Cable entries are where most IP failures happen, and orientation is the reason. A gland on the underside of an enclosure sees almost nothing in a drip test and a great deal in a jet test. The same gland on the top face sees the opposite. This is not a gasket problem; it is a geometry problem, and it is decided at the design stage.
What each test actually delivers
The three water tests differ in how the water arrives, and each arrival mode interacts differently with a gland:
| Test | Water arrives as | Worst orientation for a gland |
|---|---|---|
| IPX1 / IPX2 | vertical drips, 1 to 3 mm/min | facing up, exposed thread |
| IPX3 / IPX4 | spray at 60° to 180° | side-facing, near the seam |
| IPX5 / IPX6 | jet at 12.5 to 100 L/min | facing the jet, any angle |
A gland facing up passes IPX2 only if the thread and the gasket seat shed water faster than it accumulates. Add a splash from the side and water runs down the cable and into the gland body — the failure mode changes from "leak at the gasket" to "water tracking along the cable".
Why water follows the cable
A cable is a wick. Water landing on the cable jacket runs down it by gravity, reaches the gland, and enters the enclosure along the gap between jacket and gasket if the gland relies on compression alone. This is why a gland specified as IP68 can still leak in an IPX4 test: the IP68 rating was established with the cable vertical and the gland below, which is the easy case.
Two design responses work. The first is a drip loop: route the cable so that it enters the gland from below, with a low point before the entry. The second is a gland with an internal sealing cone that grips the jacket rather than compressing it flat. Neither works if the gland is on the top face and the loop cannot be formed.
What the test does not tell you
An IPX4 pass on a sample with the gland in its delivered orientation says nothing about the same enclosure installed with the gland rotated. Installers rotate things. A product designed for one orientation and installed in another is a common source of field failures that never appear in the type test.
The practical answer is to test the worst orientation the installation instructions permit, and to write that orientation into the instructions. If the instructions cannot constrain it, the design has to survive any orientation — which usually means the gland goes on the bottom face and stays there.
A checklist that catches most of it
Gland on the bottom face, or a drip loop. Either one removes the tracking path. Both is better.
Gasket compressed by a shoulder, not by torque alone. A gasket that depends on the installer's wrist is a variable, not a seal.
Thread sealed, not just covered. A gland body threaded into a plastic wall passes water along the thread unless a sealing washer is fitted.
Cable jacket within the gland's clamping range. An oversized jacket that the cone cannot grip behaves like an unsealed hole.
These four items are cheap at design time and expensive after a batch has been potted. The test only reports the outcome; the orientation is what decides it.
