IP ratings: what IP65, IP67 and IP69K mean

IP65 and IP67 differ by one digit and by a completely different test. Buyers treat the number as a quality ranking - higher is better - and that is how food plants end up with IP69K sensors that fail in a cold store, and how outdoor cabinets get specified as IP68 with no definition of what "8" means for that product.

The rating comes from IEC 60529 and it has two independent parts.

The two digits

The first digit is solids, the second is liquids. They do not interact: a product can be dust tight and barely splasheproof, or completely open to dust and sealed against immersion.

IP65 is therefore "dust tight, water jets", and IP67 is "dust tight, temporary immersion". Neither is a superset of the other on the liquid side: a jet at 100 kPa can force water past a seal that a static 1 m column of water cannot, because the jet has velocity.

What "8" really means

IPX8 has no fixed test. The manufacturer states the depth and duration, and the rating is only meaningful with that statement attached. A connector rated IPX8 at 2 m for 24 hours and another rated at 30 m for 1 hour both read "IP68" on the label.

That is why IP67 is often the more honest specification: the test conditions are defined by the standard, not by the supplier. If a data sheet says IP68 with no depth and duration, treat it as IP67 at best.

IP69K is not "better IP67"

IP69K comes from ISO 20653 (road vehicles) and describes a specific cleaning test: water at 80 °C, 80-100 bar, from a nozzle 100-150 mm away, from four angles at 0°, 30°, 60° and 90°, for 30 seconds each.

It exists for food and beverage plants that hose equipment down at the end of a shift. A component can pass IP69K and still fail IP67, because the tests stress different things: IP69K is about a seal surviving high-pressure hot water hitting it at speed; IP67 is about a static head of water for 30 minutes.

In practice, most IP69K sensors also carry IP67, and the useful question is not which is higher but which one matches the cleaning regime. A washdown hose at close range is IP69K. A component submerged in a tank is IP67 or IP68. A component in a cold store is neither - it is a condensation problem.

The failure nobody tests for

Seals that hold water out also hold moisture in. An enclosure heats up during the day and cools at night; the air inside contracts and draws moisture through the cable glands and door gasket, and it condenses on the coldest surface - usually the metal backplate and the terminals.

IP65 does not prevent this. It limits how much liquid water gets in during a defined test; it says nothing about vapour transport or about the humidity already inside when the door was closed.

Three things reduce it:

• Pressure compensation elements - a breathable membrane that equalises pressure without passing liquid water.
• Anti-condensation heaters - a few watts of enclosure heater keeping the internal surface above dew point.
• Correct gland sizing - a gland that is one size too large is worse than no gland, because the cable seal is not compressed.

This is the real reason outdoor cabinets fail in autumn and not in summer.

Matching the rating to the environment

For socket outlets and industrial plugs, the relevant ratings are usually IP44 (splashing) and IP67 (immersion), and the rated current matters more than the second digit in most workshops.

A word on connectors

M12 and M8 connectors on sensors are commonly rated IP67, some IP68 or IP69K. The rating applies to a mated pair, correctly tightened. A connector left finger-tight, or a cable end with the nut cracked, is not IP67 - it is an open port. Field-assembled connectors are the usual weak point, and they are worth the extra five seconds with a spanner.

The short version: read both digits separately, ask what the "8" was tested at, and remember that sealing against water and preventing condensation are two different engineering problems.