Partial Discharge Detection for Switchgear Failure Prevention

Home » News » Partial Discharge Detection for Switchgear Failure Prevention

Partial discharge detection has become the cornerstone of condition-based maintenance programs for medium and high voltage switchgear, providing the earliest possible warning of insulation deterioration before complete failure occurs. PD activity in switchgear typically originates from manufacturing defects, installation errors, or aging-related degradation at cable terminations, busbar insulation, and circuit breaker interfaces.

Transient earth voltage sensors represent the most widely deployed PD detection technology for switchgear applications. These capacitive coupling sensors are mounted on the external surface of switchgear panels and detect the electromagnetic pulses that propagate to the earthed metal enclosure when PD occurs internally. The TEV measurement technique is non-invasive, can be performed with the switchgear energized and doors closed, and provides good sensitivity for detecting internal PD activity in the critical frequency range of 3-100MHz.

Ultrasonic detection provides complementary diagnostic capability, particularly effective for surface discharge and corona detection. Airborne ultrasonic microphones and contact ultrasonic probes detect the acoustic emissions generated by PD in the 20-100kHz range. While ultrasonic signals attenuate significantly through solid barriers, making them less sensitive to deeply embedded PD sources, they excel at locating surface tracking on busbar supports and insulator surfaces that TEV sensors may miss. Combined TEV and ultrasonic surveys provide the most comprehensive PD assessment.

Permanent online PD monitoring systems extend the benefits of periodic surveys by providing continuous surveillance. These systems employ permanently installed TEV sensors, high-frequency current transformers on cable earth connections, and UHF sensors within switchgear compartments, all connected to a centralized monitoring unit. Automated PD pattern recognition algorithms classify discharge types and severity, while trend analysis over months of operation distinguishes between stable harmless PD and progressive deterioration requiring intervention.

Related Products

Scroll to Top