Rewind or Replace? A Twice-Rewound 200 hp Pump Motor with Current Unbalance
A wastewater plant's 200 hp, 480 V, 1,780 rpm TEFC motor drives a centrifugal pump 6,000 h/yr. Nameplate efficiency is 95.0% at full load. A power analyzer at steady load records 480 V average, phase currents of 232/221/208 A, 122 kW input, and 0.86 power factor. A motor shop quotes $9,800 to rewind or $21,500 for a NEMA Premium replacement (96.2% efficient at this load). The utility pays an $8/hp Premium-efficiency rebate. The shop notes the stator has already been rewound twice and expects a third rewind to land near 93.5% efficiency. The plant engineer wants to rewind because the motor 'still runs fine.' The department's simple-payback hurdle is three years.
1. Holding shaft load constant at the measured operating point, compare a third rewind against a NEMA Premium replacement. Does either option clear the 3-year hurdle on energy alone? Show your numbers.
Read the worked answer
Basis: shaft output is constant; only input kW changes with efficiency.
Shaft load = 122 kW x 0.95 = 115.9 kW (using nameplate 95% is generous to the rewind - a twice-rewound core is realistically 1–3 points below nameplate, so every saving below is conservative).
| Option | Efficiency | Input kW | Change vs today |
|---|---|---|---|
| Current | 95.0% (assumed) | 122.0 | - |
| Third rewind | 93.5% | 124.0 | +2.0 kW (worse) |
| New Premium | 96.2% | 120.5 | -1.5 kW |
New vs rewind = 124.0 - 120.5 = 3.5 kW
Energy: 3.5 kW x 6,000 h = 21,000 kWh/yr; at $0.075/kWh = $1,575/yr
Demand: 3.5 kW x $14/kW-mo x 12 = $588/yr → $2,163/yr
Incremental cost: $21,500 - $9,800 - $1,600 rebate = $10,100 → 4.7-year payback
New vs do nothing: 1.5 kW → 9,000 kWh ($675) + $252 demand = $927/yr on $19,900 net = 21 years - fails outright.
Verdict: no single option clears 3 years on energy alone. The decision that actually matters is new versus rewind, which lands at 4.7 years - just outside the hurdle. It clears if the electric rate is above ~$0.10/kWh, if run hours exceed 6,000, or if the cost of one unplanned failure (downtime, emergency rental, overtime) exceeds roughly $6,000 per event for one event inside the three-year window. Note the rewind does not save energy at all; it adds 2 kW.
2. The phase currents are 232/221/208 A. Compute the unbalance and explain what it means for this decision, including what you would measure next.
Read the worked answer
Average current = (232 + 221 + 208) / 3 = 220.3 A Maximum deviation = 232 - 220.3 = 11.7 A Current unbalance = 11.7 / 220.3 = 5.3%
Why it matters:
- NEMA MG-1 requires derating above 1% voltage unbalance and states motors should not be operated above 5% voltage unbalance.
- Current unbalance typically runs 6–10x the voltage unbalance, so 5.3% current unbalance implies roughly 0.6–0.9% voltage unbalance - enough to raise winding temperature rise by roughly 5–10% and shorten insulation life. That is a plausible explanation for two prior rewind failures.
- Load check: 122 kW / 0.746 = 155 hp = 78% of nameplate. If a measured voltage unbalance requires derating to ~75% of nameplate (150 hp), the motor is running at or above its derated capability.
What to measure next (in this order):
- Phase-to-phase voltages at the motor terminals under loaded conditions - not at the MCC, not at the service. Compute %V-unbalance the same way.
- If unbalance is real: look for a blown fuse in a power-factor-correction capacitor bank (classic single-phasing symptom), a loose lug or corroded termination, unbalanced single-phase loading on the upstream transformer, or a bad contact in a bypass/starter.
- Thermographic scan of the MCC bucket and motor peckerhead; winding resistance check across all three phases.
Implication for the decision: if the supply is unbalanced, a new Premium motor will run hot and fail the same way. Correct the unbalance first; otherwise you are buying a more expensive motor to burn up.
3. Write the recommendation you would put in front of the plant manager: what to do immediately, what to do at failure, and what monitoring you would add.
Read the worked answer
Do not rewind now to solve a symptom. Rewinding costs $9,800 and produces a machine that draws 2 kW more than the one already in service.
Immediately (next 2 weeks, before any motor work):
- Commission a voltage-unbalance survey at the motor terminals under load, plus a 7-day power-quality log. This is the leading root-cause candidate and it is nearly free to check.
- Thermographic scan of the MCC and motor terminations; record a vibration baseline.
- Verify the pump operating point - at 155 hp on a 200 hp frame the pump may be oversized, in which case a trimmed impeller or VFD removes load permanently and changes the motor-sizing question entirely.
At failure - install NEMA Premium. Rationale to state plainly: the rewind would save $10,100 up front but cost $2,163/yr more to operate (4.7-year crossover), and a third rewind on a core that has already been hot twice carries a materially higher probability of a repeat failure inside the 3-year window. Add the production cost of a day of pump downtime to the rewind side of the ledger.
Monitoring to install:
- Permanent three-phase power meter on the motor with a current-unbalance alarm (set at 3%).
- 30-day logging to confirm the 6,000 h/yr and the 78% load factor used in this analysis.
- Winding or bearing temperature trending into the plant historian.
Caveat to state in the memo: the economics flip toward immediate replacement if the voltage unbalance exceeds 2% and cannot be corrected at the transformer, because the replacement then restores derated capability as well as efficiency.