the victron manual talks about connecting the midpoints on page 21 and 22. Which is exactly the way I had it connected and yes I did have the balancer connected as well. It ended up frying all 40khw of expensive AGM batteries. later I hook up the same configuration with the lithium batteries and measured the current at all points and discovered current moving very disproportionate. As soon as I disconnected the midpoints all strings have the same current. then I studed the diagram here in the victron manual and if you visualize the path of least resistance you can see how it would flow through one battery jump to the midpoint and then shoot down through the other batteries. so it's the resistance of the batteries isn't absolutely exactly the same this will occur. However disconnecting the midpoints causes batteries to work as sets together and if there resistance is different than they will deliver it slightly different current but they won't be disproportionate to each other in the series set.Interesting. The diagram for the 4 channel balancer linked by Sycamore showed the midpoints connected, and I've also seen it recommended in other places. Like in this Victron PDF for example:
https://www.victronenergy.com/upload/documents/Wiring-Unlimited-EN.pdf (p21)
I assumed connecting the midpoints would be ok, especially if there is a balancer connected. But I guess any minor difference in cable length and connector resistance could throw the whole thing out. The Valance batteries do have an extremely low internal resistance, which probably makes any cabling differences more significant.
So a balancer channel needed for every battery then. If using 2 of those 4 channel balancers for a 6 battery system, would you parallel 2 of the channels of each balancer?
At the time I was running a 24 volt setup. Eventually one of my AGM's shorted out one of its cells and it's voltage dropped but because it was paralleled to 9 other batteries it brought the voltage of those down with it. The charge controller for the lead acid setup didn't see individual cell voltage so even though half the battery Bank was refusing to rise to the correct voltage it didn't stop bringing the voltage up to the set point. So the voltage went into the other half that wasn't shorted out causing it to go way higher than it was supposed to which caused it to cook and short out as well then I ended up with 11 shorted out batteries and 9 that were severely undercharged followed by extra severely overcharged. It took about two months for it to catch my attention and by the time it did all the batteries were toast.
That was the second huge lead-acid loss that led me to lithium. I really wanted a battery management system that new the voltage of every single cell. Despite popular belief I think this is even vital with lead acid batteries but nobody creates such a system and most of the batteries don't even allow access to the cell. I dreamt of buying single cells for submarines like 2000 amp hour single cell lead acid batteries and putting them all in series and switching to a higher voltage setup that way I would have access to monitor each cell and if one ever shorted out it wouldn't screw with any of the others too badly. and if I had alarms and safety shut-off controls yada yada and eventually I thought you know what they're already doing all this with lithium I need to bite the bullet.