Current IntelPower LibraryShop main store

Explainer

Why any two phase swaps reverse a three-phase motor

Motors and motor starters Published 6 min read

Short answer

Swap any two of the three supply conductors — A and B, B and C, or A and C — and a three-phase induction motor reverses identically, because every two-conductor exchange flips the phase sequence to its mirror; no pair changes timing. Heat after a reversal traces to voltage unbalance above NEMA MG-1's 1.0% limit, load, or a one-directional cooling fan.

On this page
  1. Does it matter which two wires you swap to reverse a three-phase motor?
  2. Why does swapping any two conductors reverse the rotation?
  3. Can swapping the wrong pair throw off the motor's timing?
  4. How do you reverse a three-phase motor safely?
  5. Can every motor run in both directions?
  6. Why would a motor run hotter after you reverse it?
  7. How much voltage unbalance is too much for a motor?
  8. When does it matter which phase sits on which terminal?
  9. Questions asked

No, it does not matter which two you swap. Exchanging A and B, B and C, or A and C each flips the phase sequence of a three-phase induction motor from A-B-C to its mirror, so the rotating field and the rotor turn the opposite way — identically for every pair, with no change to torque or speed. What can change how the motor behaves after a reversal is supply unbalance above the 1.0% NEMA MG-1 limit, mechanical load, or a cooling fan that only works in one direction. Those you measure; the pair you pick you can forget.

Does it matter which two wires you swap to reverse a three-phase motor? #

No — all three possible pairs, A-B, B-C and A-C, reverse a three-phase induction motor identically; each exchange turns the A-B-C phase sequence into its mirror, and no pair is gentler, slower, or electrically different from another.

Three conductors give six arrangements at the terminals. Three of them are cyclic rotations of A-B-C — A-B-C, B-C-A, C-A-B — and keep the sequence. The other three each result from exchanging exactly two conductors, and each reverses it. That is the entire arithmetic behind the shop rule "swap any two": every reversing move is a two-conductor exchange, and every two-conductor exchange reverses.

Change at the terminals New order Rotation
Swap A and B B-A-C Reverses
Swap B and C A-C-B Reverses
Swap A and C C-B-A Reverses
Roll all three cyclically B-C-A Unchanged

Why does swapping any two conductors reverse the rotation? #

The three phase windings produce a magnetic field that rotates in the order the phases reach their peaks, and swapping two conductors exchanges that order, so the field sweeps the air gap the opposite way at the same speed.

A three-phase supply has exactly two sequences, A-B-C and its mirror C-B-A, and the rotor follows whichever one it sees. Nothing about the swap changes frequency or pole count, so synchronous speed is untouched: a 4-pole motor on 60 Hz runs at 120 × 60 / 4 = 1800 rpm synchronous in either direction, and slip simply re-establishes itself on the other side of that speed.

Can swapping the wrong pair throw off the motor's timing? #

No — there is no motor timing to throw off: torque, slip and speed relationships mirror exactly, so locked-rotor torque, breakdown torque and full-load speed come out the same in either direction of rotation.

The devices around the motor are another matter. A phase monitor or reversal relay asserts a sequence, so after a swap it reports the new order as a fault until its setting or wiring is updated. Direction-sensitive interlocks and mechanical backstops on the load react the same way. Nothing in the motor itself cares.

How do you reverse a three-phase motor safely? #

Lock out the disconnect, prove the circuit dead at the terminals you will open, then interchange any two conductors on the load side and jog the motor to confirm rotation before you couple or load it.

  1. Confirm the driven equipment tolerates reverse rotation. Screw and centrifugal compressors, some pumps and fans, and geared loads with backstops can be damaged by even a short backward spin.
  2. Lock out and tag the disconnect, open it, and test for absence of voltage at the terminals you are about to open. Swapping leads on an energized circuit is the one move in this job that can hurt you; do not make it.
  3. Mark the three conductors as found, then interchange any two — at the motor terminal box or on the starter's load terminals, one end only, so you do not build a hidden second swap back into the feeder.
  4. Close up, jog the motor, and confirm the shaft turns the way the load needs, against the motor's direction arrow if it has one.
  5. Confirm any phase monitor, reversal relay or direction interlock in the circuit accepts the new sequence before the machine goes back to production.

Can every motor run in both directions? #

Most can, but not all: some cooling-fan and drainage arrangements are one-directional, and a rotation arrow stamped or stickered on the frame is the manufacturer telling you the machine must turn only that way.

Where an arrow exists, obey it and get the reverse motion mechanically instead — a re-oriented load or a different motor. Where there is no arrow, verify against the motor's documentation that the cooling design is bidirectional before committing to continuous reverse running, and watch current and temperature through the first loaded hour.

Why would a motor run hotter after you reverse it? #

Measure, then judge: take all three line-to-line voltages and all three line currents, compare the currents with nameplate full-load amps, and find out whether the machine is running at, or into, its service factor.

A hand on the frame is not a temperature measurement — a Class H machine can be uncomfortably hot to the touch while still inside its insulation class. A motor loaded right up to nameplate, or dipping into its service factor, runs hot in either direction, and voltage above nameplate raises magnetizing current on top of that; see why motor current rises when voltage exceeds nameplate. If the imbalance pattern repeats on every motor fed from the service, the source is upstream: diagnosing current imbalance across every motor on a service walks that trace.

How much voltage unbalance is too much for a motor? #

NEMA MG-1, the industry's Motors and Generators standard, draws the line at 1.0% voltage unbalance measured at the motor terminals; above that, its Figure 20-2 curve (2011 edition) requires derating the motor, and most manufacturers void the warranty there.

The standard's method, on the three line-to-line voltages:

  • Average = (Vab + Vbc + Vca) / 3
  • Unbalance % = 100 × (largest deviation from average) / average

Worked example — a 480 V service measures 472 V, 479 V and 481 V phase-to-phase:

  • Average = (472 + 479 + 481) / 3 = 1432 / 3 = 477.3 V
  • Largest deviation = 477.3 − 472 = 5.3 V (the 481 V leg is only 3.7 V away)
  • Unbalance = 100 × 5.3 / 477.3 = 1.12%

Decision: 1.12% is above the 1.0% limit, so this motor must be derated per Figure 20-2 of NEMA MG-1-2011 or the supply balanced — and the derating grows steeply as unbalance rises, so fixing the service is the better spend. Note the total spread, 9 V, is 1.89% of the average; that is not the MG-1 number, because the standard uses the single largest deviation, not the spread. Your motor's nameplate and manufacturer documentation set the enforceable terms; MG-1 is the baseline. To stop sustained unbalance before it cooks the windings, setting voltage imbalance limits on a 480 V phase monitor gives the trip points.

When does it matter which phase sits on which terminal? #

On an open-delta utility service it can: there the supply is inherently unbalanced, so rolling the three conductors — a cyclic move that keeps rotation — can pair the line's unbalance against the motor's own winding unbalance and change current balance and heating.

An open-delta bank derives one leg from the other two, so its voltages arrive slightly unequal, and the motor's windings add their own small imbalance. Rolling the leads changes which supply leg lands on which winding, so the two imbalances can offset instead of add. The procedure is measurement, not guesswork: record line-to-line voltages and per-phase line currents at steady load for each of the three cyclic placements, and keep the placement with the smallest current spread. Then get your direction with a plain two-conductor swap. For the transformer side of the problem, diagnosing open-delta voltage unbalance in machine shops goes deeper. On a balanced service, terminal placement is electrically interchangeable and none of this applies.

Questions asked

Does swapping two phases at the starter work the same as swapping them at the motor?

Yes. Any two of the three conductors feeding the motor reverse it, whether you exchange them at the motor's terminal box or on the starter's load terminals. Do the swap at one end only, and confirm any phase monitor or metering in between still sees the sequence it expects.

How do you confirm rotation before coupling the load?

Uncouple or unload the driven machine, jog the starter, and watch the shaft against the motor's direction arrow. On a first-time installation, a phase-sequence meter on the motor leads, compared with the sequence the load requires, lets you land the correct pair before the motor ever spins.

Can you reverse a motor through a drive instead of swapping wires?

Yes. A variable frequency drive reverses its output electronically through its direction setting or a forward/reverse command, and exchanging two drive output conductors does the same thing electrically. Use the setting rather than the wire swap when the conduit is sealed or when the drive's phase and rotation functions reference specific terminals.

Why are the line currents more unbalanced than the voltages?

A small voltage unbalance produces a much larger current unbalance, because the impedance a motor offers to reversed-sequence voltage is far lower than its normal running impedance. The amp clamp therefore finds a marginal supply long before the voltmeter does, which is why the current check comes first when a reversed motor runs hot.

Keep reading

  • How-to

    Wiring a three-phase coolant pump starter from a 120 V signal

    Run the machine's 120 V coolant output through the overload relay's normally closed contact to a contactor with a 120 VAC coil, and set the overload dial to the pump's nameplate full-load current — 1.8 A on a typical 1/2 HP three-phase pump, inside a 1.3–2.1 A adjustable range. Then bump-test rotation and swap any two motor leads if the pump runs backward.

    Motors and motor starters

  • Explainer

    Running a three-phase motor on single-phase 208 V

    Yes — a spinning three-phase motor fed across two terminals regenerates the third leg, but winding heating caps it near 30-40% of rated horsepower, so it is a bench trick, not a supply. For real loads, use a rotary phase converter with an idler at least the next size above the largest motor, or a single-phase-input VFD for motors only — never a drive for amplifiers or power supplies.

    Motors and motor starters

  • Troubleshooting

    Setting voltage imbalance limits on a 480 V phase monitor

    Keep voltage unbalance at the motor terminals at 1% or less to run full rated power; above that NEMA MG 1 requires derating, reaching 75% of rated power at 5%, with operation above 5% not recommended. A monitor set to 15% voltage imbalance is far too loose to protect a motor. Measure A-B, B-C and C-A at the starter stopped and running, and at the main panel, and call the utility only when the main panel itself is unbalanced.

    Power quality and harmonics