When it comes to detecting partial discharge (PD) in medium-voltage and high-voltage switchgear, two non-intrusive sensing technologies have emerged as the industry standard: Transient Earth Voltage (TEV) and Acoustic Emission (AE). While both methods are widely deployed in modern condition monitoring systems, they operate on fundamentally different physical principles and exhibit distinct sensitivity profiles across the four PD types — internal discharge, surface discharge, corona, and floating electrode discharge. This article provides a technical deep-dive comparison of TEV and AE, explains why the dual-detection approach has become the preferred architecture, and examines how these methods are implemented in field-deployed monitoring systems available through Qingdao Britop’s integrated PD monitoring solutions.
TEV Detection: Principles and Physics
Transient Earth Voltage detection exploits an electromagnetic coupling phenomenon unique to metal-enclosed switchgear. When a partial discharge event occurs inside a switchgear compartment, the rapid movement of charge — typically lasting between 1 and 10 nanoseconds — generates an electromagnetic wave that propagates through the internal space at near-light speed. The metal enclosure of the switchgear acts as a Faraday cage, but it is not perfect: the EM wave induces transient surface currents on the interior walls of the enclosure, and at any joints, gaskets, or openings in the metalwork, a small portion of this current flows to the exterior surface, creating a measurable voltage pulse between the enclosure surface and true earth.
This voltage pulse — the Transient Earth Voltage — typically ranges from fractions of a millivolt to several hundred millivolts depending on the discharge magnitude and the distance between the PD source and the enclosure wall. TEV sensors, which are essentially capacitive coupling plates magnetically attached to the switchgear panel exterior, detect these pulses in the frequency range of 3 MHz to 100 MHz. The sensor output is conventionally expressed in dBmV (decibels relative to 1 millivolt), with typical background noise floors around 0-5 dBmV and active PD signals ranging from 15 dBmV to above 40 dBmV.
The key advantage of TEV is its broad sensitivity to internal discharges and corona — PD types that generate strong electromagnetic transients but produce relatively weak acoustic signatures. Because the TEV signal propagates electromagnetically rather than mechanically, it reaches the enclosure wall within nanoseconds and is largely unaffected by the physical construction of internal components. However, TEV has a critical limitation: it is susceptible to electromagnetic interference (EMI) from external sources such as radio transmitters, variable frequency drives, welding equipment, and even nearby mobile phones operating at GSM frequencies. Without proper filtering and validation, EMI can generate false-positive TEV readings that mimic genuine PD activity.
Acoustic Emission Detection: The Ultrasonic Approach
Acoustic Emission detection is based on the fact that every partial discharge event releases a burst of energy that creates a mechanical pressure wave in the surrounding medium. In air-insulated switchgear, this pressure wave propagates as an ultrasonic signal in the 20 kHz to 150 kHz frequency band — a range deliberately chosen because it lies well above the audible spectrum (avoiding interference from human conversation and machinery noise) but below the frequencies where atmospheric attenuation becomes prohibitive.
AE sensors are typically constructed with piezoelectric ceramic elements — commonly lead zirconate titanate (PZT) — that convert mechanical strain into an electrical signal. The sensor resonant frequency is tuned to approximately 40 kHz, which represents an optimal compromise between sensitivity, directionality, and immunity to low-frequency mechanical vibration from switchgear operating mechanisms and cooling fans. Modern AE sensors incorporate built-in preamplifiers with 40-60 dB gain to boost the inherently weak acoustic signals (often in the microvolt range at the transducer output) to levels suitable for digital processing.
The standout capability of AE detection is its sensitivity to surface discharges and floating electrode discharges — PD types that produce strong acoustic emissions because the discharge occurs at a gas-solid interface or in an open gap where acoustic coupling to the surrounding air is efficient. AE is also inherently immune to electromagnetic interference, making it an ideal validation channel when TEV readings are ambiguous. The trade-off is that AE is relatively insensitive to internal discharges, where the solid insulation material (epoxy, XLPE) strongly attenuates the acoustic wave before it can reach the air medium where the sensor operates.
TEV vs. AE: A Head-to-Head Comparison
| Parameter | TEV | Acoustic Emission (AE) |
|---|---|---|
| Physical principle | Electromagnetic coupling | Mechanical pressure wave |
| Frequency range | 3-100 MHz | 20-150 kHz (typically 40 kHz center) |
| Sensor mounting | External, magnetic attachment to panel surface | Internal or external (directional) |
| Best for PD type | Internal discharge, corona | Surface discharge, floating electrode |
| EMI immunity | Moderate (requires filtering) | Excellent (inherently immune) |
| Vibration immunity | Excellent | Moderate (requires mechanical filtering) |
| Localization capability | Poor (panel-level only) | Good (time-of-flight triangulation) |
| Typical sensitivity | 10-50 pC (apparent charge) | 50-100 pC (dependent on acoustic path) |
This complementary profile is precisely why modern PD monitoring systems — such as those deployed in Britop’s power grid online monitoring solutions — combine both sensing modalities in a single integrated platform. When TEV and AE readings are cross-correlated, the system achieves far higher confidence in PD classification than either method alone could provide.
Signal Processing and PD Classification
The raw sensor outputs from TEV and AE channels require sophisticated signal processing before meaningful PD classification can be performed. The processing pipeline typically includes:
Bandpass Filtering: The TEV channel is filtered to 3-100 MHz to reject low-frequency power system harmonics and high-frequency radio interference outside the PD band. The AE channel is filtered to 20-100 kHz, with a notch filter at 50/60 Hz and its harmonics to reject power frequency vibration.
Pulse Extraction: Individual PD pulses are extracted from the continuous waveform using a threshold-crossing algorithm. The threshold is dynamically adjusted based on a rolling noise floor calculation to maintain consistent sensitivity despite varying ambient EMI conditions.
Phase-Resolved Partial Discharge (PRPD) Pattern Generation: Each detected pulse is tagged with its phase angle relative to the 50/60 Hz power frequency cycle, creating the PRPD pattern — a 2D histogram of discharge magnitude versus phase angle. PRPD patterns are the “fingerprints” of PD activity: internal discharges typically appear as symmetric clusters in the first and third quadrants (rising portions of positive and negative half-cycles), while corona is strongly asymmetric, appearing predominantly near the negative peak.
Cross-Correlation: The TEV and AE pulse trains are compared in the time domain to identify coincident events. A pulse detected simultaneously by both TEV and AE channels — within a coincidence window of approximately 100 microseconds — provides strong evidence of a genuine PD event. Pulses detected by only one channel may be EMI (TEV-only) or mechanical noise (AE-only) and are flagged for lower-confidence classification.
Practical Deployment Considerations
Deploying TEV+AE monitoring across a substation with dozens or hundreds of switchgear panels requires careful attention to sensor placement, calibration, and threshold configuration:
TEV Sensor Placement: For each switchgear panel, the TEV sensor should be placed on the front panel as close as possible to the busbar compartment — typically the upper portion of the panel where cable terminations and CTs are located. On panels with internal arc fault containment systems, the sensor should avoid the pressure relief vent area to prevent damage during an arc event. The SCYC-CW30 passive wireless temperature monitoring system demonstrates similar careful sensor placement engineering for ring main unit applications.
AE Sensor Placement: AE sensors are ideally mounted inside each cable compartment and busbar compartment, attached to a non-structural wall that provides good acoustic coupling to the internal air volume. For existing switchgear where internal mounting is not feasible, directional AE sensors with parabolic reflectors can be aimed at panel joints and ventilation grilles from the exterior, though with reduced sensitivity.
Threshold Calibration: PD alarm thresholds must be calibrated per installation rather than applied from a universal template. A switchgear panel in an electrically quiet rural substation may have a noise floor of 2 dBmV TEV, warranting an alarm at 15 dBmV; the same panel in an industrial facility with VFD-driven motors may have a 12 dBmV noise floor requiring a 30 dBmV alarm threshold. The monitoring system should support per-channel threshold configuration with hysteresis to prevent alarm chattering.
Integration with Temperature and Leakage Current Monitoring
PD activity does not occur in isolation. Thermal stress is often the root cause that creates the conditions for PD initiation — for example, thermal cycling that opens micro-cracks in aged epoxy insulation, or overheating that accelerates chemical degradation of insulating materials. Similarly, increased leakage current through contaminated or moisture-affected insulation surfaces often precedes the onset of surface PD. This is why the most comprehensive condition monitoring platforms, such as the DT801 surge arrester monitor, integrate PD detection with continuous temperature and leakage current monitoring to provide a holistic view of insulation health.
Conclusion
The choice between TEV and AE for switchgear partial discharge detection is, in practice, a false dichotomy. The two technologies are not competing alternatives but complementary tools that together provide comprehensive PD detection coverage across all four PD types. TEV excels at detecting internal discharges and corona that produce strong electromagnetic transients; AE excels at detecting surface discharges and floating electrode discharges that produce strong ultrasonic signatures. By deploying both sensors and cross-correlating their outputs through intelligent signal processing, asset managers can achieve detection sensitivity and classification accuracy that neither technology can deliver alone — transforming PD monitoring from an uncertain diagnostic exercise into a reliable, data-driven maintenance tool.
Related Products
- Eagle-beak Shear Excavator Attachment – QINGDAO BRITOP QUALITY PRODUCT
- Auger Excavator Attachment – QINGDAO BRITOP QUALITY PRODUCT
- Steel Structure Shear Excavator Attachment – QINGDAO BRITOP QUALITY PRODUCT
- Jaw Crushing Pliers Excavator Attachment – QINGDAO BRITOP QUALITY PRODUCT
- Jaw Crushing Bucket Excavator Attachment – QINGDAO BRITOP QUALITY PRODUCT
