Why bus bars?

[Roswell Bob]

I am thinking to use wire with ring terminals rather than bus bars. Tin plated terminals I think will work with cell studs??? I'm concerned with bloating of cell packs with rigid bus bars. I will most likely compress cells with 1/4" thread rod and springs.

Comments please.

thanks
Roswell Bob

 

[A.Justice]

I think that the biggest issue with those is the resistance from crimps and miniscule differences in wire length. Even small differences in resistance can screw with single cell voltage over time. That being said, if you do a great job crimping, and test resistance after, it should be fine.

Flexible bus bars are definitely the way to go, but there are limited options available, and most of them are DIY.

 

[Roswell Bob]

Yes, thank you. I agree some attention should be paid to the resistance.

I generally crimp and solder, but I know there are those that say not to solder too.

My household loads are quite small at about 6kWh/day. I am usually at about 5A, and occasionally may see 50A if microwave and toaster on at the same time. I intend to charge the 272/280Ah pack with about 75amps. The losses could add up quickly at about 6W/milliohm.

 

[camelCase]

I assume you have a 48 volt battery if a (small) microwave and toaster on at the same time only pulls 50 amps?

 

[jwelter99]

You really want the resistance of the interconnects between cells to be identical as any differences will end up being perceived by the BMS to be cell voltage differences, particularly when under higher loads so resistance voltage drops become larger. This is because the BMS systems used by most on here are not absolute measurements per cell but relative to the prior cell, so it includes the interconnect losses between cells.

As to it being a real issue or theoretical comes down to how well the interconnects are identical, corrosion free, and properly tightened to remain that way.

 

[Roswell Bob]

You really want the resistance of the interconnects between cells to be identical as any differences will end up being perceived by the BMS to be cell voltage differences, particularly when under higher loads so resistance voltage drops become larger. This is because the BMS systems used by most on here are not absolute measurements per cell but relative to the prior cell, so it includes the interconnect losses between cells.

As to it being a real issue or theoretical comes down to how well the interconnects are identical, corrosion free, and properly tightened to remain that way.

Yes, I am aware the wiring impedance gets lumped in with the cell voltage measurement. Just a couple of milliohms could cause problems. Maybe differential amplifiers with wire pairs would be a better choice.

I'm probably overthinking it.

 

[kenryan]

As to it being a real issue or theoretical comes down to how well the interconnects are identical, corrosion free, and properly tightened to remain that way.

This is the important question that has not been answered. If care is taken to make the cables as similar as possible, is there any practical reason not to use them? Shunts are carefully calibrated by grinding off small amounts of material. Is this also done with bus bars? Are bus bars typically calibrated by grinding? If not, I would suspect that carefully crafted cables should work as well as bus bars. But I've never done it, so maybe I'm wrong.

 

[jimbob420]

I connected my cells together using 50mm csa wire and appropriate ring crimps.. seems just fine to me, tho I'm not drawing crazy amperage in all fairness. Used a hydraulic crimper rather than a hammer, manual, or the time honored vice method.

 

[jwelter99]

This is the important question that has not been answered. If care is taken to make the cables as similar as possible, is there any practical reason not to use them? Shunts are carefully calibrated by grinding off small amounts of material. Is this also done with bus bars? Are bus bars typically calibrated by grinding? If not, I would suspect that carefully crafted cables should work as well as bus bars. But I've never done it, so maybe I'm wrong.

No, but it's usually less error prone to cut a bunch of bar stock to the same length and then drill the terminal holes. But yes, that could also be screwed up if someone tried hard enough.

The only worry I would have using cable instead of bus bars is getting really good crimps on the lugs. With the right tools/practice/attention to detail this is not be a problem. Or order cables from someone who knows how to properly assemble.

 

[shadowsteve]

The bus bars that came with my 280's measure out to 42mm cross sect area which is the same as #1 wire and are 75mm between terminals. At 150A flow that's 4.6mV drop between cells.

If you used #2 wire for interconnects and went to 100mm spacing for padding then you'd have 7.8mV drop on the connection. I don't know how large the drop can be before it matters.

As mentioned making good bars is easier than making good cables.

 

[pollenface]

One of my lifepos just used copper cable with eyelets crimped and soldered on the them. It worked.

I don't know what the awg specs were (Australian) but it was 14mm2

 

[kenryan]

I think I am going to try using cables. Should I size them the same as the cables that go to the inverter (4/0) ? Or should I make them equivalent to what is on the BMS (2 x 2 AWG) ?