Making a split battery system: bms wiring questions.

[marka-ee]

I'm building a new battery pack out of some salvaged 60Ahr LG automotive cells. These are fairly bulky and the only way I can see fitting it on my bike(s) is by splitting the system into two halves. Right now I'm considering a 12S total system because it makes at least both halves be the similar enclosure size that I would need to make. But my main question is for the BMS wiring. Since the BMS will be attached to one of the halves, is it really important to keep all of the BMS wires the same length? So should I coil up the extra length wires that are close to the one half and thereby keeping them the same? The BMS is an ANT type. I do believe there might be some settings for compensating for things, but I'm not certain. Any advice is appreciated.

 

[docw009]

Balance wires can be unequal lengths. They're just measuring DC.

 

[lnanek]

I know unequal lengths is more serious for AC because there's synchronization and load balance issues.

But even for DC, wouldn't a longer balance wire have more resistance, so would make the voltage measurement be a few mV off when compared to a shorter balance wire to a p-group at the same voltage? So the BMS may balance the groups with long wires vs. short a little less perfectly.

Probably not an issue if it's a Bluetooth BMS and you can adjust the max V difference allowed between p-group's before it trips, though. E.g. some of them refuse to run if the p-group's aren't all within .1V. Bluetooth BMS often let you force that value to something else with an app.

Tough to imagine the capacity loss from the p-group's not being perfectly balanced would be large. Especially not if the wires are thick anyway, so not much resistance to throw off voltage measurements.

 

[bengxe]

I personally think you're overthinking this issue, but it wouldn't be difficult to match resistance. If you don't have space to just loop extra wire length somewhere, you can use a larger gauge wire to compensate for extra length. I just found that 1 ft of 18ga matches the resistance of 1.5 ft of 16ga and 2.2 ft of 14 ga, for example.
Wire Resistance Calculator

 

[amberwolf]

But even for DC, wouldn't a longer balance wire have more resistance, so would make the voltage measurement be a few mV off when compared to a shorter balance wire to a p-group at the same voltage? So the BMS may balance the groups with long wires vs. short a little less perfectly.

The voltages are only different if current is flowing to create a voltage drop. It would take a fair bit of current to cause a significant drop; ohm's law formula will show you what that drop would be for any specific current and wire resistance.

The only time there should be any current flowing in the balance wires is during balancing, and even then it shouldn't be more than a few dozen mA at most (some BMS do less than that).

As current drops and cells become more balanced, any voltage differences from that should even out.

You can measure the current in a balance wire at any particular condition, from completley empty while charging to totally full and discharging.

 

[marka-ee]

Balance wires can be unequal lengths. They're just measuring DC.

The balance wires also are used for balancing.... sounds a little funny doesn't it? but yes, there's current going through those wires and depending on the BMS up to two amps and I'm worried that if the Chinese software doesn't turn off the current and take a voltage reading now and then in a rest state that it could be a problem But I don't know how their software works.

 

[docw009]

18G stranded wire is rated around 5 amps, and 20G is 3.5 amps per Google, The thinner wire is around .010 ohms per foot, That means 2 amps will drop .02 volts if it's a foot long,

If you have a BMS that pulls 2A thru the wires, that's great as you're still OK with 20G wire, and the voltage drop is trivial for a two foot wire,