Bafang HD 1000W cuts out under load - dies completely

[rgvkid]

hi all,

i've got a boardman full sus with the battery mounted on underside of downtube (which i suspect may have led to the issue due to getting all the road crap thrown at it, even with a cover).
Bafang HD motor and Hailong 20ah 52v battery.
Last week it just died during use, no power at all. wouldn't start up from the button, power cycling the battery with battery toggle switch would sometimes let it start, though this often wouldn't work for minutes on end. lots of switching battery off and on, trying power, battery off and on, power switch etc, then it would eventually start, often to die within seconds, especially if under load.
i took battery off and connections were really crappy, opened battery and cleaned that up, got a new battery baseplate after cleaning them up and problem still there.

got multimeter on it today, with battery switch off, it measures 3.4v, with battery on it's 56v.
i could make it fail within seconds by putting it under load, full throttle in driveway and it would die almost instantly. it deffo is load related, as i was using it for work earlier (after thinking it was fixed, it had done 3 journeys of 15 mins after cleaning up connections, and all SEEMED fine), and on PAS 1 it would stay on, pas 2 or 3 and it would die.
when it dies, i've slapped the meter on battery outputs and they show 3.4v, even with battery on. after one or two toggles of battery power rocker, it usually then outputs 56v, plug it back into bike and it'll start, but again fails within seconds if under any sort of load.

i suspect maybe the battery BMS is faulty, though any advice would be much appreciated. Inside of the battery was dry, no damage at all, but the battery and baseplate connections were utterly awful, pics attached. maybe this has shorrted something out in the BMS ?
weird in my mind that when i cleaned it all up, it then performed with no issues for 3 15 min journeys (mix of PAS 2 and 3), then became unuseable again.

thanks for any advice,
cheers,
col

 

[RunForTheHills]

But isn't dying under load the classic sign of voltage sag? How old is the battery?

 

[rgvkid]

Hi, the battery was bought same time as the motor, about 1 1/2 years ago. Motor has done just over 2000 miles

 

[Apartunis]

3.4 volts is leakage across FETs in the BMS that have cut off discharge. I would trust that the BMS is doing its job. When they go wrong they almost never turn on at all.

Crusty connectors like yours can cause voltage drop at the controller which might be enough to trip the BMS. Normally in that case there are signs of heat damage near the nasty terminals. If the drop is fast enough and deep enough then there may not be enough time to cook the terminals.

 

[E-HP]

Hi, the battery was bought same time as the motor, about 1 1/2 years ago. Motor has done just over 2000 miles

Can you fully charge the battery to 58.8V?
You could have one or more cell groups that are weak and out of balance so the BMSis tripping on low voltage from the weak group(s).

 

[rgvkid]

thanks all, i'll charge it fully and see what it measures on the multi meter

 

[rgvkid]

Hey folks, charged battery fully and it measures 58.3v.
I don't know enough about batteries and bms to know if the above reading could be fine, yet under load (it is deffo load related when it dies, literally with a second or so of load) cells are failing, and bms is killing all power.
Or could it be a faulty bms?
Again, I've not got any experience with bms, and all my previous self build bikes (1000w hub, then 2 x bafang HD incl this one ) have never had any battery issues.
Thanks

 

[Apartunis]

Hey folks, charged battery fully and it measures 58.3v.

That discrepancy between charger voltage and charged battery voltage leaves enough room for a cell group to be having problems. It only takes one weak or damaged cell group to hit minimum voltage and trigger BMS protection.

If you are willing to open the battery and probe the voltages of each series group, you may find one that doesn't match. It's worth checking both after a full charge and with the battery partly discharged if it will still run the bike.

 

[rgvkid]

I've had the battery open a few times since this all started,so not an issue to check each group of cells, thanks for suggestion.
Yep will try it on bike fully charged, the other day when it maybe had 70% charge, it would die almost instantly on pas 3 (of 5), but when I was trying to limp it home from work other day, pas 2 would trigger shutdown, pas1 didn't seem to be giving me much (never changed pas settings with the cable, to be honest,so.its factory settings) but it stayed powered on with pas1.
Cheers

 

[rgvkid]

hi all, apols for delay, just dug bike out today and tested it with a fully charged battery, and same as last time i tried it, it dies literally within a second of throttle being applied on pas 3 (of 5). i can try it again without throttle if anyone thinks that maaaay make a difference, but when it was failing before it was doing so when just using pedal assist, and i also tried it before with throttle unplugged and it still died.
will try to check battery cell groups tonight.
i'm loathe to buy another battery if there's any chance it's not the battery, as money is tight, hopefully there will be a discrepancy between cell groups which will confirm it's the battery.
cheers

 

[rgvkid]

hi, opened battery up and there is damage inside, i'm assuming water damage maybe, as it looks like corrosion.
all banks of cells measured approx 4.1/4.2 apart from the block of 4 where the damage is, which measured about 2.8/2.9.

i am assuming this allowed the bike to start, but when put under any sort of load presumably the BMS would kill the battery power output (willing to be corrected on this).

question being, do people think the cells themselves in the bank of 4 are potentially damaged, or could the battery be fixed by resoldering the connecting strips (presumably with new ones) ?
i would guess to test the 4 cells indiviually, i'd need to disconnect them from each other, which would then allow me to measure the voltage at each cell, but again not sure if this may not be an accurate test, as if the cell/s are damaged it could potentially show voltage fine, then fail under load ?

bottom line being a new battery is about £230, and in Edinburgh there are a few shops which do ebike and battery repairs, so my plan was to contact a few of them, and get a ballpark price for whether it's worth getting it fixed.
battery was bought new with the motor, so both have done about 2000 miles.
thanks again for any advice.

 

[E-HP]

question being, do people think the cells themselves in the bank of 4 are potentially damaged, or could the battery be fixed by resoldering the connecting strips (presumably with new ones) ?

If it were me, I wouldn't trust the cells. You were able to fully charge the pack, with the bad cell group, which means that group is, or appears to be charging. Damaged cells will appear to charge faster due to higher IR and their reduced capacity, but not maintain a real charge, which is why the voltage would drop when load is applied.

 

[Apartunis]

question being, do people think the cells themselves in the bank of 4 are potentially damaged,

Yes definitely.

or could the battery be fixed by resoldering the connecting strips (presumably with new ones) ?

Do not do that. Replace the affected cells with ones of equal or higher rating in both Ah capacity and amp limit.

 

[rgvkid]

Thanks for that, I see what you mean about it charging but cells could still be goosed.
A couple of the shops I looked at earlier (will email tomorrow) said they replace cells etc, but would really depend on price. No point in me spending £100 on it if a new one with warranty is £230, so it would have to be a relatively cheap repair for me to consider it (and I can't imagine it would be £50 cheap).
Cheers

 

[rgvkid]

Yes definitely.

Do not do that. Replace the affected cells with ones of equal or higher rating in both Ah capacity and amp limit.

Thanks for that, I've not really seen inside of many batteries, do people think it's due to moisture getting in (all seemed totally dry.... But it is mounted to underside of downtube through necessity of frame and suspension), or electrical failure of some description?
Thanks

 

[Apartunis]

Thanks for that, I've not really seen inside of many batteries, do people think it's due to moisture getting in (all seemed totally dry.... But it is mounted to underside of downtube through necessity of frame and suspension)

Battery cases that are designed to sit on top of the down tube where a water bottle would otherwise go tend to collect water when mounted upside down. Sometimes that is the only option. In such cases it is probably wise to drill a couple of drain holes at the low points in the battery case. Mark the spots first, remove the battery, disassemble the case and then drill so you do not risk poking any cells.