EG4 LifePower4 48V Battery showing 27V and High Pitch Buzz

[girfold]

I've got a support case with SS that I'm working with but wondered if anyone else had this issue. Out of the blue, one of the LifePower4 batteries in my array went completely dark - no Run, Alarm, or SoC light showing. Upon removing from the array, I could hear a high pitched noise, and tested the voltage with the breaker on to 27.36V - nearly exactly half of the expected voltage for the battery.

Flipping the breaker doesn't do anything, holding reset does nothing. The high pitch noise is constant, whether the breaker's on or not.

Anyone else seen this issue, or know what would cause it? Thanks!

 

[marionw]

Out of curiosity, can you communicate with the battery using a computer, RS485 to USB converter and the BMS_Tools software?

 

[EG4_Jared]

I've got a support case with SS that I'm working with but wondered if anyone else had this issue. Out of the blue, one of the LifePower4 batteries in my array went completely dark - no Run, Alarm, or SoC light showing. Upon removing from the array, I could hear a high pitched noise, and tested the voltage with the breaker on to 27.36V - nearly exactly half of the expected voltage for the battery.

Flipping the breaker doesn't do anything, holding reset does nothing. The high pitch noise is constant, whether the breaker's on or not.

Anyone else seen this issue, or know what would cause it? Thanks!

I would recommend removing this battery from the rack and utilize the BMS Test software to see if any errors are being reported with this battery. If there are no errors, I would attempt to charge it separately with a 48v Lithium battery charger.

 

[DougfromdaUP]

Have you got an update for us @girfold ?

 

[girfold]

Not long after I posted, got an email from SS about trying BMS Test and if no luck there, I can take the case off and inspect for any sign of damage. Going to try the BMS test shortly and let thread know how it goes.

 

[girfold]

BMS tool utility was not able to communicate with the battery. I gave it some time as I have had to do before with upgrading the BMS for it to connect, but no luck.

Opened the enclosure and everything looks good from what I can tell, I'm trying to see if any bulging cells, but they all look good to me. No burnt boards either. High pitched, low decibel constant noise persists, despite battery breaker off.

 

[girfold]

With the voltage pretty much exactly half of expected value (last state it was in was probably in the 90%+ range, and recently top balanced), my spidey sense is telling me it's that half of the batteries in series are not seen/linked properly. Is there a nice, safe way, and non warranty-voiding way to use my voltmeter on each of these cells to see what voltage is, or I would need to disconnect the cells first if I get that deep?

 

[42OhmsPA]

Is there a nice, safe way, and non warranty-voiding way to use my voltmeter on each of these cells to see what voltage is, or I would need to disconnect the cells first if I get that deep?

Yes, simply meter each cell individually.
You could then keep one lead on the main negative or positive and work your way through the series chain to see if voltage steps up / down each cell.

I do find these witness marks interesting since you are only getting half the voltage... I'm pretty sure it's just an ink smear but it's also the midpoint connection.

 

[girfold]

Thanks, I can take a higher res picture of that particular part if it helps out anyway.

 

[girfold]

I took a meter reading right at the image below and it was 53.25V, so there's something going on with one of the boards or precharge resistor? Seems like cells are fine.

 

[hwy17]

It could be in shutdown mode due to a low cell, and the 27v is like a phantom voltage reading through the mosfet that is trying to allow only charge and not discharge.

If that is the case, you should put a charger on it to wake it up out of shutdown. And fully charge and float it, so that it can balance out the low cell.

 

[girfold]

It could be in shutdown mode due to a low cell, and the 27v is like a phantom voltage reading through the mosfet that is trying to allow only charge and not discharge.

If that is the case, you should put a charger on it to wake it up out of shutdown. And fully charge and float it, so that it can balance out the low cell.

Hmm, I understand that some cells could be higher and some cells lower, but 53.25V / 16 cells = 3.328V which seems like a pretty reasonable number for all cells to be around +/-, also considering no load and no charge. Along with the high pitched noise coming from some component, wouldn't that point to an issue between the cell array and the terminals? i.e. the BMS / resister / that other small board. I checked the inverters historical page right before the battery went kaput and it's showing that it was at 80% for the whole battery array. At that time, the historical data is showing MaxCellVoltage 3.325V and MinCellVoltage 3.319V. Not sure if that reading is for the primary/main battery, but seems close to this particular trouble battery.

 

[hwy17]

Hmm, I understand that some cells could be higher and some cells lower, but 53.25V / 16 cells = 3.328V which seems like a pretty reasonable number for all cells to be around +/-, also considering no load and no charge.

I'm talking about the voltage after the BMS on the battery case terminals.

Along with the high pitched noise coming from some component, wouldn't that point to an issue between the cell array and the terminals?

The high pitch noise could be the sound of the mosfet operating to limit discharge.

I checked the inverters historical page right before the battery went kaput and it's showing that it was at 80% for the whole battery array. At that time, the historical data is showing MaxCellVoltage 3.325V and MinCellVoltage 3.319V. Not sure if that reading is for the primary/main battery, but seems close to this particular trouble battery.

I follow you in all that, but let's not chase bread crumbs and just get a charger on it instead.

A common problem is if you now tell us you don't have a standalone charger. Then it gets harder to troubleshoot. You would need to get a standalone charger or try paralleling it back into the bank.

 

[girfold]

I do have the 48V 18A EG4 charger, the old low amperage one and I'm not opposed to using it, just thinking things out.

So it faulted on 3-15 for about 5 hours, before I was around to see and remained in breaker on state while the rest of the array was charging, so my thought is that it would have gotten juice to wake itself up during that time. Along with my initial reset attempts were all done while it was connected to the array.

Additionally, emails with SS said that there can be a low hum for over-voltage issues. If the BMS is in an inoperable state, adding a charger could lead to overcharging since it will see thw low overall voltage and keep going?

 

[42OhmsPA]

There should be no overcharge conditions if the BMS is still in place.
Based on the series of events you describe I suspect the BMS is bad or failing.
Has SS gotten back to you?

 

[hwy17]

I do have the 48V 18A EG4 charger, the old low amperage one

Nice, sorry I assumed wrong. I'm getting a bit surly with people without a charger lately, lol.

If the BMS is in an inoperable state, adding a charger could lead to overcharging since it will see thw low overall voltage and keep going?

I never really thought about the idea that a mosfet bms couldn't block current if they are broken and the mosfets are closed (pass current) when unpowered. Seems like a big safety risk for all common BMS, I wonder if I misunderstand it or there is a backup safety.

Anyway, personally I see that as a good reason not to leave the charger on a dead battery unattended, but I would defiantly try it in your situation, while attended, for up to 15-30 minutes.

 

[girfold]

Nice, sorry I assumed wrong. I'm getting a bit surly with people without a charger lately, lol.

I never really thought about the idea that a mosfet bms couldn't block current if they are broken and the mosfets are closed (pass current) when unpowered. Seems like a big safety risk for all common BMS, I wonder if I misunderstand it or there is a backup safety.

Anyway, personally I see that as a good reason not to leave the charger on a dead battery unattended, but I would defiantly try it in your situation, while attended, for up to 15-30 minutes.

I'm certainly no Electrical Engineer, and don't know the details of the BMS. It's possible, and ideal, that upon fault an inoperable BMS will physically block charging to avoid bricking a battery, at least until the BMS can be replaced or fixed. I'm definitely newer to the scene so if that's common for faulted BMS, then I'm less worried, but yeah, to me it's like a black box.

Yeah, I'm all good with trying it out as long as SS says it won't void the warranty, which I don't see how it could since it's an EG4 charger and that's typical behaviour for a battery - to be charged.

 

[hwy17]

Yeah, I'm all good with trying it out as long as SS says it won't void the warranty, which I don't see how it could since it's an EG4 charger and that's typical behaviour for a battery - to be charged.

Nothing wrong with checking with them first. But I'm 99% percent sure they will prescribe it as the first step.

 

[hwy17]

@SignatureSolarJess Can you confirm generically that it is always ok to try connecting an EG4 18A charger to an EG4 battery in shutdown or error state? Or do you prefer that Signature support make a recommendation to the user on a case by case basis?

 

[girfold]

There should be no overcharge conditions if the BMS is still in place.
Based on the series of events you describe I suspect the BMS is bad or failing.
Has SS gotten back to you?

We've been having some daily email exchanges so far. If this was the only battery I owned, I would probably be calling them for live troubleshooting sessions, but this isn't too urgent at the moment. Posting here to get some thoughts and to help any others who run into this same issue in the future. If replacement BMS ends up being the solution, or really whatever solves the issue, will try and post the steps here that led to the battery functioning again.

On a related note, does any stock up on extra BMSes as a backup measure for their batteries? Is it possible to buy the same BMS, without the battery in case of future failures? I have a backup inverter solution since the life expectancy on it is significantly lower than the panels and batteries, but curious about backups for other components as well.