Marking is the only part of a plug or socket that tells the user what the product can do. If it can be removed with a thumb and a solvent, then legally and practically the product has no rating, no model number and no manufacturer.
The rub test is short: fifteen seconds with water, fifteen seconds with petroleum spirit, on the same sample, one marking after the other. It is also the cheapest test in the whole type test, which is why it is the one most often done from memory rather than from the clause.
Which markings are covered
Not every printed surface is a marking. The clause applies to the markings that carry a safety or identification claim.
| Marking | Usually required | Rub test applies |
| Rated current and voltage | Yes, on the body | Yes |
| Manufacturer or trade mark | Yes | Yes |
| Model or type reference | Yes | Yes |
| Terminal identification (L, N, earth) | Yes where accessible | Yes |
| IP rating where claimed | Only if claimed | Yes |
| Date code, batch | Not always | Only if it carries the claim |
| Decorative printing, instructions on a sticker | No | No |
The distinction matters because a product can pass the test and still be non-compliant: the marking that faded was the one nobody tested. Sample the marking that carries the highest current claim, since that is the one the test is protecting.
Doing it the same way twice
- Cloth, not paper. The specification is a cloth, and a paper towel abrades differently.
- Pressure by hand, not by jig. A jig either presses too hard or too lightly, and both change the result.
- Fifteen seconds each, both media. Water first, spirit second, on the same area.
- Judge legibility, not contrast. A marking that is still readable but blurred has passed; one that is readable only because you know what it said has not.
- Photograph before and after. The judgement is subjective, and the photograph is what makes it reviewable.
Why moulded markings beat printed ones
A marking moulded into the tool is part of the surface, so the rub test can only fail it by polishing the edge of the character. A printed marking depends on adhesion, and adhesion depends on the flame treatment of the plastic, the ink and the cure. Two mouldings from the same tool can behave differently if one was demoulded cold.
Where the design has to print, the practical defence is a shallow recess: the marking sits 0.2 mm below the surface and the cloth cannot reach it with full pressure. For the mechanical checks that follow the rub test on the same sample, see our plug and socket compliance equipment, and the clause order in the technical notes.
Standards referenced: IEC 60884-1 clause 7 (marking) and clause 7.3 (durability), IEC 60529 clause 5 for IP marking claims.
Cost of getting it wrong
A marking failure is a paperwork failure with a physical cause. A batch of 5,000 pieces with a 16 A marking that wipes off is 5,000 pieces that cannot be shipped with that rating, whatever their electrical performance. The ink costs a fraction of a cent per piece; the rework costs 30 minutes per 100 pieces.
- Ask the moulder for the surface energy value, typically above 38 mN/m for a printable polycarbonate.
- Test the first 3 shots of every tool, not the first 3 of the first tool.
- Keep the rub cloth and the spirit bottle labelled with the date they were opened.
Geometry helps more than ink. A 0.2 mm deep recess holds the cloth off the characters at 15 s and still leaves them legible at an arm length of 500 mm; a 0.05 mm recess is a printing surface with no protection at all. On a tool running 1,200 shots a day, the recess wears 0.01 mm per 100,000 shots, so a 0.2 mm depth lasts about 2 million pieces.
