“Best” is the wrong question — and here is the right one
Ask a supplier for the best insulating varnish and you will get a product that is genuinely excellent at something. Whether it is excellent at your job is a separate matter.
A varnish that performs beautifully in a vacuum pressure impregnation tank can be useless in a repair shop with no vacuum plant. A high-temperature silicone can give worse mechanical bonding than a cheaper polyester. A high-solids product with excellent fill can be too viscous to penetrate a tight winding by dipping.
The useful question is not which product is best. It is whether you can write down what you need in a way a supplier can quote against. IEC 60464 gives you exactly that structure, and most people buying varnish have never used it.
Table of Contents
What the IEC 60464 series actually covers
The series has three parts: definitions and general requirements in Part 1, test methods in Part 2, and specifications for individual materials in Part 3. Part 3 is split into sheets — Sheet 1 covers ambient curing finishing varnishes, Sheet 2 covers hot curing impregnating varnishes.
One point worth getting right, because it is widely repeated wrong. The original 1998 text of Part 1 said that all varnishes contain solvent. Amendment 1:2006 deleted that sentence and added water-based and emulsion varnishes to the designation system, along with definitions for emulsion, co-solvent and volatile organic compound content. So the current scope is simply that these varnishes may be used for finishing or impregnating, and may be dried, or dried and cured, at ambient or elevated temperature.
Be aware that some standards-reseller listings still reproduce the deleted sentence in their abstract. Work from the consolidated edition, not the abstract.
The definition in Part 1 is a solution or emulsion of one or more resins in a solvent or carrier liquid. Other components may be present, such as driers, catalysts, reactive diluents, dyestuffs, pigments or co-solvents. Solvents and by-products are released during drying and curing, while the active components polymerize and/or crosslink to form a solid product.
That still separates varnishes from the other family. Resin-based reactive compounds used for electrical insulation are covered by IEC 60455. These are typically solvent-free reactive systems, classified on glass transition temperature, and are treated differently from varnishes under IEC 60464. They are widely used in VPI processes, particularly for solventless resin systems. On the shop floor everybody calls both “varnish”. In a purchase specification, say which one you mean.
The designation system nobody uses
Part 1 sets out a code that describes a varnish in a few characters. It is worth knowing because it forces you to be precise about three separate things.
Application:
| Application | Code |
|---|---|
| Finishing varnish | FV |
| Impregnating varnish | IV |
The definitions matter here. A finishing varnish goes on the surface of equipment to improve resistance to environmental influences or to improve appearance. An impregnating varnish penetrates windings and coils to fill interstices and voids, and so to protect and bond the winding. Those are different jobs. Buying one when you need the other is the most basic specification error there is.
Base resin:
| Resin | Code |
|---|---|
| Acrylic | A |
| Epoxy | EP |
| Melamine-formaldehyde | MF |
| Phenol-formaldehyde | PF |
| Polyurethane | PUR |
| Saturated polyester | SP |
| Silicone | SI |
| Unsaturated polyester | UP |
The designation is based on the composition of the resin content or its major reactive portion.
Varnish type:
| Type | Code |
|---|---|
| Organic solvent based | S |
| Water based | W |
| Emulsion | E |
That third table is the one that reflects the amendment. Water-based and emulsion products are inside the scope, and for those the standard treats VOC content, pH and water content as properties in their own right.
The type number is the question you should be asking
For hot curing impregnating varnishes, Sheet 2 of Part 3 identifies products by a type number that is a temperature in degrees Celsius: 130, 155, 180, 200, 220.
The purchase contract is meant to carry the material designation in the form IEC 60464-3-2 type 130.
That single line does more work than a page of description. It ties the product to a defined set of requirements at a defined temperature, and it gives you something to hold the supplier to.
Note what the type number is and is not. It is the temperature for which the property levels are specified, and Sheet 2 sets a matching minimum temperature index for each type — type 130 carries a minimum index of 130, type 155 a minimum of 155, and so on up the range.
It is not a thermal class for the finished machine. Type 180 is a classification of the varnish, not a declaration that the complete insulation system is Class H. The thermal classification of the equipment or its insulation system must be considered separately, under IEC 60085.
What the standard fixes, and what you have to agree
This is the practical heart of it. Sheet 2 fixes some requirements outright and explicitly leaves others to be agreed between supplier and purchaser. If you do not agree them, you have not specified the product.
Fixed by the standard for hot curing impregnating varnishes:
| Property | Requirement |
|---|---|
| Viscosity | Within ±10 % of the nominal value, which must be stated in the purchase contract |
| Non-volatile content | Within ±2 % of the nominal value, also stated in the contract |
| Stability in an open vessel | Viscosity increase limited to a defined multiple of the nominal value |
| Curing in a thick layer | Smooth top, non-tacky underside, interior no worse than rubber-like with at most five voids, uniform throughout |
| Effect on enamelled wire | Not less than pencil hardness H |
| Bend test | No cracks visible after bending around a 3 mm mandrel |
| Resistance to solvent vapour | No change in adherence, no peeling, blistering, draining or tackiness |
| Volume resistivity | A minimum value before water immersion and a lower minimum after it |
Two of those deserve comment.
The viscosity and non-volatile content requirements are tolerances around a nominal value you and the supplier write into the contract. The standard does not tell you what the viscosity should be. It tells you the delivered product must stay within ±10 % of whatever you agreed. If the purchase contract does not state the nominal value, the ±10 % and ±2 % requirements have no useful contractual reference point.
The solvent vapour requirement carries a note limiting it to countries where regulation requires it for increased safety “e” type equipment. Do not specify it reflexively.
Left to agreement, when required:
Before drying and curing — density, dilution ability, and for water-based and emulsion products, pH, water content and volatile organic content.
After curing — tackiness, bond strength, cupping test, resistance to liquids including water, dissipation factor and relative permittivity, breakdown voltage and electric strength, resistance to mould growth, and for water-based and emulsion products, flash rusting of steel panels.
Look at that second list again. Bond strength, breakdown voltage, electric strength, dissipation factor — the properties most people assume are guaranteed by the standard are on the agree-with-the-supplier list. If bond strength matters to your machine, and for anything with vibration or frequent starting it does, it goes in the contract or you do not get it.
Flash point sits in the same category, and matters enough for solvent-based products to be worth stating explicitly:
| Property | When specified |
|---|---|
| Flash point | Agreed between supplier and purchaser, subject to applicable national safety requirements |
Where national safety regulation sets a minimum flash point for the application, that minimum applies regardless of what the parties agree.
Temperature index: two criteria out of four
Sheet 2’s temperature index requirement is worth understanding, because it explains why two products with the same type number can behave differently.
The temperature index is determined against any two of four criteria, chosen by agreement between supplier and purchaser:
- Bond strength, using the IEC 61033 method B test, with a defined end-point in newtons
- Proof voltage, per IEC 60172, on an enamelled wire substrate of class not less than 180 to IEC 60317-8 or IEC 60317-13
- Breakdown voltage, on a glass fabric substrate, with a defined kV end-point
- Loss of mass, on a glass fabric substrate, with a defined percentage end-point
Whichever two are chosen, the result must meet the minimum temperature index for the type.
So the practical question when comparing two type 180 products is which two criteria were used. A product qualified on bond strength and proof voltage has been demonstrated on a different axis than one qualified on breakdown voltage and mass loss. Ask.
The type number therefore does not describe every performance characteristic of the varnish. It establishes a minimum thermal-endurance level according to the selected criteria, and nothing more.
It is a periodic conformance test, and it does not need repeating unless the manufacturer significantly changes the composition or the production method. That is also worth asking about if a familiar product suddenly behaves differently.
Reading a data sheet: the test basis behind the numbers
Part 2 of the series holds the test methods, and it is worth a look before you compare two suppliers’ figures. Numbers only mean the same thing if they were produced the same way.
Reference conditions. Unless a specification says otherwise, tests run at around 25 °C with relative humidity between 45 % and 70 %, with the sample pre-conditioned until stable. Viscosity is a tighter case: measured at 23 °C, to a fraction of a degree, using either a rotational method or an efflux flow cup. Two suppliers quoting viscosity by different methods are not quoting the same quantity.
One precedence rule worth knowing. Where Part 2 and a Part 3 specification sheet disagree, the specification sheet wins.
The thick-layer cure result, decoded. Sheet 2 asks for a result of S 1, U 1, no worse than I 4.2, uniform. That looks like gibberish until you have Part 2 open. The specimen is roughly 4 mm thick, cured in a small aluminium foil mould, then assessed on three axes:
| Code | Meaning |
|---|---|
| S 1 / S 2 | Top side smooth / wrinkled |
| U 1 / U 2 | Underside non-tacky / tacky |
| I 1 to I 6 | Interior: rigid, horny machinable, leather-like, rubber-like, gel-like, liquid |
| .1 / .2 / .3 | No voids / not more than five voids / more than five voids |
Read the interior code as two separate things. I 4.2 means an interior condition of category 4, rubber-like, with the .2 void classification corresponding to not more than five voids. The 4 is the condition; the 2 is the void count.
So the Sheet 2 requirement reads: smooth on top, not tacky underneath, interior no softer than rubber-like, not more than five voids, uniform through the section. That is a plain-language statement about whether the product cures properly in bulk — exactly what you care about at the bottom of a slot.
Stability in an open vessel is measured after four days at 50 °C, topping up solvent daily to replace what evaporates. That is a shop-realistic test of what happens to an open tank in a warm workshop, which is why the requirement is framed as a viscosity multiple rather than an absolute.
Effect on enamelled wire is assessed as pencil hardness of the wire coating after treatment, on straight pieces of wire. This is your compatibility check, and it is why the requirement exists at all.
Solvent vapour resistance exposes specimens for seven days to vapour from a named set of solvents — acetone, xylene, hexane, methanol and carbon disulphide. If your machine lives near a specific chemical that is not on that list, the standard test does not cover you and you need a separate agreement.
Volume resistivity is measured before and after seven days in demineralised water, at a field strength of 1 000 V/mm, with the reading taken at a fixed interval after electrification. The before-and-after pair is the point: it is a moisture-resistance test dressed as a resistivity measurement.
Breakdown voltage and electric strength use a ball-to-plate arrangement under a dielectric fluid, with the voltage raised at a controlled rate. Part 2 notes that ball-to-plate reads slightly higher than a plate-to-plate arrangement, so a figure obtained one way is not directly comparable with the other.
Temperature index testing follows thermal endurance principles, with exposure periods running out to many weeks and several specimens per exposure temperature. It is a slow, expensive test. That is why it is a periodic conformance test rather than a batch check, and why a supplier’s willingness to state which criteria were used tells you something.
One aside from Part 2 that catches people in a lab: copper panels are rejected as substrates for unsaturated polyester based impregnating resins, because copper can act as an inhibitor or accelerator. Steel is the standard substrate.
The application method narrows the field faster than the chemistry
<cite index=”7-1″>The common workshop methods are dip and bake, trickle impregnation, vacuum pressure impregnation, and sealed-winding processes.</cite> They do not accept the same products.
Dip and bake. Winding into the tank, then into the oven. Wants viscosity low enough to penetrate and a cure schedule that suits your oven. <cite index=”7-1″>A new winding is typically dipped twice, and there is a common shop error here — dipping again while the winding is still hot from the first bake, so the varnish thins and runs off instead of staying put.</cite>
Trickle impregnation. Resin fed onto a slowly rotating, preheated winding. Good fill, less waste, no large tank. Needs a product formulated for the process, with the right viscosity and gel behaviour at trickle temperature.
Vacuum pressure impregnation. Evacuate, flood, pressurise. <cite index=”7-1″>Typically used on medium-voltage machines and form-coil systems, because conventional methods do not fully saturate those coils and their insulation tapes.</cite> Many modern VPI systems use solventless reactive resin systems covered by IEC 60455 rather than solvent-containing varnishes under IEC 60464. The absence of solvent matters because the VPI process is built around resin impregnation and subsequent curing, not around solvent evaporation.
Brush and spray. Field repair. Usually an ambient curing finishing varnish, chosen because you have no oven and no tank. Penetration and bond will be lower than a proper impregnation. Treat it as a repair.
Chemistry families, as orientation only
Formulations vary widely within each family, and the manufacturer’s data sheet governs.
| Family | Generally strong at | Watch out for |
|---|---|---|
| Alkyd and modified alkyd | Cost, handling, penetration | Thermal and chemical capability |
| Saturated and unsaturated polyester | Bond strength, thermal capability, cost balance | Some chemical environments |
| Polyester-imide | Higher thermal capability with good bond | Cost |
| Epoxy | Moisture and chemical resistance, adhesion, mechanical strength | Pot life and process discipline |
| Silicone | High temperature service | Bond strength, and silicone contamination of nearby painting or bonding operations |
| Polyurethane | Flexibility, impact and moisture resistance | Thermal capability relative to imides |
A specification that actually works
Putting the above together, a defensible purchase line looks like this:
- The designation — for example, hot curing impregnating varnish to
IEC 60464-3-2 type 155, plus the resin and type codes if you need to constrain the chemistry. - Nominal viscosity and nominal non-volatile content, so the ±10 % and ±2 % tolerances have something to bite on. State the viscosity method too — rotational or flow cup — or you are comparing quotes on different bases.
- The agreed properties from Table 2 that matter for your machine — bond strength at minimum, plus breakdown voltage and electric strength for anything above LV.
- Which two temperature index criteria the type qualification is based on.
- Flash point, where your national safety regulation sets one.
- Application method and cure schedule your shop can actually run.
- Container size and shelf life, with storage temperature and final date for use — Part 1 requires that marking on the container, so it is fair to ask for it up front.
Written out as a specification line, that becomes something like:
Hot-curing impregnating varnish, IEC 60464-3-2 Type 155, resin type [specify], varnish type [S / W / E], viscosity [nominal value and test method], non-volatile content [nominal value], application method [dip / trickle / other], curing schedule [temperature and time], plus the agreed electrical and mechanical properties listed below.
That is a document a supplier can quote against and a document you can hold them to. “Use a good Class F varnish” is neither.
What goes wrong most often
Buying a class you do not have. A type 180 varnish in a system whose slot liner is Class F does not give you a Class H machine. Buy for the system.
Assuming the standard guarantees bond strength. It does not. It is on the agreed-properties list.
No nominal values in the contract. Then the viscosity and non-volatile tolerances are meaningless, and batch-to-batch variation is your problem.
Ignoring the cure schedule. Undercured varnish stays soft, retains solvent, and gives poor bond and poor electrical performance.
Skipping the compatibility check. Sheet 2 sets a minimum for the effect of the varnish on enamelled wire for a reason. On a rewind where the wire, liners, tapes and varnish come from different suppliers, check it.
Not testing after. Insulation resistance and polarization index can be useful checks before and after impregnation, but they should be read alongside the curing process, visual inspection and the other applicable winding tests rather than on their own. A winding that reads worse afterwards is worth investigating.
FAQ
Which insulating varnish is objectively the best?
There is no single answer. Fix the application (finishing or impregnating), the type number, the process, and the environment, and the field narrows to a handful of products that are all defensible.
Do all varnishes contain solvent?
Not under the current text of IEC 60464-1. That statement was in the 1998 edition and was removed by Amendment 1, which also brought water-based and emulsion varnishes into the type designations. The standard now defines a varnish as a solution or emulsion of resins in a solvent or carrier liquid.
What is the difference between varnish and impregnating resin?
Varnishes fall under the IEC 60464 series. Resin based reactive compounds — solvent-free, classified on glass transition temperature — fall under IEC 60455. Both are called varnish in the workshop, which is why written specifications should say which is meant.
What does “IEC 60464-3-2 type 155” mean?
A hot curing impregnating varnish whose property levels are specified at 155 °C, meeting the minimum temperature index for that type. It is a product designation, not a thermal class for the machine.
What does a result like “S 1 – U 1 – I 3.2 – uniform” mean on a report?
It is the thick-layer cure assessment from Part 2. Smooth top side, non-tacky underside, leather-like interior containing not more than five voids, uniform through the section. The letter codes describe top, underside and interior; the number after the interior code counts voids.
Does varnishing improve insulation resistance?
A properly cured impregnation can improve insulation resistance by reducing moisture ingress and closing conductive paths through voids and contamination. The result depends strongly on winding condition, moisture content, varnish compatibility and the curing process. A decrease after varnishing should be investigated rather than automatically blamed on the varnish.
Is an ambient curing varnish acceptable for a full rewind?
Part 3 splits ambient curing finishing varnishes and hot curing impregnating varnishes into separate sheets precisely because they are specified against different requirements. For a complete rewind, select the impregnation system according to the winding design and the manufacturing process. An ambient-curing finishing varnish is generally not equivalent to a hot-curing impregnating varnish or a solventless VPI resin.