2026-11-16
An aging 10 kV three-core armored cable linking two distribution rooms in a Expressway tunnel system developed a low-insulation fault on two phases. The cable route was known, but the joint positions were not. The XZH TEST field team used insulation resistance testing, low-voltage pulse pre-location, high-voltage flashover pre-location, pinpointing, and post-isolation insulation checks to locate and verify the fault. The case shows why one pre-location method was not enough, and how excavation confirmed the final diagnosis.
| Item | Project information |
|---|---|
| Project | 10 kV cable fault detection in the Expressway tunnel |
| Application | Expressway tunnel power distribution cable |
| Cable under test | Three-core armored cable, rated 10 kV, total length approximately 1,307 m |
| Route | From the Tunnel 1 distribution room to the Tunnel 2 distribution room |
| Site conditions | Route generally clear and along the tunnel roadbed; the two tunnel portals are about 200 m apart |
| Cable history | Old cable in service for more than ten years; joint locations not recorded |
| Test end | Tunnel 1 distribution room |
| Testing objective | Locate the cable fault, confirm the fault type, and check for other faults |
The cable was a 10 kV three-core armored cable, approximately 1,307 m long, running from the Tunnel 1 distribution room to the Tunnel 2 distribution room. It had been in service for more than ten years. The route followed the tunnel roadbed, and the two tunnel portals were about 200 m apart. The main uncertainty was joint location: no reliable joint record was available.
An insulation test on phases A and B produced only 0.1 MΩ, while phase C remained normal. This indicated a serious insulation fault on two phases, but it did not show where the fault was. The team also had to work carefully near the highway roadbed, where several tens of centimeters of concrete covered the cable. Unnecessary excavation would be slow, disruptive, and risky for the tunnel infrastructure.
The team first used a 5 kV insulation resistance tester on its 2,500 V range. Phase A measured 0.1 MΩ, phase B measured 0.1 MΩ, and phase C was normal. The result confirmed a serious insulation fault on phases A and B and established the need for cable fault pre-location. It did not provide a fault distance.
![]()
Figure 0. Insulation resistance test at the Tunnel 1 distribution room: phases A and B measured 0.1 M惟; phase C was normal.
The next step was low-voltage pulse pre-location from the Tunnel 1 distribution room using an XHGG502A1 cable fault locator. The waveform showed a far-end reflection corresponding to a total cable length of approximately 1,307 m. It also showed a possible joint reflection near 630 m, but the waveform did not provide a definitive fault distance. The 630 m event was therefore treated as a suspected joint, not as the fault location. This distinction mattered because a joint reflection can be mistaken for a fault if the waveform is read without confirmation from another method.
![]()
![]()
Figure 1. Low-voltage pulse waveform used to check the total cable length, approximately 1,307 m.
Because the low-voltage pulse method did not resolve the fault distance, the team switched to high-voltage flashover pre-location. An operation box and test transformer were used to raise the voltage, and the pulse generator reached a peak voltage of approximately 21 kV. The fault broke down fully under the applied voltage. Analysis of the flashover waveform placed the fault at approximately 619.19 m from the test end. This became the working pre-location result for the field pinpointing stage.
![]()
![]()
Figure 2. High-voltage flashover waveform showing the pre-located fault distance of approximately 619.19 m from the test end.
The 503E cable fault pinpointer was then used between the two tunnels. The fault was located in the area between Tunnel 1 and Tunnel 2, about 2 m from the Tunnel 2 portal. During excavation, the discharge was audible to the human ear, and the fault was found at a depth of approximately 1 m. Several tens of centimeters of concrete covered the cable, making excavation difficult. Once the cable was exposed, the defect was confirmed as a joint fault.
![]()
![]()
Figure 3. Pinpointing area between the two tunnels; the fault was about 2 m from the Tunnel 2 portal.
After the faulted section was cut, the team tested the cable in both directions from the opened point. The insulation resistance was normal on both sides. On the basis of those readings, no other fault points were identified. The supplied report does not state that the cable was repaired or returned to service, so this article does not claim those outcomes.
![]()
Figure 4. Excavated joint fault, approximately 1 m below the roadbed surface.
| Test stage | Result |
|---|---|
| Insulation resistance, 2,500 V range | Phase A: 0.1 MΩ; Phase B: 0.1 MΩ; Phase C: normal |
| Low-voltage pulse pre-location | Total cable length approximately 1,307 m; no definitive fault distance; possible joint reflection near 630 m |
| High-voltage flashover pre-location | Peak voltage approximately 21 kV; fault fully broken down; pre-located distance approximately 619.19 m from the test end |
| Pinpointing | Fault area between the two tunnels, about 2 m from the Tunnel 2 portal |
| Excavation | Fault at approximately 1 m depth; confirmed as a joint fault |
| Post-isolation insulation test | Both cable directions normal after the faulted section was cut; no other fault points indicated |
The fault was a joint failure on the 10 kV cable. High-voltage flashover placed it at approximately 619.19 m from the Tunnel 1 test end; field pinpointing placed it about 2 m from the Tunnel 2 portal, and excavation confirmed the joint defect at a depth of approximately 1 m. The low-voltage pulse waveform showed a possible joint reflection near 630 m, but that feature was not used as the final fault distance. After the faulted section was cut, insulation tests on both cable directions were normal, indicating no additional fault points in the tested sections.
The source report confirms the fault location and fault type. It does not document the repair method, the replacement joint, or the cable‘s return to service. Those items remain outside the scope of this case study.
| Equipment | Role in the test |
|---|---|
| 5 kV insulation resistance tester | Initial insulation resistance measurement on phases A, B, and C |
| XHGG502A1 cable fault locator | Low-voltage pulse and high-voltage flashover pre-location |
| Operation box and test transformer | High-voltage source for flashover testing |
| 40/6 capacitor bank | Energy storage for the high-voltage flashover circuit, as reported |
| XHDD503E cable fault pinpointer | Field pinpointing; reported in the source as the 503E unit |
| XHGX507 cable route tracer | Route tracing and path confirmation; reported in the source as the 507 unit |
![]()
![]()
![]()
Figure 5. Post-isolation insulation test after the faulted section was cut; both directions tested normal.
XZH TEST is a manufacturer of electrical testing instruments based in Xi'an, China. Since 2013, the company has been developing and manufacturing testing equipment for power utilities, electrical contractors, testing professionals, and industrial applications.