Extremely overheated BMS, fused Breaker and dead Inverter. What could be the root cause?

[zedconnor]

EDIT : All the 16 FETS in the Inverter got blown. And replaced. The setup seems to be running fine with a new BMS at the moment.


So this happened today.

Inverter is MUST 5.5kW
Battery LFP 51.2V 100Ah (DIY)
BMS Daly Smart BMS 150A
ONCCY DC Breaker 2Pole 125A(non-polarized)

The breaker tripped. He tried to switch it to ON again but breaker tripped again. There was smell of burning from battery pack. So he called me.

When i arrived, I immediately disconnected all wires between inverter, breaker and battery, and then checked the battery. The foams used as top cover protection on the battery got heated and burned.
The BMS was extremely hot that it burned my finger as soon as i touched it.

BMS is still ON. I can see the cell voltages and stuff on the LCD display of battery.

Every wire connected between the bms and the cells are all fine. There is not a single sign of a short-circuit burn. The only thing happened is the BMS got very very heated and burned the protective foam around it.

Now onto the breaker,
The breakers contacts are fused. So it is CLOSED and cannot be OPEN by the switch.
Somehow this blown the mosfets of the inverter and caused error 09 on the Inverter.

I ended up swapping a new bms for the pack.

What i am wondering is what caused what in this event. Again i cannot find any sign of a short-circuit. All contact points and terminals, lugs are all in very clean condition. Not even a single mark of burn or anything.

Also the BMS did not shutdown when it should as it is way beyond overheating. It melted its own rubber coating.

The battery pack is less than a month old. So is the DC breaker.

what could be the root cause for this?

EDIT :
The system has been working fine for about 20days.

 

[zedconnor]

2P 200V 63A125A Non-polarity DC Mini Circuit Breaker MCB

<strong>Product Benefits</strong> <ul> <li>Overload and the short circuit protection function</li> <li>Rated short circuit breaking capacity up to 10 kA</li> <li>Non-polarity and Polarity both available</li> <li>Rated Current In up to 125A</li> <li>Rated Voltage up to 200V</li>...

onccy.com

This is the link of the breaker that malfunctioned. I thought it is a decent quality breaker.

 

[zedconnor]

This is the inverter.
Exact model is
MUST PV18 5548 VHM

 

[solar8484]

The inverter FET's likely shorted first causing over current. But that doesn't excuse the failure of the breaker and BMS to cut off the battery.

 

[zedconnor]

The inverter FET's likely shorted first causing over current. But that doesn't excuse the failure of the breaker and BMS to cut off the battery.

There was an unusual rain yesterday with a few thunder, and this house has no grounding system.
Could it be a cause?

 

[Hedges]

Somehow this blown the mosfets of the inverter and caused error 09 on the Inverter.

Shorted FETs would cause all the rest. As Solar beat me to (failed breaker is it's own fault, unless not rated high enough for available fault current.)

Lack of grounding probably doesn't cause damage, but lightning could. Typically due to grid connection leading in to electronics from overhead powerlines, but can be inductively coupled to your wires if close enough.

Might not be the lightning. Maybe bigger loads. Or, not running long, infant mortality or just weak design.

 

[solar8484]

Lightning could certainly have caused the damage. It's probably a good idea to add a ground rod for the inverter.

 

[zedconnor]

Lightning could certainly have caused the damage. It's probably a good idea to add a ground rod for the inverter.

I checked all the cells before re-assembly. There was no sign of damage. Even the blue plastic films are intact. No swallowing or buldging.
Cell voltages of all 16 cells are within 50mV delta.
And one important thing I forgot to mention is all of the cell voltages are around 3.35V only when I arrived to the scene. No overvoltage or undervoltage.


So based on your reply,

- lightning caused damage to the Inverter,
- which pushed alot of current to the breaker
- causing its contacts to fuse.
- the breaker tripped but the welded contacts allowed the current to continue to flow, thus overheating the BMS?

If that is the case, wouldnt that also destroy the cells or atleast overcharge it?
(There is no current going into the battery when I arrived. The bms itself, hot as hell.)
Or is the BMS taking all those currents and heated itself without releasing them into the cells?
That bms is still functional btw. Altho i have no plan of using it any further.


I want to avoid further incidents like this.

 

[pabla007]

Can you check what is your firmware on your BMS

 

[Alkaline]

So this happened today.

Inverter is MUST 5.5kW
Battery LFP 51.2V 100Ah (DIY)
BMS Daly Smart BMS 150A
ONCCY DC Breaker 2Pole 125A(non-polarized)

The breaker tripped. He tried to switch it to ON again but breaker tripped again. There was smell of burning from battery pack. So he called me.

When i arrived, I immediately disconnected all wires between inverter, breaker and battery, and then checked the battery. The foams used as top cover protection on the battery got heated and burned.
The BMS was extremely hot that it burned my finger as soon as i touched it.

BMS is still ON. I can see the cell voltages and stuff on the LCD display of battery.

Every wire connected between the bms and the cells are all fine. There is not a single sign of a short-circuit burn. The only thing happened is the BMS got very very heated and burned the protective foam around it.

Now onto the breaker,
The breakers contacts are fused. So it is CLOSED and cannot be OPEN by the switch.
Somehow this blown the mosfets of the inverter and caused error 09 on the Inverter.

I ended up swapping a new bms for the pack.

What i am wondering is what caused what in this event. Again i cannot find any sign of a short-circuit. All contact points and terminals, lugs are all in very clean condition. Not even a single mark of burn or anything.

Also the BMS did not shutdown when it should as it is way beyond overheating. It melted its own rubber coating.

The battery pack is less than a month old. So is the DC breaker.

what could be the root cause for this?

EDIT :
The system has been working fine for about 20days.

Ohh Look its a Daly... What a surprise

 

[wpns]

No ventilation in the battery box?

 

[zedconnor]

Ohh Look its a Daly... What a surprise

Can you suggest me which brand to use?

No ventilation in the battery box?

Yes bro. Maybe a few tiny holes. Aside from that there is no vents. Chinese-branded packs also do not have any vents. They even advertise them with IP ratings.

 

[DIYrich]

I'm guessing the problem started at the inverter, and the control section of the bms sacrificed itself to break the connection. What is the Batt voltage at the battery, and at the battery connection of the inverter?

If it was over current, Class T fuse may or may not have saved the bms. Enough current could have flowed to fry the bms before the class t fuse blew. If it were over voltage (lighning strike voltage), that could fry everything.

Note: the monitoring side of the bms seems to be protected by the control side of the bms. That would explain why the monitoring side still works. If the bms failed, it would either short out, and nothing gets to the inverter, or it passes the battery without control. Either way, it shouldn't cause the inverter problems.

 

[Brucey]

Can you suggest me which brand to use?


Yes bro. Maybe a few tiny holes. Aside from that there is no vents. Chinese-branded packs also do not have any vents. They even advertise them with IP ratings.

Even the Chinese packs don't sit the BMS on top of the cells like that I don't think? often attached to one of the sides for passive cooling thru the chassis. All that heat feeding into that bank of cells...

 

[Brucey]

Do you have insulating material between, under and where the cells could touch the enclosure?

 

[OffGridForGood]

5500W inverter
if pulling max current on a 48v battery is approx 115A - well under the 150A rating
The BMS overheating may be poor connection or undersized wires connecting it to the loads.
The breaker tried to trip, but fused closed - so something pulled higher than normal current, and once the breaker fused there was nothing to cut off the circuit - heating the BMS up more and more until you arrived. (foam around the cells and BMS = bad idea, the BMS relies on passive cooling)

I have no idea about lightning as a cause - but I would start with looking into what loads were operating at the time and checking all the connections, cable sizes looking for why 115A load would overheat the BMS - the foam pading seems like a likely contributing cause.

 

[Alkaline]

Can you suggest me which brand to use?

JBD and supplement with an active balancer.

 

[solar8484]

If that is the case, wouldnt that also destroy the cells or atleast overcharge it?
(There is no current going into the battery when I arrived. The bms itself, hot as hell.)
Or is the BMS taking all those currents and heated itself without releasing them into the cells?
That bms is still functional btw. Altho i have no plan of using it any further.

Don't know your BMS specs but some BMS can control charging and discharging separately. So, perhaps your BMS disabled charging (based on voltage) but not discharging?

 

[zedconnor]

Even the Chinese packs don't sit the BMS on top of the cells like that I don't think? often attached to one of the sides for passive cooling thru the chassis. All that heat feeding into that bank of cells...

Do you have insulating material between, under and where the cells could touch the enclosure?

Yes I have put a layer of double-sided tape (about 3mm thickness) as well as a PP-board between the bms and the cells. (The PP Board was able to withstand the heat)


I have attached the re-packed pictures. This time I screwed down the bms above the cells.
There will be no more foams.

5500W inverter
if pulling max current on a 48v battery is approx 115A - well under the 150A rating
The BMS overheating may be poor connection or undersized wires connecting it to the loads.
The breaker tried to trip, but fused closed - so something pulled higher than normal current, and once the breaker fused there was nothing to cut off the circuit - heating the BMS up more and more until you arrived. (foam around the cells and BMS = bad idea, the BMS relies on passive cooling)

I have no idea about lightning as a cause - but I would start with looking into what loads were operating at the time and checking all the connections, cable sizes looking for why 115A load would overheat the BMS - the foam pading seems like a likely contributing cause.

They only have a single 1HP Aircon and lighting to use. The normal usage may not be the problem.
I am guessing it's the Inverter.

 

[zedconnor]

JBD and supplement with an active balancer.

What about JK BMS. I am liking the latest model PB2A16S (the one on youtube reviewed by Andy)
What would you prefer between JK and JBD?