question

Hi @autimagic

Nice setup!!

TLDR: the batteries are unbalanced.

how many batteries (3 banks?) and how are they connected?

I see you have midpoint sensing, but this can only be on one bank?

Problem here might be batteries that weren't charged to the same SOC (normal procedure: put all batteries parallel and fully charge for a couple of days to all get to 100%)

Once batteries are unequal, these problems get worse as you have noticed.

Normally best practice is to use interconnects to group all batteries together, minimizing differences, like in this diagram: (I do realize this can be difficult if you have multiple battery boxes)

Adding battery balancers can help, but they also have limits, once batteries are 'unequal' they might not have enough power to compensate.

Hi @Daniël Boekel (Victron Energy Staff)

Thank you very much - and thank you for your input!

Initially, we were provided with a 12V charger by the distributor, with instructions to charge each battery till the charger said "storage" and move on to the next.

Banks are connected like this.

At all + poles where the midpoint is connected a resistor sits as well (don't know its size - can't remember and it's wrapped in heat shrink).

So basically, the one BMV measures the midpoint of all three banks. Ish.
Not my design :-)

The batteries are sitting like this in a big shelf, so interconnecting them would be relatively easy, but please note the shunt and busbar is sitting ~5 meters away - if that has anything to say.

How come there is no interconnection between the two rightmost batteries in the middle and bottom string in your drawing?

The first simple action that I would try is to give the batteries a deep discharge, say down to ~50% SOC and then to give it a high powered charge back up to 100% - ideally with a charge current of ~20% (at least 10%) the battery bank C20 capacity.

Monitor the charge current and keep the charger in absorption phase until the tail current has dropped down to ~1% the battery bank C20 capacity.

You should see some improvement after the first cycle but you will probably need ~3 cycles (maybe 1 per day) to see a good improvement.

Before you do this first check individual resting voltages to confirm that there is not an obvious issue with a single battery.

If that doesn't work sufficiently then take all the batteries out and wire them all in parallel, then conduct a ~12h charge at the appropriate absorption voltage - to help equalize things. If your charger/power supply doesn't have enough capacity to charge the entire battery bank in parallel then do it in groups. You could also do this first if you have the time and ability.

Depending how long it has been like this, it's possible that one/some batteries could have some permanent damage.

Also note that the mid point voltage is exactly that - one/some of the individual batteries could be even worse when measured separately.

I didnt know you could have one batterybank to more then one inverter/charger.

When i read your post it sounds like you have Three independent quattros connected to the same batterybank (3x48v strings)?

This may also be being caused by a bad battery, even a single cell. After all, the reason why midpoint deviation is actually measured.

When being worked, this may also manifest itself in some batts getting hotter than others. Often detectable by just touching them, but a handheld IR sensor may help confirm more accurately.

Don't overlook. I've seen batts come from the supplier stone-dead.

@Daniël Boekel (Victron Energy Staff)

I got hold of a Phoenix 12V 50A charger and a breakout cable to 12x battery connections (all same length) from our distributor.

Left the charger in "normal" mode and had it charging for three days.

Disconnected all cables and left the batteries for ~4 hours to settle and had our distributor to measure the unloaded voltage of all batteries; Anything between 13.09 VDC and 13.27 VDC was the result!

Our distributor will try and get hold on a difference battery bank.
They suspect something is wrong with the batteries - who are all brand new.

Any ideas?

I still have the charger and breakout - I can do a lot of testing if you are curious - and I just want this to work.

An order for 12 new batteries has been placed and I was told just to connect the batteries for now (so we have light etc. in the truck).

I connected only the first string - and powered on only the first quattro;

Any great ideas? :-)

@Daniël Boekel (Victron Energy Staff)

Hi Daniel and all you others :-)

I totally forgot to get back to you on this one.

Thank you very much for all your ideas, advice etc!

Things were tried and tested.
We tried - once more - to balance the banks mentioned above.
No succes.

We removed all banks and got a bank of four brand new batteries of the same kind.
We had our supplier to balance them and install them. Exact same issue.

Then, your distributor sourced batteries of same capacity and type but a different brand (if you want details, brands etc. I'll happlily send you and email). They were balanced the same way and has been running without problems at all for about two months now.

Basically, I'm happy.
But also curious - and so is the distributor.
Neither of us can't explain what happened.

Hi there I had similar issues with my 48/1000/140 and I have 16 EFB yuasa, one of my batteries as mentioned above had a bad or lazy cell as the voltage difference matched exactly one cell plate in the battery, as mentioned also I discharged them all and charged again but not with the MPPT but with the Quattro it’s self I left all 16 in parallel for a week then connected them all again with the shortest possible means of cable and lugs and I haven’t had a problem since.

david