2.7 Insulation Power Factor and Dissipation Factor Instrumentation: GST, GSTg, and UST Modes

Insulation Power Factor and Dissipation Factor Instrumentation: GST, GSTg, and UST Modes

Quick Answer: Level III task 3.1a.2 requires the technician to "measure the insulation power-factor/dissipation-factor on all windings" and "measure the power-factor/dissipation-factor on bushings." The whole technique rests on three switch positions. GST (Grounded Specimen Test) measures every path from the energized lead to ground. UST (Ungrounded Specimen Test) measures only the path to a second ungrounded terminal, ignoring ground entirely. GST-guard measures to ground while electronically discarding a nominated path.

---

1. What the instrument actually measures

Healthy insulation is nearly a pure capacitor. Apply AC voltage and the current leads voltage by almost exactly 90°. But real insulation has a small in-phase resistive component from conduction, dielectric absorption, and losses in contaminants and moisture.

• Power factor (PF) = the ratio of the resistive watts-loss current to the total current: cosθ, where θ is the phase angle between applied voltage and total current.
• Dissipation factor (DF), also called tan δ = the ratio of the resistive current to the capacitive current: tanδ, where δ is the complement of θ.

Below about 10 %, PF and DF are numerically almost identical, which is why the terms are used interchangeably in the field. They diverge at high loss levels. Both are expressed as a percentage.

The diagnostic power of the test is that it detects distributed, bulk deterioration — moisture ingress, aging, carbonization, contamination — that an insulation resistance test can miss entirely. A winding can read a healthy 5,000 MΩ on a megohmmeter and still show a badly elevated power factor, because IR is dominated by surface leakage while PF integrates loss through the whole dielectric volume.

The set also reports capacitance and watts loss. Capacitance change is the single most sensitive indicator of a shorted bushing capacitance layer, and watts loss is the raw quantity from which PF is computed.

2. Why test at 10 kV

Standard field sets apply up to 10 kV at 60 Hz. Two reasons:

1. The signal is large enough to resolve microamperes of loss current above the noise floor.
2. Substations are electrically noisy environments; many sets use a slightly offset frequency and filtering to reject 60 Hz interference from adjacent energized equipment.

Test voltage is kept below the rated line-to-ground voltage of the insulation being tested. On low-voltage windings the set is stepped down accordingly.

3. The three measurement modes

Consider a two-winding transformer with high-voltage winding H, low-voltage winding L, and the grounded tank. There are three insulation paths: C_H (H to ground), C_L (L to ground), and C_HL (H to L).

How the guard works. The guard circuit is held at the same potential as the energized lead but its current is routed back to the source without passing through the measuring circuit. Current flowing into a guarded path is supplied but not counted. This is the same principle as the guard terminal on a megohmmeter — it removes an unwanted path from the measurement without physically disconnecting anything.

The standard overall transformer test set is a sequence of GST, GST-guard, and UST measurements from which the individual C_H, C_L, and C_HL values and their power factors are extracted. The internal consistency of the set is itself a check: GST should equal the guarded measurement plus the UST measurement. A discrepancy points to a lead or connection error rather than a transformer problem.

4. Bushing tests: C1 and C2

A condenser bushing is a capacitor built from concentric foil layers between the center conductor and the grounded flange. Most have a capacitance tap (also called a power factor tap or test tap) brought out at the flange, which splits the bushing into two measurable capacitances:

• C1 — the main insulation, from the center conductor to the tap. This is the bulk of the bushing insulation and the primary diagnostic. It is measured in UST, energizing the conductor and taking the return through the tap. UST is essential here because it excludes the grounded flange and everything else in the transformer.
• C2 — from the tap to the grounded flange, a small capacitance. It is measured in GST, energizing the tap.

Interpretation:

• A rising C1 capacitance means one or more capacitance layers have shorted. Because the layers are in series, losing one raises total capacitance. A change of a few percent from nameplate or baseline is significant and is a removal-from-service indicator on many units.
• A rising C1 power factor signals moisture ingress or contamination, commonly through a failed top gasket.
• The tap cover must be securely re-installed after testing. A bushing left with an open or poorly grounded tap will fail in service, because the tap is designed to operate solidly grounded and the voltage stress distribution depends on it. This is a genuine field failure mode and a favourite exam item.

5. Corrections and interpretation

Temperature correction to 20 °C is mandatory. Dielectric losses are strongly temperature dependent, and comparing an uncorrected summer reading against an uncorrected winter baseline produces a false trend. Correction factors come from the test equipment manufacturer's published tables, differentiated by apparatus type — the factor for a bushing is not the factor for a winding.

NETA's own instruction for power factor acceptance is to compare against manufacturer's published data, and in its absence against the test equipment manufacturer's published data. NETA's Table 100.3 gives dissipation factor / power factor values at 20 °C for liquid-filled transformers, regulators, and reactors.

Trending beats limits. A single power factor number in isolation is weak evidence. The strong evidence is:

• Change from the unit's own baseline — typically taken at commissioning.
• Comparison between phases on the same unit, which cancels most common-mode error.
• Comparison against sister units of the same design, age, and duty.

Field conditions that corrupt readings:

• Surface contamination and moisture on bushing porcelain produce elevated readings that are not internal defects at all. Clean and dry the surface, then repeat.
• High humidity and dew on any exposed insulation surface — test when relative humidity is low where possible, and record it.
• Electrostatic interference from adjacent energized equipment — use the set's noise-rejection mode and, where the reading remains unstable, reverse polarity and average.

Exam trap: A question asks which mode measures only the insulation between the high-voltage winding and the low-voltage winding, excluding all paths to the grounded tank. The answer is UST. GST includes ground paths by definition, and GST-guard still measures to ground while excluding a nominated path — it is the mirror image of what the question asks for.