The all allusive 100% SoC - Is it fiction ?

[sunshine_eggo]

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

You've got a bit of a runner there. That's an imbalanced battery. You can also do a custom widget with 5 parameters pulled from any source. Fortunately, you can pull all of them from the BMS:

Voltage
Current
SoC
Min Cell
Max Cell



I manually set each y-axis based on the expected operating range. Note that sometimes the colors are jacked up. I just don't feel like re-doing it. The list at the top left is correct on order and color. The popup window is correct on order, but wrong on color. It's a known bug.

Something like this might help you establish what's going on.

Try 54.4V limit in DVCC and see what happens with cell voltages.

 

[Snapperhead]

Thanks for sharing your brain cells Solar Guru dudes..... I appreciate it very much btw ??

Ok, so I've been testing all day with MUCH coffee after Night Shift. Here is my results thus far >>

I've increased the BMS "Pack Full Charge Voltage" (as per parameter pic attached) to 56.3v and it is now charging and balancing slowly. The BMS recorded Voltage is increasing slowly above and slightly beyond 56.3v with the average Cell Delta slowly improving below 65mV (32 cells total)
There is no battery current recorded in VRM, as I'd assume it's too low for VRM to record.

Using multimeter current is bouncing around between 0 - 1.3A balancing current. Voltage measured at terminals is bouncing around between the BMS voltage +/- 0.5v

Testing Conditions >>
I have selected 'keep battery charged' within the Victron ESS environment, which I believe apply's all available DC PV power to the battery, and no longer sharing this DC power with the loads on AC-In. This is just a precaution to avoid any undue interuption to the charge curve. (Plenty of sunshine here today

Stangely.... Whatever preset voltage I enter into the BMS.... It is always reflected into the Victron VRM and Cerbo etc as +0.5v ?? Is this normal ? Am I missing something ? Is there an automatic offset applied to these settings or some form of correction factor ? Mmmm ?

Thoughts>>
I'm now wondering if the length of my battery cables are causing a voltage disparity between the Battery and Multiplus/Cerbo that may be outside of an accepted range ? Could this be the cause of my system not reaching 100% even under perfect conditions ?
i.e. Battery Terminal Voltage is about 0.7v higher than Cerbo/VRM reported Voltage. Is this normal or too large a disparity ?
It should be noted that my battery cables appear way too long (see attached pics) and are both wired separately/individually from each battery back to a bus bar, that then feeds both the Multiplus Inverter and the MPPT Charger. The individual cable sizes are .25mm (4 AWG).
Could this influence my voltage this much ? Or am I barking up the wrong tree ? ?

I'm still somewhat confused as to how the 100% is actually decided, and by what hardware and under what conditions. Is it the BMS that reports 100% or the Cerbo ? I know some BMS's need a cdertain voltage to be maintaned for a certain time, along with a current threshold.... Then pop into 100% SoC. Some other BMS's even need a cell to hit the Over Voltage Protection value along with low current to report 100%. Any thoughts ?

Plz let me know what you guys think, and thanks again for your input. It's greatly appreciated ?

 

[Snapperhead]

Here are the cable pics …. Cable data sheet is available if it helps ??

 

[sunshine_eggo]

I've increased the BMS "Pack Full Charge Voltage" (as per parameter pic attached) to 56.3v and it is now charging and balancing slowly. The BMS recorded Voltage is increasing slowly above and slightly beyond 56.3v with the average Cell Delta slowly improving below 65mV (32 cells total)
There is no battery current recorded in VRM, as I'd assume it's too low for VRM to record.

Fairly common. BMS is the culprit. Some just don't report below some X amps.

Using multimeter current is bouncing around between 0 - 1.3A balancing current. Voltage measured at terminals is bouncing around between the BMS voltage +/- 0.5v

Testing Conditions >>
I have selected 'keep battery charged' within the Victron ESS environment, which I believe apply's all available DC PV power to the battery, and no longer sharing this DC power with the loads on AC-In. This is just a precaution to avoid any undue interuption to the charge curve. (Plenty of sunshine here today

Stangely.... Whatever preset voltage I enter into the BMS.... It is always reflected into the Victron VRM and Cerbo etc as +0.5v ?? Is this normal ? Am I missing something ? Is there an automatic offset applied to these settings or some form of correction factor ? Mmmm ?

Not normal as far as I'm aware.

Thoughts>>
I'm now wondering if the length of my battery cables are causing a voltage disparity between the Battery and Multiplus/Cerbo that may be outside of an accepted range ? Could this be the cause of my system not reaching 100% even under perfect conditions ?

Possible, but this sort of thing just extends charge time as the voltage drop becomes less severe as the current decreases.

Always worth checking all your equipment with a separate meter. BMS reports X Volts, meter reports Y volts. MPPT reports X Volts, meter reports Y volts, etc.

i.e. Battery Terminal Voltage is about 0.7v higher than Cerbo/VRM reported Voltage. Is this normal or too large a disparity ?

Too large.

It should be noted that my battery cables appear way too long (see attached pics) and are both wired separately/individually from each battery back to a bus bar, that then feeds both the Multiplus Inverter and the MPPT Charger. The individual cable sizes are .25mm (4 AWG).
Could this influence my voltage this much ? Or am I barking up the wrong tree ? ?

Yes, but only under high current conditions.

I'm still somewhat confused as to how the 100% is actually decided, and by what hardware and under what conditions.

That's the trick. Just can't know with some BMS.

Is it the BMS that reports 100% or the Cerbo

BMS telling Cerbo. Cerbo telling VRM.

I know some BMS's need a cdertain voltage to be maintaned for a certain time, along with a current threshold.... Then pop into 100% SoC. Some other BMS's even need a cell to hit the Over Voltage Protection value along with low current to report 100%. Any thoughts ?

Plz let me know what you guys think, and thanks again for your input. It's greatly appreciated ?

Some BMS require some cycling to eventually incorporate 100% into the realm of possibilities. Balance appears to be improved.

 

[Snapperhead]

Fairly common. BMS is the culprit. Some just don't report below some X amps.

@sunshine_eggo >> As suspected..... Even though it reports as 'Idle' 99% most of the time, I can see it slowly balancing in the BMS and the voltage increasing.

Then .... the voltage peaks at my desired value (56.8v on this occasion) with all 32x cells eventually within 30mv of each other ?

Then .... after about 15 minutes the pack voltage starts to drop for no reason I can figure out. And it slowly settles down to a much lesser voltage and keeps falling very slowly ?

Shouldn't it maintain this charge voltage for a period of time ?..... Like all day / night long ?

I also changed the Cerbo ESS settings to "Keep Batteries Charged" and then tried a Scheduled Charge etc.... It just won't maintain any specific voltage.

In fact the only way I can get it to report 100% is to apply a large load to the Batteries (either from critical loads or via AC-In or both) then it starts a quick recovery charge cycle and reports 100% for less than 5 minutes before the voltage again starts to drop. (see chart)

There must be a setting somewhere I'm missing. I also changed the VE.Config Charger Tab voltages in the vain hope they may have an influence (i.e. setting absorption & float to 56.8v) knowing it would be ignored due to the BMS and ESS setup .... But was worth the try

Not normal as far as I'm aware. Mmm I'l have to explore this voltage BMS / Cerbo offset more me thinks

Possible, but this sort of thing just extends charge time as the voltage drop becomes less severe as the current decreases.

Always worth checking all your equipment with a separate meter. BMS reports X Volts, meter reports Y volts. MPPT reports X Volts, meter reports Y volts, etc.

Too large. I noticed later in the day as the Battery voltage gets closer to the charge voltage, that the Cerbo voltage value is almost identical to the BMS voltage.

That's the trick. Just can't know with some BMS. Handy. Makes me wish I'd purchased the Jakiper with the PACE BMS perhaps ?

BMS telling Cerbo. Cerbo telling VRM.

Some BMS require some cycling to eventually incorporate 100% into the realm of possibilities. Balance appears to be improved. Yes, within the BMS setting of 30mv (32 cells in total)

 

[n2aws]

Then .... the voltage peaks at my desired value (56.8v on this occasion) with all 32x cells eventually within 30mv of each other ?

Then .... after about 15 minutes the pack voltage starts to drop for no reason I can figure out. And it slowly settles down to a much lesser voltage and keeps falling very slowly ?

Shouldn't it maintain this charge voltage for a period of time ?..... Like all day / night long ?

I also changed the Cerbo ESS settings to "Keep Batteries Charged" and then tried a Scheduled Charge etc.... It just won't maintain any specific voltage.

In fact the only way I can get it to report 100% is to apply a large load to the Batteries (either from critical loads or via AC-In or both) then it starts a quick recovery charge cycle and reports 100% for less than 5 minutes before the voltage again starts to drop. (see chart)

There must be a setting somewhere I'm missing. I also changed the VE.Config Charger Tab voltages in the vain hope they may have an influence (i.e. setting absorption & float to 56.8v) knowing it would be ignored due to the BMS and ESS setup .... But was worth the try

Not normal as far as I'm aware. Mmm I'l have to explore this voltage BMS / Cerbo offset more me thinks

View attachment 169650

You're getting into victron-specific settings that I obviously can't help with. But just wanted to comment real fast and say: in this post you mentioned seeing 100% for a couple minutes, which absolutely sounds like progress to me! Kudos

 

[Snapperhead]

You're getting into victron-specific settings that I obviously can't help with. But just wanted to comment real fast and say: in this post you mentioned seeing 100% for a couple minutes, which absolutely sounds like progress to me! Kudos

Thanks mate. With the help of this amazing crew …. I’m sure I’ll solve this one as I did the Grid Code issue supplying the house loads. Thanks again to you and others ???

 

[n2aws]

Thanks mate. With the help of this amazing crew …. I’m sure I’ll solve this one as I did the Grid Code issue supplying the house loads. Thanks again to you and others ???

When you've got these things fully balanced out, I'd love to see the before and after graphs for cell voltage differences over time (you posted the before earlier in this thread, I'd love to see the "after", after it's been balanced a few days and has returned to normal working duty.

 

[HighTechLab]

Didn't read whole thread, but in response to OP - 100% can be elusive because SOC is reported to the Cerbo as a 3 digit unsigned integer with no decimal place. I.E. 99.9999% (which is calculated as a floating point) is transmitted as 99%. It is common with the BMS in this battery to hit full then something simple as the idle consumption of the BMS uses 0.0001Ah instantly dropping the SOC shown on the Cerbo to 99%

 

[Snapperhead]

Didn't read whole thread, but in response to OP - 100% can be elusive because SOC is reported to the Cerbo as a 3 digit unsigned integer with no decimal place. I.E. 99.9999% (which is calculated as a floating point) is transmitted as 99%. It is common with the BMS in this battery to hit full then something simple as the idle consumption of the BMS uses 0.0001Ah instantly dropping the SOC shown on the Cerbo to 99%

Thanks for the info…. Do u know what type of BMS it is ? And where I can get some more detailed specs ? What could be causing the gradual decline in voltage ? Even when in ‘Keep battery charged’ mode ? Any thoughts ?

 

[Snapperhead]

This mornings Data ?

 

[Snapperhead]

Some voltages this morning during absorption (if this is actually hapenning ?)

BMS Reported Voltages>>
Batt_1 = 54.22v (Master)
Batt_2 = 53.63v

VRM = 53.63v

MultiMeter Batt Terminals>> (DC Voltage bouncing arround between 55v - 59v - Trusty Fluke has AVG reading ability ? )
Batt_1 = 56.8v (Average)
Batt_2 = 56.8v (Average)

VRM & BMS Voltages appears WAY off Battery Terminal Voltages (over 3 Volts !!) however VRM + BMS voltages appear close and starting to increase very slowly

Battery Cell Delta is only 10mv Batt_1 / 9mv Batt_2 However between both battery's is 46mv (see chart)

Any thoughts as to the cause of the Voltage disparity ? Or does the BMS need calibrating somehow ? (see BMS calibration page) Perhaps getting ahead of myself

 

[Snapperhead]

Yes, but only under high current conditions.

From your experience, is it best practice to have both batteries cabled separately back to a buss bar (via two pairs of 3x metre 4AWG cable lengths), which is then cabled separatley to both Inverter & MPPT ? (It could be wired with one 2 metre 2AWG)

What happens if these two separate battery feeds are slightly different lengths ? (they are of course wired in parallel at the buss bar side)

Or whould this only have an effect at high current...and what effect ? Could this be imbalancing my battery's upon high current morning charging ?

What could I do to negate this as a cause, without re-wiring the battery's using a single 2AWG etc ? (or even have both pairs re-directed to master battery and parallel jumpers to slave?)

What would be a way to eliminate this as a cause ?
i.e. Low current charging experiment ?

 

[Supervstech]

I mean… do you really WANT to hit 100% SOC?

LFP cycle life drops off quickly when sitting fully charged…

 

[Snapperhead]

I mean… do you really WANT to hit 100% SOC?

LFP cycle life drops off quickly when sitting fully charged

Yes, It's not that I want it to sit at unhealthy voltages all day. My end game is to cycle the battery's down to a safe SoC overnight and simply charge them and use them to assist high loads throughout the day only. But I'm simply testing the new setup at this point, and wanting to top balance the batteries and achieve at least a weekly 100% SoC.


But as you may have seen above, it just won't maintain any real charge voltage for some reason.
In it's current config (ESS - Keep battery's charged etc) I'd expect the batteries to be maintained at whatever voltage level and just stay there (not forever - it's a test) ....

However the voltage drops off as soon as it finishes it's peak charge each morning (without reaching 100%) then dwindles away throughout the day until the sun goes down (like right now here in Oz) then it drops off once again.

Isn't the Multiplus II supposed to also maintain a charge voltage to the Battery's when there is no MPPT power ? Or am I missing something ?
I'm in search of at least one "true" voltage point that I can rely on to assist me to maintain the health of the batteries long term.
However it appears that all my voltage reference points disagree for some reason as seen above

 

[Steve_S]

The LFP chemistry will ALWAYS settle / drop voltage as soon as the charging cycle ends. This is absolutely normal & expected with LFP.
LFP is actually at 100% SOC when the cells are at the top of the "working voltage range" which is 3.400V per cell (54.4V @ Pack).
LFP working Voltage Range which is what delivers the actual Amp Hours is from 3.000-3.400 Volts. EVERYONE CONFUSES the Allowable Voltage Range which is from 2.500-3.650 which is the allowable voltage spread that does NOT harm the cells. Voltages above or below this range HARM THE CELLS !

If you wish to attain 100% SOC at Charge Completion then DO NOT OVERDRIVE the cells as they will deviate and cause HVD (High Volt) Disconnects as some cells will run and reach 3.650 and cause a cutoff. FYI, a BMS typically takes 3 full cycles to "learn" its thresholds, some BMS brands allow you to manual set those params.

Set your Charging to 55.2V (3.450V per cell) for Absorb/Bulk.
Float @ 55.1V (3.440Vpc)
Your EndAmps/TailCurrent is 5A. This is when the Charger must transition from Absorb/Bulk to Float.
As the Resistance increases and the charge taken drops to 5A, then the charger must switch to FLOAT which is Constant Voltage Variable Current. This will then top off the cells as needed while providing system power simultaneously.

BALANCING NOTES !
Passive Balancing only burns off voltage from high voltage cells. This is slow & wasteful and is only appropriate for fully matched & batched cells, with lower AH capacities.
Active Balancing actually moves energy from Hi cells to Lo cells with minimal loss. The general guideline is to have 1A Active Balancing per 100AH of Capacity.
Active Balancing does NOT USE POWER - it shifts it only and only when the conditions are met, otherwise, it is totally passive.
It can also be set to start Active Balancing @ 3.300Vpc when there is greater than 20mv difference. This in effect reduces balancing time as teh cells charge and will allow for a full charge on cells faster than without. Some BMS (Pace & TDT) allow for supplemental external Active Balancers. There are MANY Variations.

Here are some of my Resource Files to help you.

 

[Snapperhead]

The LFP chemistry will ALWAYS settle / drop voltage as soon as the charging cycle ends. This is absolutely normal & expected with LFP.
LFP is actually at 100% SOC when the cells are at the top of the "working voltage range" which is 3.400V per cell (54.4V @ Pack).
LFP working Voltage Range which is what delivers the actual Amp Hours is from 3.000-3.400 Volts. EVERYONE CONFUSES the Allowable Voltage Range which is from 2.500-3.650 which is the allowable voltage spread that does NOT harm the cells. Voltages above or below this range HARM THE CELLS !

Well said.... Yes this is so very true and often misinterpreted. However I'm pretty sure my setup doesn't have, and isn't supposed to have, any bulk/abs/float cycles at all due to it's ESS environment. So, I would normally sit the charge voltage at 55.2v and just cycle them down to a safe SoC every night..... But shouldn't my charge curve then just slowly climb to 55.2v and stay there ? (without loads of course)

If you wish to attain 100% SOC at Charge Completion then DO NOT OVERDRIVE the cells as they will deviate and cause HVD (High Volt) Disconnects as some cells will run and reach 3.650 and cause a cutoff. FYI, a BMS typically takes 3 full cycles to "learn" its thresholds, some BMS brands allow you to manual set those params.

Ok, Ive been unable to find out what exact BMS type and specs these BSL battery's contain. But I'm still trying Let's hope that at the right charging voltage these BMS's learn their curve over time ?

Set your Charging to 55.2V (3.450V per cell) for Absorb/Bulk.
Float @ 55.1V (3.440Vpc)
Your EndAmps/TailCurrent is 5A. This is when the Charger must transition from Absorb/Bulk to Float.
As the Resistance increases and the charge taken drops to 5A, then the charger must switch to FLOAT which is Constant Voltage Variable Current. This will then top off the cells as needed while providing system power simultaneously.

Once again..... See above lack of this charge cycle. The charge cycle is NOT controlled by the MPPT.

BALANCING NOTES !
Passive Balancing only burns off voltage from high voltage cells. This is slow & wasteful and is only appropriate for fully matched & batched cells, with lower AH capacities.
Active Balancing actually moves energy from Hi cells to Lo cells with minimal loss. The general guideline is to have 1A Active Balancing per 100AH of Capacity.
Active Balancing does NOT USE POWER - it shifts it only and only when the conditions are met, otherwise, it is totally passive.
It can also be set to start Active Balancing @ 3.300Vpc when there is greater than 20mv difference. This in effect reduces balancing time as teh cells charge and will allow for a full charge on cells faster than without. Some BMS (Pace & TDT) allow for supplemental external Active Balancers. There are MANY Variations.

Here are some of my Resource Files to help you.

Yes, I had a good look at the BMS whilst passive balancing today. I changed it's threshold to 20mv and 3.4vpc to activate..... It worked a treat and I managed to get it to balance better than over the last 3 weeks, from a 120mv Delta to less than 20mv... however both battery's are not yet within sync of each other with one pack voltage higher than the other at the moment.
I assume they are still not properly top balanced just yet.

 

[n2aws]

however both battery's are not yet within sync of each other with one pack voltage higher than the other at the moment.
I assume they are still not properly top balanced just yet.

I wouldn't worry about voltage differences between packs.

Focus on the cell voltages within an individual pack, once those are good, the packs will take care of themselves (via the BMS)

 

[Snapperhead]

I wouldn't worry about voltage differences between packs.

Focus on the cell voltages within an individual pack, once those are good, the packs will take care of themselves (via the BMS)

Thanks n2aws . . Good to know ??
I was wondering if the voltage imbalance between packs may be contributing to my weird range of voltage reference points … or confusing the Cerbo ?
But if it’s not a factor…. I’ll keep looking ??

 

[n2aws]

Thanks n2aws . . Good to know ??
I was wondering if the voltage imbalance between packs may be contributing to my weird range of voltage reference points … or confusing the Cerbo ?
But if it’s not a factor…. I’ll keep looking ??

Earlier in this thread, someone explained the voltage disparity you are seeing.
I forget who, but they explained the differences when measured at various points in the victron ecosystem