Guard Ring Electrodes for Tan Delta Bar & Coil Testing

By | August 31, 2026

When a stator bar is installed in its slot, the conductive slot coating is in electrical contact with the grounded core and the geometry is defined by the slot. Pull the bar out for a laboratory test, and you have to reproduce that slot-electrode arrangement with temporary electrodes. Get that reconstruction wrong and you measure the electrode system, not the insulation — the reading can swing more from a bad electrode setup than from a real difference in the coil.

That is why IEC 60034-27-3 spends so much of its bar-and-coil procedure on electrodes and guarding. Two electrode-related effects can corrupt a bar or coil reading, and the electrode system exists to kill both. This is the electrode side of the dissipation factor test.

Two ways a bar or coil reading goes wrong

Poor contact to the slot coating. The conductive slot coating serves as the outer measuring electrode. Contact it at a few points and the contact resistance is far higher than it would be inside a real slot — and that high resistance reads as unrealistically high tan δ. The fix is to contact the coating along its entire length with metal plates, wrapped metal foil, or spiralled wire, minimizing voids between the electrode and the coating surface. The contacted length runs the full slot portion — not less than the core length. Because the result depends on this electrode arrangement, comparing one bar to another only works if identical electrode systems are used on both.

That poor-contact effect is also a common cause of a negative tan delta tip-up — the reading starts falsely high at low voltage, then falls as contact points ignite and the resistance drops. It is an electrode artifact, not the insulation.

Surface currents off the ends of the slot coating. At each end of the slot coating, leakage currents run along the surface toward the high-voltage side. These end effects add to the measured loss and corrupt the reading. There is no way to contact them out — you have to divert them. That is the job of the guard ring.

What the guard ring does

A grounded guard ring electrode is placed near each end of the conductive slot coating. It intercepts these additional surface currents and diverts them through the grounded guard circuit, instead of letting them contribute to the measuring current. What is left in the measured current is the loss through the main ground-wall insulation — the quantity you actually want.

This is also why the bar/coil guarding arrangement cannot be transferred to a complete winding: the standard states that guarding techniques are not practicable for complete windings. Their readings are instead dominated by the nonlinear resistance of the stress-grading coatings relative to the slot length, which no guard ring removes — so bar/coil results and complete-winding results are not directly comparable.

Every strand in the bar is bonded together first, so no single strand’s insulation sees the full test voltage, and the bar is insulated from earth for the measurement.

Two guarding methods

The standard gives two ways to place the guard ring, and the choice turns on whether cutting the slot coating is acceptable.

Method 1 — guard ring with an insulation gap. A small temporary gap, not more than 4.0 mm wide, is created in the slot coating about 10 mm from each end of the visible start of the end-winding stress-control coating. The grounded guard ring — recommended as metal foil with conductive adhesive — is applied on the short remainder of slot coating adjacent to that gap, without overlapping the stress-control coating.

Two cautions come with this method:

  • The gap width matters. A wider gap enhances the local field and can trigger surface discharge across it, which shifts the measured value. Keep it small and consistent — this is why 4.0 mm is a ceiling, not a target.
  • The gap has to be repaired. After the measurement, the cut is carefully repainted with conductive slot-coating varnish so the coating system is left homogeneous, and the repair needs full cure time before any further testing.

There is a bigger catch on taped insulation: cutting the gap can damage the main insulation if it is not well controlled. For that reason, cutting a gap is not the preferred method for production coils and bars with taped coatings.

Method 2 — guard ring on top of the stress-control coating. For new bars and coils with taped slot and stress-control coatings, the guard ring is fixed on top of the stress-control layer, near where the slot coating and stress-control coating overlap — no gap cut into the slot coating at all. This avoids the risk of damaging the main insulation, which is what makes it the safer choice for taped production objects.

When the slot coating and stress-control coating are painted rather than taped, the guard rings can simply be created as part of the painting process, and temporary grading tapes or capacitive grading electrodes can stand in before the final stress-grading system is applied.

Why the detail is worth the trouble

All of this care exists because the acceptance limits assume it. The Table 1 values are defined for guarded single bars and coils in new condition. Because they apply to measurements made with guard ring electrodes, results obtained with a materially different electrode arrangement should not be compared directly with those limits, or with results taken using the specified arrangement. The guard ring is not an optional refinement on the bar-and-coil test; it is part of what makes the number mean anything.

For the complete-winding case, where guarding isn’t possible and the stress-control coating stays in the measurement, see the dissipation factor pillar.

FAQ

What are guard ring electrodes?

Grounded electrodes placed near the ends of a bar or coil’s conductive slot coating during a tan δ test. They intercept surface leakage currents coming off the ends of the slot coating and divert them to earth, so those end effects don’t corrupt the measured dissipation factor.

Why do stator bars and coils need guarding but complete windings don’t?

A guard ring needs the accessible end of a slot coating to sit on, which an individual bar or coil has. A complete winding’s geometry doesn’t allow it, so complete-winding readings include the stress-control coating and end effects and are judged differently.

Why can’t the slot coating be contacted at just a few points?

Because the contact resistance would be far higher than inside a real slot, and that high resistance reads as unrealistically high tan δ. The coating must be contacted along its full length with plates, foil, or spiralled wire.

What is the maximum insulation gap for guard rings?

4.0 mm. The gap is cut about 10 mm from each end of the visible start of the stress-control coating. Wider gaps enhance the local field and can cause surface discharge, shifting the reading, so the gap is kept small and consistent.

Which guarding method is preferred for production coils?

For taped insulation, placing the guard ring on top of the stress-control coating is preferred, because cutting an insulation gap can damage the main insulation if not well controlled. Painted-coating objects can form the guard rings during the painting process.

Does the guard ring gap need to be repaired after testing?

Yes. When the gap method is used, the cut is carefully repainted with conductive slot-coating varnish to keep the coating homogeneous, and the repair must fully cure before any further testing.

Author: Zakaria El Intissar

Zakaria El Intissar is an automation and industrial computing engineer with 12+ years of experience in power system automation, electrical protection, and SCADA systems. He founded InsulationTesting.com to share practical, field-tested guides on insulation resistance testing, equipment, and industry standards. His writing is used by electricians and maintenance engineers worldwide. Based in Morocco.

Leave a Reply

Your email address will not be published. Required fields are marked *