On one of my Shoto SDC10-Box5 16s 5.12kWh batteries I have one cell (Cell 5) whose voltage is always lower than all the other cells when battery is reaching full charge. After battery has discharged for a while its voltage falls inline with the other cells and stays inline until the next day (or time) when charging nears full again. All the other cell voltages are +- the same all the time. This cell still holds a good charge, 3.4V, but not as a high as the other cells at charge end. 

Currently my charge voltage is set to 56.4V (asserted via a workaround, previously the charge voltage was 57.8V (still need to get to bottom of this high charge voltage)).

Is this a bad cell? And is this a precursor to problems down the line?

Please see graph showing cell voltages and SOC over a 10 hour period. Cell 5 is the blue line, all the other cells have there lines cluttered over each other as they are +- the same.

PS: Forgot to mention that my batteries is two months old. And this specific battery is the slave battery. DOD is typically +-50% in the morning. With the lowest being 31% only once.

6 minutes ago, jgdt said:

57.8V (still need to get to bottom of this high charge voltage)).

Was that the voltage that the pack requested? If so, change it back.

10 minutes ago, jgdt said:

On one of my Shoto SDC10-Box5 16s 5.12kWh batteries I have one cell (Cell 5) whose voltage is always lower than all the other cells when battery is reaching full charge. After battery has discharged for a while its voltage falls inline with the other cells and stays inline until the next day (or time) when charging nears full again. All the other cell voltages are +- the same all the time. This cell still holds a good charge, 3.4V, but not as a high as the other cells at charge end. 

Currently my charge voltage is set to 56.4V (asserted via a workaround, previously the charge voltage was 57.8V (still need to get to bottom of this high charge voltage)).

Is this a bad cell? And is this a precursor to problems down the line?

Please see graph showing cell voltages and SOC over a 10 hour period. Cell 5 is the blue line, all the other cells have there lines cluttered over each other as they are +- the same.

That cell may simply have a very slightly higher internal resistance or the terminal connection may be just slightly off perfectly torqued.
That should not be too much of a concern and you will most likely only notice it at the point where the battery is close to or at 3.65v or close to the bottom end of 2.5v

You should set your voltages a bit lower down to about 3.5v or 56v pack voltage. There is no need or reason to charge higher than that for normal cycling. Find out what voltage your bms needs to start balancing and set the voltage as close to that as possible.
Mine is set all the way down to 3.4v but that's due to my BMS allowing balancing at lower voltages.

Edited March 21, 20232 yr by WannabeSolarSparky

35 minutes ago, jgdt said:

On one of my Shoto SDC10-Box5 16s 5.12kWh batteries I have one cell (Cell 5) whose voltage is always lower than all the other cells when battery is reaching full charge. After battery has discharged for a while its voltage falls inline with the other cells and stays inline until the next day (or time) when charging nears full again. All the other cell voltages are +- the same all the time. This cell still holds a good charge, 3.4V, but not as a high as the other cells at charge end. 

Currently my charge voltage is set to 56.4V (asserted via a workaround, previously the charge voltage was 57.8V (still need to get to bottom of this high charge voltage)).

Is this a bad cell? And is this a precursor to problems down the line?

Please see graph showing cell voltages and SOC over a 10 hour period. Cell 5 is the blue line, all the other cells have there lines cluttered over each other as they are +- the same.

I have a similar situation since day one with LBSA battery also 1 cell slightly lower that the other 15. 

42 minutes ago, P1000 said:

Was that the voltage that the pack requested? If so, change it back.

Thanks. Not 100% sure where the original 57.8V comes from, or why its so high. Not sure if its send directly by the BMS (pack requested) to inverter or whether the inverter calculates it from other communications/parameters with the BMS? The Shoto batteries uses SeplosBMS. From another thread it would seem that the consensus is that if the inverter shows 57.8V charge voltage on LiBMS page then it receives it directly as is from the BMS. In which case it feels that if the BMS requested this 57.8V to begin with then just put it back and let it be. But, the batteries data specification sheet specifies 56.4V max. So which one is it? I've been trying to get hold of Shoto's technical support to clarify this out, but no success in getting support channel information. I guess I can try SeplosBMS support, but they'll probably just state that I should contact Shoto instead.

Others, on the other hand, suggest to dial at back even further:

46 minutes ago, WannabeSolarSparky said:

You should set your voltages a bit lower down to about 3.5v or 56v pack voltage

48 minutes ago, WannabeSolarSparky said:

ind out what voltage your bms needs to start balancing and set the voltage as close to that as possible.

Seplos BMS, and specifically my model/version, starts equalization if a cell voltage is over 3.5V and if there's a 50mV or higher difference and stops when difference has reached 30mV or less.

In my case I can clearly see that it starts to equalize cells at 3,5 V and stops charging at 3.6 V, but I have to say I run my system on voltage settings, no BMS coms with inverter.

Unless you know what the balance start voltage is, use the requested voltage from the pack. Since it is a seplos, you could try to use their software to see what the BMS configuration is - if your charge voltage is lower than the balance voltage, the pack will never do any balancing.

The Seplos Balancing is passive & it's shockingly bad. 

16 hours ago, P1000 said:

Since it is a seplos, you could try to use their software to see what the BMS configuration is - if your charge voltage is lower than the balance voltage, the pack will never do any balancing.

Battery Monitor screenshots attached. Shows param 8, 62 and 63 as 'Equalization opening voltage': 3.5V, 'Equalization opening pressure difference': 50mV and 'Equalization closing pressure difference': 30mV respectively. I've seen the cell balancing kick in with the lower 56.4V charge voltage, see last screenshot where the higher cells starts flickering. As Steve87 has pointed out its passive balancing.

16 hours ago, jgdt said:

Thanks. Not 100% sure where the original 57.8V comes from, or why its so high. Not sure if its send directly by the BMS (pack requested) to inverter or whether the inverter calculates it from other communications/parameters with the BMS? The Shoto batteries uses SeplosBMS. From another thread it would seem that the consensus is that if the inverter shows 57.8V charge voltage on LiBMS page then it receives it directly as is from the BMS. In which case it feels that if the BMS requested this 57.8V to begin with then just put it back and let it be. But, the batteries data specification sheet specifies 56.4V max. So which one is it? I've been trying to get hold of Shoto's technical support to clarify this out, but no success in getting support channel information. I guess I can try SeplosBMS support, but they'll probably just state that I should contact Shoto instead.

Others, on the other hand, suggest to dial at back even further:

Seplos BMS, and specifically my model/version, starts equalization if a cell voltage is over 3.5V and if there's a 50mV or higher difference and stops when difference has reached 30mV or less.

Correct the seplos BMS requests the 57.8V. 

The only way to change it is with battery monitor software. Now the seplos BMS uses passive balancing and it's slow but it does work. You just need to keep the battery at 100% for a day or 2. How the Seplos works is it will micro cycle from 100% down to 96% then back to 100% over and over if you instruct the inverter to keep it at 100%. This is where the balancing happens. 

3.46 vs 3.51 means that either the cell has slightly higher capacity than the rest (1 or 2 ah more) or was not properly top balanced. This is nothing to get concerned about, you are probably missing out on 1% (or less) capacity and it is hardly worth the concern.

My lowest cell in my DIY battery is 3.4 vs top one reaching 3.55 and the battery still performs great. 

I have a 15 cell battery less than 1 year old. One cell is out of balance. At the top end, if the other cells is above 3.5, that cell will only reach 3.45. On the lower end, if the other cells is still above 3v, the problem cell is already on 2.5 volt and the BMS shuts down.

The pack is rated at 100 Amp Hour, but I only get 87 Amp hour. I already send it back to the supplier. They send it back and said nothing wrong.

What battery is it Jaco?

The sun pays

2 hours ago, Jaco Slabbert said:

I have a 15 cell battery less than 1 year old. One cell is out of balance. At the top end, if the other cells is above 3.5, that cell will only reach 3.45. On the lower end, if the other cells is still above 3v, the problem cell is already on 2.5 volt and the BMS shuts down.

The pack is rated at 100 Amp Hour, but I only get 87 Amp hour. I already send it back to the supplier. They send it back and said nothing wrong.

3.45v at the higher end is still good and nothing odd with it even if the other cells are higher.

When the cells start discharging they should all fall to very similar voltages and then stay at a similar level with very similar voltage differences all the way down to a very low SOC.  At a very low SOC you can then expect the cell voltages to drift apart again.

3v is already pretty low SOC (approximately 10%). 2.5v is approximately 0% SOC.  It might still be within what the manufacturer accepts as normal at the top end and bottom end of the range.

What is the reported SOC when the battery shuts down?

Yep it's still a good shift out of the cells but not ideal. Looks like that 1 cell is not in sync with the others. Could have been a assembly error when they were top Balancing. Maybe the oke took lunch & forgot his sequence.