Battery Distances and Layout

Hi all! Bit of a conundrum I can't seem to make ground on...

Due to some unfortunate geometry, I'm going to have to fit my batteries in two banks, about two metres apart. Each bank will have two 200Ah 12v batteries (4 cell LiFePo4). I'll need these to all end up in parallel, for 800Ah at 12V. The max constant discharge will be 300A.

The problem I'm having is how to wire these to achieve that, bearing in mind that all of the electronics (chargers/inverters etc) will be adjacent to one of the banks (and hence 2m from the other bank).

My thinking is thus:

Naming the batteries 1, 2, 3 and 4 (where 1 and 2 are adjacent to the electronics and 3 and 4 are far away), if I take the positive from 1, I should be returning the negative to 4 in an ideal situation. This involves running three 2m cables (to connect 2 and 3 and a negative wire from 4 to the electronics). (Option A)

I could potentially wire them such that 1 connects to 3, 3 to 4, and 4 to 2. This would involve using four 2m cables (two from 1 to 3, two from 4 to 2). This would then allow me to use a short negative cable from 2. (Option B)

The question then, is what these distances/layouts do to voltage drop and wire thickness consideration? To try to ask for some specifics rather than leave it that wide...

1) When considering the 'circuit distance' from battery to inverter (etc) back to battery, what consideration is given to the wiring between batteries/cells in a parallel setup?

2) In Option A, assuming 25cm from Battery 1 to an inverter, and 2m from Battery 4 to the inverter, the total distance is either 2.25m or 4.25m or 6.25m depending on the answer to 1. In Option B, assuming the same 25cm from Battery 1 and 25cm from Battery 2, the total distance is either 0.5m, 4.5m or 8.5m. Which is it?

3) Is there a better way to wire this? :lol:

Many thanks in advance for any help anyone can offer!

If you size (or oversize) all your wiring for at least the max current draw it would ever see, and they're all in parallel, then the best way is probably to use equal wire lengths for the two cables running from bank 1/2 to bank 3/4, as short as you can manage them, and run as closely in parallel along their length as you can manage for as much fo the run as possible.

The idea is simply to lose as little power as possible in the wires themselves, so the thicker they are, the less watts are lost in them as heat. Oversizing them won't hurt, other than cost, and a little bit of weight. If you can't get (or afford) larger cable, then double up the cable you have to double the area of conductor.

If you severely undersize them, they'll lose power to heat, and also allow voltage to develop across them such that 1/2 will be at slightly lower voltage than 3/4, while under load. This will even out shortly after the load is removed as 3/4 recharge 1/2 thru the paralleling wires.

You don't need more than two cables, one for + and one for -, to connect 1/2 to 3/4, because they are all in parallel. You'd have a separate paralleling pair between 3 and 4, and a separate pair between 1 and 2.


If you are not running them in parallel, but rather in series, you still only need two cables, one for + to the - of 1/2, and one for - to be the main in/out -, and then a cable between the other + and - of 3/4.


I don't see the reason for a third long cable, unless 3 is also very far from 4 as well as from 1/2. In that case, you'd need to run two pairs of long cables, one to 3 and one to 4.

amberwolf said:

I don't see the reason for a third long cable, unless 3 is also very far from 4 as well as from 1/2. In that case, you'd need to run two pairs of long cables, one to 3 and one to 4.

Click to expand...

Thanks for your reply. I've pulled this bit out as I think it sums up our differing thought and where I'm stuck on this.

My understanding of how to set up the batteries in parallel (how it was on our boat and how I was taught to in school) is that with all positives connected sequentially (1 to 2 to 3 to 4) and all negatives connected sequentially, the positive current should be taken from the positive terminal of 1, and the negative should return to the negative of 4. This stops 1 doing a majority of the work (if everything is connected to 1, since it has less voltage drop due to less cable than the rest of the batteries).

Hence the 'long' negative, to run from device (close to 1) to 4 (far from 1).

A thing I can't work out though is that if 2 and 3 are connected by long wires and 1 and 2 + 3 and 4 are connected by short wires, would this mean the set can't ever be balanced because there'll always be more forward voltage from 1 and less voltage drop to 4 than there is from 2 or to 3. Wow that's a complex sentence

Your school connection reference is correct ! in this way both sets of batteries will always share properly. If you do your charging that way as well, all should be fine. Proving that connection method is correct is easier to try rather than do the maths ! The method should work for two up to many batteries or even loads. I used to have lots of fun interconnecting low ohm load banks where the load resister elements were only one order of resistance higher than the connecting cables. Strict attention to the order and the path of interconnection was required to make sure the load was correctly shared.

Bob

Making the parallel connections such that they all fan out from the main connection, rather than putting them one to the next (as I suggested) would technically be the "better" way.

To do the "fan out" method correctly, so that all resistances from all packs (and thus all loads and all voltage drops and losses) are the same, you would want to make every cable that comes from the main connection to any of the four packs exactly the same length, even though the 1/2 packs are much closer than the 3/4 packs.

If you aren't doing that, and are using the shortest possible cable run for each pack location, then you might as well just use the second method, because the resistances and current flows wouldn't be significantly different, as long as you use thick enough wire.


However, if the resistance of the connections and interconnect cables is low enough at the currents needed, there will be so little difference in pack voltage drops as to be ignorable.


Except at higher currents, the resistances of the cables (if they are sized for the higher currents correctly to start with) will be so low that all the voltages will be effectively the same everywhere in the circuit, so that when low or no current is flowing, (like near end of charge), all banks will be the same voltage, "balanced" as far as the total bank voltage is concerned.

(voltages of cells *within* the banks would have be balanced by whatever internal BMS the bank has, since you do not have interconnects between all the cells of all the banks, just the ends of each bank).