress.

Agree

 

[Danidl]

Agree

Hi
If it is the keyswitch, you could just replace it with a 20A toggle switch from a Maplin . They have them in their car accessories aisle. Quicker than buying from abroad. That is what I did when my keyswitch locked in the on position.

 

[Subspace]

Ok, an update. I think the battery is to blame.

I took out the control board, soldered some wires onto the power pins and connected directly to the battery. The same problem as before, I measure 35 volts direct from the battery but when I connect the control board it drops to 3.7v.

Occasionally it works and the voltage stays at 35v. The battery is the lithium ion variety. I did check the current at the battery and sparked and blew the fuse in my multimeter. Is there a protection circuit in the battery that might be cutting out? I thought the powabyke lithium battery was pretty "dumb", just a bunch of 18650 cells welded together and a few fuses. I wonder if it has something to do with the cold weather as I noticed the battery is running down faster recently.

If I connect the control board to a spare SLA battery that I have it seems to work fine (I cannot use this battery in my bike as it is the "shopper" style).

I have put a video up to illustrate my problem.

 

[Danidl]

Ok, an update. I think the battery is to blame.

I took out the control board, soldered some wires onto the power pins and connected directly to the battery. The same problem as before, I measure 35 volts direct from the battery but when I connect the control board it drops to 3.7v.

Occasionally it works and the voltage stays at 35v. The battery is the lithium ion variety. I did check the current at the battery and sparked and blew the fuse in my multimeter. Is there a protection circuit in the battery that might be cutting out? I thought the powabyke lithium battery was pretty "dumb", just a bunch of 18650 cells welded together and a few fuses. I wonder if it has something to do with the cold weather as I noticed the battery is running down faster recently.

If I connect the control board to a spare SLA battery that I have it seems to work fine (I cannot use this battery in my bike as it is the "shopper" style).

I have put a video up to illustrate my problem.

Connecting a multimeter in current measuring mode across the pins of a 36 v battery was not the wisest action, hopefully all it needs is a replacement fuse.



I would like to eliminate the battery pack as a potential problem and means that you need to put a suitable load across it which can draw say 10 amp. E.g 3 to 4 ohm . Hence my suggestion of a suitable length of thin hookup wire E.g telephone or modem extension cable . The insulation will probably melt but that is not a problem. . Alternatively two car 55w 12v headlights bulbs wired in series. if these light up brighter than in a car and remain lighting for 10 15 minutes, then the problem is not the battery. if your multimeter has a 10a scale then you could measure it. The battery would need recharging afterwards.

Li ion battery packs are relatively simple they consist of a stack of 3.6v cells wired in series, with a fuse and should also have a battery management circuit board included. Schematic s for these are available and might be either a single if or multiple discrete components . It's role is to sample the voltage drop across each cell and to prevent the this voltage rising above 4.2 or below 3.2v these values will depend on the chemistry used . Bad things happen beyond these limits. This is a set of voltage comparators connected and switches .

It is possible that the 3.7v is an auxiliary voltage supply from the battery pack, say from the lowest cell to energise additional circuits elsewhere on the bike

 

[Danidl]

Have looked at your video again. Almost certain that you have a fault in one of the battery cells. Your circuit board is probablytrying to draw 20mA ... A few leds . If even one of the cells is discharged it could have an internal resistance up to 1 Mohm so that the overall voltage is dragged down ... Your multimeter has a resistance of 10 Mohm.
Note that when you use the SLA batteries, that there is a spark on contact. This is the current surge charging up the small electrolytic capacitor on the controller board.
Conclusion an uncharged or high resistance in at least one lion cell

 

[VictoryV]

Connecting a multimeter in current measuring mode across the pins of a 36 v battery was not the wisest action, hopefully all it needs is a replacement fuse.

When using a multimeter in "current mode" it is effectively a short circuit through the meter. It is always good practice that if you leave your leads connected to the multimeter after use that you ensure they are connected to the "volts" terminal. That way if you are in a hurry and want to check a batttery then you will never short circuit the battery. A lifetime ago when in RAF training we were being taught to use the "Avometer" or AVO for short. We had some Royal Malaysian airforce apprentices with us. One wanted to measure the "mains current" so he set the AVO to current and plugged it into the 250 volt mains. There was a very large flash and the tip of the connector probe burnt off, but the over load thermal protector tripped and protected the AVO from damage!

 

[Danidl]

Yep the model 7 AVO was some piece of kit. I knew of researchers using it as a programmable resettable fuse. Depending on the range, when ever the needle went beyond full scale. It would trip the cutout. ...

 

[Subspace]

Thanks for the advice. I let the battery drain down and recharged it and it seems to be behaving itself again. The battery must be around 8 years old so I wouldn't be surprised if it was on its last legs.

 

[Subspace]

There's no BMS inside the Powabyke battery, just several packs of 18650 cells wired together.

 

[Danidl]

There's no BMS inside the Powabyke battery, just several packs of 18650 cells wired together.

View attachment 17059[/QUOTE
I count 60 cells in blocks of 20 and a number of fuses . It looks as if there are spacers at each end to accommodate another 2 blocks of cells . Silly question but how sure are you that these are Li ion cells and not some other technology. It would be written on the outside of the battery box for instance. If they are then it is possible that there is a small BMS fitted within each of the blocks before it is sealed and the cells are configured as 10 series 2 parallel.

 

[Subspace]

The spacers are because the same case is also used to hold lead acid batteries. They are Sony US18650V cells aka "konion".

 

[Danidl]

The spacers are because the same case is also used to hold lead acid batteries. They are Sony US18650V cells aka "konion".

Then according to some websites viewed, they are a lithium class battery, of a highly respected type and may not have BMS fitted because the cells are so well matched. Apparently the factory manufacturing these was wiped out during the japenese tsunami.

 

[Deleted member 4366]

Did you check the key-switch, like I suggested a month ago?

 

[Subspace]

I finally found the problem, it was the external fuse holder. It was making poor contact with the fuse causing a voltage drop and charging issues. Fortunately it is a common part so I replaced it.