Battery voltage conversion to % charged while lightly loaded at say 5a

[Fuss]

In another thread Dockhead gave us a great little table with state of charge at various voltages with the house batteries lightly loaded.

I thought this was great as normally it is not realistic to switch all off and let the batteries settle before calculating the % of charge against the displayed voltage.

More realistic is to have the batteries lightly loaded (5a draw for example) for a while and then check the voltage.

I think we would all agree that at the end of the day its a bit rough and depends on this and that a bit... but I think its a very good idea.

Here is the table....
25.2 volts (equivalent to 12.6 on a 12v system) is 100%
24.8 volts (equivalent to 12.4) is 75%
24.4 volts (equivalent to 12.2) is 50%
24 volts (equivalent to 12) is 25%
under 24 volts and you are screwed.

The only problem I have with this is that I dont quite agree with all the figures.
Here is my attempt to slightly improve the table...
25.2 volts (equivalent to 12.6 on a 12v system) is 100%
24.6 volts (equivalent to 12.3) is 75%
23.8 volts (equivalent to 11.9) is 50%
23.6 volts (equivalent to 11.6) is 25%

What does the panel think

[Jbaffoh]

Originally Posted by Fuss

In another thread Dockhead gave us a great little table with state of charge at various voltages with the house batteries lightly loaded.

I thought this was great as normally it is not realistic to switch all off and let the batteries settle before calculating the % of charge against the displayed voltage.

More realistic is to have the batteries lightly loaded (5a draw for example) for a while and then check the voltage.

I think we would all agree that at the end of the day its a bit rough and depends on this and that a bit... but I think its a very good idea.

Here is the table....
25.2 volts (equivalent to 12.6 on a 12v system) is 100%
24.8 volts (equivalent to 12.4) is 75%
24.4 volts (equivalent to 12.2) is 50%
24 volts (equivalent to 12) is 25%
under 24 volts and you are screwed.

The only problem I have with this is that I dont quite agree with all the figures.
Here is my attempt to slightly improve the table...
25.2 volts (equivalent to 12.6 on a 12v system) is 100%
24.6 volts (equivalent to 12.3) is 75%
23.8 volts (equivalent to 11.9) is 50%
23.6 volts (equivalent to 11.6) is 25%

What does the panel think

Well, as Dockhead also said, Trojan considers 12.73 to be fully charged, give or take a small percentage. Further, this should be corrected to temperature. My batteries routinely read 12.84 (actually 6.42 per 6v batt) after 24 hours. So 12.6 is too low for my taste.

Further, Trojan considers 12.1 to be 50%

Since I use Trojans, I stick to their chart.

[sailorboy1]

of limited use as you can only use voltage values when the battery has been under no load and is cold

[Jbaffoh]

Originally Posted by Don Lucas

of limited use as you can only use voltage values when the battery has been under no load and is cold

Did you read Dockhead's thread?

[hellosailor]

Fuss, the first time I saw voltages charted into percentages of charge was in Practical Sailor and iirc all the new charts read the same way. Assuming 12.6 is 100% charge, for every1/10th of a volt lost you have lost 1/10th of your charge, and 11.6 volts represents no useable charge. (Even if you can keep a flashlihgt bulb lit with it.)

I know, engines can be started with less, but the simple way is to fogure whatever is your "full" voltage and charged and settling and the float charge are all done. Then figure one volt below that is "dead".

That may be more conservative than reality but let's face it, once you are under twelve volts, you really want to stop doing math and start charging batteries.

[Astral Blue]

There is a complex algorithm used to determine percent charge on a battery. Voltage is just one factor; and not the only one. The Xantrex charge monitor is worth its weight in gold -- as it is able to calculate it instantly.

[Adelie]

The answer depends on the size of your battery bank and what the draw is.

See the 3rd page of this link for a graph showing the state of charge vs voltage at various draw rates (c/x) where c is the capacity of the bank in amp-hr and 1/x is the fraction of that capacity used per hour.

At very low use rates you can be well above 12.0v and at the same time be more than 50% DOD (Depth of Discharge). 5amp from a 200ah bank would be c/40, 5a from a 400ah bank would be c/80.

I would think that if you had an ammeter and a very accurate volt meter, knew your bank size and posted the curves next to the meters this would be a cheap and simple alternative to a Link monitor or similar.

Looking on-line I found some simple analog ammeters for RV's for about $40 and digital voltmeters for about $80.

[daddle]

Temperature has more effect than you might think. Voltage is an unreliable indication of charge anyway. Only good for a general idea, or perhaps after a long intimate relationship with the batteries. Less than perfect connections also subtract from the voltage, which should really be measured directly at the battery studs. Specific gravity is the way to go.

[Jbaffoh]

Ok, well this thread has drifted far afield of Dockhead's original concerns, and the responses totally ignore the contextual limitations originally posed. Can't really blame the commenters for not cross-referencing the original thread, though.

Bottom line: I believe Fuss was trying to refine Dockhead's attempt to roughly estimate state of charge when it is (1) impractical to conduct open circuit testing after a 24-hour rest period, and (2), there is reason to believe the existing battery monitor is incapable of delivering untrustworthy readings. The idea was to measure under small loads to roughly simulate an open circuit after 24 hours, and, if there is any validity to that method, then what would be an appropriate scale for estimating SOC based on measured voltages.

The responses remind me of the old economist joke, where an engineer, a priest, and an economist are stranded in a canoe without a paddle. After the engineer and priest offer their intricate solutions, the economist begins his with, "assume we have a paddle."

[Fuss]

Originally Posted by Jbaffoh

Ok, well this thread has drifted far afield of Dockhead's original concerns, and the responses totally ignore the contextual limitations originally posed. Can't really blame the commenters for not cross-referencing the original thread, though.

Bottom line: I believe Fuss was trying to refine Dockhead's attempt to roughly estimate state of charge when it is (1) impractical to conduct open circuit testing after a 24-hour rest period, and (2), there is reason to believe the existing battery monitor is incapable of delivering untrustworthy readings. The idea was to measure under small loads to roughly simulate an open circuit after 24 hours, and, if there is any validity to that method, then what would be an appropriate scale for estimating SOC based on measured voltages.

Yes, thats all true.
I liked Dockheads table and I am going to start using it and if I find the numbers slightly out then I might modify them a bit over time.

[sailorboy1]

Originally Posted by Adelie

Looking on-line I found some simple analog ammeters for RV's for about $40 and digital voltmeters for about $80.

That $120 would almost get you a monitor!

Far as using voltage as an indicator under small loads as an indication of charge: on the boat this weekend I was logging SOC, amp ahours out, voltage, amps; and with less than 1 amp the voltage would indicate the battery was less than 50% charge when it was in the 80s. So to me unless the load has been off the voltage is not worth much as 1 amp was enough to effect the voltage reading.

[btrayfors]

Originally Posted by Don Lucas

That $120 would almost get you a monitor!

Far as using voltage as an indicator under small loads as an indication of charge: on the boat this weekend I was logging SOC, amp ahours out, voltage, amps; and with less than 1 amp the voltage would indicate the battery was less than 50% charge when it was in the 80s. So to me unless the load has been off the voltage is not worth much as 1 amp was enough to effect the voltage reading.

If it did, it means one of three things:

1. your voltage measuring device is way off;
2. you didn't measure voltage directly at the batteries; or
3. those batteries are shot.

FWIW,

Bill

[GordMay]

I don’t believe there is any practical use in trying to reduce State of Charge to a simple voltage table. There’s too many variables.
The following paper may explain the difficulties in modelling battery performance:
➥ http://doc.utwente.nl/64556/1/BatteryRep4.pdf

[btrayfors]

While I generally agree with that, Gord, I believe that there is nevertheless value in such a table IF the user knows his batteries, his measurement devices, and his boat.

Voltage measurements CAN be a valid, useful, and sometimes ONLY reliable way to estimate SOC (e.g., gels and AGMs where you cannot use a hydrometer), especially since many battery monitors are uncalibrated and misleading re: SOC over time.

Consider the following graph made during a two-year battery testing activity. Note especially that it is typical of fully-charged flooded batteries when put under a C/20 load or higher (10A in this case) to show a considerable voltage drop early followed by a recovery and relatively stable voltage level. Gels and AGMs tend to have a more direct voltage decay.

Note also the difference equalization can make (red and blue curves...same battery bank).

First20minEqual

Bottom line: if you know your boat, your voltage measurement equipment, etc., then measured voltage can be a good indicator of SOC.

Bill

[hellosailor]

Gord, there's always a value in q&d ballpark calculations. That's all the voltage is supposed to be.

As to temperature compensation, sure, that's important too when you're not sailing at 20C/68F. But when was the last time you saw a temperature compensated hydrometer, than automatically took the temperature into account? Right then, same shortcoming as voltmeters. You still have to know how to use them, still have to make the same compensation. Still have to wait for the electrolyte to stir itself up and unstratify.

There's more than one way to turn a cat into furry mittens.

Bill-
"especially since many battery monitors are uncalibrated " You mean, a Xantrex or Victron (or worse yet, an MPPT controller with integrated display and monitor) can't be trusted any more than the rest? (I haven't had a chance to take the calibration source out exploring.)
And I couldn't help noticing your test chart on the echo-charge seems to indicate that it reduces voltage (to the starting battery) about .4-.5v compared to the house battery voltage?? Shouldn't that be a Very Bad Thing?