How to size battery cable for 48v systems?

[nyyankees588]

Looked around online and on this forum and did not find a straight-forward answer, but I'm sure it exists. Feel free to link to another post/article if you know of one.

Question: how does one size cables between batteries and output loads in a 48v system? I am very familiar with 12v systems, but this is my first time working with a high amperage 48v system, so I want to be certain that I am doing things correctly.

System overview:

• 8x 12v batteries: 4x batteries wired in series to achieve 48v, 2x of these sets will be wired in parallel (similar to this diagram)
• Main load will be a 48v electric motor that is rated for continuous 150A draw

Specific questions:

1. Sizing of wiring connecting batteries in series (6-12" lengths): planning on 4g
2. Sizing of wiring connecting battery banks in parallel (12-24" lengths): planning on 2/0g
3. Sizing of wiring connecting complete battery bank to motor controller, then to motor (TBD, obviously as short as possible): planning on 2/0 OR 4/0, depending on length of run

Many thanks in advance...

 

[Bobert]

I am interested in responses from others on this forum in case I am wrong but I believe you can simply size the wire as you would for 12 volt remembering that the amps will be 4 times less than it’s 12volt counterpart for the same wattage. If you have converted your wattage load to amps the size will be the same for 12 or 48 volts. So 1200 watts is 120 amps at 12 volts but only 30 amps at 48 volts but 30 amps at 48 volts takes the same size wire as 30 amps at 12 volts.

 

[BentleyJ]

Current carrying capacity of wire is based on material (Copper or Aluminum Etc), cross sectional area and stranding vs. solid. Voltage is only a factor for choosing wire with regard to the insulation value being suitable for the application. Many DIY'ers on this forum use Class K or M welding cable for battery connections due to its superior flexibility and higher ampacity ratings due to the very fine stranding.

 

[rmaddy]

Let's assume you chose 4/0AWG and not 2/0AWG from the battery bank for this discussion.

Start with the series connections. Let's say you have 4 12V batteries in series. The amps going from the main wires through the series connections is all the same. Since all of the wires need to handle the full amperage then all of the wires need to be the same size large enough for those full amps. So your plan to use 4/0AWG for the main wires but only 4AWG for the series connections is not going to work. Use 4/0 for all of the series connections.

The parallel connections share the amperage. Let's say you have 2 batteries in parallel. Then each battery will handle 1/2 of the amperage. If you had 3 in parallel then each would handle 1/3 of the amperage. So one would think that for the parallel connections you could use smaller wire appropriate for 1/2 or 1/3 the amps. But here's the issue - what if one battery fails? Now the full amperage is being distributed across the remaining batteries. In the case of 2 batteries originally splitting the amps, suddenly the one remaining battery needs to handle all of the amps. So to handle this possibility it is safer to use the same full sized wire for the parallel connections.

But there is an extra constraint to worry about here. That is the max continuous discharge current rating for each battery. Let's assume your batteries can each handle 100A. Two in parallel means your system can handle 200A from the batteries. If you wire and fuse everything for the 200A then if one battery fails then suddenly the one remaining battery is asked to handle the 200A load. But it can't. If these are LiFePO₄ batteries then the BMS will likely shutdown after a few seconds due to the excessive current draw.

So if each of your series strings can each handle the full amperage then you can wire the parallel connections with the same large wire. A failed battery (and hence a failed string) will leave the remaining strings splitting the full current.

But if each series string can't handle the full amps on its own (which is likely in this case) then you have two options:

1 - Wire the parallel connections with the large wire and let the BMS shutdown a battery if too much current is asked of it. This option only applies to LiFePO₄ batteries.
2 - Wire the parallel connections with wire and fuses (each string gets fused) appropriate for the max continuous discharge current and let each string's fuse blow if the string is asked to handle too much current. This avoids the BMS needing to disconnect the battery in such a case.

 

[rmaddy]

Main load will be a 48v electric motor that is rated for continuous 150A draw

If the max current load on your batteries is 150A then you only need 1AWG or 1/0AWG assuming the length isn't too long. 2/0AWG would be for 200A loads. 4/0AWG is for roughly 300A.

 

[Roswell Bob]

Current carrying capacity of wire is based on material (Copper or Aluminum Etc), cross sectional area and stranding vs. solid. Voltage is only a factor for choosing wire with regard to the insulation value being suitable for the application. Many DIY'ers on this forum use Class K or M welding cable for battery connections due to its superior flexibility and higher ampacity ratings due to the very fine stranding.

How does stranding -vs- solid effect the current carrying capacity? Does insulating material influence capability?

 

[BentleyJ]

How does stranding -vs- solid effect the current carrying capacity? Does insulating material influence capability?

I believe it has something to do with the ratio of surface area to cross sectional area of the conductor. Alternating current tends to concentrate on the surface, its called the "skin" effect. Based on this phenomenon, stranded wire as more surface area at an equivalent AWG size.
Insulating material has what I would call only an indirect effect on ampacity based on its tolerance to heat. Silicone wire can handle more amperage than a PVC coated wire simply because the insulation will not melt until well above 300 degrees. For the purposes of this forum and Solar in general, we always try and oversize the wiring so it does not produce unwanted heat.

 

[nyyankees588]

If the max current load on your batteries is 150A then you only need 1AWG or 1/0AWG assuming the length isn't too long. 2/0AWG would be for 200A loads. 4/0AWG is for roughly 300A.

rmaddy - thank you so much. Your long post explained everything that I was having questions about... very much appreciate it. I typically reference this chart when evaluating wire size for the amperage. Based on your explanation, it sounds like that chart is accurate, regardless of voltage, is that correct? Eg. 150A @ 48 volts requires the same size wire as 150A at 12 volts. I know there's a fundamental electricity rule in there that I'm probably glossing over, but just wanted to confirm that.

 

[rmaddy]

Voltage only affects the amount of voltage drop in a wire given its length. If you look at the chart you referenced, the lengths are based on 12V. At 48V you can use 4 times the wire length shown in the chart for the same voltage drop. Or put another way, take your wire length, divide by 4, and use that length when looking at the chart to choose the proper wire gauge.

 

[BentleyJ]

Voltage only affects the amount of voltage drop in a wire given its length. If you look at the chart you referenced, the lengths are based on 12V. At 48V you can use 4 times the wire length shown in the chart for the same voltage drop. Or put another way, take your wire length, divide by 4, and use that length when looking at the chart to choose the proper wire gauge.

Your statement is correct with regard to voltage drop as a Percentage which is the usual way VD is specified. However, to clarify the formula for voltage drop is V = A x R. Which means the VD value in Volts is dependent on Current and Resistance of the wire ONLY. Its just that loosing 2 volts on a 400V DC solar string is negligible but 2 volts of loss on a 12V system is significant in percentage terms.

 

[Roswell Bob]

I believe it has something to do with the ratio of surface area to cross sectional area of the conductor. Alternating current tends to concentrate on the surface, its called the "skin" effect. Based on this phenomenon, stranded wire as more surface area at an equivalent AWG size.
Insulating material has what I would call only an indirect effect on ampacity based on its tolerance to heat. Silicone wire can handle more amperage than a PVC coated wire simply because the insulation will not melt until well above 300 degrees. For the purposes of this forum and Solar in general, we always try and oversize the wiring so it does not produce unwanted heat.

Thank you.

We hope to be dealing with DC currents so that there is no skin effect. There are some higher frequency components with dc link side of inverter so that wiring will see some AC. I'm not sure if you would be able to measure a temperature difference even if you used Litz wire. It's probably a good idea to put inverter input very close to battery array to minimize copper losses and reduce emi emissions. Some good HF caps at inverter input is a good idea. Some introduced inductance in the way of a few turns of input wiring may be good for emi, but probably not so good for losses. I'm fighting that battle now with new pack. A few microhenry of inductance could also reduce inrush currents when powering up inverter.

Yes on the higher performance jacket. The welding cables seem to have a nice cover and are popular.