diy solar

diy solar

State of charge chart

Beavo

Beavo

Didn't, don't, won't know what I'm doing.
Joined
Jan 28, 2024
Messages
70
Location
Bristol
I printed off this chart yesterday just to use for quick reference for my LiFePo4 batteries.
Now there seems to be many of these charts on the net and lots of them give conflicting figures.

There seems to quite a few figures missing.

I wondered if there was a "Gold Standard" more accurate chart that people use for reference.

1000010903.png

Cheers
 
Beavo said:
I printed off this chart yesterday just to use for quick reference for my LiFePo4 batteries.
Now there seems to be many of these charts on the net and lots of them give conflicting figures.

There seems to quite a few figures missing.

I wondered if there was a "Gold Standard" more accurate chart that people use for reference.

View attachment 193758

Cheers
Click to expand...
These charts aren’t very useful for LFP batteries. A shunt is the only reliable way to measure State of Charge.
IF the cells in your battery were well balanced, these two values would be correct:
13.8V = (about) 100%
10.0V = 0%

Everywhere else voltage and SoC just don’t correlate all that well. You might as well try to estimate SoC by measuring the width of the battery.
 
LakeHouse said:
These charts aren’t very useful for LFP batteries. A shunt is the only reliable way to measure State of Charge.
IF the cells in your battery were well balanced, these two values would be correct:
13.8V = (about) 100%
10.0V = 0%

Everywhere else voltage and SoC just don’t correlate all that well. You might as well try to estimate SoC by measuring the width of the battery.
Click to expand...
I would agree. I haven't installed a shunt yet and to be honest it's quite far down the list of things I'm hoping to add to my setup.

Just wondered if there was a more accurate quick reference guide out there.

Just went out and they're showing 13.17v.
As you can see on the chart there's quite a gap between 13.1v and 13.2v, 40% to 70% respectively.

I could obviously stand there and figure it out but then that wouldn't be "quick".

Cheers.
 
Beavo said:
I would agree. I haven't installed a shunt yet and to be honest it's quite far down the list of things I'm hoping to add to my setup.

Just wondered if there was a more accurate quick reference guide out there.

Just went out and they're showing 13.17v.
As you can see on the chart there's quite a gap between 13.1v and 13.2v, 40% to 70% respectively.

I could obviously stand there and figure it out but then that wouldn't be "quick".

Cheers.
Click to expand...
You’re sort of missing the point. You could interpolate between those two incorrect points on the chart and you’d get 61%, but that wouldn’t be a good estimation of the actual SoC. It could be 40% or 70% at that voltage. Or if there’s a large load it could be 80%. Or if you’re charging it could be 30%.

Voltage isn’t a good indication of SoC.
Not: the charts aren’t accurate enough.
Not: the charts just don’t show enough intermediate values.
 
Beavo said:
I would agree. I haven't installed a shunt yet and to be honest it's quite far down the list of things I'm hoping to add to my setup.

Just wondered if there was a more accurate quick reference guide out there.

Just went out and they're showing 13.17v.
As you can see on the chart there's quite a gap between 13.1v and 13.2v, 40% to 70% respectively.

I could obviously stand there and figure it out but then that wouldn't be "quick".

Cheers.
Click to expand...
A shunt should be a priority when dealing with Lifepo batteries going by voltage is a total guess only on the high end and low end of the voltage range it is close to being accurate
 
rodrick said:
A shunt should be a priority when dealing with Lifepo batteries going by voltage is a total guess only on the high end and low end of the voltage range it is close to being accurate
Click to expand...
In that case I'll promote it to number 475 on my list.
Cheers.
 
LakeHouse said:
You’re sort of missing the point. You could interpolate between those two incorrect points on the chart and you’d get 61%, but that wouldn’t be a good estimation of the actual SoC. It could be 40% or 70% at that voltage. Or if there’s a large load it could be 80%. Or if you’re charging it could be 30%.

Voltage isn’t a good indication of SoC.
Not: the charts aren’t accurate enough.
Not: the charts just don’t show enough intermediate values.
Click to expand...
Thanks for taking the time for that explanation.

It's not a big issue for me, I was just wondering if I could make life a bit easier.

My setup is working fine as it is, albeit I need far more panels.

Cheers
 
littleharbor2 said:
If you have Bluetooth accessibility on your battery you get a (somewhat accurate) soc.
Click to expand...

Unfortunately not. I did look at Bluetooth batteries but I just can't justify the cost of them.

To be honest, I can't justify the cost of any of it with what I'm using it for.

Still going to buy it all though. 😀

I was just about to have a look at shunts as it's quite clear that I should be adding one.

I want one with Bluetooth capability to add to victron connect but I also want a screen of some description so that I can instantly see the state of charge.

I'll start investigating now.

Edit** Looks like I'll need the BMV-712.

Cheers
 
Last edited:
The BMV 712 has a built in relay that can use state of charge or voltage to trigger events, like maybe turning a charger on when the batteries get too low.
 
Danke said:
I have to concern myself with that level of precision WRT SOC, it means my system isn’t large enough yet.
Click to expand...
I wouldn't say I'm concerned, I was just curious as to the "rough" soc of the batteries.
Just nice to glance up and see something rather than connect via app or whatever.
I'm unlikely to ever be in a position where I want exact figures.
Hence the little chart.
Cheers
 
Q-Dog said:
The BMV 712 has a built in relay that can use state of charge or voltage to trigger events, like maybe turning a charger on when the batteries get too low.
Click to expand...
I was just looking at that.
I was wondering if I could get it to turn my Blue Smart charger on and off.
We've talked about something similar in one of my other threads.
Cheers
 
LakeHouse said:
You’re sort of missing the point. You could interpolate between those two incorrect points on the chart and you’d get 61%, but that wouldn’t be a good estimation of the actual SoC. It could be 40% or 70% at that voltage. Or if there’s a large load it could be 80%. Or if you’re charging it could be 30%.

Voltage isn’t a good indication of SoC.
Not: the charts aren’t accurate enough.
Not: the charts just don’t show enough intermediate values.
Click to expand...
Voltage is a useful reference. Especially when you couple it with observational awareness. Shunts are not necessarily accurate either.
 
Beavo said:
I was just looking at that.
I was wondering if I could get it to turn my Blue Smart charger on and off.
We've talked about something similar in one of my other threads.
Cheers
Click to expand...
I am aware of your other thread. I don't know what model charger you have, but if it has a remote port, a relay can turn it on or off.
 
Cell voltage is reliable to determine SoC but cell must be rested and unloaded. For LPF, it must be zero current for a few minutes. Since LPF cells have a flatter discharge curve you also need a voltmeter with better than 0.05% accuracy. A low cost DVM that does pretty good is Uni-T UT61E or ANG AN870, but their accuracy degrades outside of 20-25 degs C room temp range.

It is the cell overpotential voltage with cell current that causes all the confusion in cell voltage vs. SoC.

LFP cell open circuit rested voltage has very small temperature dependency but overpotential voltage slump with load current is very dependent on cell temperature. The overpotential voltage slump with load current has an exponential decay and recovery time which is 1-3 minutes. The overpotential voltage slump also depends on aging condition of cell. For LFP, overpotential discharge slump and overpotential charging bump is fairly symmetrical for equal discharge and charging currents.

High SoC greater than about 95% also has the possibility of a capacitance surface charge build up due to higher absorb charging voltage which contributes almost nothing to cell capacity but bumps up cell voltage. A fully charged, rested no-load current, LPF cell voltage without surface charge has a rested open circuit voltage of 3.43v.

LFP typ Open Circuit Voltage vs SoC.png

Cell Overpotential Chart.png
LF280 overpotiential curve.png

LF280 AH battery dischg 0.1C-1.0C.png
 
I use a voltmeter in the van as a 'rough fuel gauge' for the battery banks state of charge (LYP lithiums)- my 'night load' is usually pretty much the same every night, and from experience I know know the what the voltage is in relationship to the SOC at that load level...
It isn't accurate to three decimal places' but I can readily gauge it to about 10% of the actual SOC just looking at the volts...
1707225270751.png
Nothing flash (just a $2 Aliexpress special, but it is actually surprising accurate against my Fluke- and if it is out, has a small trimpot that can be 'tweaked' to get it closer...
(mine runs from 13.1v after sunset, usally hitting a minimum of 12.9 with the nighttime loads running all night (2 fans, led light, laptop, routers and hubs, 4 security cameras and the fridge
 
RCinFLA said:
Cell voltage is reliable to determine SoC but cell must be rested and unloaded. For LPF, it must be zero current for a few minutes. Since LPF cells have a flatter discharge curve you also need a voltmeter with better than 0.05% accuracy. A low cost DVM that does pretty good is Uni-T UT61E or ANG AN870, but their accuracy degrades outside of 20-25 degs C room temp range.

It is the cell overpotential voltage with cell current that causes all the confusion in cell voltage vs. SoC.

LFP cell open circuit rested voltage has very small temperature dependency but overpotential voltage slump with load current is very dependent on cell temperature. The overpotential voltage slump with load current has an exponential decay and recovery time which is 1-3 minutes. The overpotential voltage slump also depends on aging condition of cell. For LFP, overpotential discharge slump and overpotential charging bump is fairly symmetrical for equal discharge and charging currents.

High SoC greater than about 95% also has the possibility of a capacitance surface charge build up due to higher absorb charging voltage which contributes almost nothing to cell capacity but bumps up cell voltage. A fully charged, rested no-load current, LPF cell voltage without surface charge has a rested open circuit voltage of 3.43v.

View attachment 193775

View attachment 193776
View attachment 193778

View attachment 193779
Click to expand...

That's exactly what I thought.
Cheers
 
Bop said:
I use a voltmeter in the van as a 'rough fuel gauge' for the battery banks state of charge (LYP lithiums)- my 'night load' is usually pretty much the same every night, and from experience I know know the what the voltage is in relationship to the SOC at that load level...
It isn't accurate to three decimal places' but I can readily gauge it to about 10% of the actual SOC just looking at the volts...
View attachment 193784
Nothing flash (just a $2 Aliexpress special, but it is actually surprising accurate against my Fluke- and if it is out, has a small trimpot that can be 'tweaked' to get it closer...
(mine runs from 13.1v after sunset, usally hitting a minimum of 12.9 with the nighttime loads running all night (2 fans, led light, laptop, routers and hubs, 4 security cameras and the fridge
Click to expand...
That's what I have, and have always used in the past, but that was with lead acid.

The thing is, now that I have LiFePo4 I was trying to be a little more accurate seeing that between 13.1v and 13.2v is a difference of 30%.

The charge controller works to 2 decimal places but that doesn't really correlate percentage wise.

I'm not bothered enough to go out and spend a load of money on it just yet (though I will at some point), I just wondered if there was a quick chart for reference similar to the one I posted.
Seems as though there isn't and can't be.

Cheers
 
All battery types have overpotential voltage versus current. Lead-acid needs 15-25 minutes to recover from discharge current to equilibrium. Their open circuit voltage also varies a little depending on acid concentration in electrolyte.
 
RCinFLA said:
Cell voltage is reliable to determine SoC but cell must be rested and unloaded. ...
Click to expand...
I like the term "Slump" for describing battery voltage drop under load. I have used "Droop" in the past instead of drop or sag but I think I will now use slump instead.

Thanks for the charts. Someone has done a lot of work involved in coming up with them.
 
Just to throw another one in the ring.
The beatings will continue until accuracy prevails!
In fairness this was designed for a 48 volt system. Everything was divided down to the other voltages.
 

Attachments

  • LiFePO4 SOC Charge Chart Rev C.jpg
    LiFePO4 SOC Charge Chart Rev C.jpg
    172.7 KB · Views: 22
Beavo said:
I wondered if there was a "Gold Standard" more accurate chart that people use for reference.

Cheers
Click to expand...
This chart works fine if you have sufficient battery. Easy to determine when to start the generator or to start extreme conservation.
I assume 48 to 72 hour battery as a minimum.
 
I can only give numbers for our 4 cell 300Ah Sinopoly LiFePO4 battery that has survived 10 years of full time travel.
Two paralleled shunt based battery monitors, Victron SmartShunt and $50 Junctek agree with each other. They are either relatively accurate or less likely both inaccurate by similar degrees.
In daily use I now rarely keep regular track of SOC as the numbers are very similar throughout the days.

If both monitor were to ever fail only a single number would be of concern to me. It may vary slightly between battery makers?
13.1 - 13.3V idling with light loads - everything is fine.
12.7V idling - you have let the battery deplete too far. Time to find a decent charge source now.
Our battery apart from the 700A peak engine starting current has NEVER reached 12.7V.
12.8V has occured once or twice in those 10 years

Click to enlarge
 

Attachments

  • 20230520 NY.jpg
    20230520 NY.jpg
    348 KB · Views: 1
You must log in or register to reply here.

Similar threads

MountainmanBill
48-volt battery bank SOC chart
Replies
12
Views
4K
sunshine_eggo
sunshine_eggo
T
Solar on single phase of 3 phase house
2
Replies
33
Views
1K
Tarquinbob
T
JBertok
FYI: maximizing FLA performance with Schneider using "Custom" battery calculations
Replies
0
Views
90
JBertok
JBertok
radiomean
Limit state of charge of GBLI 6532
Replies
3
Views
161
sunshine_eggo
sunshine_eggo
S
230ah battery w/200BMS and 2000w inverter question
Replies
4
Views
494
smieglitz
S

diy solar

diy solar
Back
Top