Current IntelPower LibraryShop main store

Troubleshooting

Three-speed motor trips breaker on high speed

Motors and motor starters Published 8 min read

Short answer

The motor is healthy; the sequence is the problem. Switching from 1,750 rpm to the 3,450 rpm connection reconnects the winding at about 51% slip and pulls roughly three times its 13.3 A nameplate current through the 30 A feeder to the 10 hp rotary phase converter, which trips. Start at the 850 rpm connection, step up with pauses, and move the motor-protector lead off the ground wire.

On this page
  1. Why does the breaker trip only on high speed?
  2. Should you be able to start this motor on any speed?
  3. Is switching speeds while the motor is running good practice?
  4. Why is the middle speed drawing less than half of nameplate amps?
  5. Where do you clamp the meter to see the current that trips the breaker?
  6. What is the overload lead doing on the ground wire?
  7. What is the fix for the high-speed trip?
  8. Questions asked

Your motor is almost certainly healthy, and the breaker is doing its job. Selecting 3,450 rpm reconnects the winding at about 51% slip and pulls start-class current through the single-phase feeder to your 10 hp rotary phase converter, a feeder that already carries roughly 22 A at a full 5 hp load. Start at the 850 rpm connection and step up, and move the motor-protector lead off the ground wire before anything else. The speeds, the currents and the decisions follow.

Why does the breaker trip only on high speed? #

Because the step up from 1,750 rpm is a restart, not a load change: the two-pole winding closes at 51% slip and draws roughly three times its 13.3 A nameplate, about 66 A at the converter input, against a 30 A breaker.

The nameplate speeds are pole connections. Synchronous speed is n_s = 120 × f / P, so at 60 Hz eight poles give 900 rpm, four give 1,800 and two give 3,600. Your 850, 1,750 and 3,450 rpm readings are those three connections running a few percent below synchronism, which is normal.

Slip sets the current. At 1,750 rpm on the four-pole connection, slip is s = (1,800 − 1,750) / 1,800 = 0.03. The instant the two-pole connection closes on the same shaft, s = (3,600 − 1,750) / 3,600 = 0.51. Rotor current rises with slip and the stator follows; at half slip a typical NEMA design B winding sits between full-load current and the 5–7 times locked-rotor band, roughly 3 × 13.3 A = 40 A, for the whole acceleration.

A standard clamp meter updates a few times a second and averages, so your 12.x A reading is a valley between peaks, and the first-cycle peak when the contacts close runs higher still — the same mechanism behind troubleshooting intermittent instantaneous trips on motor starts. The breaker answers the peak; the meter answers the average.

Should you be able to start this motor on any speed? #

No — start on the 850 rpm connection every time: a start directly on the two-pole connection is a locked-rotor event on the 13.3 A winding, typically 5–7 times FLA, about 80 A at the motor and 130 A at the converter input.

The low-speed start works for a plain physical reason: reaching the top speed stores about 16 times the kinetic energy of reaching 850 rpm, because stored energy scales with speed squared — the square of that roughly four-to-one speed ratio is 16. Same inertia and a similar order of current, but the high-speed start lasts many times longer, and duration at start-class current is what separates a tolerated start from a trip on this feeder.

Multispeed machines are also routinely specified to start on the low connection. Ten leads and a drum switch are not enough information to guess at the others: get the connection schedule from the motor manufacturer's data or a motor shop, and never energize combinations that are not on it.

Is switching speeds while the motor is running good practice? #

On a utility supply, yes — multispeed switching is normally done on the run; on this converter-fed circuit, treat every step as a start: unload the machine, step, and let the current settle before the next one.

Both upward steps land near half slip — 850-to-1,800 rpm is s = (1,800 − 850) / 1,800 = 0.53 — so each is a small restart even when it succeeds. The difference at the top step is that the winding, the load torque at speed and the converter's voltage sag under start-class current all arrive at once.

Two checks keep stepping honest:

  • Verify the drum switch continuity position by position against the connection schedule; worn contacts can bridge two circuits and energize two windings during the transition, doubling the event.
  • Give each step the spacing you would give a start; the winding heats like one.

Why is the middle speed drawing less than half of nameplate amps? #

Because a clamp meter at no load reads magnetizing current, not load current: 6.7 A against 13.3 A is a normal idle for the four-pole winding, and slower windings idle proportionally higher, which is why the eight-pole connection sits at 14.6 A.

Magnetizing current is set by the winding and the applied voltage, and it flows at zero load. As a fraction of rating it grows with pole count — typical no-load figures run near 30–40% of FLA on two-pole machines and climb toward 60–70% on eight-pole machines; those are typical ranges, not a guarantee for your motor.

Two follow-ups on the 14.6 A low-speed reading:

  • If it was taken at the motor, check the supply voltage first; magnetizing current rises steeply above nameplate voltage — why motor current rises when voltage exceeds nameplate.
  • If it was taken on the converter input, it includes the idler's current and says little about the motor alone.

The health check is three simultaneous motor line currents: balanced within about 10%, normal speed and quiet running mean the winding is fine. One line near zero while the other two climb is what an open winding actually looks like. Expect the converter's generated leg to read differently from the pass-through legs; when every motor on a service shows the same pattern, the source is the place to look — diagnosing current imbalance across every motor on a service.

Where do you clamp the meter to see the current that trips the breaker? #

On the single-phase input conductor ahead of the converter, because the breaker protects that conductor: motor-side clamps explain winding heating, but only the input current predicts this trip.

Worked example, with the inputs stated: high-speed connection, 13.3 A nameplate; assumed 230 V three-phase at the motor fed from a 240 V single-phase supply — read your nameplate and measure yours.

  1. Motor apparent power at full load: S = √3 × V × I = 1.732 × 230 V × 13.3 A = 5,298 VA, about 5.3 kVA.
  2. Idealized input current carrying that power: I = S / V = 5,298 VA / 240 V = 22.1 A. Losses and reactive current only raise it, so 22 A is a floor. Cross-check by watts: 5 hp × 750 W/hp = 3,750 W, plus about 500 W of idler bearing and windage loss, at a sagging 200 V input gives 4,250 W / 200 V = 21.3 A. Two methods agree: roughly 21–22 A, or 70–74% of the breaker rating, at full load.
  3. Scaling any motor-side reading to the input: I_in ≈ (√3 × 230 / 240) × I_motor = 1.66 × I_motor. The transition estimate becomes 3 × 13.3 A = 40 A motor-side, then 40 × 1.66 ≈ 66 A input, 2.2 times the breaker rating, for the acceleration time.

That is the same apparent-power conversion used when converting a soft-start kVA limit to starting current. Treat the results as estimates and check them against the governing code and the equipment documentation.

Use a meter with an inrush or peak-hold function for the transition; a slow sampling clamp reads the valleys. Record with each reading: the speed selected, the preceding speed, the load state, and which conductor the jaw is on.

What is the overload lead doing on the ground wire? #

It is a motor-winding protector contact, and it belongs in series with the starter coil so an overload drops out the contactor; a lead tied to the equipment grounding conductor is a wiring error that must be corrected before further running.

Lock out the converter and the machine first, test for absence of voltage at the drum switch and the converter terminals, and follow the converter's capacitor-discharge procedure before touching any lead.

Two leads from inside a three-phase motor are almost always a thermal protector. Its contacts go in the control circuit, in series with the coil — the same pattern as wiring overload relay contacts to the coil and a PLC input — so a winding overload opens the coil path and the contactor drops out. NFPA 70 (2023), Article 250, treats current on the equipment grounding conductor during normal operation as objectionable: that conductor exists to clear faults, not to return control current.

As now wired, the protector cannot interrupt the coil circuit, so the motor runs without its winding protection, and any control current returns through the grounding path. Identify the two leads from the motor's markings, the manufacturer or a motor shop; land them in the control circuit; keep the grounding conductor separate and continuous. The rewiring was worth doing, but copying the old connections copied any old error too.

What is the fix for the high-speed trip? #

Sequence the starts: begin at 850 rpm, step to 1,750, wait for the current to fall to its idle value, then take the top speed unloaded; that costs nothing and keeps the feeder inside its curve.

Decisions by case:

  • Machine unloaded or lightly loaded: run the sequence above, and treat each step as a start for winding heating.
  • Unloaded top-speed step still trips after the protector lead is corrected: the converter and feeder are the limit, not the motor. Rotary-converter makers commonly call for at least twice the largest motor's horsepower when starts are hard — confirm against the converter documentation — and NFPA 70 (2023), Article 455, which covers phase converters, bases supply conductor ampacity and overcurrent protection on the converter's marked input rating. Confirm the installation with the authority having jurisdiction.
  • Never upsize the overcurrent device alone; the conductor and the converter's marked input rating have to move with it.
  • Overload settings: size any external overload from the nameplate current of the connection it protects — 14 A at 850 rpm, 13.3 A at 1,750 and 3,450 — which is how NFPA 70 (2023), Article 430, sizes motor overload protection, from nameplate current. With the internal protector in the coil circuit, the winding itself is protected at all three speeds.

Nothing in your readings points at a motor fault: the speeds match 8-, 4- and 2-pole connections, the idle currents are winding behavior, and the trip is a supply event. Fix the ground-tied lead, start low, step up, and the machine should run at its top speed on the feeder you have.

Questions asked

Why does the 7.5 hp tablesaw start on the same converter when the 5 hp three-speed trips?

Duration and stacking, not nameplate horsepower. The saw accelerates a free-spinning blade through its start in well under a second, which the feeder tolerates, and it never reconnects a winding while the shaft is turning. The top-speed step accelerates the same inertia for longer at start-class current on top of the current already flowing, so the same breaker that shrugs off the saw trips on the step.

Will repeated trips like this hurt the motor?

Each 51% slip transition heats the winding like a start, so count steps as starts. NEMA MG 1 gives an ordinary polyphase motor two starts in succession from ambient temperature and one from rated operating temperature unless its nameplate states more; confirm your motor's basis in its documentation. Spaced, sequenced steps stay inside that duty; repeated back-to-back attempts at the two-pole connection do not.

Can I fit a bigger breaker on the converter input?

Not on its own. The input conductors and the converter's marked input rating set the overcurrent protection; NFPA 70 (2023), Article 455 covers phase converters and bases supply ampacity and protection on the converter nameplate. Upsizing the device without upsizing conductor and converter rating is not a compliant fix. Sequence the starts, or size the converter for the duty and the feeder to match, and confirm with the authority having jurisdiction.

What does an actual open winding look like on this motor?

Unequal line currents you can see on a simultaneous three-phase clamp: the open phase reads near zero while the other two climb above nameplate, the motor growls and loses speed, and it will not restart on that connection. Balanced currents within about 10% at normal speed and quiet running mean the winding is healthy, whatever the idle value reads.

Do I still need the drum switch if the machine has a magnetic starter?

Yes. The starter applies and removes line power on one fixed connection; the drum switch does the winding reconnection between pole counts that a single contactor cannot. Building the speeds into contactor control means a proper multispeed scheme: separate contactors per speed, interlocks, and a low-speed-first sequence, with the motor protector wired into the coil circuit like any overload contact.

Keep reading

  • Explainer

    Why motor current rises when voltage exceeds nameplate

    A supply 4% above nameplate sits inside the ±10% band NEMA MG1 Part 12 expects motors to tolerate. The power equation gives a lower current only if efficiency and power factor are held fixed; rising magnetizing current near saturation can push measured current back up, so keep the nameplate full-load rating and treat the recalculated figure as an estimate.

    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

    Troubleshooting intermittent instantaneous motor start trips

    An intermittent instantaneous trip on motor start is usually the random closing angle of the contactor, not a failing motor. A first peak can reach about 2.8 times the symmetrical locked-rotor current, so only starts that close near a voltage zero hit magnetic pickup. Capture all three phase currents, rank contactor pole, winding, rotor and breaker causes, and let the trip follow the part you roll.

    Motors and motor starters