PD Test Setup Time on Site: Why It Runs Long and How to Cut It
1. Where the time actually goes
The pattern is consistent: most of the excess sits in preparation and assembly, not in the measurement itself.
2. The biggest lever: preparation before mobilising
Setup time is decided before anyone leaves the depot.
- Cable records first. Obtain the route drawing, joint positions, as-built length and the commissioning baseline (if one exists). A test team that knows the cable has four joints at known distances sets up differently from one that discovers it on site.
- Access and isolation pre-arranged. Confirm switching, earthing points, permit-to-work and the person authorised to issue it. Late permit approval is the single most common cause of a whole morning lost.
- Cable ends left ready. Where the test follows installation or repair work, agree in advance that the end is cleaned, dressed and left accessible — not buried under a temporary cover or scaffolding.
- Equipment pre-check. Power up the VLF source and PD unit the day before, confirm the calibrator is within calibration validity, and confirm spares (fuses, leads, connectors, earthing sets) are in the van.
3. Standardise the test circuit
A non-standard circuit costs time twice: once when it is assembled and again when it is fault-found.
- One connector family for HV leads, measurement leads and earthing — no adapters on site.
- Leads cut to known lengths and labelled at both ends, so routing is decided rather than improvised.
- Coupling capacitor position determined by the circuit drawing, not by what fits.
- Grounding established first, and only then signal and HV connections made.
- A circuit diagram inside the lid of the test case, so any team member can rebuild the same arrangement.
4. Calibrate immediately after assembly
Calibration is often deferred to the end, then redone when someone realises the setup changed. That doubles the step.
- Inject the calibrator pulse immediately after assembly, before HV is applied.
- Record the scale factor in the test record for that specific circuit.
- Re-calibrate only if the test circuit genuinely changes: cable length, terminations, coupling configuration.
- If a joint or termination is re-made mid-programme, treat the circuit as new.
5. Establish the noise baseline before energising
Time is lost later when the team cannot decide whether what they are seeing is PD or interference.
- Measure and record the background noise floor before applying test voltage.
- Identify local noise sources by inspection: drives, welding, fluorescent lighting, nearby HV equipment, mobile phones in the hatch.
- Set gating and filters against that measured baseline, not against intuition.
- With a baseline recorded, the later "is it PD?" discussion is a comparison rather than a debate.
6. Sequencing multiple circuits in one visit
When several circuits are tested in one shift, setup time per circuit dominates.
- Group circuits by the equipment configuration they need — same test voltage, same coupling arrangement.
- Keep the common elements (earth, instrumentation, power) connected between circuits where isolation rules allow.
- Re-terminate only what changes; do not dismantle the whole circuit to move one lead.
- Test the circuit with the most joints first: if it consumes time, you want it early, not at the end of the shift.
7. What not to cut
Four steps are never time optimisation targets, because each one has caused fatalities or false results in the field:
- Isolation, lock-out and verification of absence of voltage
- Discharge and earthing of the cable under test before connection and after the test
- PD circuit calibration
- Documenting the result with its test conditions
A test that finishes early because one of these was skipped is not a faster test — it is a test that has to be repeated, or one that puts the crew at risk.
8. Common delays and their fixes
FAQ
How long should an on-site PD test setup take? For a typical MV cable with one or two joints, 60–90 minutes from arrival at the cable end — assuming isolation and access are pre-arranged and the end is prepared. Two to four hours indicates preparation gaps rather than an inherently slow task.
Can one calibration serve several circuits? No. The scale factor depends on the specific circuit: cable length, terminations and coupling configuration. Calibrate each circuit after it is assembled, and record the factor with the circuit identity.
What is the single biggest avoidable delay? Late or unconfirmed isolation. It is also the most predictable: switching arrangements are known days in advance, so waiting on the day is a planning failure rather than bad luck.
Does setup time differ between joints and terminations? Yes. Terminations usually take longer to prepare because they are the PD-relevant component and must be clean and correctly assembled. Joints affect setup mainly through the number of measurement points to be correlated.
Is it faster to test several circuits in one visit? Typically yes, if they are grouped by test voltage and coupling arrangement, because the earth, instrumentation and power infrastructure stay connected. Per-circuit setup time falls, but the total shift length still depends on the number of measurement points.
Related reading
- PD Testing for Commissioning — what to measure once the circuit is set up
- PD Testing Equipment for Commissioning — the four functional blocks and the calibration hardware
- Common Mistakes in On-Site PD Testing — ten field errors that produce false results
- Cable Commissioning Tests: The Complete Checklist — where PD fits in the wider acceptance programme