A branch circuit is a constraint system that happens to have a table in the middle of it. You have a load that varies with time, a conductor with a thermal limit, terminations with their own lower limits, an overcurrent device that has to protect all of it, and an environment that quietly changes the numbers you looked up. Most DIY wiring mistakes are not arithmetic mistakes. They are modeling mistakes: someone reads a single cell out of NEC Table 310.16, treats it as the answer, and never asks which of the other four constraints was supposed to override it.
This note is about building the right mental model. It is educational material, not a design approval, and it is not a substitute for your local electrical code or a licensed electrician. The authority having jurisdiction always has the final word, and local amendments are frequently stricter than the model code. But if you understand the structure, you will at least know which questions to ask.
People expect ampacity to behave like resistance: plug in the numbers, get a value, done. It does not, because ampacity is not really an electrical property. It is the current at which a conductor reaches its maximum permitted temperature and stops. That maximum is set by the insulation chemistry. The current that gets you there depends on how fast heat can leave the conductor, which depends on the surrounding air, the conduit, the number of neighboring conductors, and how tightly everything is packed.
So the table is not derived from first principles in any way you could reproduce with Ohm's law. It is an empirical consensus, tabulated for a specific set of baseline conditions. NEC Table 310.16 states those conditions in the title block: an ambient temperature of 30 °C (86 °F), and not more than three current-carrying conductors in a raceway, cable, or direct burial. Every other situation is handled by taking the baseline value and multiplying it by correction or adjustment factors.
That structure matters enormously, because it means the number in the table is a starting point, not an endpoint. If you remember only one thing from this article, remember that the table is a base case plus multipliers, and that the multipliers can be severe.
Table 310.16 gives three ampacity columns per conductor size for copper, and three for aluminum: 60 °C, 75 °C, and 90 °C. A very common misreading is that these columns describe how hot the wire will actually get in service. They do not. They are indexed by the temperature rating of the weakest component in the current path — the termination and the insulation system together.
The governing idea is in NEC 110.14(C): the ampacity of a conductor must not exceed the lowest temperature rating of any connected termination, conductor, or device. A 90 °C conductor landed on a 60 °C breaker lug is a 60 °C conductor for ampacity purposes, no matter what the insulation is rated for. The reverse is also true and is the reason the 90 °C column exists at all: 110.14(C) permits conductors with higher temperature ratings to be used for ampacity adjustment and correction calculations, provided the final result is still checked against the termination limit.
In practice that makes the 90 °C column an intermediate variable rather than a usable answer. You may start your derating arithmetic at 90 °C when the insulation is rated for it, but the number you actually ship has to survive the 60 °C or 75 °C ceiling. This is why a full NEC wire ampacity table is most useful when you read it across a row rather than down a column: the row for 10 AWG copper reads 30, 35, 40, and those three numbers answer three different questions.
For residential work there is an extra wrinkle that surprises a lot of people. Most modern residential breakers and receptacles are in fact rated for 75 °C terminations. That would suggest you could use the 75 °C column for ordinary house wiring. You usually cannot, because of the cable rather than the terminals: NEC 334.80 limits the ampacity of Type NM, NMC, and NMS cable to that of a 60 °C conductor, regardless of the 90 °C rating printed on the conductors inside the jacket. Romex is a 60 °C product for ampacity purposes. That single sentence resolves more confusion about residential ampacity than any other fact in the code.
The practical way to model a branch circuit is as a sequence of gates. Each gate can only lower the answer. Run them in order and you will not miss one.
Notice that gates 1 and 5 are not about the wire at all. Gate 1 is about the load's duty cycle, and gate 5 is about the breaker. A circuit can fail either one while passing the ampacity lookup comfortably, which is exactly why reading a single table cell feels sufficient and is not.
There is a related trap in NEC 240.4(B), the "next size up" rule. It lets you round up to the next standard overcurrent device when a conductor's ampacity falls between two standard ratings, subject to conditions. It is genuinely useful for feeder and motor work. It does not rescue you from 240.4(D), because 240.4(D) is a specific cap on small conductors, not a general ampacity statement. If you are trying to put a 25 A breaker on 12 AWG copper because the 75 °C column says 25 A, the next-size-up rule is not the tool you want and the inspector will say so.
Here is a case that looks trivial and is not. A 240 V appliance draws 18 A and runs for hours at a time. The instinct is 12 AWG on a 20 A breaker, because 12 AWG is "the 20 amp wire." Watch what happens.
The correct answer is 10 AWG copper. At 60 °C that is 30 A, which covers the 22.5 A requirement and supports a 25 A or 30 A device within the 240.4(D) cap of 30 A. The lesson is that the binding constraint was 240.4(D) and the 125 percent multiplier together, and neither of them appears anywhere in the ampacity column.
This is the sort of arithmetic that a copper wire ampacity by AWG lookup makes fast, because you need the 60, 75, and 90 °C values side by side in one row while you walk the gates. Doing it from three separate lookups is how people skip gate 4.
The first is ambient. The table assumes 30 °C. An unconditioned garage in August is not 30 °C, and an attic is wildly hotter. At 36 to 40 °C the 60 °C correction factor is 0.82; at 41 to 45 °C it is 0.71. Apply 0.82 to a 12 AWG copper conductor at 60 °C and 20 A becomes 16.4 A, which fails an ordinary 20 A continuous load before you have even counted conductors. Attics and rooftop raceways are their own problem, and the code has a separate provision that adds 60 °F to the design ambient for raceways exposed to sunlight on or above a roof. If your cable run spends thirty feet in a 140 °F attic, the ampacity question is no longer the one the table answers.
The second is bundling. NEC 310.15(C)(1) requires adjustment when more than three current-carrying conductors share a raceway or cable. Four to six conductors get 80 percent, seven to nine get 70 percent, ten to twenty get 50 percent. Equipment grounding conductors do not count, and neither does the neutral of a multiwire branch circuit that only carries the unbalanced current. But every hot conductor counts, and in a conduit carrying three 120 V circuits, each with its own neutral, that is six current-carrying conductors and an 80 percent factor. Fill up a half-inch conduit because the pull is easy and you have silently derated everything in it.
The third is metal substitution. Aluminum has roughly 61 percent of copper's conductivity by volume, so the same gauge carries meaningfully less current: 6 AWG is 65 A in copper at 75 °C and 50 A in aluminum. Aluminum also needs terminals listed for it, and small aluminum branch-circuit sizes are a historical artifact rather than a sensible modern choice. If you are replacing a run and the existing conductor is aluminum, the gauge that "looked right" in the panel may not be the gauge you need.
The reason this is worth modeling carefully is that the failure mode is invisible. An undersized conductor does not announce itself; it warms up, ages its insulation, and waits. The table in the code book is a compact encoding of a century of testing, and it rewards being read as a system rather than as a lookup. If you want a fast way to walk the gates above, a wire ampacity chart that shows copper and aluminum across all three temperature columns, with the ambient and conductor-count factors in the same place, will save you from the most common error of all: confidently using the 90 °C number because it was the biggest one in the row.