IEC 60034-27-3 Explained: Clause-by-Clause Guide

IEC 60034-27-3 is the standard that governs dielectric dissipation factor — tan δ — measurement on the stator winding insulation of rotating machines. This is a section-by-section walk through the document, in the order the standard itself uses, so you can find and understand any clause quickly.

For the test in practice rather than the document structure, start at the dissipation factor pillar. This page is the map of the standard.

What the standard is, and where it came from

The document is the first edition, published in 2015, prepared by IEC technical committee 2 (rotating machinery). It is a full International Standard, not a technical report. It replaces the older technical report IEC TR 60894:1987, and the move from report to standard brought four substantive additions: digital measurement of dissipation factor and capacitance, published limit values, a detailed description of the measuring techniques, and an extension of scope to complete windings.

That history matters for one reason: before this standard, tan δ acceptance limits were largely private, set inside individual manufacturers and utilities. This document made a public set — which is why its limits are both useful and, as we’ll see, debated.

Clause 1 — Scope

The standard gives guidelines for both the test procedure and the interpretation of results for tan δ measurements on stator winding insulation. It applies to machines with conductive slot coatings rated 6 kV and above.

It covers three test-object states: individual form-wound bars and coils outside a core (uninstalled), individual bars and coils installed in a core, and complete form-wound stator windings — in new or aged condition. It applies to vacuum-pressure-impregnated (VPI) and resin-rich fully-loaded taped insulation. It explicitly does not apply to non-impregnated bars, coils, or windings. Limit values are given for bars and coils rated 6 kV and above, tested at 50 Hz or 60 Hz.

Clause 2 — Normative references

Two documents are indispensable to applying this one, both from the high-voltage test-technique family: IEC 60060-1 (general definitions and test requirements) and IEC 60060-2 (measuring systems). The AC supply and the measuring chain have to conform to these.

Clause 3 — Terms and definitions

Four definitions carry the standard:

• Rated voltage (U​N) — the line-to-line voltage assigned to the machine. Every test voltage is expressed as a fraction of it.
• Dielectric dissipation factor (tan δ) — the tangent of the dielectric loss angle δ, the complement of the power factor angle, at a defined voltage, frequency, and temperature. A note records that tan δ and power factor are nearly identical when tan δ stays below 100 × 10⁻³, which holds for all modern windings.
• Delta tan delta (Δtan δ) — the difference in tan δ between two successive test voltages, in steps of 0.2 U​N.
• Tan delta tip-up — the difference in tan δ between 0.6 U​N and 0.2 U​N.

A note warns that if you use voltage steps other than these, the limits in the standard no longer apply. The definitions and the limits are locked together at specific voltage points. For what these values reveal, see tan delta tip-up explained.

Clause 4 — Theory and measuring techniques

4.1 Dielectric dissipation factor measurement. The clause sets up the physics: insulation current splits into a capacitive component leading the voltage by 90° and a resistive component in phase with it, and tan δ is their ratio. It can be modelled with a parallel or series equivalent circuit. The key diagnostic idea lives here — two loss types respond differently to voltage. Solid-insulation loss (polarization, conductivity) barely changes; partial discharges in voids rise as voltage pushes more voids into ionization. The measured curve is their sum.

4.2 Analogue Schering bridge. The classic high-voltage bridge, balanced against a low-loss standard capacitor (typically 100 pF or 1000 pF, with a dissipation factor below 0.01 × 10⁻³). It is sensitive to stray capacitance to earth, so it needs screened cabling and a Wagner earth circuit.

4.3 Transformer ratio arm bridge. An alternative using a current comparator around a magnetic core. It rejects stray-capacitance effects inherently, so it does not need the Wagner earth the Schering bridge requires.

4.4 Digital phase shift measurement. Modern computer-controlled systems that measure the phase difference between the currents through the test object and a reference capacitor. With fiber-optic isolation, the measuring electronics can sit at high potential, which lets these systems measure permanently earthed machines in the field.

Clause 5 — Test procedures

5.1 General. The core procedure: raise the voltage stepwise, typically in 0.2 U​N intervals from 0.2 U​N to rated voltage, holding at each step until the reading stabilizes. Unstable readings signal a measurement problem, not the insulation. Tests run at ambient temperature in air, and winding temperature, ambient temperature, and relative humidity are all recorded, because the curve shifts with temperature.

5.2 Winding bars and coils. The largest sub-clause, because individual bars and coils are where the test does its main quality-control job.

• 5.2.1 Test object preparation — the slot coating must be contacted along its full length with plates, foil, or spiralled wire, not at a few points, or contact resistance inflates the reading. Guard electrodes are required.
• 5.2.2 Guarding techniques — guard ring electrodes strip out surface currents leaking off the end of the slot coating. Two arrangements are described: with a small temporary insulation gap (maximum 4 mm), or placed on top of the stress-control coating. A cut gap is repainted afterward.
• 5.2.3 Measuring procedure — test a statistically meaningful sample from a production run, with some conditioning to stabilize discharge activity before recording.

5.3 Complete windings. Here guarding is not practicable, so the reading is influenced by the stress-grading coatings, end windings, and surface condition. For the most information, measure each phase separately and all phases together, with idle phases tied to the frame.

Clause 6 — Test results

6.1 General. The interpretation rules. The baseline tan δ at 0.2 U​N indicates cure state and depends on the insulation components; Δtan δ as a function of voltage indicates void content and delamination, influenced also by impregnation, contamination, humidity, stress-control coating, and aging.

6.2 Winding bars and coils. This clause holds the limit values — the maximum tan δ0.2 of 20 × 10⁻³, and maximum Δtan δ per step and tip-up of 5 × 10⁻³ each, for new guarded bars and coils up to 21 kV. It also carries the crucial application detail: the limits are compared against the computed value including its measurement spread, above 21 kV values are set by agreement, and the same limits hold for global-VPI objects. New-coil and aged-coil interpretation guidance follows. The full detail, and why these numbers are contested, is on the IEC 60034-27-3 limits page.

6.3 Complete windings. No limits are provided, because the end-winding stress-control coating strongly affects the reading. New windings are measured to control manufacturing quality and set a baseline; aged windings are judged by trending, with a consistent upward trend triggering inspection and further diagnostics. A reduced test voltage down to 0.6 rated voltage can be agreed for aged windings.

Clause 7 — Test report

7.1 General. The report must carry everything needed for both quality analysis of a new winding and future trend analysis of aging. Each individual coil or bar keeps the same identifier in the report as it carries on the object itself, and the three Table 1 characteristic values are recorded for future reference.

7.2 New coils, bars and winding. A detailed report content list, grouped into machine data (manufacturer, type and serial number, ratings, insulation class and system, cooling), test conditions and equipment (test specification, circuit drawing, equipment and its accuracy, responsible organization, location and date, ambient temperature, humidity and pressure, frequency, guarding technique used, personnel), and measurement results (the tan δ and capacitance values at each voltage step, the table of characteristic values, and the tan δ-versus-voltage graph).

7.3 Operational aged winding. Reporting for an aged winding measured on site during a maintenance outage, oriented toward condition assessment and trending rather than acceptance.

Annex A (informative) — Power factor versus dissipation factor

The one annex resolves a common confusion. Power factor (cos φ) and dissipation factor (tan δ) are physically different quantities, related by PF = tan δ / √(1 + tan²δ). The annex tabulates the two across a range of values and shows the difference is negligible below 100 × 10⁻³ — about 0.5 × 10⁻³ of difference even at 100 × 10⁻³ itself. For modern windings, the two are interchangeable in practice, which is why field usage treats “power factor,” “tan delta,” and “dissipation factor” as the same test.

Bibliography

The references point to the standard’s lineage — notably EN 50209 (insulation testing of bars and coils of high-voltage machines) and IEEE 286, the recommended practice for power factor tip-up measurement on stator coil insulation. IEEE 286 is the North American counterpart most likely to come up alongside this standard.

How this standard fits with the others

IEC 60034-27-3 sits in the 60034-27 family covering stator winding insulation diagnostics, alongside the partial discharge parts of the same series. Read next to your partial discharge testing and IEEE 43 insulation resistance work, it completes the AC-loss corner of the machine-insulation picture. For the full standards landscape, see the IEC insulation standards overview.

FAQ