# Battery charging while in use in a bank

I have fifteen 12V 150Ah flooded batteries in series as part of a 10kVA off-grid system. Every once in a while, I test individual battery voltages. I've noticed that a few of them are, regularly, lower than the rest. For example, here's a reading I took today (#1 thru #15):

12.18   12.23   12.29   12.17   12.07   12.17   12.15   12.27   12.24   12.21   12.08   12.17   12.03   12.23   12.15


As you can see, batteries #5, #11 and #13 are "behind" the rest. And, here's the average of about two dozen readings taken over a period of 18 months:

12.19   12.28   12.29   12.12   12.08   12.15   12.02   12.25   12.21   12.16   12.07   12.12   12.03   12.23   12.13


So, this is a chronic problem.

Btw, here's the battery voltage history.

I've tried "equalization" a number of times, but given the nature of off-grid systems and the lack of a suitable generator, equalization hasn't followed the typical protocol (raise individual battery voltage to 16.5, continue charging at 3-6 amps for 12 hrs) ... and it hasn't achieved much.

So, here's what I'd like to do: Using a 12V battery charger, charge each of #5, #11 and #13, in turn, while they are in use. The plan would be to run the charge between sun down and sun up, doing one battery each night.

Can someone point out the holes in this plan?

Also, can somebody answer the key question: is it technically sensible to charge a single battery within a bank of 15 while the bank is connected to the load? [My worry is the fact that the charger will be connected to 2 circuits; one, the battery being charged; the other, the rest of the batteries and the load ... in the reverse direction]. Some explanation of why or why not would be appreciated.

• That would work but it might be that these 3 batteries are simply the "weak brothers" in the pack and after a while they will again have a lower voltage. Also a fully charged Lead-Acid battery can have a voltage above 13 V. If you charge a battery overnight it might reach that value, that's not directly an issue but will again result in imbalance. It would be better to charge until the battery's voltage is about 12.2 V (or the average voltage of the other batteries). – Bimpelrekkie Feb 15 at 7:24
• @Bimpelrekkie. Responding to the first part of your comment, "after a while they will again have a lower voltage". Possibly. But take a look at #7. Its average (12.02) puts it as the lowest; but not any more (currently 12.15). Its history is that it was taken offline (using a replacement) and fully charged over 48 hours before being put back online, some 8 months ago. My hypothesis is that it will be easier to get the benefits of equalization if the batteries start off with voltages close to each other -- hence the attempt to "lift" 5, 11 and 13. – RudyF Feb 15 at 8:08
• Is it actually giving you a problem that those batteries are slightly lower voltage? Are they limiting your max discharge, when the others in the string still have plenty? If so, they may need replacing, not messing around with. In the limit, what you're doing by providing charge to them in use is changing the charge balance of the whole string. That's fair enough, if you can't replace dud batteries, and must keep the bank going in a life or death situation, but it's not the way to run a battery bank. Test capacity, and replace if actually low, otherwise ignore. – Neil_UK Feb 15 at 10:43
• What do these batteries measure, when NO CURRENT is demanded? Are the "low" batteries simply colder, because of physical location? – analogsystemsrf Feb 15 at 14:37
• How is the water level of the batteries? Do some need topped off more often than others. How does that correlate with the batteries that are low? – vini_i Feb 15 at 16:05