I have a kodak 7.2 inverter with 4x 5.12 kw Shoto batteries. I recently started to use solar assistant to monitor my system. I notice that the batteries discharge enevenly. Battery pack 1&4 was installed in March 2023.

Battery pack 2&3 was installed in Dec 2022.

What could be the problem?

Edited August 12, 20232 yr by Andries Steyn
Added shoto battery

1 hour ago, Andries Steyn said:

I have a kodak 7.2 inverter with 4x 5.12 kw Shoto batteries. I recently started to use solar assistant to monitor my system. I notice that the batteries discharge enevenly. Battery pack 1&4 was installed in March 2023.

Battery pack 2&3 was installed in Dec 2022.

What could be the problem?

Was the picture taken during a discharge or charge?

What is the max charge voltage that your batteries reach? 57.6V?

You probably know this already but it is not a good idea to mix old and new batteries as you will get the performance of the oldest battery in your system.

1 hour ago, HendrikBigChief said:

Was the picture taken during a discharge or charge?

What is the max charge voltage that your batteries reach? 57.6V?

You probably know this already but it is not a good idea to mix old and new batteries as you will get the performance of the oldest battery in your system.

Not true with Lithium batteries, you can mix old and new 

11 minutes ago, hoohloc said:

Not true with Lithium batteries, you can mix old and new 

Nope,
 

"but be aware that as a lithium-ion battery ages its internal resistance increases.

When you wire an old and new cell in parallel the new cell will initially supply more of the load current. This causes it to discharge faster until both cells share the load equally. When load current is reduced the old cell will charge the new cell for a while until they equalize again. When charging a similar effect occurs."

 

 

I'm surprised that in 4 months usage beyween the old and new, the batteries are so different. What are they like when fully charged and balanced?

Try making no 1,4 batteries 2,3 in the parallel chain. It's your first and last battery thst are taking more load. Swap 2 ,3 to 1, 4.

Are you able to check individual cell voltages? 

Edited August 12, 20232 yr by FixAMess

I do have a similar issue. I bought 4 LEOCH TB48100 batteries gradually within a few months and connected them all in parallel. But the last one added looks quit different. It appears to have significant lower internal resistance than the others, it has higher charge and discharge currents. I attribute this to  improved manufacturing, confirmed by the BMS reporting 103Ah capacity as compared to 100Ah of the others. It looks however that slowly the issue improves, the differences are getting less. It could well be that due to higher stress it ages faster and internal resistance is rising until it finally matches the others. My guess.

For information here the cycles counts: 992, 907, 819, 677.

Edited August 12, 20232 yr by Beat

2 hours ago, FixAMess said:

 

 

I'm surprised that in 4 months usage beyween the old and new, the batteries are so different. What are they like when fully charged and balanced?

Try making no 1,4 batteries 2,3 in the parallel chain. It's your first and last battery thst are taking more load. Swap 2 ,3 to 1, 4.

Are you able to check individual cell voltages? 

Hi, yes I can check the individual cell voltages. The original picture was taken during discharging. 

See the latest picture as the batteries are currently charging. Even then there is a difference

Interesting..  I have 4 pylontechs, all of different ages and I don't see such big differences...I need to check! Maybe they are different.

When last were these top balanced?

The wire lengths are also very important. A battery can only do as good as it can do but the wire lengths can cause all sorts of issues. 

2 hours ago, Andries Steyn said:

Hi, yes I can check the individual cell voltages. The original picture was taken during discharging. 

See the latest picture as the batteries are currently charging. Even then there is a difference

Interesting, makes sense that the two new ones charge faster.

On 2023/08/12 at 1:42 PM, Andries Steyn said:

See the latest picture as the batteries are currently charging. Even then there is a difference

FYI this is not the best time to look at the difference as batteries 2&3 will currently be in balancing mode (over 98%) and the BMSs will be physically limiting the current to those 2 batteries which is why there is a lot less power flowing into them. At this stage batteries 1&4 should charge faster to catch up, but I'm not sure what happens when batteries 2&3 reach 100% as their BMS will stop charging and I'm not sure if that will also stop batteries 1&2 from charging before they reach 100%.

On 2023/08/12 at 3:17 PM, mzezman said:

When last were these top balanced?

I agree with @mzezman, the batteries seem to be quite far apart, were they all at 100% when originally hooked up? I would suggest disconnecting batteries 2&3, charge batteries 1&4 to 100% and then disconnect them and connect only batteries 2&3 and charge them to 100% and then hook up all the batteries again. Then observe from there to see if they start to drift apart again.

30 minutes ago, jumper said:

FYI this is not the best time to look at the difference as batteries 2&3 will currently be in balancing mode (over 98%) and the BMSs will be physically limiting the current to those 2 batteries which is why there is a lot less power flowing into them. At this stage batteries 1&4 should charge faster to catch up, but I'm not sure what happens when batteries 2&3 reach 100% as their BMS will stop charging and I'm not sure if that will also stop batteries 1&2 from charging before they reach 100%.

I agree with @mzezman, the batteries seem to be quite far apart, were they all at 100% when originally hooked up? I would suggest disconnecting batteries 2&3, charge batteries 1&4 to 100% and then disconnect them and connect only batteries 2&3 and charge them to 100% and then hook up all the batteries again. Then observe from there to see if they start to drift apart again.

No need to charge them separately. Once some of the batteries are fully charged, the other ones will continue charging until all 4 are charged as long as the charging voltage is high enough. The ones that are full will simple stop taking any extra current. The cool thing about parallel batteries is that they auto balance. With serial that is not the case.

This is an interesting video showing a similar outcome.

https://youtu.be/JcV6IIFqKt0

3 Pylontech U3000 installed 4years ago. When charging SOC is aqual when discharging to 50% there is a difference of up to 4% of the SOC. 

   Amount of cycles over 4 year     No1  536                                      No2   524                                        No3   569 The difference is slowly increasing over time .They say this could be the current sensor and can be calibrated.

SOH 95%    capacity 210.1Ah The system ie OK so will leave it as is.

On 2023/08/12 at 3:17 PM, mzezman said:

When last were these top balanced?

How do I balance them? 

43 minutes ago, Andries Steyn said:

How do I balance them? 

You shouldn't have to.

 

In you last picture it's not surprising to see large differences in current as 2 and 3 are nearly full. In you first post I wouldn't expect to see such differences.

Batteries 2 and 3 are certainly being worked harder.

Can you post a photo of your wiring from the batteries to the inverter?

Edited August 14, 20232 yr by Tinbum

1 hour ago, Tinbum said:

You shouldn't have to.

 

In you last picture it's not surprising to see large differences in current as 2 and 3 are nearly full. In you first post I wouldn't expect to see such differences.

Batteries 2 and 3 are certainly being worked harder.

Can you post a photo of your wiring from the batteries to the inverter?

As requested herewith a photo of the battery connection

10 minutes ago, Andries Steyn said:

As requested herewith a photo of the battery connection

That looks ok from a balancing point of view. (It looks like a Seplos BMS in them).

If you are worried you could try using just the new batteries for a little while (for about the same length of time you used the old ones). That may equalize them a bit.

On 2023/08/12 at 1:42 PM, Andries Steyn said:

Hi, yes I can check the individual cell voltages. The original picture was taken during discharging. 

See the latest picture as the batteries are currently charging. Even then there is a difference

Take a screenshot of the battery voltages. Do they differ also from each other?

4 minutes ago, Jacques Ester said:

Take a screenshot of the battery voltages. Do they differ also from each other?

The first post shows them all the same.

Sorry missed that one. It seems the way these BMS are balancing between different batteries is via voltage and not State of charge. I have seen the same thing happen to my 3 batteries. It will close the gap as the batteries become more fully charged. From about 90% the lagging batteries will catch up and all will show 100% at almost the same time.

Edited August 14, 20232 yr by Jacques Ester

3 minutes ago, Jacques Ester said:

Sorry missed that one. It seems the way these BMS are balancing between different batteries is via voltage and not State of charge. I have seen the same thing happen to my 3 batteries.

Their is a way to reset the SOC on a Seplos BMS but I don't have time to look it up at the moment. It varies between model numbers.

 

Edited August 14, 20232 yr by Tinbum

Pylontech batteries does the same thing.

16 hours ago, Jacques Ester said:

It seems the way these BMS are balancing between different batteries is via voltage and not State of charge.

Yes, there is no balancing between batteries, they are just connected in parallel, so they have the same voltage. What theoretically happens is that they will all be at 100% when charging ends, and they should be close to the same at their lower cutoff, so near the low voltage cutoff. In between the extremes, the variations can be quite big between batteries.

It does not quite work that way in practice, which is one of the reasons pylontech lowered their minimum SoC to 5%.

Edited August 15, 20232 yr by P1000