I have always used the same voltage vs SOC at all ambient temperatures. Is that proper?
The AGM loses 15% capacity at freezing vs at rating temp of 77F, so is it still 13.0v when full? BUT at freezing temp, your charging rate of 14.8 is now supposed to be adjusted upwards to 15.2 volts to compensate for the cold. Voltage going opposite directions there maybe.
Does that matter for voltage vs SOC? The 100% full SOC is at the existing capacity, so "full" is a variable (with temperature changes). So is "full" 13.0 volts no matter what the real full capacity?
--When doing my capacity testing with Wets at the 20 hr rate, where I run the test for up to ten hours, checking SG, I am looking for the 50% "mark" according to the voltage/SG vs SOC table for the battery. Eg 12.1 volts is Trojan's 50%. Some batts use 12.2. My AGMs are more like 12.4 at 50%.
Say it gets down to that SG in 9 hours, I stop, wait for voltage bounce back for an hour (it only goes up a little more if I wait a day) to confirm the 50% voltage "matches" what the SG says, and that's my 50%. I can now say that using the time of 9 hours, the battery is at 90% of rated as new capacity.
I also use the AH counter to compare the count with the 50% mark using SG. It all comes out "right" by cross checking the SG, resting voltage, AH count, and temperature adjusted capacity.
But as Mr Wiz says, with AGMs you have no SG, so you can only go by the AH count from your estimated capacity (which is the thing you are trying to measure !!!)
What I do there is to go to say 8 hours, wait for bounce back and if it goes higher than the 50% value (12.4v for my AGMs --sort of anyway)then I restart the draw and check again at 9 hours. I check more often near the end.
So it all works the same but now can only cross check AH count with voltage after bounce back. Still, it gives you the idea of where your batts are compared with the last time you did it doing the test the same way. (Doing it the same way each time is important. Even if the capacity you get is "wrong", it is how much lower than last time that matters to you)
The glitch in the whole test whether AGM or Wets, is what 20 hr rate do you use? You don't know the capacity (yet). If you use the "as new" capacity for the rate, and the capacity is now less than that for your older battery, do you use a slightly lower amp draw? How much lower? I use the as new rate, which will give me a lower capacity measurement for the test result (Peukert)
I figure my test result at some lower capacity than it really is, will be on the good side for any testing inaccuracies, and by doing it the same way each test, I can tell how the batteries are doing. I run this test once a year to keep track.
BTW it is a problem to run the constant amps 20hr rate for 10 hours using my equipment. Using enough 12v lamps to hit the rate initially will leave you with fewer amps as the test goes on. Lamps dim with lower DC voltage and draw fewer amps. So you have to keep turning on more lights.
Inverter draw using some steady 120v draw is also a problem. The inverter will pull more amps as its DC voltage input declines, so now your initial amps will get too high.
Either way, you can anticipate that amps movement and average it out by starting a bit too high or low and it ends up lower or higher but averages out to the 20 hr rate, sort of. It comes out close enough--as long as you do it the same way as last time that's all you can do with the equipment you have for testing.
--You cannot use the AH counter by zeroing it at the start of the test and watch the count, because you don't know yet what the count will be when down to 50%. You have to use the bounce back voltage (and SG if wets) to say when you are at 50%, then you can just double your AH count to get the capacity (and compare that with the time it took out of 10 hours as a cross check)
