How to know accurate state of charge

bziarnick

Hi everyone, I have a question that I can't seem to find the answer to.  I am new to solar.  I read a few books, and watched hundreds of videos and got the courage to spend about $2,000 on building my own off grid system.  I went with a 24 volt system.  My question has to do with getting an accurate state of charge.  I know I don't want to run my batteries down below 50% due to the risk of damaging them.  The problem is how do I know when they are at 50% when there is a load on them.  I have a chart that shows at 24.20 volts I am at a 50% state of charge.  The issue I have is when I plug in even a small load of 100 watts the volts drop immediately to about 24.7 (or 70% state of charge).  When I unplug that load it immediately goes back above 25.5 volts (100% charged).  Anyone know of a battery monitor that has found a way to give an accurate state of charge even when under a load?  I have a small system of 500w of solar panels, (4) 6 volt 235ah batteries, mppt Renogy charge controller.  Thanks.

Mountain Don

What type of batteries?  Hopefully you have flooded lead acid batteries because then you can use a hydrometer to read the specific gravity of all cells.  A great one is this.   Sp. Gr. is the only accurate method. 

BB.

Just to give you an idea of what a Lead Acid battery can look like under various conditions (charging/discharging/various states of charge):

http://www.scubaengineer.com/documents/lead_acid_battery_charging_graphs.pdf

Besides measuring battery bank specific gravity--The next best method is to get a battery monitor. Basically, it measures the Amp*Hours in and out of the battery bank. Take 100 AH from a 200 AH battery bank, you have 100 AH left and 50% state of charge.

For the most part, lead acid batteries are near 100% efficient when keeping track of state of charge--Until the battery gets over ~80-90% state of charge--At that point, the battery bank begins to gas during charging (gassing is 100% loss of AH). So the monitor needs to monitor bank voltage and time at voltage to prevent the meter from drifting too much... Plus a "hard charge" to ~100% every vew days/week will "reset" the meter to 100% state of charge

http://www.solar-electric.com/batteries-meters-accessories/metersmonitors/tr20mosy.html (our host)
http://www.victronenergy.com/battery-monitors/ (good reviews from some posters here)

And, there are monitors that estimate state of charge based on battery bank voltage only (no shunt):

http://www.solar-electric.com/batteries-meters-accessories/metersmonitors/mnbcm.html (our host)
http://smartgauge.co.uk/smartgauge.html (they have a very extensive website about batteries and their meter's function)

In any case, these meters can help estimate battery bank state of charge for non-technical users (spouse, kids) and make some instructions on system operation easier (turn on genset at 50% state of charge + bad weather, off at 80% state of charge; call you if battery bank falls to 40% state of charge or less, and turn off all optional loads until problem fixed).

However, any of the above meters can also "lie" to you too... Hence checking battery electrolyte SG (flooded cell batteries) and keeping track of battery bank voltage history (only option for sealed batteries).

-Bill

zoneblue

As bill says battery monitors are only an estimate. You can never know the true SOC unless you use SG for FLA, or load test, the latter of which is destructive (both to the SOC you are trying to measure, and the battery).

I find with time you adapt to the first thing in the morning quasi "rest" voltage. If your overnight loads are fairly consistent, then this morning voltage will be more or less corollated to SOC. Now that midnites SOC model is fairly stable it will be interesting to graph SOC v the 5am battery voltage and see what emerges.

scrubjaysnest

As the others have said it's a guesstimate. A couple of links which say the same thing: http://solarhomestead.com/battery-state-of-charge/  and http://www.batteryfaq.org/

SandyP

We use one of the Victron Battery Monitors and, once we had set the charge efficiency to reflect our battery bank's specifics, it has worked well and correlates with what we see with regard to the end of the absorb stage when the charging current reaches ~1% of the battery amp capacity (ie SoC being ~100%).

zoneblue

Even if they are set up perfect in the beginning, one of the other hazards of battery monitors, is that 99% of them arent smart enough to account for the battery as it ages. A nasty spiral occurs, as the actual capacity of the cells decline, the percent SOC gets more and more wrong the lower you go.

If we take a hypothtical 100Ah battery that is 5 years old, (and has had a bit of a hiding). Unbeknown to us, the real capacity is now 75% of its original capacity. Now say one day I draw the bank down to 50% on the monitor, but is actually 75-50 or 25% SOC. This hurts the bank, and so it looses a bit more capacity, now say  (exagerating to speed up the story) 60%. You recharge it, then again draw it down to 50% on teh monitor. Now its actually, 60-50 or 10%. This is critically low for a lead bank, and maybe that one bad cell actually goes to 0, then negative. Now the entire bank is toast.

SandyP

zoneblue said:

Even if they are set up perfect in the beginning, one of the other hazards of battery monitors, is that 99% of them arent smart enough to account for the battery as it ages.

As with all battery monitoring systems none are set and forget, whether it is manually monitoring SGs or updating battery capacity based on battery age, one must always keep an eye on your batteries.