Burnt Star Terminals: High Resistance, Not Insulation

Three of six 44 kW submersible pump motors show two burned leads at the star connection after about one year, while insulation readings remain good and similar to the unaffected motors. Start at the star-point joints: a loose, contaminated, damaged, or poorly crimped connection can generate concentrated I²R heat even when operating current is only 70 A against an 80 A nameplate rating. Good insulation resistance does not clear a series-resistance fault.

Read the burn pattern before testing

The location and symmetry of the damage narrow the fault. Record each motor before disturbing the terminals: which two leads burned, whether the same lead positions recur, how far the insulation receded, and whether the nut, plate, lug, strand bundle, or an internal splice shows the hottest point.

Start here. Do not begin by condemning the winding because its terminal insulation is charred. The heat source is normally found where discoloration is most severe, but heat conduction can move visible damage away from the defective interface.

Understand why a star point can burn

A balanced star point has little potential relative to an ideal neutral, and the instantaneous vector sum of the three phase currents is zero. That does not mean the individual leads carry zero current. Full winding current flows through each phase lead into or out of the node; only the vector sum at the common point cancels.

A connection resistance therefore produces heat according to P = I²R. At the reported 70 A operating current, each 1 mΩ of connection resistance dissipates 4.9 W at that joint. This calculation assumes 70 A is the RMS current through the affected lead. A few milliohms concentrated in a small lug, strand bundle, internal thermal device, or plated interface can create a severe local hot spot without raising total motor current above the 80 A nameplate value.

Heat also accelerates the fault. The joint oxidizes, contact pressure falls, resistance rises, and heating increases again. A nut that feels tight after the event does not prove that the electrical interface was sound before it heated; thermal cycling and damaged surfaces can leave tight hardware around a poor current path.

Inspect the star connection first

1. Isolate the motor circuit, prove it de-energized, and document the terminal positions before removing bridges, plates, nuts, or leads.
2. Inspect all three star-point interfaces under equal lighting. Look for oxide, arcing marks, softened insulation, fretting, discolored plating, cracked lugs, broken strands, reduced strand count, and hardware that bottomed before clamping the conductor.
3. Find the heat center. Damage concentrated at the contact face points to interface resistance; damage beginning inside the cable or crimp points to strand contamination, a poor crimp, or a defective internal splice.
4. Check the assembly sequence against the motor documentation. Washers, plates, lugs, and nuts must present full contact area and maintain pressure through thermal cycling.
5. Clean or replace heat-damaged conductive parts. Do not reuse a lug, plate, or fastener whose contact surface has pitted, lost plating, softened, or distorted.
6. Terminate the conductors using the specified preparation and tightening method. Obtain the tightening value from the motor or terminal-block documentation; do not invent a torque from fastener size alone.

Pay close attention to varnish or epoxy within the conductor strands. Vacuum-pressure impregnation can draw insulating material into flexible leads. A mechanically tight crimp or bolted joint can then have limited metal-to-metal contact. Tightening the nut again will not remove insulation trapped between strands.

Test resistance and phase balance under load

1. Before energizing, measure the resistance of corresponding connections with a four-wire low-resistance instrument when access permits. Zero ordinary test-lead resistance is not adequate for milliohm-level comparison.
2. Compare like-for-like paths across all three phases and, where practical, against an unaffected motor of the same construction. Investigate repeatable asymmetry rather than relying on a generic resistance limit.
3. Energize the repaired motor under its normal hydraulic load. Measure all three line-to-line voltages at the motor circuit and all three line currents at the same operating point.
4. Check the motor protection or upstream controller for loss-of-phase and current-imbalance indications. If one supply phase is open or intermittent, two current paths can be heavily loaded while the third differs sharply.
5. When safe access exists, measure millivolt drop across each comparable termination while current flows. A higher drop at similar current identifies higher connection resistance directly.
6. Trend temperatures at identical locations with the same instrument, surface condition, load, and elapsed run time. A single surface reading cannot clear a buried joint.

If the reported 70 A is an average or a measurement from only one line, repeat the test on all three lines. One acceptable reading can hide a severe imbalance. Record minimum, maximum, and steady readings rather than reporting only one current.

Open the motor when external checks pass

Remove the motor for internal examination when the accessible terminal hardware is sound but the lead continues to heat. Trace each star lead back to the winding connection and any embedded thermal device. Look for a high-resistance splice, partially severed conductor, heat-damaged joint, or insulating resin inside the conductor strands.

A star-point thermal overload or embedded connection can fail as a resistive element. Its location may place the true hot spot inside the motor while leaving only the external lead damage visible. Compare the construction of a failed motor with one of the three unaffected units before changing the design or deleting any protective component.

Phase reversal is not, by itself, an explanation for two burned star connections. Reversing phase sequence changes motor direction; it does not inherently create a high-resistance joint. Verify rotation because a pump operating backward can have abnormal hydraulic performance, but keep the electrical investigation focused on measured current balance and connection voltage drop.

Verify the repair at normal operating load

1. Record all three line currents and line-to-line voltages after the motor reaches stable operation. Compare them with the pre-repair results and the healthy installations.
2. Recheck the three termination temperatures using the same measurement method. Judge temperature rise relative to comparable connections, ambient conditions, and load—not from the 57 C value alone.
3. Inspect for odor, discoloration, insulation softening, and movement at the star-point hardware after the controlled run.
4. Repeat the low-resistance or millivolt-drop comparison. The repaired phase paths should no longer show a distinct outlier.
5. Confirm that loss-of-phase and overload protection detects the conditions it is configured to detect. Use the protection manufacturer's test method and recorded settings.
6. Schedule an early follow-up inspection because three of six units have already developed similar damage. Trend current and connection temperature so rising resistance is found before insulation burns again.

Use the unaffected motors as controls only when they operate under comparable supply, pump load, cable length, cooling, and measurement conditions. A clean-looking terminal is a useful comparison point, not proof of a correct assembly.

Avoid the fixes that waste time

• Do not treat good insulation resistance as proof of a healthy connection. It does not measure the series path through the lug or star joint.
• Do not clear the motor because 70 A is below the 80 A nameplate current. Local I²R heating can occur below rated motor current.
• Do not retighten burned hardware and return it to service. Heat-damaged contact surfaces and relaxed conductors can retain high resistance.
• Do not diagnose single phasing from two burned leads alone. Prove it with simultaneous three-line current and voltage measurements or protection records.
• Do not compare temperatures taken at different loads, locations, elapsed times, or surface conditions. For submerged equipment, use installed sensors or accessible measurement points whose relationship to the buried joint is known.
• Do not replace a complete winding before locating the heat origin. Inspect the bolted interface, crimp, strands, internal splice, and any star-point device in sequence.

Frequently asked questions