Recell my Battery?

[relkan]

Hi Everyone,

What a wonderful community!

I wonder if I could ask a question?

I bought a good quality used electric bike for my wife, an Axcess Exmoor, The bike itself is 16 months old and literally never been used however the battery was only charged up 16 months ago and left to deplete. It will not take a charge and cost of replacement has been quoted at £580.

It's a 36v 6ah bottom fed battery. I have opened it and it looks like 18650 battery cells.

I wonder if anyone knows whether it can be recelled, hopefully here in the UK and save us a bit of money as well as doing a bit for the environment?

 

[Fordulike]

http://pedelecs.co.uk/forum/threads/bga-reworking-for-batteries.27494/#post-370668

 

[Nealh]

The reason it won't charge is because the cell voltage is to low and the bms won't switch on to allow charging this is called the lvc (low voltage cutoff), the voltage will be under the minimum so needs to manually be raised above the lvc threshold. With the battery opened you need to find the JST block with sense wires that go to the cell groups, with a volt meter check the voltage via the sense wires. With low cell voltage sometimes it is possible to raise the voltage of each cell group to above the lvc and then the bms will allow charging to take place.
Take some clear pics and post them here, we can then guide you on the next step.

 

[Deleted member 4366]

I think I'd have a go at jumping the BMS. Get a battery charge connector from Maplins and wire on two crocodile clips, then connect the charger directlt to the main positive and negative wires from the cell-pack for abou half an hour. After that, there's a fair chance that it'll charge normally.

Alternatively, as you have the battery open, make a temporary bridge from the charge socket negative to the cell-pack negative.

 

[Danidl]

Hi Everyone,

What a wonderful community!

I wonder if I could ask a question?

I bought a good quality used electric bike for my wife, an Axcess Exmoor, The bike itself is 16 months old and literally never been used however the battery was only charged up 16 months ago and left to deplete. It will not take a charge and cost of replacement has been quoted at £580.

It's a 36v 6ah bottom fed battery. I have opened it and it looks like 18650 battery cells.

I wonder if anyone knows whether it can be recelled, hopefully here in the UK and save us a bit of money as well as doing a bit for the environment?

.. don't necessarily assume that the battery is dead. It might be dead or might be sleeping. , Since you have opened it up , you should be able to access the individual strings of battery cells. In the fully charged state they should be 4.1 V accross any cell and in the depleted state about 3.0v . With a little care you might tickle it awake. But according to the manufacturer, the battery is still in warranty.. so that is where I suggest you make your first inquiry.
If no satIsfaction then try the following and report.

What is the measured voltage accross the full pack. What is the voltage out from the supplied charger.
I note that there was a two year warranty on the battery pack have you consulted the manufacturers .

 

[Danidl]

I think I'd have a go at jumping the BMS. Get a battery charge connector from Maplins and wire on two crocodile clips, then connect the charger directlt to the main positive and negative wires from the cell-pack for abou half an hour. After that, there's a fair chance that it'll charge normally.

Alternatively, as you have the battery open, make a temporary bridge from the charge socket negative to the cell-pack negative.

.. great minds !!! Etc.

 

[relkan]

Thanks for the quick responses. I am attaching a couple of photo's hoping they can help, I suspect I may have to do some further dismantling though.

 

[relkan]

I do have one of these chargers with crocodile clips. I enclose a picture... but where do I attach them to bypass the BMS?

 

[relkan]

I also have three different e-bike chargers all capable of charging this battery, the one that came with it, my own battery charger and a spare so I know the problem is not with the charger.

 

[relkan]

The Battery is still under warranty but as it has been poorly maintained and I have now opened it, I assume that would invalidate any warranty.

 

[relkan]

Another picture with the cardboard removed

 

[Danidl]

I also have three different e-bike chargers all capable of charging this battery, the one that came with it, my own battery charger and a spare so I know the problem is not with the charger.

See the black lead in your picture and the number of parallel white leads all going towards some printed circuit board under the white mastic material. You could remove carefully a small piece of insulation from each of these white leads ..To expose the copper conductor... offset them so that there is no danger of accidentally shorting one to another . These are the feeds from each tier of the battery to the BMS and there should be a difference of about 3 to 3.6volts between each one. .. measure the voltages . Eg if measured from the black lead the voltages should be 3.6 7.2,10.8 etc they are probably, but may not in in the same row sequence . If one of these is decidedly lower than the others , you could inject a voltage between it and it neighbour of 3.6 volts .. say 3 ordinary torch batteries in series. It should not take a lot of current to bring it up to the 3.1v level at which state it should start accepting real current. . Best if you report back on the measurements obtained. Before that stage.

But maybe the best option, before doing this is to contact the manufacturer if it is within the two year warranty.

 

[relkan]

I count 30 cells in the battery pack and was told it was an upgraded 16 amp hour battery pack as opposed to the standard 10 amp hour. Not quite sure of the maths here, could 16 amp hours be created by 30 x 18650 cells?

 

[relkan]

See the black lead in your picture and the number of parallel white leads all going towards some printed circuit board under the white mastic material. You could remove carefully a small piece of insulation from each of these white leads ..To expose the copper conductor... offset them so that there is no danger of accidentally shorting one to another . These are the feeds from each tier of the battery to the BMS and there should be a difference of about 3 to 3.6volts between each one. .. measure the voltages . Eg if measured from the black lead the voltages should be 3.6 7.2,10.8 etc they are probably, but may not in in the same row sequence . If one of these is decidedly lower than the others , you could inject a voltage between it and it neighbour of 3.6 volts .. say 3 ordinary torch batteries in series. It should not take a lot of current to bring it up to the 3.1v level at which state it should start accepting real current. . Best if you report back on the measurements obtained. Before that stage.

But maybe the best option, before doing this is to contact the manufacturer if it is within the two year warranty.

So I will cut/scrape a little piece of insulation of each cable (offset) and using my voltmeter take a reading from right to left, they should increment by approx 3.6v. I should then be able to identify which group/cable has the low voltage. Once identified you can help me understand how to charge it, can i use my li-on universal battery charger and if so where do i touch the crocodile clips on to (the low voltage cable and it's neighbour)? I take it i can just use a tiny piece of electrical tape to cover the exposed copper?

 

[Danidl]

I count 30 cells in the battery pack and was told it was an upgraded 16 amp hour battery pack as opposed to the standard 10 amp hour. Not quite sure of the maths here, could 16 amp hours be created by 30 x 18650 cells?

... If these are standard li ion cells then there needs to be 10 tiers at nominal 3.6v per tier to get the overall 36 to 41 volts of a fully charged 36v pack. If you have 3 cells in parallel in each tier , then they would have needed to be 5 amperes hour cells. I don't think they were available at that time

 

[Danidl]

So I will cut/scrape a little piece of insulation of each cable (offset) and using my voltmeter take a reading from right to left, they should increment by approx 3.6v. I should then be able to identify which group/cable has the low voltage. Once identified you can help me understand how to charge it, can i use my li-on universal battery charger and if so where do i touch the crocodile clips on to (the low voltage cable and it's neighbour)? I take it i can just use a tiny piece of electrical tape to cover the exposed copper?

Yes that's the idea

 

[Hugh]

I can certainly recommend Insat (see post with link above). I posted a similar query a few weeks ago, and they were suggested. Did a great job at a very good price, upping my old 8ah Bosch battery to 14ah in the process.

 

[Nealh]

I just don't see the need to scrape or exposed the wire under any protection them have to bodge a repair 10X to prevent any shorts.
Easy simple option is to find the jst connector or what ever it may be that connects the sense wiresand disconnect it, you then simply apply your probes to the individual metal pin/lugs to get the accumulative cell counts.

 

[Deleted member 4366]

There's no need to cut anything. The BMS (circuit board) is nothing more than a switch. It switches the black wires. Sometimes they switch the charge and discharge together, and sometimes seperately. The switching depends on what they sense from the cells.

Set your charger to 2 amps and 42v. Connect the red crocodile to anywhere where you can get it on a red wire. All reds are common. Connect the black crocodile to the place where the thick black wire from the cell-pack joins to the circuit board or where it joins to the cell-pack - whichever you can get at. That's where it's marked 0v or B- in this schematic:

 

[Danidl]

I just don't see the need to scrape or exposed the wire under any protection them have to bodge a repair 10X to prevent any shorts.
Easy simple option is to find the jst connector or what ever it may be that connects the sense wiresand disconnect it, you then simply apply your probes to the individual metal pin/lugs to get the accumulative cell counts.

... If you look at his pictures you will see that the BMS is covered in a white jelly.. i would assume this is a silicone sealant. Rather than asking him to scrape this off, removing a bit of Insulation seems easier to me. It is also easier to attach a croc clip to a piece of wire than to a crowded PCB where the prospect of as short is greater.
Should the repair be successful, I would then have suggested an additional smear of silicone over the exposed copper. Removing insulation with either a small craft knife, an old fashioned Razor blade or a Stanley knife, and leaving the copper intact, is straight forward.