How to know where to stop when you battery is under a load

[oldcat]

when my lead Acid battery are fully charge at 12.7 volts- under a 10 amp- load -it reads- 12.1 volts how do i no where to stop at 50% when the battery chart says stop at 12.10 ..

 

[gnubie]

Battery voltage as a state of charge indicator is really only a guide as there are many variables. What you want is a device that can measure power drawn from the battery and power put back into the battery. Often this is done in amp hours. Search for coulomb meter.

 

[Fred]

gnubie, could you please elaborate on that? I can understand how counting amps would tell you how many you have removed from a full battery. But, how do you know when the battery is full? More importantly, how would you know when the battery is 80% charged since that is the preferred level to charge to for extended battery life (for LiFEPO4)?

If you only used the amp counter to track amps in, amps out then over time the count would be less and less accurate due to cumulative small losses, inaccuracies, etc. At some point the counter needs to be reset right? Is the idea to charge to 100% once a month and reset the counter then?

 

[offgriddle]

I have often wondered that myself, how does any battery monitor that gives a percentage of charge indication know? If the battery percentage of charge reading is based on the no load voltage reading of the battery, then I guess you would have to remove all loads, (except for the nominal load placed upon the battery by the monitoring device itself), and use the no load reading as your indication for battery percentage of charge.

 

[Fred]

I think that is the way to really know the state of charge however not only do you have to disconnect loads, you are supposed to wait as long as 24 hours for the voltage to stabilize to get a true reading. Of course that is usually not practical.

Would it be possible to develop a correction factor for the apparent voltage reading that is a function of the load?

You gave one data point- at 100% SOC and a 10 amp load the correction to the apparent reading would be +0.6 volts. I wonder if someone were to test this at various SOC levels and loads could they develop some tables or formulas that could do a pretty good job of correcting the apparent voltage?

 

[offgriddle]

I think that is the way to really know the state of charge however not only do you have to disconnect loads, you are supposed to wait as long as 24 hours for the voltage to stabilize to get a true reading. Of course that is usually not practical.

Would it be possible to develop a correction factor for the apparent voltage reading that is a function of the load?

You gave one data point- at 100% SOC and a 10 amp load the correction to the apparent reading would be +0.6 volts. I wonder if someone were to test this at various SOC levels and loads could they develop some tables or formulas that could do a pretty good job of correcting the apparent voltage?

Yep Fred I was thinking about the requirements of a stabilization delay time period, (at least in lead acid chemistry not sure about other battery chemistries), in order to let the movement of electrons through the elrctrolyte between poles settle down to their most static state, and then take a voltage reading. By introducing a delay period we are splitting hairs a bit for practical, at home state of charge readings but if a person was housebound for example, owned a lab coat and had plenty of time on their hands .... ?

 

[Kcp]

I dare anyone to search for "Coulomb Meter" and make any sense out of what it is and repeat it in normal-person speech. Better to search "Bogart Trimetric battery monitor" or "Expion360 battery monitor" for the universal battery monitors or for Midnite controllers the "Whizbang Jr." All will come with or need a shunt.

These devices need to be programed to the amp hours capacity of your bank and you can see capacity from there. But. there is always a but with lead acid and I won't go into it involves older batteries and such.

While I find it odd that Oldcat drops from 12.7v to 12.1v under a simple 10a load I'll assume it's a tiny battery, a small 10ah sealed thing that looks 'cute'.

 

[offgriddle]

I dare anyone to search for "Coulomb Meter" and make any sense out of what it is and repeat it in normal-person speech. Better to search "Bogart Trimetric battery monitor" or "Expion360 battery monitor" for the universal battery monitors or for Midnite controllers the "Whizbang Jr." All will come with or need a shunt.

These devices need to be programed to the amp hours capacity of your bank and you can see capacity from there. But. there is always a but with lead acid and I won't go into it involves older batteries and such.

While I find it odd that Oldcat drops from 12.7v to 12.1v under a simple 10a load I'll assume it's a tiny battery, a small 10ah sealed thing that looks 'cute'.

The clamp part of this meter is a coulometer. https://www.amazon.com/FLUKE-362-Tr...=2025&creative=165953&creativeASIN=B00L7KZYSA

 

[offgriddle]

PS: I speak fluent, "normal person", because I'm not!

 

[oldcat]

I dare anyone to search for "Coulomb Meter" and make any sense out of what it is and repeat it in normal-person speech. Better to search "Bogart Trimetric battery monitor" or "Expion360 battery monitor" for the universal battery monitors or for Midnite controllers the "Whizbang Jr." All will come with or need a shunt.

These devices need to be programed to the amp hours capacity of your bank and you can see capacity from there. But. there is always a but with lead acid and I won't go into it involves older batteries and such.

While I find it odd that Oldcat drops from 12.7v to 12.1v under a simple 10a load I'll assume it's a tiny battery, a small 10ah sealed thing that looks 'cute'.

i have 2 - 6 volt 205 AH crown batterys i have on money for a shunt and monitor right now if it wood work......

 

[Kcp]

What those meters I mention do is not only tell the SOC% (I always ignore that and have a Whizbang Jr- there is no Sr) and voltage but also current amps in and out and amp hours remaining in real time. Real time means they track both charge AND discharge from the batteries. They're the bees knees to really know what's going in and out of your batteries. For example, they'll show I'm using -5 amps, turn on a light, now -7 amps so that light uses 2 amps to be on. If I was charging at 10 amps, and using 5 amps the meter would show +5 amps going to the battery. So Oldcat your 2 6v @ 205ah you'd program the meter for 102ah as total capacity. When the ah=0 you're at 50%.

I'd recommend the Expion360 meter over the Trimetric for 'works on any battery configuration meters' simply due to the simplicity of Expion360 over Trimetric. There's a thing in programming- never allow the programmers to make a UI as it will won't make sense to an end user. Trimetric didn't heed that wisdom- the install sheet is confusing, way too many wires and using the unit means a small learning curve. The Expion however is a piece of cake to understand how to hook up, the wires have 1 connector to the shunt that goes to the unit and 1 other wire hitting battery + for power, programing is simple and there's some other perks in the display- only problem is the size of the display unit is too small.

For Midnite controllers the Whizbang is simple to hook up and the controllers already have a screen built-in for the information. Unfortinatly the 30a Kid info pales to a Classic, plus the Classic keeps the Whizbang screen on all the time while the Kid don't.

 

[fat_old_sun]

i have 2 - 6 volt 205 AH crown batterys i have on money for a shunt and monitor right now if it wood work......

If a 205 Ah battery quickly goes from 12.7 V to 12.1 V on a 10 Amp current, it is very likely that it is not fully charged or sulfated (or both). Charge it to 14.4 V and hold it there until the charging current drops below 4 Amps.

I have a soon expiring $35 off code for a Bogart Trimetric monitor & shunt. PM me if interested.

 

[oldcat]

fat_old_sun ------- sorry my batterys are good and new and thay are not sulfated i keep them charged i just don't no ware to stop when under a load when looking at my battery with a multimeter... and the voltage come back up when they sit a bit . ...... i can't afford a Bogart Trimetic monitor........

 

[Freddmc]

Or you can get a Blamar Smart Gauge (not for Lithium though). It connects directly to the battery(no shunt needed) with two wires and takes into account all the variables. Also Balmar has a similar gauge for lithium batteries (with a shunt)

 

[JIMMY]

The manual the battery manufacturer sent me for the old flooded lead acid batteries in my off grid beach house has this graph.

I found it pretty handy as a quick reference in a working system and reduces how often I am measuring the specific gravity of the electrolyte to obtain a true SOC.

Jimmy

 

[Donald Siegel]

Like many here I want to keep it simple, get some lengthy cycles out of my batteries but also not let them occupy my full attention with every use. I have a MPP Solar (PIP) with a relatively steady load daily using Renology gel 200amp batteries. Life would be so simple if I could just follow the manufacturer recommendations and charge to 12.8v and discharge to 12.0v. But as the author of this thread observed your voltage measures high when charging and low under a load compared to a no load 24 hour rested battery, so simple is not so simple. I decided to figure out how high to set charge voltage so that the disconnected, rested battery lands at 12.8v and how low to set the battery cut off so the no load rested battery is at 12.0v and leave it at that. In essence I decided to believe the no load, rested voltages as an accurate state of charge/discharge and call it a day. That may not be the best way, but it’s simple and I suspect not killing too many cycles.

 

[Red Squirrel]

In telecom we pretty much just go by the voltage. Float is at 54 volts (2.25v per cell) for a 48v system when AC power is on and rectifiers are running. So when power goes out and the load is running off the batteries, the voltage will typically drop to around 49.x volts fairly fast. It will hold there for quite a while, sometimes a day or two depending on how big of a battery plant is installed. At some point it will drop to 48.x. When it's at 48.0 that is basically where you start to plan if you need to send a generator. We check the Hydro outage map to see when their ETR is and play it by ear based on how long it's already been out for. Once it hits the 47.x volt range then we start to act. This is not precise by any means, but good enough. Though every now and then some sites with super dead batteries will catch us by surprise and drop to 47 within an hour.

Of course the load can have influence on the voltage, but it's a safe enough rule of thumb to go by. Essentially, when it hits 47 consider it as almost depleted and take whatever action is needed. I've seen some go down to like 42 volts and the equipment is still running, but that is very hard on the batteries.

The nice thing with lead acid batteries is when charging they don't care about current. You can stick 2.25v per cell to it and they'll charge. You don't need to balance charge them either. Very good for telecom application where the load and batteries are in parallel. I think at some point we might see lithium ion in telecom but I think it will be a very long time just because lead acid is so much simpler, no need for BMS or anything.

 

[fat_old_sun]

You don't need to balance charge them either.

Lead acid batteries do need balancing, but it goes by a different name - equalizing. In telecom applications of lead acid batteries, the batteries don't see much use. They are kept in float most of the time and rarely discharged to any significant level and the voltage is a good indicator of the batteries' state of charge.

In applications where the batteries are deeply discharged regularly (fork lift trucks), voltage is simply not a good indicator of the SoC. The only reliable way is to "count the electrons" going in/out of the batteries.

 

[Red Squirrel]

True but equalizing is simply giving it a higher than float voltage for a little while. You don't need to do anything fancy like monitoring the current going into each cell, which is hard to do for an online application. Typically we do measure midpoint voltage but that's only an alarm point in case something is wrong with a cell.

It is a good idea to run an equalize once in a while though to prevent sulfation. Ours typically run one after an extended power outage.

 

[fat_old_sun]

The effect of applying a higher than normal voltage to the cells makes sure all the cells have the voltage - exactly what cell balancing does in lithium chemistry packs. Higher voltage stirs up the electrolyte (otherwise, the electrolyte stratifies) and converts lead sulfate in to lead/lead oxide. To do it right, current during equalizing *must* be limited/monitored. Or else, you'd lose a lot water to electrolysis.