Request a Quote for High-Voltage Components&Equipment

Tell us your requirements — rated voltage, model, quantity, and destination — and our XBR Electric team will prepare a detailed quotation within 24 hours.
Contact Form Demo
Epoxy part quality inspection showing crack, void, and tracking defect types with inspection tools for MV switchgear

Epoxy Part Quality Inspection: Cracks, Voids & Tracking—Field Guide

Cast epoxy resin serves as the backbone of solid insulation in medium-voltage switchgear. It encapsulates vacuum circuit breaker poles, supports bus conductors, and forms the bushings that transition power between compartments. When epoxy fails, the equipment fails with it—often during peak demand when thermal and electrical stresses combine.

This field guide provides systematic inspection methods for detecting cracks, voids, and tracking in epoxy components rated 12 kV through 40.5 kV. The techniques apply to incoming inspection, commissioning verification, and periodic in-service assessment.

How Cracks, Voids, and Tracking Form in Cast Epoxy

Understanding defect origins sharpens inspection focus. Each defect type follows a distinct formation pathway.

Void Formation

Voids form during casting when entrapped air or volatile compounds cannot escape before the resin gels. In quality assessments across 200+ epoxy insulator batches, voids larger than 0.5 mm consistently concentrate at resin-filler interfaces where wetting is incomplete. The exothermic curing reaction generates temperatures of 120–180°C, creating pressure gradients that nucleate gas bubbles.

The dielectric consequence is severe. Virgin epoxy withstands 20–25 kV/mm. A void drops local breakdown strength to approximately 3 kV/mm due to Paschen’s law effects in enclosed gas cavities. Partial discharge within these voids generates temperatures exceeding 500°C, progressively enlarging the defect.

Crack Initiation

Thermal cycling drives crack formation through coefficient of thermal expansion (CTE) mismatch. Unfilled epoxy exhibits CTE of 50–70 × 10⁻⁶/°C while copper conductors measure 17 × 10⁻⁶/°C. This mismatch generates interfacial stresses exceeding 15 MPa at temperature differentials of 80°C. Cracks propagate from sharp corners, filler clusters, and conductor interfaces.

Field experience with outdoor switchgear shows that temperature cycling between -25°C and +55°C produces micro-cracks at stress concentration points within 8–12 years of service.

Tracking Development

Surface tracking represents progressive carbonization under sustained electrical stress combined with moisture and contaminants. When surface contamination creates conductive moisture films, leakage current flows. The current heats the surface unevenly, creating dry bands where resistance concentrates. Arcing across these dry bands carbonizes the epoxy, forming permanent conductive paths.

Coastal substation inspections demonstrate that salt fog contamination accelerates tracking initiation, reducing surface insulation resistance below 10 MΩ within 18 months of exposure.

Cross-section diagram showing void, crack, and tracking defect formation mechanisms in cast epoxy insulation systems
Figure 1. Epoxy defect formation mechanisms—internal void with field concentration (left), thermal stress crack at conductor interface (center), and surface tracking carbonization path (right).

[Expert Insight: Manufacturing Quality Indicators]

  • Yellowed epoxy suggests overcure or UV exposure during storage—inspect more carefully for internal stress
  • Flow marks on surfaces indicate mold filling problems that correlate with internal void clusters
  • Sink marks near thick sections often overlay subsurface voids
  • Batch-to-batch color variation warrants supplier quality discussion

How to Inspect Epoxy Parts for Cracks

Visual inspection catches most crack defects when performed systematically with proper lighting.

Equipment Required

  • 10× magnifying loupe
  • LED inspection light (5000K color temperature minimum)
  • UV-A lamp (365 nm wavelength)
  • Removable marking tape

Inspection Procedure

Start with the part at arm’s length under diffuse ambient light. Rotate slowly through 360°. Surface cracks longer than 3 mm appear as shadow lines even without magnification.

For detailed examination, position the LED light at 15–30° grazing angle to the surface. This low angle creates shadows that reveal crack depth and extent. Scan systematically from one end to the other, overlapping inspection zones.

UV inspection reveals cracks invisible under white light. Many manufacturers add fluorescent tracers to epoxy formulations. Under 365 nm illumination, cracks appear as bright lines against a darker background. This technique excels at finding hairline cracks in complex geometries.

Critical Inspection Zones

Concentrate attention on high-stress areas:

  • Metal insert interfaces where CTE mismatch is greatest
  • Mounting flange corners and bolt hole perimeters
  • Geometric transitions from thick to thin sections
  • Areas near vacuum interrupter mounting surfaces
Grazing light inspection technique diagram showing LED positioning at 15–30° angle for epoxy crack detection
Figure 2. Grazing light inspection technique—LED positioned at 15–30° angle creates shadows that reveal crack depth and extent in epoxy surfaces.

Crack Acceptance Criteria for MV Epoxy Components

Not every crack warrants rejection. Location and size determine the appropriate response.

Crack TypeLocationMaximum AllowableAction
Hairline <0.1 mm widthNon-stressed surface5 mm lengthAccept with documentation
Hairline <0.1 mm widthNear metal insert2 mm lengthReject or consult manufacturer
Visible ≥0.1 mm widthAny locationNot acceptableReject
Through-crackAny locationNot acceptableReject immediately

Cracks near conductor interfaces demand strict interpretation. The electric field enhancement at a crack tip accelerates partial discharge inception. A hairline crack that might survive decades on an unstressed surface can progress to failure within months near a high-voltage conductor.

Document all accepted cracks with photographs and dimensional measurements. This baseline enables trending during subsequent inspections.

Void Detection Methods for Epoxy Insulation

Internal voids require detection methods beyond visual inspection. Three techniques apply to field and factory settings.

Ultrasonic Testing

Pulse-echo ultrasonic inspection detects voids ≥0.3 mm diameter in epoxy up to 80 mm thick. The technique works because voids create acoustic impedance mismatches that reflect ultrasound energy.

For field application:

  • Select 2–5 MHz transducers based on part thickness (higher frequency for thinner parts)
  • Apply glycerin or water-based couplant liberally
  • Scan at 50% overlap between passes
  • Flag any reflection amplitude exceeding 20% of back-wall echo

Partial Discharge Testing

PD testing identifies electrically active voids—those that will cause progressive damage. Apply voltage at 1.5–2.0 × rated phase-to-ground voltage and measure discharge magnitude.

PD testing during incoming inspection requires specialized equipment and controlled conditions. Many facilities reserve this method for high-value components or random sampling from large batches.

X-Ray Radiography

Industrial X-ray inspection reveals voids regardless of electrical activity. This method suits high-value components where internal void location matters as much as void presence. The technique identifies voids near conductor surfaces—the highest-risk locations—that ultrasonic methods may miss due to geometric complexity.

Void Location Risk Assessment

Position determines consequence. Voids near conductors experience field enhancement that accelerates partial discharge.

Void LocationRelative RiskField Enhancement
Within 5 mm of conductorCritical3–5× average field
At metal insert interfaceHigh2–4× average field
Bulk material centerModerate1–2× average field
Near grounded surfaceLower1–1.5× average field

Reject any component with voids within 5 mm of conductor surfaces, regardless of void size.

Void location criticality map showing risk zones radiating from conductor surface in epoxy insulation cross-section
Figure 3. Void location criticality assessment—voids within 5 mm of conductor surfaces (red zone) experience 3–5× field enhancement and require rejection regardless of size.

[Expert Insight: Field PD Testing Realities]

  • Background noise in operating substations often exceeds 5 pC—use gating and filtering
  • Temperature affects PD magnitude; test at stable thermal conditions when possible
  • A single PD test provides a snapshot; trending over time reveals degradation rate
  • Correlation between UT void detection and PD activity runs approximately 70%—some voids remain inactive for years

How to Identify and Prevent Tracking Damage

Tracking damage leaves visible evidence on epoxy surfaces. Recognition enables intervention before flashover occurs.

Visual Identification

Look for these indicators:

  • Brown or black dendritic patterns branching across the surface
  • Roughened texture along discharge paths
  • Pitting at arc root locations
  • White powder deposits indicating decomposition products

High-risk inspection areas include outdoor bushings, components in industrial environments with conductive dust, and parts near cable entries where condensation collects.

Comparative Tracking Index Requirements

CTI quantifies tracking resistance. The IEC 60112 test applies ammonium chloride solution drops between electrodes while increasing voltage until tracking occurs. Results guide material selection:

CTI ValueClassificationApplication Suitability
≥600 VCTI 600Outdoor, contaminated environments
400–599 VCTI 400Indoor, normal environments
<400 VNot recommendedAvoid for MV insulation

Quality epoxy parts from established switchgear component manufacturers specify CTI ≥600 for any surface exposed to environmental contamination.

For components destined for outdoor VCB installations, verify CTI rating against site contamination severity. Coastal, industrial, and desert environments demand CTI 600 minimum.

Complete Incoming Inspection Checklist for Epoxy Parts

Systematic documentation transforms inspection from subjective assessment to defensible quality record.

Visual Examination (100% of received parts)

  •  No visible cracks under 10× magnification with grazing light
  •  Uniform surface finish without flow marks or sink marks
  •  Embedded metal parts correctly positioned per drawing
  •  Consistent color throughout (no yellowing or discoloration)
  •  Legible markings: date code, batch number, manufacturer ID

Dimensional Verification (sample basis per lot size)

  •  Critical dimensions within drawing tolerance
  •  Mounting hole locations and diameters correct
  •  Creepage distances meet or exceed specification

Electrical Testing (100% for critical components, sample for routine)

  •  Partial discharge below acceptance threshold
  •  Insulation resistance >10 GΩ at 5 kV DC

Documentation Package

Record for each inspected lot:

  • Part number and manufacturer lot code
  • Inspection date and inspector identification
  • Test equipment serial numbers with calibration dates
  • Pass/fail determination with supporting measurements
  • Photographs of any anomalies, even if accepted

This documentation supports root cause analysis if field failures occur and provides evidence for warranty claims.

Incoming inspection workflow flowchart for epoxy parts showing visual, dimensional, and electrical test decision points
Figure 4. Incoming inspection workflow—systematic progression from receiving through visual, dimensional, and electrical verification to documented acceptance or rejection.

Source Quality Epoxy Components for Switchgear Projects

Epoxy component quality begins at manufacturing. XBRELE applies incoming material testing, vacuum casting process control, and 100% partial discharge verification to every epoxy part.

Standard specifications include:

  • CTI 600 for all surfaces exposed to environment
  • Void content verified by sample radiography per production lot
  • Full dimensional inspection with CMM documentation
  • Traceability from raw material batch through finished component

Application engineering support addresses VCB pole assemblies, vacuum contactor housings, bus support insulators, and custom switchgear frame components.

Contact XBRELE for epoxy component specifications, material certifications, and sample evaluation.


External Reference: IEC 60071 — IEC 60071 insulation coordination

Frequently Asked Questions

What causes most epoxy insulation failures in switchgear?

Voids from manufacturing defects cause the majority of in-service failures, with thermal cycling cracks as the second most common mechanism. Environmental tracking failures occur primarily in outdoor or contaminated installations where CTI ratings were inadequate for site conditions.

How can I tell if an epoxy crack is structurally significant?

Location matters more than size. Cracks within 5 mm of any conductor surface or at metal insert interfaces pose dielectric risk regardless of visible dimensions. Cracks on unstressed external surfaces may be acceptable if documented and monitored.

What partial discharge level indicates an epoxy defect requires action?

PD magnitudes above 10 pC at 1.2 times rated voltage generally warrant investigation, though acceptable limits vary by component type and manufacturer specification. Trending is more valuable than single measurements—rising PD over time indicates active degradation.

Can ultrasonic testing detect all void types in epoxy?

Ultrasonic methods reliably detect voids ≥0.3 mm in accessible geometries but may miss defects near complex metal inserts or in thin sections. Combining UT with PD testing improves detection confidence for critical components.

How does altitude affect epoxy insulation inspection requirements?

Higher altitude reduces air density, lowering external flashover voltage but not affecting internal void behavior. Creepage distance becomes more critical above 1000 m elevation. Internal defect acceptance criteria remain unchanged.

What inspection frequency is appropriate for in-service epoxy components?

Annual visual inspection catches gross degradation. PD testing every 3–5 years, or following significant fault events, provides quantitative condition assessment. Harsh environments with temperature cycling or contamination exposure may warrant more frequent evaluation.

Hannah Zhu marketing director of XBRELE
Hannah

Hannah is the Administrator and Technical Content Coordinator at XBRELE. She oversees website structure, product documentation, and blog content across MV/HV switchgear, vacuum breakers, contactors, interrupters, and transformers. Her focus is delivering clear, reliable, and engineer-friendly information to support global customers in making confident technical and procurement decisions.

Articles: 99