question

codefoster avatar image
codefoster asked

BMV-712 voltage level versus state of charge in old AGM batteries

I just installed a BMV-712 for my bank of 2 Victron AGM 6V-225AH batteries. I used to use the voltage level on my old meter to determine my SOC. I followed the values in this chart followed advice to try to never run my batteries below 50% (12.05V). I did a test by leaving on a ~3A DC load for about 24 hours. After doing so, my BMV-712 told me I was at 77% and 12.25. That's approximately a .2V discrepancy from the chart (where 80% = 12.5V)... not too big a deal. Then I removed the load and waited 25 minutes and the voltage was up to 12.47V. That almost directly lines up with the chart... all good. I'm just wondering if I'm reading this right. These batteries are pretty old (~6-7 years) and seem to be in great shape, but how would I track them for degradation. Would the V to SOC numbers stop lining up so well or will it just lose energy quicker? Is the BMV smart enough to account for degraded batteries?

BMV Battery Monitorbmv power consumption
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
Alexandra avatar image
Alexandra answered ·

@codefoster

The battery monitor would only count energy in and out of the battery. So really if it is the same load everyday then energy returned, it would look right all the time. But as batteries age they loose capacity.

So really you need to work out if you still have the same capacity. And to work out capacity is to do a full discharge test at its rated c rating and then charge it up again at its rated c rating and see how much was put back in, then enter that as your new battery capacity in the BMV.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

snoobler avatar image
snoobler answered ·

The chart only applies when the battery has been completely at rest, no charge or discharge, for several hours.


Tracking is tricky as temperature is also a factor. You'd basically have to track, voltage, SoC, current and temperature for correlation. It's a mess and is often inconclusive.


I periodically run a reserve capacity, RC, test. This is the number of minutes a battery can provide 25A and stay above 10.5V. In most cases, this is notably higher than the C20 rate, so you're not actually testing them down to 0%. Some batteries have other capacity ratings for higher currents. Testing to these values also help you avoid taking them down to 0%.


If the RC is 100 minutes, and I get 90 minutes, I assume the batteries meet 90% of rated capacity. I then adjust my capacity in the BMV to reflect the tested capacity.


There are also other times where I might do a deeper than normal discharge. If I can leave the system offline for a few hours to allow the voltage to rebound, I'll use the voltage to SoC charge for correlation.


Another thing to watch for with batteries in series is that the battery voltages stay identical. If they aren't perfectly balanced and matched, one battery can get over charged and the other under charged.


I doubt 6-7 year old batteries meet capacity unless they've been very lightly used and always maintained at a good float voltage, but they could be at a very high state of health.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Related Resources

Victron BMV battery monitors product page

Additional resources still need to be added for this topic