Resistors on BMS heat up AFTER charge

[RLT]

help?!?!

I just ran the first test on my first 36V 20AH (4P 10S of 32650 5AH cells) Lithium Ion battery pack.
I'm using this BMS:
http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=3558

Everything seemed to be working properly; Discharge fine, Amp limit fine.... (Didn't run it down low enough to see if low voltage cutoff is correct, but I'm not really concerned about that anyway). Charged the batteries up, that seemed to work just fine; When I took it off the charger, I noticed that all the power resistors were pretty warm. Not alarmingly hot, but warmer than I'd like... Probably well within component specs though. So, I wasn't too worried.

Checked each of the serial cell groups. All were at 4.21 to 4.23 V , (individual cell balance of the BMS I'm using is supposed to be 4.22V max) and the entire pack was right at the 42.2V it was supposed to be. I noticed that the resistors were still pretty warm, but no big deal. I turned off my main power switch (between battery pack and BMS on the negative pole line) and worked on something else for a while.

About a half hour or so and when I went back, the resistors were hotter than they were right when it came off the charger.... And with the main power to the BMS off... the only connections to the board were the thin little cell balance wires that go to each series battery group and the positive wire. At that point, I unplugged the balancing socket just to be safe.

So, despite main power being off, was the BMS still trying to bring each cell into perfect balance by burning off energy???
Did I maybe wire it wrong (although it is correct according to the wiring diagram)????
Is the board maybe defective????

Assuming that there isn't a problem with the BMS or my wiring, are you SUPPOSED to leave the main power ON between the battery pack and the BMS ? According to the specifications the BMS only draws 300µA, so that isn't really a big deal, but I figured for safety, it is better to shut off main power.

Thanks in advance.

 

[fechter]

That doesn't sound right. You should be able to leave the BMS connected all the time.

The resistors are only supposed to be on when the cell voltage is above the upper limit.

It's possible the voltage where the shunts turn on is too low somehow. If the charger output voltage is higher than the combined shunt voltages, the pack might get 'overcharged' a bit and then get drained back down to the shunt voltage.

I'd suggest trying it again, charge it a bit, then leave the thing connected while periodically checking the cell voltages manually. Babysit it for a while.

I'd guess that the voltage will steadily drop until it reaches a certain point and then the resistors will turn off and get cold. Just what is that voltage?

If the resistors stay on and the voltage continues to drop below any resemblance of a normal resting voltage, then disconnect and terminate the test.

 

[RLT]

Thanks, Fechter.

I'll do that.

Plus I'll hurry up and build the other pack and see if the same thing happens with the other BMS.

 

[RLT]

Just a long overdue update:
I did as Fechter recommended, and for a while, it seemed OK.

The BMS worked as it was supposed to under actual use, and it worked OK when I recharged.
Took it off the charger and kept checking it every minute or so as I was working on some other stuff on the bench. The resistors heated up a little, but not alarmingly so... just about what I'd expect from them bleeding off a little excess voltage from the higher cells.

Then about a half hour after I took it off the charger: sudden meltdown of nearly all the little balancing wires; one moment merely warm, fifteen seconds later it about 'went super-nova';
One of the outside wires on each side of the ribbon cable survived intact, but the middle ones were really bad. a couple burned through, the rest; insulation melted off and scorched copper. The battery pack itself was OK.

Unfortunately I didn't want to install the other one on my other battery pack until I figured out what the problem was... and due to a couple of stupidities in a row, I destroyed the other battery pack on the first real test ride. $600 or so down the drain. But a lesson learned.

Posted some photos and info on the unused BMS over on the schematics and circuits thread for anyone interested.
http://endless-sphere.com/forums/viewtopic.php?f=2&t=278&p=56328#p56328

 

[safe]

...I destroyed the other battery pack on the first real test ride. $600 or so down the drain. But a lesson learned.

This is the kind of thing I've been talking about elsewhere. I'm unhappy to lose capacity on my batteries after getting over 3200 miles out of them. The idea of losing $600 because of a mistake is tough... really tough. I know that ultracapacitors are potentially very dangerous (and maybe the potential for loss is worse if it physically was broken loose in an accident) but if they aren't broken they will not wear out.

Arrrrgh! This does not make me eager to jump into the less durable chemistries. (it makes my SLA's seem sensible in comparision... they are durable that's for sure)

 

[RLT]

I just finished telling the story of what happened to the destroyed pack over on my new thread in the photos& videos section....

All would have been OK if I had been more careful in both design and operation, and maybe just a little bit less unlucky.

Expensive mistake yes, but what I'm trying to do would be physically impossible for me on SLA, borderline imossible on NiXx and prohibitively expensive on anything of the better Lithium chemistries except for maybe Ping duct tape batteries on eBay, and I don't want to get into the new bidding wars over them.

 

[Toorbough ULL-Zeveigh]

Arrrrgh! This does not make me eager to jump into the less durable chemistries.

Don't misconstrue Batteryspace offerings as an indictment of an entire chemistry.

 

[RLT]

Don't misconstrue Batteryspace offerings as an indictment of an entire chemistry.

Or my wiring skills. That is still a possibility.
I don't think so, because it isn't that complicated and I'm not completely inept, and triple checked it three times, but I can't completely rule it out yet either. I'll try the offending board again on a 1P10s pack to see if it still overheats after charging, and try the unused board on a new pack too. That ought to give me enough data to determine the cause of the problem.