BMS's damaging 52 volt batteries when left with very low current draw off.

Waynemarlow

10 kW
Joined
Jul 9, 2016
Messages
778
Location
Bucks, England
I have two packs, a 14S2P and a 13S2P ( EBike ) that have been damaged with a similar outcome, but albeit with different causes. Both batteries have been fully discharged and then left with a very low current attached over a period of time. Both packs the group of cells nearest the negative terminal have been discharged totally to around zero damaging the cells. The next group up the chain is discharged to about 0.45V and so on up the pack until the BMS cutoff of 2.75Volts is reached at about the 5 /6 pack.

One pack was in an EBike and the other attached to a down voltage convertor ( 12 volts in a camper van ).

My question is that I would have thought that the BMS low voltage circuit would have kicked in and cut the low current draw off from the pack. Have I just been unlucky and had faulty BMS's not doing their job or is there another reason that I'm not aware of allowing small currents to be drawn from the pack ?

Thanks.
 
Never trust a cheap 18.00 bms plus this happens all the time. And all different bms's worked different. Just like all 18650 cells are of different quality. Plus a lot of bms's use cells 1and 2 and or 3 channels to supply power for bms.
Now state of charge
What is full voltage ?
And now what is voltage of cell groups 1 thru 13 write just like this on paper... start at neg end

1. 3.35v
2. 4.10v
3. 0.00v


13. Xxx volt
Dothisfor both batteries.
Do not charge till you report back as this is how you make battery fire. Very hard to put out.
Never charge in house or garage.
 
999zip999 said:
Never trust a cheap 18.00 bms plus this happens all the time. And all different bms's worked different. Just like all 18650 cells are of different quality. Plus a lot of bms's use cells 1and 2 and or 3 channels to supply power for bms.
Now state of charge
What is full voltage ?
And now what is voltage of cell groups 1 thru 13 write just like this on paper... start at neg end

1. 3.35v
2. 4.10v
3. 0.00v
Thanks
1. OV
2. 0V
3. 0.6V
4. 0.7V
5-12 2.8V
13. 0V
14. 2.8V

Sorry I have already replaced low voltage batteries on the other pack.
 
On the pack you replaced cells on. You need to balance pack to same voltage of each cell group. Before installing bms as that little bms will have an impossible job. A must.
That other pack is dead and dangerous to recharge.....
 
999zip999 said:
On the pack you replaced cells on. You need to balance pack to same voltage of each cell group. Before installing bms as that little bms will have an impossible job. A must.
That other pack is dead and dangerous to recharge.....
Yes I have been balancing up the cells.
ebuilder said:
That pack is bust Wayne. Time for a new pack.
Totally agree but what concerns me is that the pack has gone from a good working pack to dead pack simply by the BMS not shutting off correctly ? Or has the voltage convertor some how created a low current situation which has got around the BMS shut down ? Just trying to prevent a recurrence and damaging another pack.

Lots of parameters here that I guess I will never know fully the cause of the failure unless someone can point me to a routine that will test the BMS to at least eliminate that part of the pack.

Thanks all.
 
Thanks

Interesting re the parasitic draw, I would have thought the BMS would have protected the pack against this. The pack I gave figures for was virtually an overnight connection to the voltage convertor, sure I had used the pack a lot to charge Ipads and lights etc but normally the BMS kicks in, this one didn't but at the time there was nothing connected other than the convertor. Need to dig a bit deeper and see what current the convertor pulls when nothing is connected.

My Ebike batteries never get the chance to sit long enough to parasitic discharge below the cutoff so this is all a bit new to me.
 
Battery Murder System

Wire so easily removable without breaking the pack apart
 
Waynemarlow said:
Interesting re the parasitic draw, I would have thought the BMS would have protected the pack against this.

It can't, if the parasitic draw is from the BMS itself, because the BMS has to be powered by the cells (at least some of them--for packs where the first few cells are drained dead, it's powered by those. When it's killed all the cells equally, it's powered by the entire pack).

All the BMS can protect against is anything connected to the charge or discharge port. When any cell reaches HVC, it shuts off the charge port. When any cell reaches LVC, it shuts off the discharge port.

If there is a device connected to the discharge port, and the BMS has turned the port off due to LVC, the device no longer receives power...except a FET is not a "perfect" switch, and may still conduct very slightly, so there may be leakage current thru it, probably only microamps, but it's still a drain, and if the pack is already empty, and sits long enough this way....

Also, power used by the BMS to power itself is only microamps to at most milliamps, but if the cells were already at LVC, they don't have much left to give. If the pack is left in that state long enough, the BMS will drain the rest of whatever was left.


But in no case can it disconnect *itself* from the cells, for the vast majority of BMSs. There probably are some advanced ones out there that detect the cells are below some critical point and actually shutdown the MCU of the BMS, possibly even it's LVPS (low voltage power supply, that converts the battery power to the voltage(s) the BMS uses), which would minimize power usage, but there could still be tiny leakage currents in the balancers, etc, that would eventually still drain the cells when they're that empty.


Another problem that can happen is a balancer can be stuck on, and that will continously drain the cell group it's attached to, regardless of the state of the battery or BMS. This is a hardware failure of the BMS itself, and is "obvious" from a voltage test of all the cell groups, where that group is "always" lower in voltage than the others. (if there is more than one stuck balancer, there will be more than one always-lower group).


So, all that said, I would guess that this:
1. OV
2. 0V
3. 0.6V
4. 0.7V
5-12 2.8V
13. 0V
14. 2.8V
assuming 1 is the most negative cell, and 14 is the most positive, then 1-4 are used to power the BMS, and killed by that, and 13 has a stuck-on balancer.

The other cells 5-12 and 14 are still at 2.8v because the BMS shut off the discharge port at that point, and then they stopped draining.
 
j bjork said:
What amberwolf said, but there is also self discharge of the cells.
Maybe a cell in group 13 for example has a little high self discharge and drain that group.
2,8V is not a lot, it dosent take very much to drain a group from that with a little time.
Thanks all guys. Sadly yes I do build my own wheel sets, do build my own batteries, have owned both a Yammie 250F and a KTM 250EXC ( and have a metal rod inserted in my femor to prove it ) among other off road bikes, so feel well aligned with j bjork :D

Just has an aside for j bjork, having ridden MX and then Enduro and now Emtb, the modern Emtb is now almost as good as a MX bike in fun factor, will take jumps as big as your cahones will allow, will up the adrenaline factor similar to your EXC and when stuck in mud, it weighs 23kgs. And you can ride it out of your door to the your nearest trail head where most people will simply look as you go past. As much as I miss Enduro ( there's now Emtb Enduro series springing up everywhere but I'm getting a bit long in the tooth for that ) the Emtbs have filled a void equally as well and some ways are much better, you need to give them a go.

All my battery building has been for EBikes so far and its always been space that dictates how they are built. Sadly its always the series bars that are the pain to remove if a cell were to go bad. So far out of the 8 batteries I have built for myself and others, this one is the first to go sour. Its probably 4 years old, done about 200 charges and been worked hard, its probably had its day. But not quite, we use 4 and 2 cell groups for our lights, guess what the good ones are going to be used for.
 
ebuilder said:
Waynemarlow said:
john61ct said:
Battery Murder System

Wire so easily removable without breaking the pack apart

Bit to cryptic for me. :(
what he meant is...
BMS's are sometimes referred to as Battery Murdering systems because they can actually promote cell destruction versus protect cells.

...

Second thing he meant was...if you plan to rebuild the battery, wire it in a manner that the battery can be serviceable without removing the nickel strips spot welded to the cells and removing cells to work on the battery. You want to be able to remove wiring like balancing leads and wires connecting the BMS to charging and discharge ports etc without having to take the cells apart.
Yes, correct

Rather the balance wiring needs to just be easily accessible, not only swapping out the BMS but measuring / monitoring voltage at the cell/group level while there is no BMS, periodically using a (different) balancer device, or 1S charge source if manual balancing...

Atomising, tearing the actual cells apart may if course be required at some point, but hopefully rare, while doing the above maintenance / testing routines may be helpful for in getting a longer lifespan.

 
Back
Top