Condition Assessment and Diagnostic Testing Methods for Surge Arresters

Without systematic condition monitoring, a degrading surge arrester will fail without warning at the worst possible moment.

1. What Condition Data Your Maintenance Program Needs to Collect

Metal-oxide surge arresters (MOSAs) operate silently—24 hours a day, 365 days a year—with no moving parts and no visible indicators of internal health. This “silent degradation” makes systematic data collection the only reliable way to detect incipient failure before a surge event exposes a weakened arrester.

A robust maintenance program begins with establishing baseline values for every arrester upon installation. Trend analysis, not absolute values, delivers the earliest warnings. The five parameters below form the core dataset that every utility should collect.

Resistive Leakage Current (I_r)

The resistive component of leakage current is the most sensitive online indicator of ZnO block health. As grain boundaries within the metal-oxide varistor degrade, the voltage threshold for significant current flow drops, increasing I_r weeks or months before catastrophic failure. Modern monitoring systems use a Rogowski coil or low-power CT to measure total leakage current and extract I_r through phase-sensitive detection referenced to the system voltage.

Third-Harmonic Component (I_3ω)

As varistor nonlinearity increases with aging, the third harmonic of the leakage current rises proportionally. Third-harmonic analysis is particularly valuable in polluted coastal or industrial areas, where total leakage current measurement may be dominated by capacitive surface currents caused by surface contamination. I_3ω > 0.3 mA warrants a warning; values above 1.0 mA indicate critical degradation.

Reference Voltage (U_1mA)

The DC voltage required to produce exactly 1 mA through the arrester—measured during offline testing—directly reflects the condition of the ZnO block stack. A decrease of more than 10% from the original factory value indicates significant block degradation or internal shorting. Offline U_1mA measurement remains the definitive assessment method when a system outage is available.

Thermal Signature

A healthy arrester operating within its ratings maintains a nearly uniform temperature distribution along its length, typically only slightly above ambient. An elevated internal temperature, detectable via infrared thermography, signals excessive I_r or poor internal connections. Thermal imaging also detects hot spots at line and ground terminals—a common problem independent of ZnO block condition.

Partial Discharge (PD) Activity

PD signals inside an arrester indicate voids, cracks, or contamination within the internal insulation system. Online PD monitoring uses high-frequency current transformers (HFCT) clamped onto the ground lead, or ultra-high frequency (UHF) sensors. PD monitoring is essential for GIS-installed arresters, where visual inspection and thermal imaging are impossible.

Figure 1: Test setup for resistive leakage current measurement. The CT measures total leakage current; phase-sensitive detection using the voltage reference separates I_r (resistive) and I_c (capacitive) components.

2. Evaluating Assessment Methods: Accuracy, Cost, and Practicality

No single diagnostic method provides a complete condition picture. The most reliable assessment combines complementary methods matched to your operational constraints, budget, and risk tolerance. The following evaluation frameworks help you select the right mix.

Online Methods: Continuous Insight Without Outages

Online (energized) diagnostic methods allow assessment without taking the arrester out of service. Resistive leakage current monitoring is the most widely adopted online technique, capable of detecting incipient ZnO degradation weeks before failure. Third-harmonic analysis adds sensitivity in polluted environments. Thermal imaging detects poor connections and thermal runaway conditions but requires favorable weather. Online PD monitoring is invaluable for GIS-installed units.

Accuracy: High for resistive current and third-harmonic methods when properly calibrated. Medium for thermal imaging (affected by weather and emissivity variations).

Cost: Medium for resistive current monitors; medium-high for thermal cameras; high for PD systems.

Practicality: Excellent for routine monitoring; resistive current monitors can be left in place permanently with IoT data transmission.

Offline Methods: Definitive Assessment During Outages

Offline (de-energized) testing delivers the most definitive condition data but requires a scheduled system outage. The DC reference voltage test (U_1mA) is the fundamental offline test—simple, low-cost, and highly sensitive to ZnO block degradation. Insulation resistance testing detects moisture ingress. Capacitance and tan δ measurements reveal changes in dielectric properties.

Accuracy: Very high for U_1mA and insulation resistance. High for capacitance/tan δ.

Cost: Low for U_1mA and insulation resistance; medium for capacitance/tan δ test sets.

Practicality: Limited by outage availability. Best performed during scheduled maintenance windows every 5–10 years.

Visual Inspection: The First Line of Defense

Visual inspection is the simplest and most frequent assessment activity. Trained inspectors check for housing cracks, seal condition, corrosion of end fittings, pollution accumulation, and lightning counter readings. While visual inspection cannot detect internal ZnO block degradation, it remains essential for identifying external damage that compromises seal integrity and leads to moisture ingress.

Accuracy: Low for internal condition; high for external damage detection.

Cost: Very low.

Practicality: Excellent as a routine screening tool; should be performed annually or after major storms.

Figure 2: Decision flowchart for condition-based assessment and replacement of metal-oxide surge arresters, combining visual, online, and offline diagnostic results.

3. Building a Condition-Based Maintenance Program That Works in the Field

Condition-based maintenance (CBM) replaces time-based replacement with data-driven decisions, extending arrester service life by 20–40% while reducing forced outages. A successful CBM program rests on three pillars: data continuity, interpretation discipline, and clear replacement thresholds.

Establishing Baselines and Trends

The single most valuable action is recording baseline diagnostic values for each arrester at installation. Without baselines, absolute thresholds are your only reference—and they vary with arrester design, system voltage, and ambient conditions. Trend analysis detects degradation far earlier than one-off measurements. At a minimum, record the initial I_r, I_3ω, and U_1mA values, then re-measure annually for online parameters and every 5–10 years for offline parameters.

Setting Replacement Thresholds

The following interpretation guidelines, based on IEC 60099-4 and IEEE C62.11 recommendations, provide actionable thresholds for field decisions:

Integrating Smart Monitoring Systems

The latest trend in CBM is the deployment of integrated smart monitoring systems that measure resistive current, third harmonic, housing temperature, and ground lead current simultaneously. These systems transmit data via cellular IoT or fiber optic links to a central asset management platform, enabling continuous condition tracking without site visits. For substations with critical transformers or in areas with high lightning density, smart monitoring delivers an outstanding return on investment by preventing unplanned outages.

Age as a Secondary Criterion

Even with good diagnostic data, age remains a useful secondary input. Arresters older than 25 years should be considered for proactive replacement, even if diagnostic values remain within acceptable limits. Polymer housing materials degrade independently of the ZnO blocks, and seal integrity deteriorates over decades. A risk-based approach weighs the cost of proactive replacement against the consequence of failure for each unit.

4. Online Monitoring vs. Offline Testing vs. Visual Inspection: Which Approach Fits Your Operation?

Selecting the right assessment strategy depends on your substation criticality, maintenance budget, outage flexibility, and risk tolerance. Many utilities adopt a hybrid approach: annual visual inspection for all units, online monitoring for critical substations, and offline testing during scheduled outages for units approaching 15+ years of service.

The optimal strategy for most utilities is a three-tier program: Tier 1—annual visual inspection for all units; Tier 2—online resistive current monitoring for arresters protecting critical transformers or located in high-lightning-density areas; Tier 3—offline U_1mA and insulation resistance testing during scheduled outages for units aged 15+ years. This approach balances cost, risk, and diagnostic value while building a multi-year dataset for trend-based decisions.

5. Summary

• Pain point: Without systematic condition monitoring, a degrading surge arrester will fail without warning at the worst possible moment.
• Finding 1: Resistive leakage current (I_r) and third-harmonic analysis are the most sensitive online methods for detecting incipient ZnO block degradation weeks before failure.
• Finding 2: A complete condition assessment program combines online monitoring (for continuous insight), offline testing (for definitive diagnosis during outages), and visual inspection (for external damage detection).
• Finding 3: Condition-based maintenance, supported by baseline data and trend analysis, can extend arrester service life by 20–40% compared to time-based replacement, while significantly reducing unplanned outages.
• Comparison conclusion: Online monitoring delivers the best early-warning capability without outages; offline U_1mA testing provides the most definitive assessment when an outage is available; visual inspection is essential but cannot detect internal aging.