A varnish carrying a temperature index of 155 has not been run at 155 °C for its service life and survived. Nobody has that much time. (The type number and the temperature index are related but not the same thing — more on that below.)
What has happened is this: specimens were aged at a set of elevated temperatures chosen according to the test procedure and the material’s behaviour, a defined property was tracked until it reached a defined end point, the times to reach that end point were plotted, and the relationship was extrapolated to give the temperature at which the same end point would be reached after a long reference period.
That extrapolated temperature is the temperature index. It is a controlled prediction built on the observation that degradation rate rises with temperature in a fairly regular way, so a short exposure at a high temperature can stand in for a long exposure at a lower one.
Understanding that it is an extrapolation explains most of what follows — why the test takes months, why two products with the same index can behave differently, and why the index is not a promise about your machine.
Table of Contents
How the index is produced
The thermal endurance work follows established principles for electrical insulating materials, with the varnish-specific arrangements set out in IEC 60464-2.
Specimens are prepared on a defined substrate. For the varnish criteria that use a fabric substrate, that is textile glass fabric specified by weave, yarn count, mass per unit area and yarn type. For the wire-based criterion, it is enamelled winding wire of a defined class.
They are aged at more than one elevated temperature. A single temperature gives you a point, not a line — the extrapolation needs several. The number of exposure temperatures and the number of specimens at each are set by the thermal endurance procedure being followed, and IEC 60464-2 recommends a minimum specimen count per exposure temperature.
Exposure runs in periods. The example sequence in IEC 60464-2 doubles — one week, two, four, eight, sixteen, thirty-two. Specimens come out at the end of each period and get measured.
A defined end point marks failure. Not “it looked bad” — a number. For each criterion there is a specific value at which the specimen is considered to have reached the end of its useful life.
The results are plotted and extrapolated. Time to end point against temperature, on the appropriate scale, giving a thermal endurance graph. The index is read off that line at the reference time.
The output is not just a number. A proper report carries the specimen details, the exposure temperatures, the times to end point at each, the plot of the property against log time, the thermal endurance graph, the index itself, and the correlation coefficient of the fit.
That correlation coefficient matters. It tells you how well the data actually fitted the straight line the extrapolation depends on. A weak fit means the extrapolation is standing on soft ground.
The four criteria, and why the pair matters
Here is the part most buyers never ask about.
For hot curing impregnating varnishes, IEC 60464-3-2 requires the temperature index to be established using any two of four criteria, chosen by agreement between supplier and purchaser:
| Criterion | Substrate | End point |
|---|---|---|
| Bond strength | Enamelled wire, to IEC 61033 method B | 22 N |
| Proof voltage | Enamelled wire, class not less than 180 | Per IEC 60172 |
| Breakdown voltage | Glass fabric | 3 kV |
| Loss of mass | Glass fabric | 30 % |
Whichever two are used, the resulting index must meet the minimum for the type.
Now look at what those four measure.
- Bond strength is mechanical — does the varnish still hold the winding together.
- Proof voltage is an electrical endurance check on a wire substrate: the aged specimen is asked to withstand an applied voltage without failing. It is a pass-or-fail withstand at a set level.
- Breakdown voltage is a different electrical criterion on a fabric substrate: the voltage is raised until the specimen punctures, and the measured value is tracked down to the end point. It reports how much margin is left, not just whether a level was survived.
- Loss of mass is chemical — how much of the material has volatilised or decomposed away.
Two products can both be legitimately type 180 and have been qualified on completely different axes. One demonstrated on bond strength and proof voltage has been shown to keep holding the winding and keep withstanding voltage. One demonstrated on breakdown voltage and loss of mass has been shown to keep withstanding voltage and not to lose much material. The first tells you more about a machine that vibrates. The second tells you more about a machine in a hot, still environment.
So the useful question to a supplier is not “what is the temperature index” but “which two criteria was it established on”. That question is often more informative than the headline figure, because it tells you which failure mode the qualification actually represents.
Three numbers that get confused
This is where most of the real-world error lives. Three different things, all temperatures, all in degrees Celsius.
| What it describes | Where it comes from | |
|---|---|---|
| Type number | A varnish product classification | IEC 60464-3-2 — types 130, 155, 180, 200, 220 |
| Temperature index | The extrapolated thermal endurance of a material | Derived by the test above; each type carries a matching minimum |
| Thermal class | The thermal capability of a complete insulation system | IEC 60085, applied to the system in the equipment |
The type number and its minimum index line up — type 155 carries a minimum index of 155, and so on up the range. That part is straightforward.
The trap is the third column. A type 180 varnish does not make a Class H machine. The thermal class applies to the complete insulation system — magnet wire enamel, slot liners, phase separators, impregnant, lead insulation, and the interfaces between them — evaluated as a whole. Another material, or an interface between materials, can limit the system below the capability of any single component in it.
Buying a higher type number does not, by itself, upgrade the thermal class of the machine. The rest of the insulation system still has to support the required class. A higher-performance varnish may well give useful additional margin on particular properties — it just does not move the system class on its own.
Why it is a periodic test, not a batch test
The arithmetic is brutal. Several exposure temperatures, several specimens at each, two criteria in parallel, and exposure periods that in the example sequence run out to around thirty-two weeks. That can be the better part of a year of oven time before a number exists.
So it is treated as a periodic conformance test. It does not need repeating for every batch. It needs repeating when the manufacturer significantly changes the composition or the method of production.
Two practical consequences.
First, when a familiar product suddenly behaves differently in your shop — different gel time, different bond, different smell — asking whether the formulation or the production method has changed is a fair question, and it has a standards-backed reason behind it.
Second, the index on a data sheet may have been established some years ago on a formulation that has since been adjusted within the limits the manufacturer considers non-significant. That is legitimate. It is also worth knowing.
Using the index when you buy
Practical checklist:
Ask which two criteria. Match them to what actually threatens your machine. Vibration and frequent starting point at bond strength. Voltage stress points at proof or breakdown voltage. High steady temperature points at loss of mass.
Ask when the qualification was done, and whether the formulation has changed since.
Ask for the correlation coefficient if the application is demanding. A well-fitted extrapolation and a poorly-fitted one produce the same headline number.
Do not buy above your system. Match the type to the insulation system class you actually have, and spend the difference on process control instead.
Remember what it does not cover. The index is a thermal endurance figure. It says nothing about moisture resistance, chemical resistance, or how well the product penetrates a tight winding. Those are separate properties, and most of them sit on the agree-with-the-supplier list rather than being fixed by the standard.
FAQ
Is the temperature index the maximum operating temperature?
Not directly. It is an extrapolated thermal endurance figure for the material, derived at a reference time. Operating limits for equipment come from the thermal class of the insulation system and the applicable equipment standard, not from a single material’s index.
Why does the same varnish sometimes carry different index values?
Because an index is tied to the specific criterion and end point it was derived from, and to that test result. Ageing a material to a bond strength end point and ageing it to a mass loss end point answer different questions about the same product, so the derived values can differ. This is not a matter of picking whichever figure suits — the criteria are agreed between supplier and purchaser, and the applicable specification sets a minimum the result has to meet whichever pair is chosen. It does mean an index should always be read together with the criterion behind it.
What is the difference between temperature index and thermal class?
The index is a property of a material, derived from thermal endurance testing. The class, under IEC 60085, applies to a complete insulation system evaluated as a whole. A material with a high index can sit inside a system with a lower class.
Does a type 180 varnish make a Class H motor?
No. The insulation system is evaluated as a whole, and the weakest element or interface can limit it. Upgrading only the varnish generally does not upgrade the machine.
How long does temperature index testing take?
Months. The exposure periods alone can run out to around thirty-two weeks in the example sequence, before the measurement and analysis work. This is why it is a periodic conformance test rather than something run per batch.
Can I compare the index of a varnish with the index of a resin?
Cautiously. Varnishes under IEC 60464 and resin based reactive compounds under IEC 60455 are treated differently — reactive compounds are classified on glass transition temperature rather than through the varnish type system. Check what each figure is describing before putting them side by side.