Resetting the bms

PRW

100 kW
Joined
Nov 10, 2011
Messages
1,044
Location
Melbourne, Australia
I haven't ridden one of the bikes for a couple of months (travelling and winter), and the battery is now dead.

Connecting the charger to the charge wires does nothing - the charger doesn't switch on. I also connected the charger to the discharge wires for about 10 minutes, hoping to charge it enough to get the bms awake, but this didn't start charging, either.

I don't think it is related to the charger - I have never had issues with it.

It is a 22s 18650 30Q battery, currently sitting at 39V.

Any advice/ suggestions would be much appreciated!
 
That's a pretty dead battery. I'd check the individual cell group voltages before attempting any charging on it at all. The charger might be refusing to charge because of the extreme low voltage, as it's really not safe to blindly charge cells that have dropped to that low of a voltage.
 
The 39v (assuming you are measuring at the output of the BMS) is probably simply because the BMS is shut off, so all you get is leakage at the output ("ghost voltage").


Measure the actual cell groups to see what voltage they are.

Most likely there's a group (the one(s) that powers the BMS, probably) that's been drained down too far below LVC, so the BMS is in protect mode to prevent a fire (which could happen if recharging a battery with a group that dead at full charger current).

If that's the case, you might be able to manually, slowly, very low current, recharge the low group(s) up to match the other ones, and then run the regular charger on teh whole pack, and leave it on until the whole thing is balanced.

It is possible that if the cells went low enough, they may be damaged in a way that makes them a fire risk, but it'd be more likely that they just wont' have the same capacity and/or c-rate that they used to, so they could sag more than the rest of the cells.
 
so, I got around to sorting this out - had to wait for a cycle satiator to arrive, which was set to trickle charge at 1 amp. pulled the battery apart, and bypassed the BMS - as can be seen in the first picture. Charged the battery up slowly to 75v, then tried charging via the charge wires - and it worked!

IMG_2893.JPG
IMG_2895.JPG

many thanks to all in this thread (and Vad from Powervelocity) for the help!
 
So what were the individual cell (group) voltages?

Was the entire pack drained down equally too far to charge, or was it just one cell?

Did the BMS already rebalance any low cells?
 
Well, the reason we asked about the cell voltages before you did the recharge is that it can be very important. If you have cells that drained really low, it can actually be dangerous to recharge a pack, at least at higher currents, and sometimes at all, depending on how low the low cells were, and their particular chemistry/specs. Some cells when discharged below their minimum allowed voltages become fire risks for the rest of their usage lifespan.

It's not a likely thing, but it can happen.

Also, not knowing for sure which cell(s) were low, it's hard to say if it's just a general "ran it down to far and left it, then the BMS drained it more", or if it's a BMS bad-channel (balancer stuck on) problem. The former is more likely, but the latter has happened to posters here before, and it means that those cells will always be lower in capacity than all the others, and get worse over time unless they get topped off manually (since the BMS isn't going to allow taht to happen if it's balancer(s) is stuck on.

Since it's not LiFePO4, then for at least some flavors of Li-Ion, 3.6 is at the low end of the flatter part of the discharge curve, and 3.8 is at the higher end. There could be quite a lot of capacity difference between one end and the other, depending on the specific cell brand/model.

So the pack is probably pretty unbalanced.

You can reconnect the BMS, and then leave it on the charger so the BMS can balance the pack. This can take from hours, to days, to weeks, depending on how unbalanced it is.

Or you can manually bring up each lower cell to match the higher ones, with a single-cell charger.

Or manually discharge each higher cell to match the lower ones, with a heater coil or "light bulb discharger" (LBD), etc.
 
Kind of important to know if you ruined your battery letting it self discharge that low. In fact, that it got that low could mean there was already damaged cells in there.

Charge it and store it outside from now on. This is no joke. P2110064.JPG
 
thanks all, I appreciate the concern.

When I opened the battery (22s6p), the starting voltage was 70v. I am not sure exactly how to measure groups - I measured each individual battery.

I trickle charged bypassing the BMS from 70v to 75v, then via the BMS up to 84v. I have stored it outside for now, thanks for the warning DD.

I need to replace the bms - it is no longer working.

Would measuring a group be from a positive to each of where the blue wires are attached? Wouldn't that just show what is going through the bms?
 
What specifically about the BMS is not doing what you expect?

Note that the BMS may not be doing what you expect because a cell group may be below it's minimum voltage to allow it to charge at all, or above it's maximum voltage to allow it to continue charging. Whether that is from a bad cell in the group, high resistance in a cell group causing high voltage readings when charging current flows (or low voltage readings when discharging current flows), a bad connection to the BMS at the sense wire, or the BMS having drained that cell group for one reason or another, will require further testing.

Just replacing the BMS may not fix the problem, and if you're ordering the BMS from somewhere you might wait for a part you may not need (and still leave you with a wait to get the part you do need).


Cells are usually wired in parallel groups, so all the cells connected to a single BMS sense wire are all the same voltage.

Write down all the individual voltages you see when measuring below.

Measure from the black main battery wire to the first BMS sense wire, usually this is closest to the black wire, but it will be the one with the lowest voltage relative to that black wire, at one end or the other of the row of BMS sense wire connections.

Measure from the first BMS sense wire to the second BMS sense wire.

Repeat that for all the BMS sense wires.

Measure from the last BMS sense wire to the main red battery wire.


Post up the list of voltages you got above. For a 22s pack, there should be 22 individual voltages. Then we can see which cells are lower or higher than the others, to determine what you might need to do (if anything).
 
You can measure the individual cell voltages along the balance wire connector on the BMS. You want to avoid shorting anything and the space is tight. I use a piece of solid strand copper 'zip wire' which is a pair stuck together. With the right size wire, you can insert the wires into the connector holes. Another approach is to leave the connector attached and place the measurement wires along side the balance wires where they enter the connector (back probing).

Single Cell Charging Hookup 2.jpg
 
You can also take down a high cell( group) with a 12v car brake light bulb. Just solder some wires on. You can put a house light blub on the battery pack to drain or I use a hair dyer.
 
amberwolf said:
What specifically about the BMS is not doing what you expect?

well, after you raised the alarm about the groups, I went down to check the "groups" - and managed to create a spark on the bms where I thought I should measure. I now have two blackened sections on the bms, and the charge point is back to measuring 40v.

This is not my natural world... what is a "bms sense wire" in my picture?

thanks!
 
ok, so I measured the groups:
3.75
3.74
3.74
3.74
3.74
3.74
3.75
3.74
3.74
3.74
3.75
3.75
3.75
3.74
3.74
3.74
3.75
3.65
3.75
3.74
3.75
3.74

overall voltage 82.6
 
At the moment, all those cells look like they're right about in the middle of their charge curve, so it's hard to say if any are really different than others or not.

But the lowest one (3.65) might be the original culprit that was preventing charge at the start of this thread. Once the pack is more fully charged, you may see more of a difference between that group and others.



PRW said:
well, after you raised the alarm about the groups, I went down to check the "groups" - and managed to create a spark on the bms where I thought I should measure. I now have two blackened sections on the bms, and the charge point is back to measuring 40v.
Ouch. :( I have trouble with my hands (and eyes) and slip with probes and tools all the time...and end up haivng to fix things that weren't broken when I started. :/

It might be fixable but it's probably easier to get a new BMS.



FWIW, there's a lot of battery repair threads that have tips and tricks like what Fechter showed above, and others I haven't thought of or seen myself yet, to do testing of packs with. Unfortunately there's no easy way to locate all that stuff, so it takes a lot of looking to find them. :(
 
Your cell voltages look great other than the one cell at 3.65v, which is not that far out of line. A little manual charging on just that one cell would get the pack nicely balanced.

It sounds like your BMS is toast now. That's happened to me before when a probe tip slipped and shorted. It's why I use the wire setup now, which makes this less likely. It may have been toast to start with.

Some part of a BMS are easy to work on and other parts are just not worth it. Finding the exact replacement parts for the cell circuits is very difficult as is actually replacing them on the board.
 
Back
Top