question

Dear @DayAndNight ,

Thank you so much for your response.

It maybe due to my internet poor connection.

Now I made it as PDF and you could check and help to me.

Batteries SOC Measurement Result ( 1_3_2023 ).pdf

How to solve cell unbalanced problem for US2000 Pylontech batteries.pdf

Hello @Kyaw Thu ,

I already had a strategy in mind for you, but then it occurred to me that you are probably really 100% OFF-GRID on the ship and in pure island mode.

Nevertheless, one thing is important for me to explain. You can charge one or more Pylontech batteries with an external charger at any time and bring them into a defined state of 100% SOC.

I use a JOY-IT laboratory power supply for this purpose, which is also intended for charging rechargeable batteries and has a special plus-pole connection for this purpose. I set the voltage on this charger to 53.2 V and use e.g. 2.0 A as the current. If you use it to charge the Pylontech battery, it is controlled by the battery technology and the internal BMS of the Pylontech.

You don't have to worry about the current and voltage value. Since I deviate from the explanation of @Matthias Lange - DE . This means that the fuller the battery becomes, the less power the battery will consume. The current will decrease to a few mA and sometimes the BMS will even switch the battery to "IDLE". You can see this by the fact that all LEDs on the Pylontech go out and only the RUN LED flashes from time to time. You can also track and monitor all of this via the batteryview. As far as I understand it, in the 100% SOC state, the BMS will also balance the individual cells in the battery to each other. But that can take some time. It will all work without you ever having to change the preset amperage or voltage.

With such a procedure, you could pre-charge individual Pylontech batteries to almost 100%. If your complete stack then reaches 100% SOC again from the sun, you could re-integrate the pre-charged batteries into the stacks.

I'm just not sure if you have the appropriate charging technology and as someone who runs his house in island mode ( as a luxury problem) and always has mains available, I find it easy for me because I can switch to mains operation at any time. All things that are probably not or only with difficulty possible for you.

I thought about the following while writing and what do you think of this proposal.

Combine all functioning batteries with matching SOCs. During this time period, there may not be 5 stacks with 5 batteries, but only 4 stacks with 5 batteries or 5 stacks with 4 batteries or similar combinations.

Then you should take care of the problem batteries separately. I am truly convinced that many or even all can be brought back to a perfectly working condition.

I just checked your table. It also looks like you're getting 100% SOC with no issues.

I think the condition actually looks good. Only the batteries A1, A2, B1, B2, E1 E2 seem to cause problems for whatever reason.

The SOC81% column means that in this state you are shown a total SOC of 81% and then you read out the individual batteries with batteryview and then listed them in the table?

That would strengthen my suspicion that one, some or all of these six problem batteries are messing everything up in terms of the overall SOC. As a test, I would make five stacks of three batteries each from the 19 good batteries and then see how the whole system behaves and balances itself

If everything is going well and balanced, the next step would be to re-integrate the most promising A1 battery and the remaining four good batteries into the stacks and operate and monitor five stacks, each with four batteries.

I would subject the remaining batteries to individual treatment, later.

How old are the batteries? Maybe you also have a guarantee about Pylontech? Have you already checked that?

Kind Regard - DayAndNight

I have some suggestions.

Personally, I believe the cabling is causing you issues. Although the manual shows connecting both extremeties of positive and negative to each parallel array, I ONLY connect one positive at one end, and one negative at the other. This helps to ensure balance. ALL battery cables should be the same length, cross section and resistance.

Try removing the extra positive and negative from each parallel array.

The firmware of each unit should ideally be the same but this is not critical as long as your Master units have the latest versions in their arrays.

Some really good suggestions above.

Be aware as well that the batteries can only balance internally if they reach a charge close to 100% and remain at that level for sufficient time. If the charge rate is too high with out of balance cells, the batteries can cut out for self protection.

Guessing that you're not regularly charging to 100% and holding them there. But this does not discount the wiring issues suggested above. Could also be a combination of things.

On the environment. It would be worth checking that you're operating within the battery specifications. Hot and humid isn't a good match for any batteries.

Hello,

Thank you for this post which was very interesting. I have a similar issue of unbalanced cells in Pylontech batteries. I also wondered if too unbalanced cells can prevent the battery parc to reach 100% SOC ?