Overkill BMS malfunction?

[AdamMillsap]

System Specs
~1600w array
Growatt SPF 300TL LVM-24P all in one for: MPPT charge controller; inverter; utility power management
Victron Smart shunt for redundant SOC monitoring
16 - 200 ah LifePo4 cells in 8s 2p configuration for 9600 Wh rated capacity - Top balanced before initial install
2 - 8s 100A 24v @OverkillSolar BMS modules

This system provides power in our Skoolie conversion. The system was first commissioned in May of 2021, and performed well, and reliably through three periods of heavy use, 5 months in the Summer of 2021, and 4 months in the Summer of 2022, and 3 months in the Summer of 2023. In the off-season, the system was left on, maintaining onboard 12v loads and kept a full-size fridge running.

The only issues I encountered during this period were:

• Occasionally the voltage would spike toward the end of a charge cycle, and kick the inverter, and charger off. Power cycling would return the system to operational status, and eventually, I reduced this occurrence by reducing my maximum charging current.
• In the spring of 2023, after the "standby" period throughout winter, I discovered that one bank was well below the cutoff voltage (I can't remember exact voltage now). The other bank was operating normally. I slowly recharged the ailing bank, and it performed normally throughout the Summer of 2023. I did not understand how this condition could have developed with the BMS in place, but I got busy and failed to delve deeper.

The current issue is:
I parked the bus 6 weeks ago and left the system running. When I checked the system two days ago, I found all 16 cells well below 1 volt! As you would expect, all the cells have significant swelling (this is new).

Questions:
1. Even if the Growatt shut off, and was not charging for some reason or another, shouldn't the BMS have gone into protection mode when the battery low voltage cutoff, and/or at the cell low voltage cutoff setting was reached? Isn't this precisely what the BMS is intended to prevent?
2. Is there any hope for saving these cells? That's a whole lot of money down the drain if not.

This was the first solar system I built, and reading through this forum, and watching @Will Prowse's YouTube channel was incredibly helpful and empowering. Thank you all for the education you provide through your contributions here! Since building this system I've built batteries for two golf carts we use on our farm, installed a direct drive solar irrigation system, and installed a 20kWh off-grid EG4 system to run our greenhouse. There are lots more projects planned for the future!

 

[sunshine_eggo]

• Occasionally the voltage would spike toward the end of a charge cycle, and kick the inverter, and charger off. Power cycling would return the system to operational status, and eventually, I reduced this occurrence by reducing my maximum charging current.

Likely caused by BMS charge protection event. Very common.

• In the spring of 2023, after the "standby" period throughout winter, I discovered that one bank was well below the cutoff voltage (I can't remember exact voltage now). The other bank was operating normally. I slowly recharged the ailing bank, and it performed normally throughout the Summer of 2023. I did not understand how this condition could have developed with the BMS in place, but I got busy and failed to delve deeper.

The BMS can draw power on its own. It can also fail to cut off. If the cells discharged to below 2.5V, it sustained damage.

The current issue is:
I parked the bus 6 weeks ago and left the system running. When I checked the system two days ago, I found all 16 cells well below 1 volt! As you would expect, all the cells have significant swelling (this is new).

Ouch! They are done.

Questions:
1. Even if the Growatt shut off, and was not charging for some reason or another, shouldn't the BMS have gone into protection mode when the battery low voltage cutoff, and/or at the cell low voltage cutoff setting was reached? Isn't this precisely what the BMS is intended to prevent?

I would conclude a BMS failure.

2. Is there any hope for saving these cells? That's a whole lot of money down the drain if not.

No. 1V + swelling = done.

 

[AdamMillsap]

No. 1V + swelling = done.

Because they are unsafe, or because they are unlikely to retain any valuable performance characteristics? If unsafe, what are the risks, and how do they present? Not being argumentative, just haven't been able to find reliable information about what happens with damaged cells.

 

[sunshine_eggo]

@RCinFLA can answer best.

 

[RCinFLA]

Electrolyte's stable range is about 0.1v to 4.3v of cell voltage. Outside this range it decomposes at an accelerated rate causing bloating of mostly carbon dioxide and carbon monoxide gases and hydro-carbon tars that clogged up graphite negative and LFP positive electrode pores reducing the migration path for lithium ions. This results in higher cell impedance and more terminal voltage slump under load current. Gases are not the major problem, the tars left behind on electrode surfaces are.

At low cell voltage there are other detrimental factors that happen.

Most damaging and dangerous issue is lithium dendrite growth that creates little metallic lithium needles that can grow to the point where they pierce the separator creating shorting paths between negative and positive electrodes. This begins to occur below 1.0 to 1.5vdc cell voltage. It takes some time to grow lithium dendrites which gives you a little grace period on over-discharge event.

Another dendrite grown is from copper current collector foil laminates. Typically, not as bad as lithium dendrites but it also corrodes the electrical connection between graphite neg electrode and copper current collector driving up cell impedance.

If a cell is below 2.0v you should try charging at low current, less than 0.01 C(A) rate. If there is significant dendrites their leakage will consume the low charging current preventing cell voltage from rising, without creating excessive cell heating. If cell rises above 2.5v there is a good chance the cell will be okay.

If the cell has significant dendrites and charged at normal moderate current it will cause the cell to overheat and possibly melt the separator plastic causing a sudden extreme cell short causing a lot of heat and bursting cell.

 

[AdamMillsap]

Man, you really provided some great and well-explained info here. Thanks!

If a cell is below 2.0v you should try charging at low current, less than 0.01 C(A) rate. If there is significant dendrites their leakage will consume the low charging current preventing cell voltage from rising, without creating excessive cell heating.

I put the cells in parallel and hooked them to my power supply @ 10 amps (oops) yesterday to see what would happen. The bank rose to 2.9v pretty quickly. I separated them today and measured each cell individually and found that they are all very close to 2.9v.

If cell rises above 2.5v there is a good chance the cell will be okay.

Does "ok" mean safe for normal cycling, or might perform close to specs again or.........? If they all rose to 2.9v should I try a higher charge rate or stick to 1a?

If the cell has significant dendrites and charged at normal moderate current it will cause the cell to overheat and possibly melt the separator plastic causing a sudden extreme cell short causing a lot of heat and bursting cell.

What happens when a Lifepo4 cell bursts?

I searched the forum for dendrites, and it looks like there's some good info about cell degradation under normal and abnormal conditions. Are there any specific resources you can recommend to learn about lifepo4 batteries?

 

[RCinFLA]

If you got to 2.9v and they are holding you are fine to do regular charging current.

Safe means there is not internal shorts and you are okay to do a full charge at normal charging current. As a secondary check you can make sure the cells are not warming up too much during the full charging.

LFP overheating and bursting violently blows out the vent port and sends out a strong blast of hot gases and electrolyte.

LFP cathode material differs from all the news reports on Li-Ion batteries catching fire. LFP cathode heating runaway does not generate the large amount of free oxygen that other Li-ion cell chemistries do. It does not mean they will not catch fire, but it is much less likely.

The electrolyte solvent is about as flammable as diesel oil. Most all Li-ion batteries use very similar electrolytes.

When electrolyte decomposes from thermal runaway a lot of carbon-dioxide gas is created that mixes and helps retard ignition of electrolyte solvent that is sprayed out. Again, the difference in LFP cell is the lower concentration of internal sourced oxygen from cathode material.

 

[garetwo]

I reduced this occurrence by reducing my maximum charging current.

What was it first set at and what did you adjust it to?