Bottle Battery Repair.

If you've got a faint heart, don't read this thread because it has graphic pictures of major surgery.

 

I was given a bottle battery to have a look at, which wouldn''t charge. I checked the voltage on both the charge socket and the main connector, which both showed about 36v.

Charger showed 41.7v -- time to open it up.

 

Opening is easy enough: Undo the screws in the bottom and top; remove the LED board from the top, and slide everything downwards and you then have the contents to look at:

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/bb1_zps74ac6d4b.jpg

 

This battery has 40 cells on a 10S4P configuration. Lets say 10 strings of four cells.

 

I pulled the plug on the BMS and checked the string voltages. This battery only had 10 wires to the BMS connector, which means either the red or the black main wire is the 11th. You need 11 wires to measure 10 strings. In this case, the 11th was the black wire. All strings were OK except number six, which showed 0v. This means four individual cells are scrap.

 

I followed the balance wire down, and it led to the middle of the pack somewhere. Oh well, time to ripit apart. It's scrap as it is, so nothing to lose.

 

 

I had to unsolder several wires because they're put on at the end, and hold it together. Then I was able to break out the middle section to see the problem. There was one cell showing signs of leakage. I had some spare cells, but the ones in this battery are welded together after being assembled into the plastic holders. The only way to get them out is to cut all the welded strips, but you can't get cutters on them because the plastic holder prevents you from getting cutters on them. I therefore had to use my trusty Dremmel to remove plastic at every cutting point. Here it is all cut and ready to pull apart. You have to do both ends. It's a lot of work:

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/bb2_zps4f56b507.jpg

I was then able to remove the four offending cells and put in four new ones.

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/bb3_zpsa906fc5f.jpg

 

Next, I had to solder some wire accross all the strips I cut. It looks messy, but all the joints\are good and strong:

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/bb4_zpsa9da7bf7.jpg

Finally I rejoined all the balance and power wires. It took a bit of figuring out which one went where, but the lengths were a good clue. Finally I checked again at the balance connector, all at about 4.1v and 41v total. I plugged in the connector and then the charger, and yippee, it's charging. Hopefully some happy chappy will be riding his bike for a lot longer before major expense. I hope this thread gives the confidence for others to have a go, and hopefully it''ll give an insight into the construction of these batteries.

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/bb5_zps422bc583.jpg

Great post Dave, how do the balance wires connect, is there a pair per 4 cell and one on pos and one neg..?

 

Thanks

Great post Dave, how do the balance wires connect, is there a pair per 4 cell and one on pos and one neg..?

 

Thanks

There's one for each string of four cells that goes from the top of the cells and the BMS. The BMS is monitoring the difference in voltage between each wire plus the difference between the first one and ground, like this:

 

http://i451.photobucket.com/albums/qq236/d8veh/batteries/battery_zps4f05b9b9.jpg

Thanks, understood. Would much prefer bottle battery, good to see they have a chance of being repaired.

nice work dave!! I'll follow this to work on my lipos

Not one for the faint hearted or the health and safety conscious, but well done Dave.

 

One thing I don't quite understand is what role did the tomato puree have to play:confused:

Hi and a good morning to all

 

well done dave Amazing what you can do with a swiss army penknife and a tube off tomato puree

 

 

Frank

You need the tomato puree for when you cut yourself on the sharp strips, which are like knife blades. You go to your wife/girlfriend/mate/customer for a bit of sympathy, but all you get is, "Its only a small cut", so you go back, smear on copious amounts of the puree, and try again. The effect is amazing, and it's only 64p in Aldi - worth every penny.

Dave you were robbed, tom puree is only 35p in my local Aldi.

 

Nice job on the battery though; I know what you mean about the nickel strips, very sharp edges and easy to cut yourself. The bloody sympathy card is always a good one to play.

 

Chris

You need the tomato puree for when you cut yourself on the sharp strips, which are like knife blades. You go to your wife/girlfriend/mate/customer for a bit of sympathy, but all you get is, "Its only a small cut", so you go back, smear on copious amounts of the puree, and try again. The effect is amazing, and it's only 64p in Aldi - worth every penny.

 

Never get any sympathy from my wife, she used to be a nurse:(

Hi John

 

My wife never was a nurse But she still has the uniform

Since you can buy all the individual parts necessary to make one of these, has anyone attempted it?: -

 

http://www.bmsbattery.com/aluminium-alloy/620-bottlle-battery-case-for-cylindrical-cell.html

 

http://www.bmsbattery.com/battery-case/621-18650-cells-protection-cover.html

 

http://www.bmsbattery.com/bmspcm/391-10s-lithium-ion-bms-for-bottle-battery-case.htmlhttp://www.bmsbattery.com/bmspcm/391-10s-lithium-ion-bms-for-bottle-battery-case.html

 

etc.

You still need a spot-welder to join the cells. Soldering is alright on a battery that's already given most of its life, but t's a bit risky n new cells. Also, you'll probably pay as much for the cells as the whole ready-made battery. If you have a load of decent re-cycled cells, it could be a worthwhile exercise.

Great thread d8veh, was especially interested in your Battery and BMS layout earlier in the thread. Can I ask a quick newbie question please?

 

My Wisper 905se battery (36v 14ah lithium polymer) will charge though runs out of juice within a few hundred meters triggering the under voltage cut-off having dropped to 17v (from 41v fully charged). I opened up the battery case to see if there was an obvious issue...nothing obvious to be seen though am struggling to locate the BMS unit.

 

There are 11 black wires going into a connector which is taped over which I take to be the balance wires? Though there is nothing connected to this. The battery is still it its blue shrink wrapping, could the BMS be under this?

 

Would appreciate your input please.

 

Oh and Hi everyone...long time reader of this excellent forum, now joined and hopefully able to contribute a little over time:)

Good thread by d8veh here. There must be lots of packs like this which only need a few cells replacing.

 

My Wisper 905se battery (36v 14ah lithium polymer) will charge though runs out of juice within a few hundred meters triggering the under voltage cut-off having dropped to 17v (from 41v fully charged).

 

There are 11 black wires going into a connector which is taped over which I take to be the balance wires? Though there is nothing connected to this. The battery is still it its blue shrink wrapping, could the BMS be under this?

Some batteries have two connectors on the balance lead for the two different sized connections to the BMS, so you see one sticking out unconnected. The extra one might also be there for testing and balancing externally without having to rip the pack apart, so You might be able to measure the cell voltages through that spare connector without any disassembly. The BMS will be under the shrink wrap normally at the top. If you cut the four corners where it overlaps the top, and then carefully lever off the top (normally a piece of GRP sheet, the BMS is underneath. Assuming it's a standard silver (black) fish type of battery, you need to slide the pack up the casing too access it. You remove the bottom piece, turn it sideways on, and use it to push the pack up the case a couple of inches.

 

Once you have access to the multi-pin connector you can measure the cell voltages. If there are 11 pins, that's all the wires you need to measure the 10 cells. If there's 10 wires, the 11th will be either the battery negative or positive. Once the BMS has cut off, the negative will be disconnected, so you have to access the cell-pack negative (pre-BMS) to measure the 10th cell if there's only 10 pins on the connector.

 

The 17v you measured is residual charge. The BMS has shut off, so you can't draw any current from it. Another time it might show a different voltage, or even one going down.

 

Measure the cell voltages after the BMS has shut off, so that you can see which one/s is/are causing the problem.

Very appreciative of your kind input on my battery issues d8veh, cheers:) Had recharged the battery so will nip out on the bike tomorrow weather permitting and as you suggest, let the BMS cut off the power then open the casing and measure the voltage on what looks to be the spare 2nd set of easily accessible balancing wires.

 

Point noted on accessing the cell pack negative pre-BMS once a cut-off has occurred, thanks for the heads-up on that and the other kind tips to make this task more straightforward. Will peel some of the battery wrapping to gain better access and see if I can locate the BMS. I will need to do this anyways if I cannot get voltages from the loose 11 wire connector.

 

Actually, thinking on, it might just be worth removing the battery pack from the casing as currently it is unusable unless it can be repaired as this will give easy access to the cells and BMS. Once I can ascertain how many cells are not performing can make a value judgement on if the pack is saveable. Be nice if it was repairable as dislike the 'it doesn't work, throw it away' way of going on. Having said that, making electrical items very hard to repair/service isn't helping much.

 

Will report back how I get on, thanks once again.

First thing is to get the cell voltages. They will determine what the next steps are.

First thing is to get the cell voltages. They will determine what the next steps are.

 

Agreed as am a little perplexed. Unlike the bottle battery you were working on which showed an under-voltage after charging this one seems to be ok voltage wise showing 41v after a charge, leave the battery a couple of days and it remains fully charged as in it will not take anymore charging from the charger unit which within a few seconds shows a green 'fully charged' light.

 

Though go out for a ride on the bike and in under a mile there has been a low voltage cut-out and its the full leg power only cycle home.

 

Will try get out on the bike today and let the battery cut-out, then come back and open the case and measure the individual voltages via the balance wires and let you know what that throws up. Am curious to see what the results will be.

 

Cheers:)

I had one battery that was totally worn out, Everything looked good after charging. Each cell was at about 4.1v and they stayed like that, but as soon as I tried to draw any current, all the cells sagged, and two went right down to LVC within a couple of minutes to cause a cut-out. When I stopped it, they all bounced back to over 40v. I had it on a test rig, so I was then able to check the voltages while it was going, which showed the problem. If you get the same, it would be worth temporarily wiring your voltmeter so that you can see how the overall voltage changes while riding.

I had one battery that was totally worn out, Everything looked good after charging. Each cell was at about 4.1v and they stayed like that, but as soon as I tried to draw any current, all the cells sagged, and two went right down to LVC within a couple of minutes to cause a cut-out. When I stopped it, they all bounced back to over 40v. I had it on a test rig, so I was then able to check the voltages while it was going, which showed the problem. If you get the same, it would be worth temporarily wiring your voltmeter so that you can see how the overall voltage changes while riding.

 

Think your 'crystal ball' on this maybe being just a totally worn out battery is going to prove spot on, on this one. I have managed to test voltages, just need to write up the results and post them here which will do asap.

Ok:) went for a ride on the bike, turned on the battery, used pedal assist for 100m then wound on the throttle, within 2 seconds the low voltage disconnects the battery, cycle home and open battery pack.

 

Check the underside of the spare balance wire connector i mentioned before, 10x black wires 1x red wire. Multimeter red lead onto the red wire in the connector then individually the multimeter black lead onto onto each of the 10 black wires.

 

Starting with the black wire next to the red and working away from the red, readings as follows:

 

1- 4.17

2- 8.34 (4.17 difference)

3- 12.52 (4.18)

4- 16.69 (4.17)

5- 20.3 (3.61)

6- 24.3 (4.00)

7- 28.0 (3.70)

8- 32.2 (4.20)

9- 36.4 (4.20)

10- 40.6 (4.20)

 

Battery charges to 41.1v maximum.

 

Interesting that the readout was cumulative. Also, reading 5 and beyond required the multimeter to be upped from 20v to 200v so lost the accuracy of the additional decimal place sadly.

 

D8veh, the BMS was where you indicated it would be at the top of the battery pack, thanks for the hint. Didn't unpack the BMS or battery pack any further as was getting the above voltage data without delving deeper.

 

Would seem there are a couple of under voltage cells though as the battery hits low voltage cut-off within a few seconds of any 'draw' being put on it, would it be fair to say this battery has just done its days? I'm not the first owner and know the bike has been well used and covered a lot of miles.

It's a bit difficult to say without a test rig and a single-cell charger. I have that stuff, so I'd charge the lowest two cells back up to 4.2v, then connect a voltmeter to cell 5 to see what happens when I draw some current from the battery. With your extra connector, it's pretty easy to do the latter. you can charge the cells with a phone or USB charger, but you have to continuously monitor because you mustn't go over 4,2v. They'll take it all the way to 5v if you let them. If the cell is healthy, it'll take all day and night to charge up like that.

 

I use this stuff for connecting to the multi-pin socket:

http://www.ebay.co.uk/itm/Pin-Header-36-Way-SIL-Straight-2-54-Gold-0-1-x-20pcs-of-36pin-strips-720-pins-/231136233590?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item35d0cb0876

 

Opening up the cell-pack will give some idea of the chance of restoring the battery. If it's 18650 cells (cylindrical, there's a good chance. it might be one or two cells have leaked, Pouch cells are difficult to find and replace. If it has the thick prismatic Phylion cells, you can chuck them, and just keep the BMS and case.