I have 12 used agm batteries that I want to configure to a single 24v batterybank. How would you wire this in the best way? Do I need balancers? If yes, how many?
asked
How do I wire 12 agm batteries to a 24v inverter?
Easy to answer: put 6 batteries parallel and the 2 groups of 6 in series then you get 24v system.
agm batteries need no balancer
Balancers would be good here. And long strings of six batteries is likely to cause imbalance, even if tapped at opposite ends of the strings or connected with busbars. It's not true that AGMs don't go out of balance, although long periods on float and low currents mean self balancing is likely.
Four strings of three in parallel would be much better.
I totally forgot this thread and didn't get any notifications.
Now almost one year later I have had some imbalanced in my bank, 2 stings in 6p2s config.
I have had these disconnected and measured individually because the two 12v strings had 1.2V difference during charge and rest. Now they are pretty much okay again after some switching around.
My questions is: Why would 4 strings of 3 in parallel be better than all six in parallel?
Right now I'm thinking about wiring these in 2s6p instead. The disadvantage of this as I see it is that I will need more sensors to monitor each 24v string. Currently I only need two sensors for each 6p string. And I can easily see the difference in voltage. Also I have heard that the advantage of parallel batteries is that they balance themselves? That was the main reason why I had them wired all in parallel. But now I'm unsure, also I have gotten my hands on more of these batteries and it would need less cables to wire these in series first.
Any thoughts?
2s6p vs 6p2s?
"Any thoughts?"
Mine border on nightmares. The trouble with paralleled Pb's is that you can have the same V over each battery yet the current can't be controlled due to differing batt resistance. That resistance can be quite low too, down at the level of wire/connector resistance, so your wiring needs to be perfect. Still you'll likely see the effects of the different currents with the Temp rising in the strings taking the most current. They may be the best of the batts, but will fail early.
The options you're considering are akin to whether or not to connect string midpoints. Like in this pic, which is ironically from a Balancer manual.. 
Those connectors need not be full diameter wiring, and could be fused, so that when a fuse blows you know you have an issue with at least one battery.
I once had a system like yours (never again!), and tested regularly for Temp by hand. Also, lowering Charge V can help control it.
All I'm really saying is that this (parallel) imbalance won't go away. The likelihood of it occurring over 6x strings of older batts is quite high. All you can do is try to control it as best you can.
That's a good tip, thanks!
But still, would it make my life any easier if I connected them by two in series with bussbars first? And then parallel with my cables? I figured that they will still be balanced as individual 24v batteries (due to the parallel connection) and that would also save me a bunch of cables.
As a bonus I also have a few more batteries from same destination that I wish to add to the bank for increased ah. As long as I don't risk mor problems ofc
Sorry for the cruelty, but 'lipsticking a pig' comes to mind. No matter how you wire them you'll still have 6x (or more!) parallel strings. That said, how you wire them is still important. Please read this (and this guy knows his batteries).. http://www.smartgauge.co.uk/batt_con.html
Look at his Amp figures particularly. Surprising stuff. And when the bigger A increases Temp in a string above the others, the resistance drops even further and A keeps rising. Thermal Runaway, yes it can happen.
I don't know what other kit you have, but I use a Smartshunt with midpoint V monitoring. Graphed on VRM, you might find such useful to pick up unusual trends. And even alarm you with an email if you set a suitable deviation limit.
You could also limit Charge A. Let's say 120A is your total (20A per string max), and one string decides to take 40 A. But limiting Absorb V is useful too. Stuff you could try, hey..
You have used batteries. Good chance of very different states of charge. And wear. Make sure all fully/equally charged before connecting.
completely agree, you can also check each battery separately with a multimeter to ensure there all within 0.1v of each other.
section 3.4
https://www.victronenergy.com/upload/documents/Wiring-Unlimited-EN.pdf
Totaly different situation, but the victron battery balancer (2pcs) do their job very good balancing my 2s2p setup on 24V. Added temperature probe to each battery, (yes it occupies the whole Cerbo temp inputs) and i have not a single issue, battery's are from the same production date though. (VMF Greensystems 105Ah) But i would like to see a balancer which can deliver more than 1A, since its really slow when there is an inbalance. If you wire the balancers to your digital inputs you can generate inbalance alarms by the balancer relay, which comes in handy.
I had a HA01 to balance the two strings I had. But it melted in the end. It's optimizong current is rated to 0-5amps.. However probably just a cheap product that I won't dare to try again even tho many have had success with it. But Im guessing that they worked in the beginning when the batteries were pretty close. After time they got worse making the balancer work 24/7 and heating it up.
Mabye this victron device would be better?
