Field Notes

A “pitch motor current high” alarm on one axis; the pitch system is not ready at morning start-up.

Axes moved one by one in service mode: one axis is stiff. Winding balanced, motor brake releases fully.

A slow oil leak at the output shaft seal; the gearbox ran short of oil for months and wore.

Gearbox replaced to the OEM procedure; the other two gearboxes checked for level and leaks.

Watch the slope, not the threshold: the current rise was visible in the SCADA trend months earlier.

Gearbox temperature warning for three windy days; the turbine keeps running.

On a temperature-versus-power plot the turbine sits in its own band above its neighbours. Oil level below minimum, oily dust around the breather.

A leaking shaft seal; with the level low, less oil passed through the cooler.

Oil topped up with the OEM-approved type, leak monitored; seal replacement and oil analysis scheduled.

Read temperature against power, and next to the neighbours — not against time.

In gusty wind the nacelle keeps reversing; rhythmic knocking, hourly yaw count far above neighbours.

Wind direction signal offset and noisy compared with neighbours; yaw brake residual pressure below its lower limit.

A loose wind vane, weak brake holding and late bearing lubrication — together.

Vane tightened and realigned, brake leak fixed and pressure set, bearing lubricated.

Hunting rarely has a single cause; check sensors and mechanics before suspecting parameters.

At high power one winding channel crosses the threshold and the turbine stops; after reset the alarm returns.

The same channel still reads high on a cooled machine. Resistance correct at the sensor, a few ohms higher at the controller input.

An oxidised, loose terminal connection adding resistance and pushing the reading up.

Terminal renewed and tightened; the channel returned to the band of the others. The alarm limit was not changed.

Never raise the alarm limit or bypass a channel; verify the measuring chain first.

Intermittent converter overcurrent trips; every check on the stopped turbine is clean.

Events matched high power. A thermal scan done with an OEM-approved method showed a hot DC busbar joint.

A DC busbar joint that had lost its torque, heating at high current and stressing the neighbouring module.

After LOTO, discharge and a verified zero-voltage test, the joint was renewed and torqued to OEM value; the stressed module was replaced.

With intermittent faults, look for the condition that triggers them — not the fault message.

The turbine is producing. Dark, metallic grease from some purge points; a rubbing noise at one nacelle position.

Grease collected point by point; metal concentrated at three neighbouring points, a magnet test confirmed steel particles.

A blocked lubrication line left one sector of the bearing without grease.

Line cleared, distributor replaced, sector greased; close follow-up at short intervals.

A running pump does not mean grease reaches every point.

Nacelle vibration only in one wind band; no visible blade damage.

Spectrum dominated by 1P with normal 3P; a physical measurement found one blade at a different angle.

After an earlier encoder replacement the blade angle reference had not been calibrated correctly.

All three blade angle references reset to the physical marks with the OEM method; 1P dropped.

Look at the spectrum before calling a blade inspection.