The all allusive 100% SoC - Is it fiction ?

[Snapperhead]

Hi Solar experts..... I need some fresh perspective on why a battery refuses to show 100% SoC even under the most favourable conditions.


First a little about the setup in question..... Two(2) parallel LiFePo4 48v 5.12kWh 100Ah BSL racks (via CAN-Bus) fed by Victron MPPT 450/100 TR (via VE.Direct) into a Multiplus II 8K/110 230v monitored by a Cerbo GX (via VE.Bus) using a Victron ET112 Energy Meter between Grid and the MCB. Batteries have their own BMS as shown in the Pics with current BMS config.

In an attempt to be kind to the battery cells and achieve 100% SoC (obviously without a float period due to the ESS setup) BMS Config (pack full charge voltage) was modified to reflect 55.2v in the Cerbo GX. As shown in the VRM graph's .... this didn't achieve anything either.

I've attached pics of the BMS Settings and VRM Details also.

Looking for info / education on where and how the 100% SoC is achieved, what conditions need to be met before it is achieved, and how and where it is reported.

I'm assuming that this ESS setup has a lot to do with the current SoC dilemma but am unsure of how to achieve 100% for any meaningful length of time. My aim is to be able to properly manage other parameters that will only come into effect once 100% SoC has been achieved.

Cheers ?

 

[sunshine_eggo]

Can you observe the individual cell voltages when battery voltage is at peak? New batteries typically arrive imbalanced and can make achieving 100% difficult.

Ultimately, 100% doesn't matter a lot. Normal use/cycling and regular charges to the 55.2V+ range will eventually balance the battery out.

 

[Snapperhead]

Can you observe the individual cell voltages when battery voltage is at peak? New batteries typically arrive imbalanced and can make achieving 100% difficult.

Ultimately, 100% doesn't matter a lot. Normal use/cycling and regular charges to the 55.2V+ range will eventually balance the battery out.

Thanks EggDude.... Yes I've kept an eye on the cell voltages, and they all appear within 30mv of each other when balanced or even as little as 10mv or less when being discharged..... The most they ever appear apart is about 100mv during the rapid charging first thing in the morning.
Do you think the VE Config values have much influence under these circumstances ? I've found 3x different sets of recommended voltage levels in three separate documents regarding BSL and Victron. They are all contradicting each other ?‍

 

[n2aws]

Thanks EggDude.... Yes I've kept an eye on the cell voltages, and they all appear within 30mv of each other when balanced or even as little as 10mv or less when being discharged..... The most they ever appear apart is about 100mv during the rapid charging first thing in the morning.
Do you think the VE Config values have much influence under these circumstances ? I've found 3x different sets of recommended voltage levels in three separate documents regarding BSL and Victron. They are all contradicting each other ?‍

Have you looked at the cell voltages when the pack voltage is high? I'd try going to something like 56.6v or something for a while, and let them balance out. Often, you'll find that at the upper ends of voltages, you'll see the imbalance show up. Allowing that to top balance brings them more inline than doing it at lower voltages.

 

[timselectric]

Never getting out of the flat part of the charging curve. You won't see much difference between the cells. Or ever get any balancing, to achieve the 100%.

 

[Snapperhead]

Have you looked at the cell voltages when the pack voltage is high? I'd try going to something like 56.6v or something for a while, and let them balance out. Often, you'll find that at the upper ends of voltages, you'll see the imbalance show up. Allowing that to top balance brings them more inline than doing it at lower voltages.

Ok great, this could be a nice start point to experiment for a solution.... So just to be 100% on the same page, I'd be adjusting this "Max Charge Voltage" within the BMS parameters right ? And then making sure the DVCC "CVL" is de-activated or above this new BMS value ? I'm also assuming that all the voltages under the charger tab of VE-Config3 have no effect whilst the BMS is managing things... Correct ?

 

[n2aws]

Thanks EggDude.... Yes I've kept an eye on the cell voltages, and they all appear within 30mv of each other when balanced or even as little as 10mv or less when being discharged..... The most they ever appear apart is about 100mv during the rapid charging first thing in the morning.
Do you think the VE Config values have much influence under these circumstances ? I've found 3x different sets of recommended voltage levels in three separate documents regarding BSL and Victron. They are all contradicting each other ?‍

I'd research how many cells are in a pack.

Ideally, you want to charge lithium iron phosphate cells to between 3.5 and 3.6v per cell.

What I tend to do is determine how many cells (I'll use 16 in this example, as thats how many are in my batteries.
I like to split the difference, halfway between 3.5 and 3.6 is 3.55v. so for me, I prefer to be 3.55 x 16 or 56.8v regardless of the battery manufacturers suggestion (as long as it doesn't violate warranty). If my battery manufacturer recommends higher, I'll set their recommendation as my "equalize" voltage, and do an "equalize" cycle occasionally. (LiFePo4 doesn;t actually need equalization, so hence the quotes. it's just a convenient way for me to charge to the manufacturers recommendations occasionally without me having to change programming)

 

[sunshine_eggo]

Thanks EggDude.... Yes I've kept an eye on the cell voltages, and they all appear within 30mv of each other when balanced or even as little as 10mv or less when being discharged.....

Voltage is only indicative of balance when it's above 3.40V. Voltages during discharge and typically in the 3.1-3.4V range rarely exhibit much deviation even if imbalance is present.

The most they ever appear apart is about 100mv during the rapid charging first thing in the morning.

That's pretty notable. Please include snapshots of all cell voltages when this is observed. Assessing balance requires looking at cell voltage data for longer than one typically desires.

Do you think the VE Config values have much influence under these circumstances ? I've found 3x different sets of recommended voltage levels in three separate documents regarding BSL and Victron. They are all contradicting each other ?‍

LOL... not uncommon. The VEConfig values will be overridden by the BMS always UNLESS they're part of the ESS assistant. To change battery parameters, you need to change the battery parameters directly or use DVCC charge voltage and current limit values.

Try running the battery up to 56.8V and observe the cell behavior.

Most BMS will report a min/max voltage in a widget in Advanced VRM. Recommend you set that up with "show range values" enabled and ensure VRM is set to log every minute. Here's the last 7 days for my Batrium:



Yes, 3.92 is normal for my battery. NCM chemistry with a 3.0-4.2V operating range.

 

[n2aws]

Ok great, this could be a nice start point to experiment for a solution.... So just to be 100% on the same page, I'd be adjusting this "Max Charge Voltage" within the BMS parameters right ? And then making sure the DVCC "CVL" is de-activated or above this new BMS value ? I'm also assuming that all the voltages under the charger tab of VE-Config3 have no effect whilst the BMS is managing things... Correct ?

Honestly, I'm not a victron guy, so I have no idea what all those abbreviations mean.

At the end of the day, you'd want to set your bulk, absorb, and float voltages to whatever setting you determine. How that maps to victorn-specific settings, I don't know. But I'd assume you can search for "How to set bulk voltage in victron cerbo" for what they call it.

 

[sunshine_eggo]

I'd research how many cells are in a pack.

Ideally, you want to charge lithium iron phosphate cells to between 3.5 and 3.6v per cell.

What I tend to do is determine how many cells (I'll use 16 in this example, as thats how many are in my batteries.
I like to split the difference, halfway between 3.5 and 3.6 is 3.55v. so for me, I prefer to be 3.55 x 16 or 56.8v regardless of the battery manufacturers suggestion (as long as it doesn't violate warranty). If my battery manufacturer recommends higher, I'll set their recommendation as my "equalize" voltage, and do an "equalize" cycle occasionally. (LiFePo4 doesn;t actually need equalization, so hence the quotes. it's just a convenient way for me to charge to the manufacturers recommendations occasionally without me having to change programming)

With this type of battery and BMS control interface, there is no float. Batteries will be held to target voltage all day, i.e., if you're charging to 56.8V, it will hold them there all day. This is regarded as stressful to the battery. 3.45V/cell results in a lower stress charge.

 

[sunshine_eggo]

Honestly, I'm not a victron guy, so I have no idea what all those abbreviations mean.

At the end of the day, you'd want to set your bulk, absorb, and float voltages to whatever setting you determine. How that maps to victorn-specific settings, I don't know. But I'd assume you can search for "How to set bulk voltage in victron cerbo" for what they call it.

With closed loop communication with these types of batteries, this is not possible. This is true for any server rack battery with "pylontech" type communication, i.e., EG4, SOK, Trophy, etc.

 

[n2aws]

With closed loop communication with these types of batteries, this is not possible. This is true for any server rack battery with "pylontech" type communication, i.e., EG4, SOK, Trophy, etc.

Sure, but closed loop communication isn't *required*, and if closed loop communications isn't allowing cells to top balance, then.. I'd disable it temporarily to allow them to top balance.

 

[n2aws]

With this type of battery and BMS control interface, there is no float. Batteries will be held to target voltage all day, i.e., if you're charging to 56.8V, it will hold them there all day. This is regarded as stressful to the battery. 3.45V/cell results in a lower stress charge.

With this type of battery and BMS control interface, there is no float. Batteries will be held to target voltage all day, i.e., if you're charging to 56.8V, it will hold them there all day. This is regarded as stressful to the battery. 3.45V/cell results in a lower stress charge.

I suspect you're taking my replies out of context. I'm not saying to do this forever. this is specifically for top balancing the batteries, which is the context for everything I've said since my original reply to this thread.

 

[Farm Boy]

1 of my LL batteries showed to be in protected mode. I checked the cell voltages as it showed 1 cell having a higher voltage than the rest. Not knowing what to do, I did nothing. The other day which was a month later I checked, it was not in protected mode and the voltages corrected it's self. so I suppose it's all good now and the bms knows what to do.

 

[Snapperhead]

Voltage is only indicative of balance when it's above 3.40V. Voltages during discharge and typically in the 3.1-3.4V range rarely exhibit much deviation even if imbalance is present.



That's pretty notable. Please include snapshots of all cell voltages when this is observed. Assessing balance requires looking at cell voltage data for longer than one typically desires.



LOL... not uncommon. The VEConfig values will be overridden by the BMS always UNLESS they're part of the ESS assistant. To change battery parameters, you need to change the battery parameters directly or use DVCC charge voltage and current limit values.

Try running the battery up to 56.8V and observe the cell behavior.

Most BMS will report a min/max voltage in a widget in Advanced VRM. Recommend you set that up with "show range values" enabled and ensure VRM is set to log every minute. Here's the last 7 days for my Batrium:

Yes, 3.92 is normal for my battery. NCM chemistry with a 3.0-4.2V

This is what I'm seeing here.... Would this be normal @sunshine_eggo ?

 

[Snapperhead]

I'd research how many cells are in a pack.

Ideally, you want to charge lithium iron phosphate cells to between 3.5 and 3.6v per cell.

What I tend to do is determine how many cells (I'll use 16 in this example, as thats how many are in my batteries.
I like to split the difference, halfway between 3.5 and 3.6 is 3.55v. so for me, I prefer to be 3.55 x 16 or 56.8v regardless of the battery manufacturers suggestion (as long as it doesn't violate warranty). If my battery manufacturer recommends higher, I'll set their recommendation as my "equalize" voltage, and do an "equalize" cycle occasionally. (LiFePo4 doesn;t actually need equalization, so hence the quotes. it's just a convenient way for me to charge to the manufacturers recommendations occasionally without me having to change programming)

Yes, I have 16x cells per pack. I'll play with the Master BMS value and see how that influences things. I've heard 3.55v to be a pretty good voltage for top balancing (temporarily of course for the sake of top balancing). Andy from Off Grid Garage has done a very good video on this and a lower float voltage also

 

[Snapperhead]

 

[n2aws]

View attachment 169530

yeah, those graphs show a pretty solid imbalance when at higher voltages (as eggo mentioned above, at lower voltages the cell imbalance is hard to notice due to how the voltage curve is for this chemistry.

It looks in this graph like you hit the ideal voltage, and then it immediately backs off. I think you need to take it out of closed loop for a while, and set it to stay at 56.8 for several hours and let those cells actually balance out. depending on how far out they are, you may need to do this over a day or two.

 

[Snapperhead]

yeah, those graphs show a pretty solid imbalance when at higher voltages (as eggo mentioned above, at lower voltages the cell imbalance is hard to notice due to how the voltage curve is for this chemistry.

It looks in this graph like you hit the ideal voltage, and then it immediately backs off. I think you need to take it out of closed loop for a while, and set it to stay at 56.8 for several hours and let those cells actually balance out. depending on how far out they are, you may need to do this over a day or two.

Yes, it appears from all angles that that would start the process of top balancing and see what it does from there..... So what's the best way to do this ? I realize not everyone has Victron, so I'm restricted to the paramteres of the mighty blue.

Do I just increase the "Max Charge Voltage" within the BMS parameters ? And then making sure the internal Victron DVCC "Charge Voltage Limit" is de-activated or above this new BMS value so as to not disable the BMS changes ?

Or should I disable the BMS input all together and just set the parameters in the MPPT for Absorption and Float etc ?

 

[n2aws]

Yes, it appears from all angles that that would start the process of top balancing and see what it does from there..... So what's the best way to do this ? I realize not everyone has Victron, so I'm restricted to the paramteres of the mighty blue.

Do I just increase the "Max Charge Voltage" within the BMS parameters ? And then making sure the internal Victron DVCC "Charge Voltage Limit" is de-activated or above this new BMS value so as to not disable the BMS changes ?

Or should I disable the BMS input all together and just set the parameters in the MPPT for Absorption and Float etc ?

I'll defer to the victron users for this. From what was being said earlier, I think the second option is going to be the only way to go.

Disable BMS comms temporarily, set bulk/absorb/float for 56.8, and let it sit there until the cells show they are balanced.

Ideally, if you can remove load from the batteries and do it in isolation, that would be preferred. if not, just let hte batteries charge to full and stay there daily. it'd correct itself over time, if allowed to sit at those higher voltages for a bit longer. (Looking at the graph you posted, it looks like it gets *slightly* better each day as it is. so it's possible that over time, they'd just balance on their own. But if you have the ability to just let it balance for a day or so, that's likely the approach I'd take. Then when they are balanced, re-enable the BMS control, and let it do it's thing automagically.