A single tan δ reading tells you how lossy a winding’s insulation is at one voltage. What the insulation does as the voltage rises tells you why. That behavior — the slope of the tan δ curve — is where the diagnostic value lives, and it is captured by two numbers: delta tan delta and the tip-up.
This is the deep dive on those two. For the full test and where it fits, start at the dissipation factor pillar.
Table of Contents
Why tan δ rises with voltage at all
Raise the voltage across stator ground-wall insulation and two different kinds of loss respond in two different ways.
The first is the loss of the solid insulation itself — polarization and conductivity in the resin and mica. This barely changes with voltage. A plot of it against voltage is nearly flat. It sets the baseline.
The second is ionization loss — partial discharges inside gas-filled voids in the insulation. This is strongly voltage-dependent. Below a certain field, the voids stay quiet. As voltage climbs, more and more voids reach their discharge inception and start ionizing, and each discharging void adds loss. So the ionization component climbs, and keeps climbing, as voltage rises.
Add the two together and you get the measured tan δ curve: a flat solid-loss floor with a rising ionization contribution stacked on top. The steeper the rise, the more the insulation is discharging internally. That is the entire physical basis for reading the slope.
The two numbers that capture the slope
Delta tan delta (Δtan δ) is the change in tan δ between two successive voltage steps, measured in 0.2 UN intervals. Walk the voltage up from 0.2 UN to rated voltage in 0.2 UN steps, record tan δ at each, and the step-to-step differences are your Δtan δ values. The largest of them, expressed per 0.2 UN, is the headline — it marks the voltage band where the insulation is ionizing hardest.
Tip-up is the simpler, single-figure version: tan δ at 0.6 UN minus tan δ at 0.2 UN. It captures how much the insulation “tips up” over the middle of its voltage range, without needing the full step-by-step set. It is the number most people quote.
Both point at the same thing — void content and delamination. A low tip-up and small Δtan δ steps mean a dense, well-impregnated insulation with few voids. A high tip-up and a steep Δtan δ mean voids that begin discharging well below rated voltage, which is the signature of poor impregnation or delamination between layers.
One important constraint: these numbers are defined for specific voltage points and steps. Tip-up is specifically the 0.6-to-0.2 UN difference. Use other voltage steps and the published acceptance limits no longer apply, because they were set against these exact points.
Reading the curve shape
The three characteristic values are really shorthand for the shape of the tan δ-versus-voltage curve. Four shapes cover most of what you will see.
Flat, low curve. Low baseline, negligible rise. A dense, well-cured, well-impregnated insulation with little void content. What you want from a new coil.
Flat but elevated. A high baseline that stays roughly flat as voltage rises. The rise is small, so voids are not the story — the solid material is. This points at cure state or resin: an inadequate cure raises the baseline through larger polarization losses, and contamination or moisture can do the same.
Gentle upward slope. Some rise is unavoidable, even on high-quality VPI coils, because mica insulation always contains some micro-delamination inside its anisotropic particles. A modest, smooth increase is normal and not a defect on its own.
Steep upward slope. A sharp climb, especially one starting well below rated voltage, is the warning shape. On a new coil it means part of the insulation was not fully impregnated. On an aged coil, a steep rise beginning far below phase-to-ground voltage signals severe delamination — between the conductor and insulation, or between insulation layers — from operational stress or from over-bending during extraction.
New coils versus aged coils
The same curve is read differently depending on the coil’s history.
On new bars and coils, the test is a manufacturing check. The baseline tells you about cure and resin; the tip-up tells you about impregnation. The capacitance at 0.2 UN is a bonus — it is a direct measure of correct impregnation and cure, and with the coil’s geometry it gives you the relative permittivity of the insulation. A smooth, gentle rise is expected; a steep one flags an impregnation problem.
On aged bars and coils removed from service, the test becomes a condition and re-use tool. A steep increase starting far below phase-to-ground voltage points to severe delamination. Tracking the change in capacitance (ΔC per 0.2 UN step) alongside Δtan δ gives a sensitive combined measure of void content and delamination. On older insulation systems built on asphalt or shellac, long-term thermal aging degrades the organic material and lifts the baseline tan δ. And when you need to decide which bars from a stripped winding are worth reusing, plotting the Δtan δ values of all the bars as a distribution helps separate the good from the bad.
The artifact to rule out first: negative tip-up
Occasionally tan δ starts high and falls as voltage rises — a negative tip-up. This is almost never the insulation. It is a contact problem.
Too high a contact resistance at the measuring electrode inflates the apparent loss at low voltage. As voltage climbs, more contact points ignite, the contact resistance drops, and the reading falls with it. The same reversal can come from an inner conductor shield whose contact to the copper strands improves under voltage.
Either way, a negative tip-up is telling you to fix your electrode contact — contact the slot coating along its full length, not at a few points — and remeasure. It is not a pass, and it is not a fail. It is an invalid curve.
What the slope maps onto
The value of separating baseline from slope is that they point at different stator winding failure mechanisms. A high baseline is a solid-material story: cure, resin, thermal aging, contamination. A steep slope is a void story: poor impregnation, delamination, internal discharge.
That void story overlaps with what partial discharge testing sees — both respond to internal discharge. The difference is that the tip-up measures the aggregate loss those discharges produce across the whole insulation, while a PD measurement can locate individual sites. A high tip-up tells you the winding is discharging internally; PD tells you where. Used together, they are stronger than either alone.
FAQ
What is tan delta tip-up?
The difference in dissipation factor between two voltages — tan δ at 0.6 UN minus tan δ at 0.2 UN. It measures how much a winding’s insulation loss rises with voltage, which indicates void content and impregnation quality.
What does a high tip-up mean?
That the insulation contains voids beginning to discharge well below rated voltage — typically from poor impregnation or delamination. The larger the tip-up, the more internal ionization is occurring as voltage rises.
What is delta tan delta?
The change in tan δ between two successive 0.2 UN voltage steps. The largest step increase, per 0.2 UN, identifies where the insulation is ionizing most and, like the tip-up, indicates void content and delamination.
Why does tan δ increase with voltage?
Because two loss types respond differently. Solid-insulation loss stays nearly flat with voltage, but partial discharges in internal voids increase as rising voltage pushes more voids into discharge. The measured curve is the sum, so it climbs where voids ionize.
What causes a negative tip-up?
A contact-resistance artifact, not an insulation property. High electrode contact resistance inflates the low-voltage reading; as voltage rises the contact improves and the reading falls. Fix the electrode contact and remeasure.
Is some tan δ rise with voltage normal?
Yes. Mica insulation always contains some micro-delamination, so a smooth, gentle rise is expected even on high-quality VPI coils. A steep rise, especially one starting well below rated voltage, is the abnormal shape.
