A quotation that says IPX4 and a specification that says IPX4K describe two different tests with the same number. Put an IPX4K sample on an IPX4 rig and it will be pronounced compliant after a test that delivered roughly an eighth of the water at a quarter of the pressure. That is the most expensive mistake in this corner of ingress protection, and it comes from reading the K as a grade rather than as a different standard.
This article sets out what the K grades actually are, the numbers that separate them from their unlettered neighbours, and the rig changes that a K test forces.
Where the K grades come from
IEC 60529 defines the IP code for enclosures in general. ISO 20653 defines it for road vehicles, and it re-states some of the water digits with harder conditions intended to represent a vehicle in service: a chassis sprayed at close range, a wheel arch hit by a pressure washer, an engine bay cleaned with hot water. The result is a set of grades that share a digit with IEC 60529 and do not share a test.
The letter K marks those re-defined grades. It does not mean the enclosure is better than the unlettered grade, and a product cannot claim IPX4K by testing to IPX4. It can be the other way round: an IPX4K sample that passes will normally pass IPX4, because the harder test subsumes the easier one. The reverse does not hold.
The numbers, side by side
| Grade | Source | Nozzle | Flow | Pressure | Duration |
| IPX4 | IEC 60529 | Oscillating tube, 0.4 mm holes | 0.07 l/min per hole | About 100 kPa | 10 min |
| IPX4K | ISO 20653 | Oscillating tube, 0.8 mm holes | 0.6 l/min per hole | About 400 kPa | 10 min, 5 plus 5 with a 90 degree turn |
| IPX5 | IEC 60529 | 6.3 mm bore | 12.5 l/min | About 30 kPa | 1 min per square metre, at least 3 min |
| IPX6 | IEC 60529 | 12.5 mm bore | 100 l/min | About 100 kPa | 1 min per square metre, at least 3 min |
| IPX6K | ISO 20653 | 6.3 mm bore | 75 l/min | About 1000 kPa | At least 3 min, 2.5 to 3 m |
| IPX9 / IPX9K | IEC 60529 / ISO 20653 | Jet, 100 to 150 mm standoff | 14 to 16 l/min | 8000 to 10000 kPa | 30 s at each of 4 angles, water at 80 degrees C |
Two rows do the most damage. IPX4K raises the per-hole flow from 0.07 to 0.6 l/min, a factor of about 8.6, and raises the pressure fourfold by opening the holes from 0.4 mm to 0.8 mm. On a 400 mm radius tube with 25 holes that is roughly 1.8 l/min for IPX4 against 15 l/min for IPX4K. An IPX4 rig cannot be run harder to reach IPX4K: the holes are the wrong size, and no amount of pump pressure fixes a 0.4 mm orifice.
IPX6K is a different machine, not a different setting
The unlettered IPX6 is a 12.5 mm bore at 100 l/min and about 100 kPa. IPX6K is a 6.3 mm bore at 75 l/min and about 1000 kPa, delivered from 2.5 to 3 m. The flow is lower and the pressure is ten times higher, so the water arrives as a hard, narrow jet rather than a broad wash. Ten bar through a 6.3 mm nozzle is beyond most IPX5/IPX6 rigs on two counts: the pump, and the hose. A 1 inch hose at 10 bar with 75 l/min will move, and a clamped nozzle that was fine at 1 bar will leak or come off.
The practical consequences are a pump rated for continuous duty at 12 bar rather than 2 bar, a pressure gauge on the nozzle side rather than the pump side, and a restraint on the sample. A 75 l/min jet at 10 bar will move a 5 kg sample on a turntable.
The 5 K rule and why it decides the verdict
Both K grades and IPX9K carry a temperature condition: the difference between the water temperature and the temperature of the specimen must not exceed 5 K, and at IPX9 the water itself is at 80 degrees C. The reason is condensation. A cold specimen in warm water forms a film on its inner surfaces, and that film migrates by capillary action to places the jet never reached. A test run with a 20 K difference produces water inside a sealed enclosure that a correctly run test would not, and the report will blame the gasket.
The mirror case is just as bad. A specimen at 80 degrees C sprayed with 20 degrees C water condenses on the outside, and the droplets run into the same seams as the test water. Laboratories that pre-heat or pre-cool a sample to make the water temperature convenient are measuring something other than the grade.
What the pass criterion actually says
For IPX4K and IPX6K the criterion is that water shall not enter in a quantity or in a place that could interfere with the operation of the equipment or impair safety. It is not "no water inside". A vehicle connector that takes on a teaspoon of water in a drainage channel and still operates is compliant. What is not compliant is water reaching a terminal, a PCB edge or a seal face that will later be submerged in its own puddle.
That wording puts the burden on the test house to know where the sensitive places are, which means the enclosure has to be opened after the test and inspected against a drawing rather than shaken out and photographed. A report that records only "no water ingress" for an IPX6K test is a report that has not looked.
Matching the grade to the vehicle position
The grades are positional. IPX4K fits an area that is splashed at close range but not pressure-washed: a lower door seam, a step, a chassis area behind a wheel without a liner. IPX6K fits an area hit by a pressure washer at distance: a wheel arch, an underbody part, an engine bay component. IPX9K fits an area cleaned at close range with a hot washer: a grille, a lamp bezel, an exposed connector on a commercial vehicle.
Choosing IPX6K for a cabin component is not conservative, it is a different test that the part may fail for reasons unrelated to its service. Choosing IPX4 where the customer's process is a hot pressure washer is the expensive direction, and it is the one that shows up as a field return rather than a failed report.
The chambers and nozzle sets for these grades are listed under ingress protection test equipment, and the wider range starts at the product index.
