Laatste zaak van het bedrijf over Xi'an Xu&Hui Electromechanical Technology Co., Ltd. Certificeringen

This case study documents a field cable fault test carried out for a large photovoltaic (PV) power generation project in Northwest China. A 35 kV collector feeder cable suffered a flashover-type sealed fault, and the owner needed an accurate fault distance and the exact excavation point before repairs could begin. The following sections summarise the background, field conditions, test equipment and the workflow used to locate and verify the fault.

Project Background

Field Conditions

• Cable route: from a 35 kV box-type transformer (Feeder 03) to cable riser BJ1 at an overhead line tower, direct-buried with cable joint manholes. The route was clear and well marked.
• Joints: two intermediate joints (each inside a manhole), one indoor termination and one outdoor termination.
• Insulation resistance: measured phase-to-earth (remaining phases grounded) with a 5,000 V megohmmeter, about 2 GΩ; absorption ratio (IR) ≈ 1.
• Preliminary evidence: AC withstand at 52 kV flashed over and tripped at 50 minutes; the voltage could not be re-raised beyond roughly 7 kV, confirming a flashover breakdown on Phase A.

Test Equipment Used

• Insulation testing: XHMR-5kV insulation resistance tester
• High-voltage source: 5–50 kV control unit with test transformer and a 40/6 pulse energy-storage capacitor
• Prelocation: Model 502 fault distance meter
• Pinpointing: Model 503D pinpointing instrument
• Withstand voltage: XHBP-1056/528 variable-frequency series-resonant AC withstand test set
• Auxiliary: 60Q cable burn-through bridge

Test Process

1. Insulation test: before the withstand test, R15s = 387 GΩ and R60s = 869 GΩ (absorption ratio 2.25); after the withstand test the insulation fell to about 2 GΩ with an absorption ratio of 1.
2. Route survey: the cable had been newly laid with marker posts present, and the installation crew was still on site, which helped confirm the buried route.
3. High-voltage detection:
   • Using the low-voltage pulse method on the Model 502, the total cable length was measured as 940 m. The joints at approximately 300 m and 600 m were initially suspected breakdown points, and the joint locations provided by the site matched the prelocated distances closely.
   • With the faulty phase isolated and all other conductors plus the transformer grounded, AC voltage was raised to about 10 kV. The 5,000 V megohmmeter could not directly reveal the fault, so a higher voltage was judged necessary for the flashover method. A burn-through bridge was therefore applied first to create a complete conductive channel: voltage held near 4 kV with current fluctuating around 10 mA; after roughly 10 minutes the voltage dropped to 0.3 kV and the current rose above 300 mA, confirming a full burn-through.
   • With the high-voltage flashover method (control unit coupled to the test transformer, shorting bar removed, HV routed through the 40/6 capacitor to the faulty phase via a spark gap, and the capacitor ground connected to the cable copper braid), the flashover waveform sampled by the Model 502 indicated a fault distance of 317.7 m.

Precise Pinpointing

No fault discharge sound was detected at the intermediate joint manholes located before and after the prelocated position. Using the Model 503D pinpointing instrument, a clear but faint discharge sound was heard around the 320 m position; the signal was weak — likely dispersed by sand — and there was no obvious ground vibration. After excavation no visible external damage was found, but on re-pressurising the cable one spot showed markedly stronger vibration, confirming the fault point. Opening the cable confirmed a single-phase joint within the cable body itself. The cable was then repaired.

Withstand Voltage Verification

After repair, the AC withstand test was repeated on the cable. It passed at 52 kV for 60 minutes, confirming the cable was restored to a healthy operating condition.

Key Takeaways

• A prelocator and a pinpointer work as a team: the prelocator narrowed the fault to a section of the route, while the pinpointer locked the exact excavation point.
• For sealed flashover faults with no direct insulation reading, a burn-through bridge can first establish a complete channel, after which the flashover method delivers an accurate distance.
• Manufacturers may treat a damaged cable section with a joint-like repair before delivery; such pre-repaired body joints can later cause the cable to fail the withstand test, so the fault should be verified by opening the cable.
• Body faults can be pinpointed by observing the strength of cable vibration through sand or insulating media, which reveals the exact fault location within the cable body.

Note: For privacy protection, all specific place names, substation/feeder identifiers and personnel names in the original report have been removed or generalised in this published version.