AWG wire gauge

26 wire gauges from 4/0 to 40, in millimetres and inches, with area, resistance and — where there is a single well-known answer — ampacity.

Choosing between two gauges? The voltage each loses over a run, and how far it reaches before the drop passes 3%.

AWG is a formula, not a table

Every number on this page comes from one line:

diameter = 0.005 × 92^((36 − n) / 39)  inches

That is why the gauge numbers run backwards. The scale counts drawing operations — wire is pulled through progressively smaller dies, and the gauge records how many passes it took, so a higher number means more passes and a thinner wire. Gauge 36 is anchored at 0.005 inches and gauge 4/0 at 0.46, with 39 steps between them.

Two rules fall straight out of the exponent, and they are the reason electricians can size wire in their heads:

Six gauges lower doubles the diameter. 92^(6/39) = 2.005.
Three gauges lower doubles the cross-sectional area, and so halves the resistance. A 6 AWG conductor carries 4.02 times the copper of a 12.
Ten gauges lower is about ten times the area. The exact factor is 10.16, which is close enough to ten to be worth remembering.

The full chart

AWG	mm	inches	mm²	Ω/km Cu	Ω/1000 ft Cu	NEC 60 °C
4/0	11.684	0.46	107.219	0.16	0.05	195 A
3/0	10.405	0.4096	85.029	0.2	0.06	165 A
2/0	9.266	0.3648	67.431	0.26	0.08	145 A
1/0	8.251	0.3249	53.475	0.32	0.1	125 A
1	7.348	0.2893	42.408	0.41	0.12	110 A
2	6.544	0.2576	33.631	0.51	0.16	95 A
3	5.827	0.2294	26.67	0.65	0.2	85 A
4	5.189	0.2043	21.151	0.82	0.25	70 A
6	4.115	0.162	13.302	1.3	0.4	55 A
8	3.264	0.1285	8.366	2.06	0.63	40 A
10	2.588	0.1019	5.261	3.28	1	30 A
12	2.053	0.0808	3.309	5.21	1.59	20 A
14	1.628	0.0641	2.081	8.28	2.53	15 A
16	1.291	0.0508	1.309	13.17	4.02	—
18	1.024	0.0403	0.823	20.95	6.38	—
20	0.812	0.032	0.518	33.31	10.15	—
22	0.644	0.0253	0.326	52.96	16.14	—
24	0.511	0.0201	0.205	84.21	25.67	—
26	0.405	0.0159	0.129	133.9	40.81	—
28	0.321	0.0126	0.081	212.9	64.89	—
30	0.255	0.01	0.051	338.53	103.18	—
32	0.202	0.008	0.032	538.28	164.07	—
34	0.16	0.0063	0.02	855.91	260.88	—
36	0.127	0.005	0.013	1360.94	414.82	—
38	0.101	0.004	0.008	2163.98	659.58	—
40	0.08	0.0031	0.005	3440.87	1048.78	—

About the ampacity column

Those figures are the NEC 310.16 copper values in the 60 °C column, which is what sets the breaker on an ordinary branch circuit — 14 AWG on a 15-amp breaker, 12 on a 20, 10 on a 30. Treat them as orientation, not as permission. The real limit depends on the insulation temperature rating, the ambient temperature, how many current-carrying conductors share a raceway, and the code in force where you are. Aluminium is derated further: it has 1.64 times the resistivity of copper, so the same load needs a larger conductor.

Copper against aluminium

Aluminium conductors appear in service entrances and feeders because they are cheaper and lighter, and the trade is resistance: 1.64× that of copper for the same cross-section. In practice that means going roughly two gauges larger in aluminium to match a copper conductor — which is exactly the "three gauges doubles the area" rule applied to a factor of 1.64.

Stranded wire is not the gauge it says

A gauge names a cross-sectional area of conductor, and stranded wire reaches that area as many small circles rather than one large one. Circles do not tile, so the bundle is wider than a solid conductor of the same gauge — typically 1.13 to 1.25 times the diameter depending on how many strands and how tightly they are laid.

That matters in three places and nowhere else: it fills a conduit faster than the solid-wire tables suggest, it needs a larger crimp or terminal than the number implies, and it will not seat in a screw terminal sized for solid wire without a ferrule. The electrical properties are the ones in the chart above — same area, same resistance, same ampacity.

What stranded buys is flexibility and fatigue life. Solid wire work-hardens and eventually snaps where it is repeatedly bent, which is why fixed building wiring is solid and anything that moves — appliance cords, automotive, robotics — is stranded.

Why the gauge numbers run backwards

The scale counts manufacturing steps rather than measuring the wire. Wire is drawn by pulling it through a die, then a smaller die, and so on; the gauge number is how many dies it passed through. More passes means thinner wire and a bigger number, which is the whole explanation for a system that otherwise looks perverse.

It also explains the sizes below 1. A wire that needed no reduction is 0, one thicker still is 00, then 000 and 0000 — written 1/0 through 4/0 and pronounced "one aught" to "four aught". There is no negative gauge because there was no way to count backwards through dies that were never used.

The exponent falls out of the same process: 2.005 is 92^6/39, which is why six gauges doubles the diameter, and squaring it is why three gauges doubles the area. Those two facts are the only ones worth memorising, because everything else in the chart can be reconstructed from them.

Nominal sizes that measure nothing — gauge is the odd one, it counts rather than measures. Unit converters · Battery sizes · Lumber sizes