BMS Li-Ion Battery died < 2.5 years usage...advice?

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
Hi all,

Hope you're all making the most of the sunny weather!

A couple of weeks ago I decided to get my Mezzo out and have a last ride on it, the day before my knee surgery as I knew I wouldn't be able to ride it for weeks or months after..

Got all ready, took the bike down the stairs, only to find the battery dead! Rather odd, as although I haven't been using the bike a lot due to my knee problem, I've keep the battery charged up regularly few 2-3 months or so.

I checked all the wiring over and everything seems fine (inside the battery too). I even got a VERY old 5Ah LiFePo4 battery I found and that worked, but only spun the wheel a little and then conked out as that battery hasn't been looked after or charged for years.

So, it looks like the BMS Battery has just died mysteriously.. surprisingly because there was no sign it was running down or showing any signs of wear when I last used it.

So my questions are:

1) Is it normal for Li-Ion batteries to just suddenly die like this?
2) Has anyone else had problems with BMS Li-Lion batteries?

I'm reluctant to buy another BMS Battery now after this, and wondering what other options I can explore? Anyone have any ideas or suggestions? Something small, light and cheap would be great, 10Ah as a minimum I think.

Thanks,

Paul

P.S. It's been 11 days since I had my knee surgery and I'm on crutches still at moment so will be at least 4-6 weeks before I can ride again. This disease I have is a nasty one with a high risk of it returning (50% chance) and when it does come back it often needs surgery and radiotherapy every year to halt it, else it destroys the joint and bones requiring bone drafts or joint replacement.. it's a bit like cancer, but very rare, affecting only 1 in 9 million people!
 
Last edited:
D

Deleted member 4366

Guest
It's not dead, but asleep. You didn't say which particular battery it is. It would help to know. Open it up, unplug the multi-pin balance lead connector, You might as well measure the cell voltages while you're there. Start at one end and check between each adjacent pair until you have 10 results, or 12 if it's LiFePO4. Plug the connector back in and it should spring back to life provided no cell is below the minimum. If you get only 9 or 11 results, the last cell will be between the red output wire and the highest on the connector, or between the black wire into the BMS and the lowest on the connector. It's always cell 1 that is the lowest, so make sure you get all results.
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
It's not dead, but asleep. You didn't say which particular battery it is. It would help to know. Open it up, unplug the multi-pin balance lead connector, You might as well measure the cell voltages while you're there. Start at one end and check between each adjacent pair until you have 10 results, or 12 if it's LiFePO4. Plug the connector back in and it should spring back to life provided no cell is below the minimum. If you get only 9 or 11 results, the last cell will be between the red output wire and the highest on the connector, or between the black wire into the BMS and the lowest on the connector. It's always cell 1 that is the lowest, so make sure you get all results.
What makes you think it's "gone to sleep?"?

At the moment I have no multi-meter so can't check the cell voltages, and it might be a bit beyond my ability as the battery is pretty well covered up (the wiring) and I'm liable to either short/damage it, or electrocute myself (or both, knowing my luck!).

I've removed the shrink wrap cover and some of the rubber sheets that cover the cells, and found the connector I think you're referring to. I unconnected it and reconnected it, but it may no difference.

I've taken some pictures of the battery which may shed some more light on this. It seems a pity the battery has stopped working suddenly after maybe less than 2 years usage.

Does anyone have any suggestions for alternatives? I'm reluctant to buy BMS again.

The battery is a BMS 36v 10Ah Li-Ion.

Paul





 
Last edited:
D

Deleted member 4366

Guest
There's no chance of electrocuting yourself, and neither can you make it any worse than not working. Measuring cell voltages is easy. You unplug the connector, and along the side, you'll see small slots with shiny metal underneath where you stick your voltmeter probes. You put your probes in the first two (1 and 2) to get the first cell voltage, then the next two (2 and 3) for the next, and so on until you have all 10. that info will tell us what's wrong with your battery.
 

jackhandy

Esteemed Pedelecer
May 20, 2012
1,820
323
the Cornish Alps
There's no chance of electrocuting yourself, and neither can you make it any worse than not working. Measuring cell voltages is easy. You unplug the connector, and along the side, you'll see small slots with shiny metal underneath where you stick your voltmeter probes. You put your probes in the first two (1 and 2) to get the first cell voltage, then the next two (2 and 3) for the next, and so on until you have all 10. that info will tell us what's wrong with your battery.
Problem, D8veh - He don't got a voltmeter at the mo'

And, as for not electrocuting himself - I have vivid memories of my old Dad (dear of him) getting me to stick my tongue on the contacts of a 4.5v battery, to see if it was ok.......

I guess it depends what you poke in them slots, but I bet 37v wouldn't taste good :)
 
D

Deleted member 4366

Guest
A voltmeter costs about £5. What would you rather do - spend £5 on a voltmeter or £250 on a new battery? You need a voltmeter anyway to check the polarity of a new battery. Each cell is only 4v, so probably tastes OK. You don't have to taste the whole 40v.

Morphix, if you want to send the battery to me, I'll have a go at fixing it, but it'll probably cost about £10 each way postage. Where about in Worcs are you, if it's not too far south, I might be able to come over and fix it.
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
Thanks Dave, I found an old multimeter but the 9v battery is dead! One screw on the back is corroded but I will try and drill it out and put a battery in, else I will order a new meter and do as you suggest, checking the voltage on the cells.

Assuming one of the cells is dead, what happens then? Can these be readily replaced with replacements in UK?

I do have a soldering iron and basic skills, maybe I can have a go at swapping out a broken cell if that's the cause.. else I have no probs couriering this to you and arranging collection, cheaper than buying a new one as you say, and better for environment.. I'd be happy to any costs of course, and pay something for your time, if it comes to it.

Paul
 
Last edited:

Zebb

Esteemed Pedelecer
Jun 13, 2012
371
12
d8veh;Morphix said:
Once again, outstanding help on this forum by d8veh.
 
D

Deleted member 4366

Guest
One cell won't be dead, but you might have one or two a bit low that need charging up. That's my guess. I'm assuming that the battery was working OK before you left it for a bit?
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
One cell won't be dead, but you might have one or two a bit low that need charging up. That's my guess. I'm assuming that the battery was working OK before you left it for a bit?
Yes it was working fine, no change in power or pull at all.. but I hadn't used the bike for about 3-4 months, and during that time I just top-up charged the battery once or twice, as it didn't need charging very long.

UPDATE:

I've just fixed the meter and removed the tape off the battery terminals... the battery cell tabs are soldered together and I think it will be tricky to unsolder them without damaging the tabs and/or battery cells themselves as there's a lot of solder been used and it will require a lot of heat and care not to get any solder drops on the cells themselves which are obviously very heat sensitive and fragile! I took a picture so you can see what I mean.



Anyway, I disconnected the big connector like you said, then read the voltage across each of the 5 grouped terminal pairs. I got a reading of exactly 3v from each (15v)... and from the actual battery kettle plug connector I'm getting 14.9v.


I'm guessing then like you said, some of the cells joined together are malfunctioning and/or won't charge? I'm guessing there should be 7v+ coming from these terminals from the grouped cells, not 3v?
 
Last edited:

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
New battery options?

Even if this battery is repairable, I need to buy a second backup battery in case it dies.

So I'm wondering what battery to get which is light and portable? Any recommendations for 36v 10Ah?

The lipo batteries sound appealing but I don't like the idea of the fussy charging and risk associated with these batteries...have things moved on a bit in that regard, or is there a safer lipo option for noobs?
 
D

Deleted member 4366

Guest
You don't have to unsolder anything. Look at the multi-pin connector in the photo. In the side underneath are slots. That's the best place to measure cell voltages. It's a bit tricky to get your probes in because the slots are small, but that's how I do it. You have to be careful not to let the probes touch while connected. Alternatively, you might be able to get your probes up the back of the connector where the wires go in to it. I need 10 results for the 10 cell groups. There's 11 wires, so you measure each adjacent pair.
3V per cell shows that the battery is empty, but the BMS should still allow charging. One or more must be lower than that. That's why I need 10 results.
 
D

Deleted member 4366

Guest
Normally, I'd recommend the battery you already have. No batteries like being left for a long time without use. Even top-up charging doesn't seem to be enough for some depending on the BMS. They often go out of balance, which is what's probably happened to yours. I've got about 20 batteries, so I often get this problem, but a quick re-balance with my lipo charger always brings them back to life.

Lipos can be a good solution, but you have to buy a charger and make/buy connectors to wire your packs into the right configuration. Charging is always more complicated and normally involves disconnecting each pack to charge. I like these ones because they have a hard case, which stops them catching fire if you drop them.
Turnigy 5000mAh 4S1P 14.8v 20C hardcase pack (UK Warehouse)
You'd need 6 to make a 12S2P configuration 44v 10aH. The extra volts gives a useful 20% increase in speed and torque, otherwise look for 4 x 5S 5000maH packs for 36v 10aH. You then need a charger. This one can fix your out of balance battery very easily:
AOK BC168 1-6S 8A 200W Super Speed LCD Intellective Balance Charger/Discharger | eBay
otherwise you can use an Imax charger:
IMax B6 Digital LCD Lipo NiMh battery Balance Charger UK Delivery | eBay
both chargers require an additional 12v power supply. I use a PC power supply for the 12v.

I personally don't believe that safety is an issue provided that you follow wiring and polarity conventions; however, it's easy to get big sparks and melt connectors when you forget to disconnect your series wiring harness when charging in parallel. As a general rule, you must never have exposed connector pins, so every battery side one must be a female.
 
Last edited by a moderator:

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
You don't have to unsolder anything. Look at the multi-pin connector in the photo. In the side underneath are slots. That's the best place to measure cell voltages. It's a bit tricky to get your probes in because the slots are small, but that's how I do it. You have to be careful not to let the probes touch while connected. Alternatively, you might be able to get your probes up the back of the connector where the wires go in to it. I need 10 results for the 10 cell groups. There's 11 wires, so you measure each adjacent pair.
3V per cell shows that the battery is empty, but the BMS should still allow charging. One or more must be lower than that. That's why I need 10 results.
Ok Gotcha, sorry for being a bit slow and thanks for your patience ;-)

I'll try and take those 10 readings now off the connection terminals and report back.
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
Normally, I'd recommend the battery you already have. No batteries like being left for a long time without use. Even top-up charging doesn't seem to be enough for some depending on the BMS. They often go out of balance, which is what's probably happened to yours. I've got about 20 batteries, so I often get this problem, but a quick re-balance with my lipo charger always brings them back to life.

Lipos can be a good solution, but you have to buy a charger and make/buy connectors to wire your packs into the right configuration. Charging is always more complicated and normally involves disconnecting each pack to charge. I like these ones because they have a hard case, which stops them catching fire if you drop them.
Turnigy 5000mAh 4S1P 14.8v 20C hardcase pack (UK Warehouse)
You'd need 6 to make a 12S2P configuration 44v 10aH. The extra volts gives a useful 20% increase in speed and torque, otherwise look for 4 x 5S 5000maH packs for 36v 10aH. You then need a charger. This one can fix your out of balance battery very easily:
AOK BC168 1-6S 8A 200W Super Speed LCD Intellective Balance Charger/Discharger | eBay
otherwise you can use an Imax charger:
IMax B6 Digital LCD Lipo NiMh battery Balance Charger UK Delivery | eBay
both chargers require an additional 12v power supply. I use a PC power supply for the 12v.

I personally don't believe that safety is an issue provided that you follow wiring and polarity conventions; however, it's easy to get big sparks and melt connectors when you forget to disconnect your series wiring harness when charging in parallel. As a general rule, you must never have exposed connector pins, so every battery side one must be a female.
This sounds very interesting, I think I'd go for the 4 x 5S 5000maH option, which works out £90 + charger? Very nice price saving to BMS Li-Lion 10Ah. And the weight is only 2.6kg too, less than BMS a bit. What you said about the wiring and potential for accidents puts me off a bit though.. I'd like a fool-proof set-up and would be willing to pay someone to do the wiring and provide instructions etc..

I had a very nasty accident with batteries in 1990's and very nearly was left blind so I'm very weary of batteries now and like safe options with no chance of accidental shorting or wrong charging!
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
You don't have to unsolder anything. Look at the multi-pin connector in the photo. In the side underneath are slots. That's the best place to measure cell voltages. It's a bit tricky to get your probes in because the slots are small, but that's how I do it. You have to be careful not to let the probes touch while connected. Alternatively, you might be able to get your probes up the back of the connector where the wires go in to it. I need 10 results for the 10 cell groups. There's 11 wires, so you measure each adjacent pair.
3V per cell shows that the battery is empty, but the BMS should still allow charging. One or more must be lower than that. That's why I need 10 results.
OK I may be a bit dense here, but I can't figure out what I'm doing here! I tried to measure the voltage from the bottom of the connector pads (the big connector which has 11 pins) but I wasn't sure where to put the other probe.. do you measure in series the pads, ie across pin 1+2, pin 3+4 etc? If you put the probe on 1 pin at a time and work along like that, where does the other probe go?
 
D

Deleted member 4366

Guest
It's best to go pins 1+2, 2+3, 3+4, etc. because it'll give the cell voltages directly.
You can go 1+2. 1+3, 1+4, etc, but then you have to do some sums to calculate cell voltages, i.e. it'll go something like 3.0v, 6.1v, 9.1v, 12.0v instead of 3.0v, 3.1v, 3.0v, 2.9v.

If you get negative values, you're going from cell 10 to cell 1, positive values means cell 1 to 10.
 

morphix

Esteemed Pedelecer
Oct 24, 2010
2,163
119
Worcestershire
www.cyclecharge.org.uk
Last edited:
D

Deleted member 4366

Guest
I charged up the cells to 3.5v with my BC168 charger that can do up to six at a time. That should have been enough to get it to start workng, but it didn't want to play. With some batteries, you can wake them up by unplugging and re-connecting the multipin connector, but I'd done that during charging. Everything that I checked was OK, put the charging and discharging FETs remained switched off. Then after a bit of brainstorming, I figured it out. There's a white wire directly from the battery +ve that powers the BMS, so unplugging the multipin doesn't reset the software. I unsoldered the white wire then resoldered it and bingo! Everything is working again. Now on final test. Should go back to Morphix tomorrow.

After that I found a thread on Endless Sphere about this BMS, where people are hacking the software todo all sorts of clever things. The BMS is quite sophisticated. Once the cells go below 3.0v, it goes to sleep and cuts off all activity to save draining the cells further. There must be a wake-up procedure, but I don't know what it is; however, there's a reset pad on the back of the PCB that would probably do it, or my method. Apparently you mustn't disconnect that white wire with the multipin in, otherwise it damages something. I was lucky to pull it out just in case. The white wire runs along the edge of the PCB at the top of the photo. It goes to P+ pad.