Pylontech US5000 Alarm sounding all lights flashing

[irmscher001]

Hi all, I need your help! Problem is Pylontech US5000 battery during discharge constant tone alarm sounds with all lights flashing. Solar Assistant shows 0 Watts, 61% state of charge, highest/lowest cell 3.3v. This has happened around 5 times, usually with a slightly increased load of about 12 Amps.

System:
DIY mostly self Installed. Batteries were charged at 49.1v SOC when I turned system on so reasonable level after delivery and storage.
22Kw - 2 x Voltronic (MPP Solar) 11Kw PIP Max in parallel.
6 x Pylontech US5000 batteries total 28.8 KWH
Independent battery cables 25mm2, via 100A NH fuses to bus bar.
Bus bars connected to inverters via 95mm2 cables through 250A NH fuses.
All input/output AC & DC power circuits equal length, inverters and batteries generally sharing load equally.
PV system being fitted soon, hence currently charging at night cheap, discharging in daytime to no lower than 55-70% generally.

I tried attaching a short video showing alarm as well as a few photos of system so far. I think video failed so the photo showing all lights on battery 4 (problem battery) are actually flashing in the photo, and sounding constant tone alarm. Ignore the solar assistant Raspberry Pie cables, not tidied yet, work in progress.

This has happened a number of times, about 5-6 times. Having read on these chat forums and manual, I read this is supposed to mean over voltage, or on forums possibly poor charge balancing between cells insufficient voltage cause.

Batteries have been charged initially each night as per Pylontech automatic settings to inverters ie 53.2v both absorption and float. I tried reducing for a few nights to reduced voltages, then settled on 52.5/51.8. I realise lower voltages were likely a bad idea for balancing. I have therefore had voltages back at Pylontech auto settings 53.2v. I'm now back at 52.5/51.8v. I have tried to give this time to ensure balancing would sort this out, yet again today it has happened yet again. I wanted to reduce SOC yo increase battery life, and thought that caused this. Now I am not sure, is this likely a faulty battery?

Usually this alarm happens when I have a load of 3000-6000 Watts, but equally spread (as proven) this is still only around 10-20A discharge when this happens. To reset I shut down batteries, start them ip again and they run fine, particularly after a charge again. It is always battery 4 the problem on top right side.

Anyone have any ideas on here what I should do? The battery is showing 7 cycles only, it has been charged slowly by grid power from inverter chargers.

Thanks for taking a look at this for me, I have built this following 12 months research and watching and reading everything I could. Generally I have tried to over engineer things and keep things safe as possible overrating cables etc. I'm not an expert, but have tried my best. Without this warning system runs great, is a complete beast, runs anything and everything I throw at it in the house. I have checked everything out with heat camera, everything running nice and cool.

Will's video's made me brave enough to try this project having seen similar, thanks Will! Now just to solve this battery issue.

 

[sunshine_eggo]

If you think this is a top balance issue, and I think it's very common with these batteries, I think you were right to switch to USE and run 51.8V. Recommend you repeat that - set Absorption to 51.8 and float to 51.7.

if this happens exclusively on discharge when the battery is at lower voltage, I don't know what to think.

 

[irmscher001]

Okay great thanks Sunshine, I'll try that again.

I have at times reduced charge amps down, I wonder if I allowed them time to balance, I think they had hours. I will try again and see if they need more time. Do you have these inverters, what is the best way to change the time period the float voltage? I didn't think that was possible, but maybe I just don't realise how yet. That's one of my final tweaking setup questions. Right now I have inverter connected to batteries but have selected user on both and 52.5/51.8v.

Am I correct to be less concerned for safety that this happens exclusively on discharge?

 

[sunshine_eggo]

Okay great thanks Sunshine, I'll try that again.

I have at times reduced charge amps down, I wonder if I allowed them time to balance, I think they had hours. I will try again and see if they need more time.

Passive balancers might only adjust SoC by 0.5-1% every 24 hours. If you can verify that all 15 cells are at 3.45V when at 51.8, then it's probably balanced.

EDIT: 51.8V was 41.8V

Do you have these inverters, what is the best way to change the time period the float voltage? I didn't think that was possible, but maybe I just don't realise how yet.

Float should be indefinite. If you mean absorption, there's no way to set a fixed time to my knowledge. I believe it uses an adaptive algorithm based on DoD and time to re-charge to X volts. When running in communication mode, these settings are ignored.

That's one of my final tweaking setup questions. Right now I have inverter connected to batteries but have selected user on both and 52.5/51.8v.

Am I correct to be less concerned for safety that this happens exclusively on discharge?

LFP safety is pretty reliable as long as the BMS are operating properly.

 

[irmscher001]

Many thanks for this. This forum is great as without it you can feel quite alone trying to figure things out. I guess I need to work out next how to connect a PC to get even more info out of this battery.

I only get so much out of Solar Assistant, but that goes down to 0.1v accuracy and reports when this happens that voltage for maximum and minimum cells are at 3.3v when the alarm sounds. That suggests that the cells are not massively at any different voltage, but maybe I need to know more accurately what every cell is doing. It also reports that max/min cell temperature is similar at 20.1-20.8°C. If it continues to do this I will have to send it back, it's barely used very new.

So after a full charge and holding batteries at 51.8v the charts and info are attached. I'll report further if I still get alarms. To be honest I expect to get it again within 48hrs fingers crossed I'm wrong. If it alarms again, I will send it back for replacement as I'm running 6 of these batteries and this is always the same one going into alarm. The others are running perfectly fine so it suggests there's something wrong with this one. Strangely when this battery has trouble it's model name in SA changes to ABSENT. Only just has it returned to US5000.

On a positive note, I am pleased at least when the alarm sounds, the BMS is disconnecting all power from/to battery, that is reassuring that it is doing its job.

 

[sunshine_eggo]

@3.5V @ 51.9V, I would not say you're dealing with a top balance issue.

I'm at a loss. An alarm at 3.3V seems very out of place.

Try swapping communication cables and see if the issue moves?

 

[irmscher001]

That's a good idea!

 

[irmscher001]

Alarm sounding and all lights flashing again after running fine all day. As soon as load increases very slightly having discharged to 55-70% SOC, away it goes.

I'll try moving the comms cables. Thanks again!

 

[irmscher001]

Battery will be being hopefully sent back.

 

[rpdom]

Many thanks for this. This forum is great as without it you can feel quite alone trying to figure things out. I guess I need to work out next how to connect a PC to get even more info out of this battery.

I only get so much out of Solar Assistant, but that goes down to 0.1v accuracy and reports when this happens that voltage for maximum and minimum cells are at 3.3v when the alarm sounds.

The batteries themselves go down to 0.001V accuracy, so it might be worth you trying to connect a PC to the main battery console port.
I use a Pi and a simple program to log the battery status from the "bat n" command to a database server every minute. The output of the bat command looks like this:

pylon>bat 1
@
Battery  Volt     Curr     Tempr    Base State   Volt. State  Curr. State  Temp. State  SOC          Coulomb      BAL         
0        3275     -3475    24900    Dischg       Normal       Normal       Normal       60%         29829 mAH      N
1        3276     -3475    24900    Dischg       Normal       Normal       Normal       60%         29830 mAH      N
2        3276     -3475    24900    Dischg       Normal       Normal       Normal       60%         29831 mAH      N
3        3275     -3475    24900    Dischg       Normal       Normal       Normal       60%         29831 mAH      N
4        3275     -3475    24900    Dischg       Normal       Normal       Normal       60%         29837 mAH      N
5        3276     -3475    25400    Dischg       Normal       Normal       Normal       60%         29831 mAH      N
6        3276     -3475    25400    Dischg       Normal       Normal       Normal       60%         29836 mAH      N
7        3276     -3475    25400    Dischg       Normal       Normal       Normal       60%         29832 mAH      N
8        3276     -3475    25400    Dischg       Normal       Normal       Normal       60%         29830 mAH      N
9        3276     -3475    25400    Dischg       Normal       Normal       Normal       60%         29836 mAH      N
10       3275     -3475    24700    Dischg       Normal       Normal       Normal       60%         29829 mAH      N
11       3276     -3475    24700    Dischg       Normal       Normal       Normal       60%         29830 mAH      N
12       3276     -3475    24700    Dischg       Normal       Normal       Normal       60%         29830 mAH      N
13       3276     -3475    24700    Dischg       Normal       Normal       Normal       60%         29831 mAH      N
14       3275     -3475    24700    Dischg       Normal       Normal       Normal       60%         29832 mAH      N
Command completed successfully
$$
pylon>

That's on a US2000C.

 

[irmscher001]

sorry for late posting a final resolution to this thread.

The end result was a return of the battery to Pylontech repairers in UK. That went smoothly actually, very easy, currier collection and return no issue.

They say they updated firmware and it solved the over voltage issue. Not sure and wouldn't know if they did anything else, but I got the battery back within around 2-3 weeks and have had no issues since. I have not updated any of the others so seems strange, but all is well. Thanks for everyone helping out above!

 

[sunshine_eggo]

sorry for late posting a final resolution to this thread.

The end result was a return of the battery to Pylontech repairers in UK. That went smoothly actually, very easy, currier collection and return no issue.

They say they updated firmware and it solved the over voltage issue. Not sure and wouldn't know if they did anything else, but I got the battery back within around 2-3 weeks and have had no issues since. I have not updated any of the others so seems strange, but all is well. Thanks for everyone helping out above!

thanks for following up!