Battery state of charge calculation

[GHA]

Absolute battery capacity isn't relevant as it only reports a percentage result. So 60% doesn't mean 300Ah left in a 500Ah bank. It just means 60% of whatever capacity your batteries can store. Older 500Ah bank might only manage 400Ah and it would be 60% of that figure.

Good writeup here, he found the same inaccuracies during charging > https://marinehowto.com/smartgauge-battery-monitoring-unit/

I actually did carry out tests on my previous set of batteries and it was very labour intensive over 4 days. I monitored Voltage, Amps load, SG and temperature hourly from very early in the morning for about 18 hours each day.

Temperature control wasn't a big deal as I stuck each battery in a big water bath overnight before testing and during the test. I don't think temp. varied by more than a degree C at most.

Load control was a bigger problem but I did manage reasonably well with a variety of small bulbs and a larger one. I noted actual current and included that in the final calculations so that Ah wouldn't be far off.

Hardcore Plans afoot here to hopefully build an arduino based constant current load which will let the house loads stay on and log all the data, might work, might not.. gold dust if it does to keep track of capacity, even if it's just relative.

 

[Mistroma]

Good writeup here, he found the same inaccuracies during charging > https://marinehowto.com/smartgauge-battery-monitoring-unit/




Hardcore Plans afoot here to hopefully build an arduino based constant current load which will let the house loads stay on and log all the data, might work, might not.. gold dust if it does to keep track of capacity, even if it's just relative.

Interested to see the results when you finish the build. I was planning to do something about recording data automatically. I do have automatic recording of daily figures and keep manual records but an automated system would be better.

I'm not too bothered about a constant load as I believe that simply recording actual input and output is more useful. I'm not too worried about proving I have an actual capacity of nnnAh when the domestic bank is fully charged. Knowing the actual capacity for my normal usage profile is sufficient.

Amps. will jump up and down as fridge, lights and instruments turn on and off. I'd be happy with an estimate of capacity based on real values from continously changing Ah, Voltage and Wh. I know that Peukert's law means that results will vary for diferent discharge rates. However, results should be fairly predictable when pattern of usage remains fairly constant.

Still a nice project to complete.