Explainer
NEC motor branch circuit #14 AWG with a 20 A breaker
Short answer
#14 AWG is the code-correct conductor for a 3 HP, 230 V three-phase motor with 9.6 A full-load current on a 20 A inverse-time breaker. The 15 A small-conductor cap yields because Article 430 of the NEC specifically permits motor-circuit sizing: 430.22 requires 12.0 A of ampacity, which #14 THHN's 21 A clears, and 430.52 allows the breaker up to 24 A — provided a separate overload relay protects the motor.
On this page
- Which wire is legally correct, #14 or #12?
- What is the 15 A cap actually protecting?
- How does 430.22 size the motor conductor?
- Does the 21 A table value survive corrections and terminations?
- How big can the branch-circuit breaker be under 430.52?
- What does the overload relay have to do with the wire size?
- How do I document this so the inspector signs off?
- When is running #12 the better call anyway?
- Questions asked
#14 AWG is the correct conductor, and the 20 A breaker is legal with it. The small-conductor cap in the ampacity-table footnote — 15 A for #14 copper — applies 'unless otherwise specifically permitted elsewhere in this Code', and Article 430 of NFPA 70, the National Electrical Code, 2020 edition, which governs motors, motor branch circuits and controllers, is that permission. For a 3 HP, 230 V three-phase motor at 9.6 A full-load current, 430.22 requires 1.25 × 9.6 A = 12.0 A of conductor ampacity, which #14 THHN copper clears at its 21 A table entry, and 430.52 permits an inverse-time breaker up to 2.50 × 9.6 A = 24 A, above the installed rating. One condition carries it all: a separate overload relay protects the motor. Confirm the requirement with the local authority having jurisdiction and the equipment documentation, since editions and amendments move these numbers.
Which wire is legally correct, #14 or #12? #
#14 AWG is the code-correct answer for this circuit: the required ampacity against a 21 A table value, and the installed breaker at 208% of full-load current under the 250% motor-circuit ceiling. An inspector's demand for #12 is a ruling to live with, not a code fact.
The cap you are reading — 15 A for #14, 20 A for #12, 30 A for #10 copper — is the general rule of 240.4(D) in recent editions; older printings carried it as the footnote under the ampacity table. Its own wording saves the motor circuit: the limits hold 'unless otherwise permitted', and the Code's overcurrent provisions for specific equipment point motor branch circuits to Article 430 for protective-device sizing. A specific rule governs a general one, so where Article 430 states how the motor branch circuit is built and protected, the general cap steps aside.
What is the 15 A cap actually protecting? #
It protects the conductor against sustained overcurrent when the breaker is the only overload device on the circuit; a motor branch circuit hands that job to the overload relay, so the cap's premise is gone from this circuit.
On a lighting or receptacle circuit, a 20 A breaker on #14 would be a real violation: that breaker is the overload protection, and the wire must never carry more current than it is protected for. The motor circuit differs in both directions. Running, the motor draws its table current — the breaker rating never becomes the conductor's load. Starting, it draws several times full-load current for the seconds of acceleration, and the breaker's job is to stay closed through that while still clearing a bolted fault or a ground fault. Anything sustained lands on the overload relay, sized to the motor, tripping long before conductor insulation is at risk. Assign each device its real job and the mismatch disappears.
How does 430.22 size the motor conductor? #
Branch-circuit conductors supplying a single continuous-duty motor must carry at least 125% of the motor's full-load current: 1.25 × 9.6 A = 12.0 A, and any conductor that still has that ampacity after the permitted column and corrections qualifies.
The current you multiply is itself a rule. Article 430 splits the currents by device: conductors and the branch-circuit breaker are sized from the full-load current tables in Article 430 (430.6(A)(1)), not from the nameplate, while the overload relay is sized from the nameplate. For a 3 HP, 230 V three-phase motor the table lists the same figure as this nameplate, so both routes agree here.
The panel as built, one line of numbers:
| Quantity | Value | Basis |
|---|---|---|
| Motor | 3 HP, 230 V, 3-phase, 60 Hz | Nameplate |
| Sizing current | 9.6 A | Article 430 full-load current table; nameplate agrees |
| Required conductor ampacity | 12.0 A | 430.22: 1.25 × 9.6 A |
| #14 THHN copper ampacity | 21 A | Ampacity table entry, column to be confirmed |
| Margin | 9.0 A | 21 A − 12.0 A |
| Breaker | 20 A, inverse-time | Installed |
| Breaker vs full-load current | 208% | 20 A ÷ 9.6 A × 100 |
| Breaker ceiling | 24 A | 430.52, Table 430.52(C)(1): 2.50 × 9.6 A |
| Overload protection | Separate overload relay | Condition of the Article 430 permission |
Treat the outcome as an estimate until it is checked against the edition and amendments your jurisdiction enforces and against the equipment documentation.
Does the 21 A table value survive corrections and terminations? #
Yes — the load uses 57% of the table entry and leaves 9.0 A of headroom, so ordinary ambient correction and a few current-carrying conductors in the raceway keep #14 qualified; only a derate that removes more than that headroom disqualifies it.
Two checks belong in the file anyway. First, the column: the terminations' temperature rating decides which column of the ampacity table sets the final ampacity, and the higher insulation column is used only to apply corrections, never as a free-standing number. Second, the adjustments: high ambient or a raceway packed with current-carrying conductors cuts ampacity, so recompute for that installation, and if the adjusted value falls under the requirement, move up a size. That is a computed decision, not a preference.
How big can the branch-circuit breaker be under 430.52? #
An inverse-time (thermal-magnetic) circuit breaker on this branch circuit may reach 250% of full-load current, so the installed rating sits comfortably inside the limit at roughly 208%.
430.52, with the percentages in Table 430.52(C)(1), sets the maximum size or setting of the branch-circuit short-circuit and ground-fault device by device type; 250% is the inverse-time breaker row, generous on purpose so the device rides through locked-rotor current without tripping on every start. Nothing about this breaker protects the conductor from overload, which is why a breaker larger than the wire's general cap is tolerated here. Two cautions: an instantaneous-only (magnetic-only) breaker is a different table row with far higher percentages — redo the check if that is what is installed — and where a calculated maximum falls between standard ratings, the next larger standard size is permitted within the article's limits.
What does the overload relay have to do with the wire size? #
Everything: the 125% and 250% arithmetic is conditioned on a separate overload device limiting sustained current, so the relay's size, range and trip class are part of the wire-size justification, not a separate topic.
Size that relay from the motor's nameplate current, per the same split of duties, and read its range and trip class from the device and its data sheet. The relay is also what turns a contactor into a starter — the motor starter vs contactor distinction the overload requirement rests on. If you are choosing the device, selecting overload and jam protection for a reversible motor walks through the range and trip-class decision.
How do I document this so the inspector signs off? #
Show the whole chain on one page beside the schematic: the ampacity required by 430.22, the table value available, the installed inverse-time breaker as a percentage of the ceiling, and the overload relay's identity and setting.
The recurring failure is citing the table entry alone, which invites exactly the 15 A cap argument. Write out the 'unless otherwise permitted' basis, the 430.6(A)(1) current source, the 430.22 multiplication, the 430.52 percentage with the device type named, and the column and corrections you applied, and submit it with the drawing package at plan review rather than debating it at final inspection. The authority having jurisdiction can still rule for #12, and that ruling governs the job even though the code basis stands.
Safety and field checks: lock the branch circuit out and test for absence of voltage before opening the controller, the wireway or the motor terminal box. After a controlled re-energization, confirm the breaker holds through acceleration, the three line currents are balanced and near nameplate on a clamp-on ammeter, and the terminations show no abnormal heating at steady load.
When is running #12 the better call anyway? #
Choose #12 when a written specification demands it, when the run is long enough for voltage drop to matter, or when the terminal's listed wire range excludes #14 — the code minimum remains #14, and the project decision may exceed it.
Long runs deserve a voltage-drop check at the running current before you commit, and many plants standardize #12 for power wiring with #14 reserved for control — a spec to honor, not a code rule. Where the engineering time to defend #14 costs more than the larger wire, pull #12 and keep the documented Article 430 basis in the file for the next job. The decision in one line: #14 THHN with an inverse-time breaker and a correctly set overload relay is code-compliant; change size for a computed or written reason, not because of the footnote.
Questions asked
Does the motor nameplate or the NEC table set the current for sizing?
Both, but for different devices. The full-load current tables in Article 430 set the current for the branch-circuit conductor and the breaker, while the overload relay is sized from the motor nameplate. For a 3 HP, 230 V three-phase motor the table lists 9.6 A, matching this nameplate, so the conductor calculation gives the same result either way. When they disagree, run each device from its own current; never swap them.
Can this breaker go above its rating if it trips during acceleration?
Yes, if it is an inverse-time breaker and the trip is truly acceleration. The ceiling is 2.50 times the table full-load current, or 24 A; because that is not a standard rating, the next larger standard size, 25 A, is permitted. Before upsizing, rule out a jammed load, a single-phasing condition or an undersized motor — a larger breaker fixes nuisance tripping, not a mechanical problem, and it never changes what the conductor must carry.
The inspector insists on #12 anyway — who is right?
On the code arithmetic, #14 with the installed breaker complies, and that is worth stating in writing with the 430.22 and 430.52 calculations attached. On the job, the authority having jurisdiction has the final word, and once the ruling is made, #12 becomes the requirement for that installation. Keep the documented basis in your file so the same panel design is argued once at plan review instead of panel by panel.
Does the same exception let #12 or #10 motor wiring exceed their 20 A and 30 A caps?
Yes, by the same mechanism. The 20 A cap for #12 copper and the 30 A cap for #10 are the same general defaults, and a motor circuit with its own overload protection sizes its breaker from the 430.52 percentages instead. Run the identical arithmetic on the table full-load current for the larger motor: 125% of that current for the conductor, 250% for an inverse-time breaker.
What changes when the motor runs on a variable frequency drive?
The sizing basis moves from the motor to the drive. With an adjustable-speed drive, the supply conductors are sized from the drive's rated input current under the adjustable-speed-drive provisions of Article 430, not from the motor nameplate or the full-load current tables, because the drive draws its own input current and manages the motor. Follow the drive manual's input rating for the branch circuit feeding it.