6.3 Bushing Power Factor & Transformer PF Configurations

What Power Factor / Tan-Delta Measures

An insulation power-factor test (also called dissipation-factor or tan-delta) applies an AC voltage to the insulation and measures the loss angle between the capacitive and resistive components of the resulting current. Pure insulation behaves like an ideal capacitor — current leads voltage by 90°, and the loss angle is zero. Real insulation has resistive losses, so the angle is slightly less than 90°. The tan of that loss angle equals the ratio of resistive to capacitive current — a direct measure of dielectric loss.

Elevated PF indicates moisture ingress, contamination, or thermal aging of the oil-paper insulation. Trending PF over years is more diagnostic than a single absolute reading.

Bushing C1 and C2 Tests

A capacitance-graded bushing has a concentric capacitor stack between the high-voltage conductor and the ground flange, with a capacitance tap brought out for testing. This creates two distinct insulation paths:

• C1 — main insulation: between the line conductor and the capacitance tap. Tested in UST (Ungrounded Specimen Test) mode at 10 kV. This is the primary dielectric path that sees service stress.
• C2 — tap-to-ground insulation: between the capacitance tap and the ground flange. Tested in GST-g (Grounded Specimen Test, guarded) mode at 500 V for a test tap or 2 kV for a potential tap. C2 protects the tap circuit and indicates tap-area moisture or contamination.

The two tests are complementary. A high C1 PF points to main insulation degradation. A high C2 PF (or abnormal capacitance, or negative PF) points to tap-area moisture, a loose flange ground, or a defective tap. Hot collar tests (10 kV, GST-GND) supplement C2 by locating low-oil or condenser faults section by section.

Acceptance guidelines (nameplate comparison)

• C1 PF > 1.5× nameplate → questionable; test more frequently.
• C1 PF > 2× nameplate → unacceptable for service.
• Capacitance change > 5% → investigate; > 10% → consider replacement (shorted condenser layers).
• C1 tip-up (10 kV PF − 2 kV PF) should be ≈ 0.00% for healthy oil-paper insulation. Values above ~0.05% warrant investigation.

Transformer Power-Factor Configurations

A two-winding transformer has three insulation paths of interest: high winding to low winding (H-X), high winding to ground (H-GND), and low winding to ground (L-GND). The Doble test set measures each using a combination of three modes:

• UST (Ungrounded Specimen Test) — measures only the current between the HV lead and the specific terminal connected to the measuring input; rejects any current flowing to ground. Isolates a single inter-winding path.
• GST (Grounded Specimen Test) — the specimen's low end is grounded; measures all current paths between the HV terminal and ground simultaneously (winding-to-winding plus winding-to-ground).
• GST-Guarded — adds the guard terminal to divert a specific stray current path around the measuring circuit so it is not counted.

The standard three measurements are:

Running all three provides a consistency check: the readings are arithmetically related, so a measured third value that does not match the value calculated from the other two reveals a connection error, a grounding problem, or external interference. CHL is not C1 — C1/C2 is bushing nomenclature; CH/CL/CHL is winding nomenclature. Confusing the two is a common exam trap.

Typical PF values (oil-filled transformers, corrected to 20°C)

Dry-type transformer values are higher — ventilated dry-type CHL up to 2.0%, CL up to 4.0%, CH up to 3.0%.

Guard Terminal Use

The guard terminal diverts surface leakage current (e.g., across dirty or wet bushing porcelain) around the measuring circuit so the meter reads only the volume (through-insulation) resistance. Proper guarding gives a cleaner reading for trending. Improper guarding — guarding out a path that is part of the intended measurement — produces falsely low (good-looking) results. Guarding must be deliberate, never default.

Temperature Correction and Trending

PF is strongly temperature-dependent. Correct all readings to 20°C using the oil temperature (not ambient air). Trending a given bushing or winding over years, with consistent temperature correction, is more diagnostic than comparing to a generic table. A sudden step-change between scheduled tests is the strongest alarm — even if the absolute value is still under 0.5%.

The Doble Test Set

The Doble M-series (and equivalent OMICRON test sets) is the field-standard instrument for insulation PF / tan-delta and capacitance testing. It performs bushing C1/C2, transformer CH/CL/CHL, CT, PT, and breaker insulation tests, and supports voltage and frequency sweeps for tip-up and DFR analysis. Trending software compares each test to the unit's nameplate and to prior readings — the diagnostic power comes from the trend.

Common PF Test Traps

• Confusing C1/C2 (bushing) with CH/CL/CHL (winding) — C1 is the main bushing insulation in UST at 10 kV; CH is the high-to-low winding path in UST. They are different tests on different apparatus.
• Guarding by default — guarding out a path that is part of the intended measurement produces a falsely low (good-looking) PF. Guarding must be deliberate.
• Comparing an uncorrected field reading to a 20°C nameplate — always temperature-correct to 20°C using oil temperature before any nameplate comparison.
• Ignoring capacitance change — a C1 capacitance change > 5% (especially an increase) indicates shorted condenser layers, a serious mechanical fault even if PF looks fine.
• Trusting a single reading over the trend — a step-change between scheduled tests is the strongest alarm, even when the absolute value is still under 0.5%.