JK bms melted balance leads

[Alex1]

Im not sure why this happened, there was no serious load 300-500W at that time, also, battery was low state charge 2.6-2.8V cells at this point.
Also other battery (paralel connected) bms reading discharge was off, charge on 22.6V, voltage, at time, inverter was on bypass.
I have bought fuses for balance leads, but never install them
Im not sure did something shorted in bms, bad connection, balance lead shorted to bussbar? cell balance was turned off by setting, at 3.42V balance turns on.

 

[BarracudaBob]

I am using PICO FUSES on my balance cables near the cells. Others were less than enthusiastic about this. That is a lot of current potential on a small wire. Hopefully you can buy a replacement wire and use the BMS.

 

[Steve_S]

WOW ! Haven't seen that before.
I think the only way to know what happened is to look inside the BMS to see what blew causing that short.
The likelihood of the USB dongle had anything to do with it is nil unless some truly freak back-voltage got through somehow, so not likely.
Although I could see it used as an excuse

Would be interested to see the insides of the BMS.

 

[tinyt]

Looks like the sense wires are one of the victims and not the cause. Maybe a picture of the complete assembly prior to the event will help.

 

[Alex1]

WOW ! Haven't seen that before.
I think the only way to know what happened is to look inside the BMS to see what blew causing that short.
The likelihood of the USB dongle had anything to do with it is nil unless some truly freak back-voltage got through somehow, so not likely.
Although I could see it used as an excuse

Would be interested to see the insides of the BMS.

I will open bms
I dont think to use it any more at least not in house, terminals are damaged, its to risky for me, smoke detector save me.
orange pi with solar assistant also doesnt work.

 

[rogerheflin]

Adding to what Tinyt said. If one (or more) of the balance leads were shorted into the BMS (or to each other at the bms input) then the spots where the balance leads are taped together (far from the BMS) tightly should have been the location that had the most heat damage (no easy way for the heat to get out ot the bundle quick enough), not the spread out location right at the BMS. This looks more like there was an external fire going up into this spot, and/or internal fire inside the BMS that caused heat on this side and cooked the balance leads.

I don't think a fuse on the balance leads would have made any difference.

A lose wire on one of the batteries would do this (under the BMS) but there should be blackening on the terminal/wires and other signs of heat at the location. Can you include a photo looking down at the batteries?

 

[Checkthisout]

Do you have a pic of this BMS as it was installed before the failure?

I have my balance leads running under the bus bars and I'm thinking that's probably a bad idea.

 

[JJJJ]

I wonder if a rodent such as a mouse saw the wires as a possible snack.

 

[Skypower]

I wonder if a rodent such as a mouse saw the wires as a possible snack.

That’s a concern with everything Inverter interconnects, battery interconnects, WiFi and solar assistant. This means strategic placement of big glue traps every 6 months. I had vole and rat damage in that area before the solar system but sure as hell won’t let that happen now.

 

[JJJJ]

Don’t know if it’s the case here but soy oil has been used in car wires and can be a mouse magnet.

 

[Checkthisout]

Don’t know if it’s the case here but soy oil has been used in car wires and can be a mouse magnet.

They chewed on wires for fun long before they made it out of soy or whatever.

Rats and mike like to gnaw on lead battery terminals for whatever reason.

Good thoughts on whoever said it might have been mouse damage. OP, any red mouse poop in the area? Urine smell, etc?

 

[Skypower]

I’m thinking that somehow a balance wire got cut or a pinch finally went through. That one glow worm then laid down across other wires and added to the flaming domino event. Some busses and two hole terminals have some sharp edges so care needs to be taken. Good routing of balance leads, tie strap mounts and a dab of Shoe Goop can help hold leads in position especially for mobile applications where vibration can be a problem. Cells potted in box and wires restrained to survive a bass boat thrashing;

 

[Alex1]

I dont have any pictures before but balance wires were not secured/glued, they were lying down on batteries but didnt touch busbars, and there was a tire on top covering whole area to prevent metal object shorting across.
Here are inside pictures of bms.

 

[Alex1]

I don't think a fuse on the balance leads would have made any difference.

A lose wire on one of the batteries would do this (under the BMS) but there should be blackening on the terminal/wires and other signs of heat at the location. Can you include a photo looking down at the batteries?

Why do you think fuse would have made any difference?
Only one of all leads was melted whole lenght, i know that because i unscrew this lead first, and than others during melting, good thing was that i didnt close batteries with plexyglass as I planned.
the smoke reached the sensor quickly

I dont know what to do?
to order another bms or it was false bms causing melting.

 

[Steve_S]

This is clearly an internal failure. You can see the scorching all along the PCB circuits going up. It's not possible to see what chip failed first causing the rest without personal close up inspection. ZPhoto's like this won;t do it... Except a LOT of speculation & Theory though ! (a new passtime).

More than likely there is no harm to the battery pack cell at all, (nothing that can't be cleaned quickly anyway). A New BMS is certainly in order. This model of JK has generally been pretty stable & reliable with only occasional duds (not like this though). I cannot Recomend another with similar features (Active Balancing etc) unfortunately.

 

[Checkthisout]

This is clearly an internal failure. You can see the scorching all along the PCB circuits going up. It's not possible to see what chip failed first causing the rest without personal close up inspection. ZPhoto's like this won;t do it... Except a LOT of speculation & Theory though ! (a new passtime).

More than likely there is no harm to the battery pack cell at all, (nothing that can't be cleaned quickly anyway). A New BMS is certainly in order. This model of JK has generally been pretty stable & reliable with only occasional duds (not like this though). I cannot Recomend another with similar features (Active Balancing etc) unfortunately.

Howdy.

Why couldn't it have been a short/failure at the balance leads VS internally?

 

[rogerheflin]

The trace(s) IN the board are burned up and black fairly far into the board. That indicates that the short was inside and that something inside the BMS itself caused the short to happen. The BMS hardware connected to the BMS wires should be very high resistance (similar to a multi-meter voltage lead) and when working correctly cannot cause very much current to flow, so something had to break inside a chip or something happened (maybe a piece of metal moving around or a some trace having limited protection shorting through).

 

[BarracudaBob]

Questions if it was internal in nature:
The burned up trace isn't B- or B+ ?
Would the balance wire coming from the cell to the load resistor and transistor carry enough current to melt the wire?
The Balance wire voltage sense wire going to the A/D circuit have low enough impedance to melt the wire?

A fuse at the source (cell) on each balance lead would not have prevented a wire overcurrent issue?

 

[k490]

Could make a fuse board with micro automotive blade fuses and a custom circuit board. You can get the plugs from any electronics vender.

 

[BarracudaBob]

Could make a fuse board with micro automotive blade fuses and a custom circuit board. You can get the plugs from any electronics vender.

Pico Fuse soldered on the spade connection where the wire comes from the cell?

 

[BigGun]

I'm wondering if the balance voltage was set very low. Therefore if it was balancing for a long time at 2 amps, things could of heated up. Just a thought.

 

[Quattrohead]

Looks like an internal failure to me.

 

[rogerheflin]

Questions if it was internal in nature:
The burned up trace isn't B- or B+ ?
Would the balance wire coming from the cell to the load resistor and transistor carry enough current to melt the wire?
The Balance wire voltage sense wire going to the A/D circuit have low enough impedance to melt the wire?

A fuse at the source (cell) on each balance lead would not have prevented a wire overcurrent issue?

A/D is high impedance. There is likely a voltage divider circuit that gets the 0-60V battery range down to what the A/D accepts (probably 0-4.5 or 0-3V). There is also a probably a multiplexer component to select between the multiple different channels/wires and if that component internally shorted (in front of the A/D) then there is the potential for current to flow/short through a number of the BMS wires. The multiplexer is typically because decent A/D is a more expensive component and the A/D needs a lot of wiring to hook up, so typically there are not a A/D per channel being measured. I would not have thought a multiplexor would fail, but if the multiplexor had an minor internal defect reducing the voltage isolation/insulation and was being used at close to its voltage limits then it could eventually fail. You might disassemble the board some more and see if the component where all of the bms wires go burned, that would likely be multiplexor though it could be combined as one component with the A/D converter. If there is no external voltage divider then the A/D/Multiplexer would have had to be running at full voltage and/or have an internal voltage divider. If there is a voltage divider then the voltage divider itself would have enough impediance that the short had to be in the voltage divider.


The fuse would need to be very small. How tiny of fuses are there? I would suspect that we would want one in the say .1A range or even smaller. And the fuse itself in the wiring harness does add extra failure points.

 

[tinyt]

I still don't think the balance wires experienced over current. In the OP's picture, I highlighted in yellow the intact portion of the balance wires, if they experienced over current, the whole lenght will be burned.

 

[rogerheflin]

The taped part should have been on fire if the balance leads shorted, the damage looks like a fire came out of inside the BMS itself by the wires, and that agrees with the heavily charred internals. So this almost had to be a short from the power lead to the negative terminal that caused it.

Which lead was the power lead for the BMS? The power lead on my BMS looks large enough that it could supply enough current (10+ Amps) without having any obvious damage and that amount would be enough to cause significant internal damage. That may be were the fuse really needs to be. The A/D and/or multiplexer and voltage dividers should be pretty high impedance and more robust than the power supply parts the power lead is connected to.