Wire size for 100 A over 15 ft on a 24 V system

Smallest wire that works

2 AWGat 2.42% drop

100 A over 15 ft one-way on a 24 V system

Circuit power100 A × 24 V	2,400W
Conductor length15 ft out and 15 ft back	30ft
Ampacity alone would allow2.42% drop — legal on heat, and fine on drop	2 AWG
Power lost in the cable2.42% of what you are sending	58.2W
— Voltage drop by size —
14 AWGnot rated for 100 A — 15 A max	—
12 AWGnot rated for 100 A — 20 A max	—
10 AWGnot rated for 100 A — 30 A max	—
8 AWGnot rated for 100 A — 50 A max	—
6 AWGnot rated for 100 A — 65 A max	—
4 AWGnot rated for 100 A — 85 A max	—
2 AWG115 A rated	2.42%0.58 V
1/0 AWG150 A rated	1.52%0.37 V
2/0 AWG175 A rated	1.21%0.29 V
4/0 AWG230 A rated	0.76%0.18 V

Current: 100 A
One-way distance: 15 ft
System voltage: 24 V

Ampacity and voltage drop are two different questions

Ampacity asks whether the wire gets hot enough to be dangerous. Voltage drop asks whether enough of the power arrives at the other end. On a low-voltage DC run they almost never give the same answer, and the bigger one wins.

For 100 A over 15 ft at 24 V, ampacity alone would allow 2 AWG. Voltage drop says 2 AWG — 2.42%, or 0.58 V of the 24 V you started with.

Why the distance doubles

Current goes out on one conductor and comes back on the other. The copper in the circuit is 30 ft, not 15 ft. Forgetting the return leg is how people arrive at exactly half the real drop.

Why it matters more than it sounds

2,400 W leaves the source. Undersized cable does not just waste a few percent — on a solar charge controller it means the battery never sees absorption voltage, never finishes a charge cycle, and sulphates its way to an early death while every meter in the system reads normal.

What is not in the table

Copper at 75 °C. Aluminium conducts about 60% as well — go up two sizes. Conduit fill, ambient temperature above 30 °C and bundled conductors all derate ampacity further.

Notes

At this length ampacity and voltage drop point at the same wire, which is unusual. Runs over about 20 ft at low DC voltage are normally decided by drop, not by heat.
3% is the NEC informational recommendation for a branch circuit, not a hard rule. On a 24 V DC system it is already generous: 0.72 V of the 24 V you started with.
Running at 24 V is doing a lot of work here. The same 2,400 W at 12 V would draw 200 A and need a far heavier cable.
Figures are copper at 75 °C. Aluminium runs about 60% of copper conductivity — go up two sizes. Conduit fill, ambient temperature above 30 °C and bundling all derate ampacity further, and none of that is in this table.

Nearby sizes

Roof Space for a 10 kW Array: Panel Area Is Not Roof AreaTwenty-five panels cover about 525 square feet. The roof they need is closer to 680, and the gap is mostly required by code rather than by geometry.
Daily Solar Output by Region, and the Seasonal Swing Behind the AverageAn annual average tells you what the system produces over a year. It tells you almost nothing about December, and December is where the sizing arguments happen.
How Many Panels for 1,000 kWh a MonthThe answer ranges from fifteen panels to thirty-four, for the same house using the same electricity. Location does most of that, panel wattage does the rest.
Converting kW to Panels, and the Three Ratings That Get ConfusedThree different numbers on a solar quote are all measured in kilowatts and none of them mean the same thing. Knowing which is which makes quotes comparable.