battery was fine, then self discharged to 7.4 volts.... ideas what is likely?

[TwisterGripper]

What you do you think most likely happened here? critically low battery self discharge.

Working on a scooter for customer - initial problem: turns on, short power pulse then nothing.

I correctly diagnosed melted fuse holder and fuse - poor connection had it heating up and arcing until so much of the fuse had "vaporized" there was not enough metal to ensure good connection.
also , water had infiltrated the battery compartment - mud rocks etc…. found some corroded connections.
- someone before me had mucked around in there - foam was missing, allowing the battery and controller to slide around, a lot - there was abrasion wear on the battery wrapping - but not all the way through
- cleaned up connections, replaced fuse holder & fuse (all good soldered joints). added foam to securely hold battery and controller in place.
- test ride 1 - worked for a short bit - maybe 100 yards then power cut out.
- opened up battery compartment: fuse intact, connections seem good - I disconnected the previously corroded connectors and reconnected just in case of poor connection still - that did nothing.
- opened up controller - despite the rust on outer casing, it was all clean and sealed inside, no evidence of damage
- put it all back together and inexplicably, it powered up and worked perfectly again.
- test ride again - this time rode 1 km. good power, no faults.
I considered this a done job and put it away

Next morning I go to do another test ride and - its dead.
battery has drained down to 7.4 volts!!!!
- tested with charger - it slowly climbs - got it up to 13 volts after a few minutes (obviously not a smart charger). disconnected battery from everything, and watched the volt meter - volts dropped quickly back down to 7.4 volts on the battery main leads.

This battery pack is self discharging now. that problem did not exist before. I have not opened up the pack to inspect. at this point I've put too much time into it and it wont even hold voltage to "wake it up" - I hate telling the client it’s a total loss, and I'll likely eat the labor and offer a few bucks for scrap as I don't feel right taking money after getting approval for repairs, only to find more problems making it a money pit for the customer.

side note: while I had the scooter opened up after the first test ride when it died - after buttoning it back up, I heard a short "bzzzzt" almost like a quick electrical arc sound - pulled cover again and found no evidence of any arcs or shorts - so either I imagined it, or it was inside the controller or battery - still had power, so maybe that was when battery started self discharging, but still had enough volts at that point to power on and test ride.

What do you think is most likely cause? I am careful and knowledgeable enough not to have done anything stupid or cause a short on this unit. Im just baffled…battery was charged and stable at 52 volts when it was brought to me.

Is it possible because the kid rode it with a loose battery pack flopping around inside the casing it got damaged internally and my little bit of handling was just the final straw before something shorted? It happened while in my possession but I did nothing abnormal to cause it...

I will tear into it if they accept the scrap offer.

 

[amberwolf]

What voltage do you read at the battery cell block itself? Between the B- wire to the BMS, and the B+ wire to the outside?

If it's normal, then the BMS has shut off either because of damage to it or there is one or more of the cell groups that is too high or too low and it has turned the output (and possibly the input) off to prevent discharge (and charge).

If it's actually at that low voltage, then the cells are not recoverable (anything below that cell's specified low voltage cutoff isn't safe to recharge from, but I really wouldn't below 2v/cell).

If you don't read a voltage across the cell block itself as a whole, it could also be broken interconnect between series sections of the pack.

Also, if there was rust on the controller, and it's in the same compartmetn as the battery, there's a likelihood of water damage inside the battery.

 

[TwisterGripper]

I havent cut into the pack at this point. it still belongs to the client and i'm way past the time I estimated for repair. I'll cut into it If i take ownership or they authorize further diagnostic work... I am curious and would open it up, but i cant go further now until client gets back to me.

So if BMS did go into protection mode, then seeing 7.4 volts could be normal and its still possible the actual cells are ok? is that what you are saying?

 

[amberwolf]

So if BMS did go into protection mode, then seeing 7.4 volts could be normal and its still possible the actual cells are ok? is that what you are saying?

Yes.

The FETs are not a perfect switch; they have some leakage current, so a voltage devlopes on the "off" side. It's just a tiny charge, so if you connect any load to it (even turning on the scooter should, but something resistive like a heater or incandescent lamp would for sure) would drop that to nothing, in these cases.

If it maintains the 7.4v even with a load on it, then the BMS hasn't turned off, which means it's either shorted (stuck on) or bypassed (wired around it) or it doesn't have a BMS, and the cells are actually damaged in some way.


If the BMS *has* turned off the output, it did it for a reason, adn so there *is* a battery cell problem, no matter which result you get. (or an interconnect problem).

 

[TwisterGripper]

Understood. Thank you Amber Wolf you have been very helpful. I'll try the power switch and see if voltage drops.

 

[TwisterGripper]

update: yes it drops to zero under load...result for the customer is same though- cost to find the right BMS for this battery and install it is about the same as just buying another cheap scooter battery...on top of investigating/testing the cell groups, so might as well just replace it.

 

[amberwolf]

It's almost certainly *not* a broken BMS if it drops to zero under load.

The BMS is doing the job it's suppossed to do, by turning off the output to protect the cells from damage that can lead to a fire.

If you get the chance to check the cell votlages, you'll probably find that at least one group is very low, below the BMS LVC. Or that a wire has broken off one or more cell groups, or the balance connector is unplugged, so that the BMS can't read the cell voltage. Or some other damage to the cell block or wiring.

It *could* be the BMS...but I doubt it.

 

[amberwolf]

Also, I don't recommend the "cheap" batteries, as they're usually poorly built and are fire hazards as often as not.

They're also not even always what they say they are.

(can't necessarily tell from the outside, requires disassembly or at least opening the case and checking construction, parts, etc).

 

[TwisterGripper]

Ahhh thank you. I was of the mistaken belief when bms cut off power, something breaks and you have to replace it lol not sure where I got that idea. I repaired a Genze ebike by replacing the BMS a couple years back, one cell group was 0.2v lower than the rest, due to sitting 2 years....charged it up to match the others & replaced BMS....maybe the bms on that one was good too lol. whoops.

Fully aware cheap batteries are cheap for a reason, sellers lie etc...

Client accepted my offer for scrap value - its mine now and I can open up the pack. I'm not a battery builder and dont' want to be, but i am curious and wanting to learn so i will get to the bottom of it, and likely harvest the good cells for other purposes.

thanks so much again!

 

[amberwolf]

Ahhh thank you. I was of the mistaken belief when bms cut off power, something breaks and you have to replace it lol not sure where I got that idea.

There are OEM integrated systems that "brick" the battery by permanently shutting the BMS off when it detects failure conditions that can lead to cell damage that can lead to a fire. It's a good idea since it means a user can't mess with it and "fix" it by recharging cell groups that have dropped lower than safe limits, only to have it catch fire later.

Mostly it's for liability reasons because of lawyers, but it still doesn't hurt to have such protections, as the only "good" way to fully repair a pack that's failed in a way that causes the BMS to have to brick it is to replace the cells anyway, and that's most of a pack's cost.

Sometimes a cheap BMS just fails, because it's cheap and poorly made. More often it fails because it can't handle the load it's put under because it's not the right one for the application, or because it was assembled into the pack incorrectly, or ESD damaged in handling somewhere along the line, or environmentally damaged (heat, water, etc).

I repaired a Genze ebike by replacing the BMS a couple years back, one cell group was 0.2v lower than the rest, due to sitting 2 years....charged it up to match the others & replaced BMS....maybe the bms on that one was good too lol. whoops.

Any good BMS *should* prevent operation with a cell group that far off--that's a huge imbalance and means that cell group is very different from the others in capacity and characteristics. Could mean that a cell in that group has failed internally and draining down the others, or that it jsut has more aged or defective cells in it.

Many BMS will prevent operation once there is more than one tenth of a volt difference between groups for that reason.