DC Wire Gauge Calculator

Anyone coming to DC wiring from household electrics finds the wire sizes absurd. A 30 amp circuit in a house runs on 10 AWG; the same 30 amps over a modest run in a 12 volt system wants 4 AWG, a conductor four times the cross-section. The reason is not that DC is different in kind — it is that the tolerance is different by a factor of ten.

Voltage drop is judged as a percentage of the supply. Half a volt lost in a 120 volt circuit is 0.4%, invisible. The same half volt in a 12 volt circuit is 4.2%, which is at the edge of acceptable and, on a charging circuit, is money.

The current is higher too, for the same reason. Delivering 400 watts at 120 volts takes 3.3 amps; at 12 volts it takes 33. Drop is proportional to current, so the twelve-volt circuit suffers ten times the drop before the percentage tolerance is even considered. Combined, the two effects mean a low-voltage circuit is roughly a hundred times less tolerant of a given length of a given wire.

This is the single strongest argument for 24 or 48 volt systems in anything above modest power. Doubling the system voltage halves the current and doubles the volts available to lose, so the wire needed falls by roughly a factor of four.

Every conductor has to pass two independent tests, and they are about different failure modes.

Ampacity is a safety limit. Current heats a conductor, and enough heat degrades the insulation, softens terminations and eventually starts a fire. The ampacity of a wire depends on its cross-section, its insulation temperature rating, and how easily it can shed heat — which is why the same wire is rated lower in conduit, in a bundle, or in a hot engine bay than it is hanging in free air. Ampacity does not depend on length at all.

Voltage drop is a performance limit. It does not endanger anything; it simply means the load does not get the voltage it expected. Drop is proportional to length, so it is the constraint that grows with distance.

Which one binds tells you what to do about it. A battery-to-inverter cable carrying 250 amps over four feet is ampacity-limited: it needs 4/0 regardless of how short you make it, and relocating the inverter saves nothing. A 40-foot run to a rear-mounted fridge drawing 6 amps is drop-limited: 16 AWG would carry it safely all day, but the fridge would see 10 volts, so it needs 10 AWG — and moving the source closer would let you go back down.

The consequences differ by circuit type, which is why the target percentage differs too.

Wire sizing and overcurrent protection are two halves of the same job, and the relationship between them is frequently misunderstood: the fuse protects the wire, not the appliance.

The appliance has its own protection, or it does not and that is its manufacturer’s problem. What the fuse in your system exists to prevent is a conductor carrying more current than it can shed as heat — a short circuit, a chafed cable against a chassis, a terminal that has worked loose and is arcing. A battery bank can deliver thousands of amps into a dead short, and a cable in that condition becomes a heating element in seconds.

Two rules follow. The fuse is sized to the wire’s ampacity, not the load’s rating — a wire rated 95 amps feeding a 30 amp load should be fused somewhere between the two, closer to the load, but never above what the wire can carry. And the fuse goes at the source end, within inches of the battery positive terminal, because a short anywhere along the cable bypasses a fuse fitted at the far end entirely.

For the main battery cable specifically, an ordinary blade or glass fuse is not adequate. Their interrupt rating — the current they can break without arcing across the gap — is far below what a large battery can deliver. Class T fuses, or an appropriately rated ANL, exist for this and are the correct choice on any bank capable of high short-circuit current, which includes essentially every lithium installation.

Beyond the arithmetic, several practical points regularly cause problems in DC installations.