Hi All,

I am running 2x Victron multiplus II 3000kva inverters in parallel coupled, to 1 Phylontech US3000 and 2 Phylontech US2000 batteries, this was commissioned in June/July 2019. 

I unfortunately have lost a Phylontech battery (US2000) due to over charge and subsequently swollen/blown battery pack. Over the last few months the frequency of the high voltage alarms and intermittent tripping of the whole system started to become more frequent. So after lots of observations I had come to the conclusion that there was a cell in balance in the one battery as the difference between the lowest cell voltage and the highest cell voltage was in my unprofessional opinion getting too big. (3.4v - 3.6v even as high as 3.7v) It appears to me that the bms only looks at the voltage of the lowest cell in the battery and keeps charging to try up that voltage and get the battery charged, and therefore causing the other cells to over charge. 

So after returning the battery to Segan Solar, they downloaded the battery logs and replied that it had been over charged to above 54v and therefore the warranty is void. I am obviously disappointed with the outcome as the Victron setup is set to 53.2v and is pretty much set and forget. The weird thing to me is that the very first entry on the battery log shows over charge of just above 54v and over the 4 years there were about 5 or 6 days that this occurs. Surely if the voltage is set too high then this should be consistent through out all the log entry's? 

The other weird thing is that before I returned the battery to Segan solar they requested a video of what the system was doing. And what's weird to me is that on this particular day that I sent video proof of what the system was doing there is no overload recorded on the battery log for that day, yet the system sent through multiple overload alarms and eventually tripped completely. If seems to that the bms and the Victron software are not aligned in the way they record the battery voltage? 

I can only seem to go back 6 months on the VRM data logs and these are all below the 54v mark. Is there any way of getting all the logs from the date of commission? 

So is this a Phylontech error or a Victron software error? Or maybe it's something I did. 

I would appreciate any help or recommendations from those of you who are more knowledgeable than me. @P1000 maybe? Or any Victron experts? 

Event Data (1).xlsx

Looking at your logs you seem to have drawn some very large currents from this pack, way above those recommended. Did you have communication set up between the batteries and the inverter?

Edited October 10, 20231 yr by Tinbum

How was these batteries wired up ? Do the logs of the other batteries also show those 50-60A discharge currents ? It almost seems like your packs were not sharing current equally as 60Ax3 would be a 180A draw which would be very unlikely from your two inverters. It doesn't look like the overcharging bothered it as much as those large currents. Your batteries are very much under spec for the system as the three of them are only rated for 87A continuous draw. Edit* I have also seen on other forums that Victron inverters can be quite Amp heavy and they recommend over-sizing the battery bank.

Edited October 10, 20231 yr by Nexuss

16 minutes ago, Tinbum said:

Looking at your logs you seem to have drawn some very large currents from this pack, way above those recommended. Did you have communication set up between the batteries and the inverter?

1 x US3000 recommended discharge 37A, 1 x US2000 recommended discharge 25A, second US2000 recommended discharge 25A, Total gives me 87A recommended discharge. Look at the logs again and see if there is any day OP came even close to 87A draw from his battery bank. Op has the same issue as the other guys were having with segen, the supplier just want a way out of honouring the warranty claim. 

@Kilowatt It looks like the last time you had over voltage was on the 17th Jan 2021 and for the past two years and eight months you were ok. There is no data for the past two months, did you change any setting on your inverter which could have led to the batteries getting swollen? Because if it was due to OV, the issue would have occurred in 2019 when you had series of OV. Not three years down the line 

1 hour ago, Kilowatt said:

 It appears to me that the bms only looks at the voltage of the lowest cell in the battery and keeps charging to try up that voltage and get the battery charged, and therefore causing the other cells to over charge

 

They don't work like that.

1 minute ago, hoohloc said:

1 x US3000 recommended discharge 37A, 1 x US2000 recommended discharge 25A, second US2000 recommended discharge 25A, Total gives me 87A recommended discharge. Look at the logs again and see if there is any day OP came even close to 87A draw from his battery bank.

The log is for a single battery and it shows DOCA error.

2 minutes ago, Tinbum said:

The log is for a single battery and it shows DOCA error.

Only seeing now🤣

OP, you were hammering this battery in the first year of use. Was it the only one you had at the time? and then added the other two later on. This is the only explanation I can give with just having a look at the logs. You will not win this one with Segen 

36 minutes ago, hoohloc said:

OP, you were hammering this battery in the first year of use. Was it the only one you had at the time? and then added the other two later on. This is the only explanation I can give with just having a look at the logs.

Ah yes that would make sense.

@Kilowattsorry to hear of your battery & BMS problem. Very fascinating that so many Pylontechs are showing up with these issues. It's a problem across inverter equipment. The BMS should really be able to take the punch & not allow higher than required voltage/ amps. It's called a Battery management system. Not battery reporting system. Seems the Pylontech has a BRS. 

Once the damage is done It reports the problem & keeps a log. If only Pylontech invested as much time & effort into what the device should actually do. Protect the dam cells. 

 

9 minutes ago, Steve87 said:

Very fascinating that so many Pylontechs are showing up with these issues.

Yip definitely a concern.

10 minutes ago, Steve87 said:

The BMS should really be able to take the punch & not allow higher than required voltage/ amps. It's called a Battery management system. Not battery reporting system. Seems the Pylontech has a BRS.

Agreed with that statement fully.

49 minutes ago, Steve87 said:

@Kilowattsorry to hear of your battery & BMS problem. Very fascinating that so many Pylontechs are showing up with these issues. It's a problem across inverter equipment. The BMS should really be able to take the punch & not allow higher than required voltage/ amps. It's called a Battery management system. Not battery reporting system. Seems the Pylontech has a BRS. 

Once the damage is done It reports the problem & keeps a log. If only Pylontech invested as much time & effort into what the device should actually do. Protect the dam cells. 

 

"The BMS should really be able to take the punch & not allow higher than required voltage/ amps."

Yes , that's my point exactly.  My take on this is that the BMS is pre-configured in order to serve its mandate , that is to protect the battery chemistry from critical over charging and over voltage.  If you even have the intention to 'hammer' your battery , then the BMS must be able to protect the chemistry from the 'hammering' . If a number of protection states have been recorded in succession , the BMS should passify and allow no more until the user is made aware of it  . 

In the end , the only way a owner should be able to 'hammer' a lithium battery , is to use up cycles quickly , but that should be it ! If cells become swollen , the BMS has not done its work , thats it.

Edited October 10, 20231 yr by BritishRacingGreen

59 minutes ago, BritishRacingGreen said:

"The BMS should really be able to take the punch & not allow higher than required voltage/ amps."

Yes , that's my point exactly.  My take on this is that the BMS is pre-configured in order to serve its mandate , that is to protect the battery chemistry from critical over charging and over voltage.  If you even have the intention to 'hammer' your battery , then the BMS must be able to protect the chemistry from the 'hammering' . If a number of protection states have been recorded in succession , the BMS should passify and allow no more until the user is made aware of it  . 

In the end , the only way a owner should be able to 'hammer' a lithium battery , is to use up cycles quickly , but that should be it ! If cells become swollen , the BMS has not done its work , thats it.

Sadly i think the Pylon BMS is just very outdated at this point,i doubt they have done much in terms of development over the years to improve it , if you look at the latest BMS's in the market they offer lots more flexibility and features . JK , Seplos v3 ect

The thing is the pylontech bms relies on the inverter following it's instructions.

9 hours ago, Tinbum said:

The thing is the pylontech bms relies on the inverter following it's instructions.

Thats potentially part of the problem , it may be relying too much on that . It should work 100% without ANY comms. Many people only use voltage without BMS comms. Some people have many different batteries operating together(Andy from off-grid garage as example) and the only way to do that is via voltage.

17 hours ago, Tinbum said:

Looking at your logs you seem to have drawn some very large currents from this pack, way above those recommended. Did you have communication set up between the batteries and the inverter?

Yes have always had communications setup between batteries and the inverter, as per the Victron recommendations. 

3 hours ago, Nexuss said:

Thats potentially part of the problem , it may be relying too much on that . It should work 100% without ANY comms. Many people only use voltage without BMS comms. Some people have many different batteries operating together(Andy from off-grid garage as example) and the only way to do that is via voltage.

That may be the case but it does say in their warranty terms that the batteries may only be used with approved inverters. I would never run them without comms being set up and working.

2 minutes ago, Kilowatt said:

Yes have always had communications setup between batteries and the inverter, as per the Victron recommendations. 

I thought the Victron system reduced the max charge voltage to 52.4v or s that just some of their equipment and not all?

Edited October 11, 20231 yr by Tinbum

16 hours ago, Nexuss said:

How was these batteries wired up ? Do the logs of the other batteries also show those 50-60A discharge currents ? It almost seems like your packs were not sharing current equally as 60Ax3 would be a 180A draw which would be very unlikely from your two inverters. It doesn't look like the overcharging bothered it as much as those large currents. Your batteries are very much under spec for the system as the three of them are only rated for 87A continuous draw. Edit* I have also seen on other forums that Victron inverters can be quite Amp heavy and they recommend over-sizing the battery bank.

The batteries are wired according to the victron manual for phylontech. I unfortunately don't have the data logs for the remaining batteries. But thank you for pointing the high discharge currents out to me. My initial setup was just the 2 phylontech us2000 batteries and 1 Victron 3000va inverter, the 3rd battery and second inverter were installed in the beginning of 2021. I have never just had this one battery running on its own so I think this is really strange. This setup runs my house and we are not an energy hungry house hold we use on average about 450 -500kw a month. But, I have just looked at the current draw for the past 6 months and I can see that there are many instances where the current draw is high, 116amps being the highest. So this could be the problem, I will look for the setting to limit this. 

17 hours ago, hoohloc said:

OP, you were hammering this battery in the first year of use. Was it the only one you had at the time? and then added the other two later on. This is the only explanation I can give with just having a look at the logs. You will not win this one with Segen 

My initial setup comprised of 2 x phylontech 2000 batteries and 1 victron inverter 3000va. It is true that the batteries have worked hard, but I have never only used just this one battery on its own. I suppose I have always thought I would not get 10 years out of these batteries, but I did not expect them to fail outright. I thought the capacity would just reduce quicker than expected time frame. Expensive lesson, but I agree i am not going to win this fight. 

11 minutes ago, Kilowatt said:

My initial setup comprised of 2 x phylontech 2000 batteries and 1 victron inverter 3000va. It is true that the batteries have worked hard, but I have never only used just this one battery on its own. I suppose I have always thought I would not get 10 years out of these batteries, but I did not expect them to fail outright. I thought the capacity would just reduce quicker than expected time frame. Expensive lesson, but I agree i am not going to win this fight. 

Before giving up, contact Pylontech directly and see if they can help. Some times Segen just repudiate the claims only for Pylontech to accept the failure and honour the claim. With Segen, you will not win because they are only focusing on the events from the first year of use, when the battery failed almost three years later. To be honest, they should focus on what happened at the time of failure. It is good to look at the past history just to have an idea of how the battery was used but they still need to find out why it failed and also if you had two of them and the other one is still ok, then the issue is clearly with the battery and not how it was being used. This is just my honest opinion 

Maybe pull the logs from the other US2000 and share with Pylontech logs from both batteries.

@Kilowatt, sorry to hear that :(.

Before going to voltage only, I was using the BMS communication to the inverter to manage the charging. The batteries got every day to 53.2 which is regarded as too high by the community. I've read on this forum that balancing happens at 3.48V (52.2V) which in case of us5000 is certainly not true, as I see a 30mV cell imbalance still after two weeks. I have 3xus5000. One battery is almost new (4cycles). I've charged it alone, and the other two, set them in parallel for 1h and then connected to the inverter. The new battery is set as master and when using BMS communication and the battery going to 53.2V, I can see that the BMS temperature of that battery is rising trying to balance the cells. I guess it's burning voltage of the higher cells while trying to charge the other(s). I can see that the higher voltage cell can reach 3.64V while the lower is trying to go above 3.5V. The cells in the other two modules are sitting at 3.54V. 

I went Pylontech for the sake of a solid BMS. I wish I didn't.

Maybe the Voltronic (Easun) bus is overreading the voltage, and the battery is not actually at 52.2V but lower, hence balancing will not happen there.

Edited October 11, 20231 yr by onobeka

35 minutes ago, onobeka said:

@Kilowatt, sorry to hear that :(.

Before going to voltage only, I was using the BMS communication to the inverter to manage the charging. The batteries got every day to 53.2 which is regarded as too high by the community. I've read on this forum that balancing happens at 3.48V (52.2V) which in case of us5000 is certainly not true, as I see a 30mV cell imbalance still after two weeks. I have 3xus5000. One battery is almost new (4cycles). I've charged it alone, and the other two, set them in parallel for 1h and then connected to the inverter. The new battery is set as master and when using BMS communication and the battery going to 53.2V, I can see that the BMS temperature of that battery is rising trying to balance the cells. I guess it's burning voltage of the higher cells while trying to charge the other(s). I can see that the higher voltage cell can reach 3.64V while the lower is trying to go above 3.5V. The cells in the other two modules are sitting at 3.54V. 

I went Pylontech for the sake of a solid BMS. I wish I didn't.

Maybe the Voltronic (Easun) bus is overreading the voltage, and the battery is not actually at 52.2V but lower, hence balancing will not happen there.

The balancing current in the BMS is very very low and it will take a long time to balance. It also only starts to balance above 30mv.

I regard 53.2v as too high and alter the CAN messages down to 52.5v except for one day a week. Pylontech must be happy with their 53.2v though.

41 minutes ago, onobeka said:

set them in parallel for 1h and then connected to the inverter.

Maybe that was too short a time to balance the batteries more would have been better.

All the pylontech batteries I have connected have had Over voltage issues which usually settle down after a few days/charge/discharge cycles.

9 hours ago, FixAMess said:

Maybe that was too short a time to balance the batteries more would have been better.

All the pylontech batteries I have connected have had Over voltage issues which usually settle down after a few days/charge/discharge cycles.

I concur. Had two Pylontech US2000 which I took to Segen for firmware update, and the logs also showed overvoltage at the beginning. A flimsy reason to repudiate a warranty claim, in my opinion.