IP65 for solenoids: what the rating actually requires, and what it does not cover

Why this happens

IP65 is two separate tests, and reading the code properly is the start of getting the specification right.

The first digit covers solids. A 6 means dust tight: the enclosure is tested in a dust chamber containing talcum powder with a partial vacuum inside, for up to eight hours, and no dust enters. Below that, a 5 permits a small amount of dust but not enough to interfere with operation.

The second digit covers water, and the steps are not evenly spaced. IPX4 is splashing from an oscillating tube or spray nozzle. IPX5 is a jet from a 6.3 mm nozzle at 12.5 litres per minute, directed from any direction at a distance of 3 m for at least three minutes. IPX6 is a jet from a 12.5 mm nozzle at 100 litres per minute, eight times the flow. IPX7 is immersion at 1 m for 30 minutes, and IPX8 is continuous immersion under conditions agreed between buyer and supplier. A unit rated IP65 has not been tested against IPX6 and will not necessarily survive it.

What the test does not cover is the longer list. There is no requirement in the standard for condensation, for steam, for washdown with hot or caustic water, or for the device still meeting its force and timing after exposure. The test checks that water does not get in. It does not check that the actuator still works, and on a solenoid that distinction matters more than it does on a junction box.

A solenoid has two ingress problems that a static enclosure does not. It has a moving shaft, which cannot be sealed with a static seal, and it has a coil that heats and cools, which makes the enclosure breathe. Both are addressed below, and neither is solved by thickening the wall.

The breathing is arithmetic rather than opinion. Take a 5 cm³ cavity sealed at 20 °C and heated to 90 °C by the coil. At constant pressure it would have to expel 5 × (363/293 − 1), which is about 1.2 cm³ of air. Sealed, it does not expel anything, so the pressure rises by the same ratio: 1.24 times, or 0.24 bar gauge. That pressure pushes outward past every seal on the way up. On cooling it draws the same 1.2 cm³ back in, and the air it draws in carries whatever humidity the surroundings have.

Condensation then follows from the same heat cycle. A 5 cm³ cavity holding air at 90 °C and 50 % relative humidity carries about 1 mg of water vapour. Cooled to 20 °C, only a small fraction of that can stay airborne, so roughly 1 mg condenses inside the unit. As an absolute quantity it is trivial. Spread over a guided surface with a 20 μm clearance it is a film that will not drain, which is why a device can pass an IP65 test on the bench and corrode in service. The chemistry of that attack is covered in corrosion in humid environments.

Check these in order

1. The lead-wire entry. This is the most common leak path, and the gland usually is not the cause. Water wicks along the strands of a stranded conductor by capillary action and travels past a gland that is sealing correctly against the jacket. Sealing the entry means sealing the strands, which is potting, a sealed splice, or a solid conductor through the wall. A compression gland alone does not do it.

2. The breather. A sealed cavity has to breathe, so decide how. A hydrophobic membrane vent passes air and holds back liquid water, typically passing hundreds of millilitres per minute per square centimetre while resisting more than 0.5 bar of water pressure. Without one, the 0.24 bar above will find its own path and will use whichever seal is weakest.

3. The plunger joint. A moving interface cannot be sealed with a static seal, so there are two workable routes. A rolling diaphragm or an isolation sleeve keeps the plunger out of the fluid entirely, which is what an isolated solenoid valve does. Alternatively an O-ring runs on the plunger inside a bore that is itself vented. An O-ring in a closed cavity does not seal; it pumps.

4. Shell and crimp joints. On a tubular design the joint between the shell and the end cap is a seam, and it is also where a thin wall meets a hard crimp. Thermal cycling works on that combination, so it belongs on the list of places to look when a unit leaks after a few hundred cycles rather than immediately.

5. The cable jacket. The jacket is part of the sealing system. PVC stiffens below about −10 °C and can crack at the gland, while PUR stays flexible to around −40 °C and silicone further down. A jacket that cracks at the entry defeats every other measure, and this is the first thing to check for equipment that fails in its second winter, as covered in agricultural equipment.

6. Confirm whether the failure is ingress or condensation. They present identically, and they have opposite fixes. Ingress is solved by sealing. Condensation is made worse by sealing without a vent, because the water is already inside in vapour form and cannot leave.

7. Ask what the actual fluid is, not just the IP number. Clean water at 20 °C, a dairy caustic wash at 80 °C and coffee machine condensate are three different problems that one rating does not describe.

What actually to change

Symptom	What to change	Why not the obvious thing
Water in the winding, gland intact	Seal the conductor strands rather than the entry	A tighter gland cannot stop capillary travel inside the wire
Water inside a fully sealed unit	Add a membrane vent	More sealing traps vapour that is already inside
Leaks after a few hundred cycles	Check the shell joint and the crimp	A static seal does not fail on a cycle count, a seam does
Passes IP65, fails in the wash	Specify IPX6 or IP69K, and state the fluid temperature	IP65 uses ambient water at a fixed jet size
Force lower than the unsealed version	Measure friction on the sealed plunger	The loss is friction, so chasing it in the magnetics wastes time
Stroke slower than specified	Check whether the plunger bore is acting as a dashpot	A sealed cavity has to move air through a restricted path
Unit hot after potting	Check for voids rather than blaming the compound	A properly wetted epoxy conducts far better than the air it replaces

On that last point the direction is usually the opposite of what people expect. A typical unfilled epoxy runs about 0.2 to 0.4 W/m·K against roughly 0.03 W/m·K for air, so filling the gaps between turns improves the winding-to-shell path and a potted coil normally settles cooler at the same power. Where it goes wrong is incomplete wetting, which leaves voids that are invisible once the unit is sealed, and the CTE mismatch between compound and enamel over thermal cycles.

When it IS the harder problem

The real environment is a hot or caustic washdown. Then the IP code is the wrong instrument and IP69K is the right one. IP69K applies water at 80 °C and 80 to 100 bar from a 12.5 mm nozzle at 100 to 150 mm, from four angles. That is written for food and dairy equipment, and specifying it changes the materials rather than the seals: FKM or EPDM rather than NBR, a coated or stainless shell, and a cable jacket that survives the temperature.

The failure mechanism is condensation rather than ingress. Then more sealing makes it worse, and the fix is a vent plus a drain path plus, sometimes, a deliberate desiccant or a conformal coating on the winding. This is the case where a lower IP rating on the outside and a better breathing strategy inside outperforms a sealed unit.

The unit is already inside a sealed panel. Paying for IP65 twice is common and avoidable. If the enclosure is rated and drained, the solenoid inside it needs corrosion protection rather than its own ingress rating.

The damage is chemical rather than hydraulic. A unit that stays dry can still fail from solvent vapour, from condensation of a process fluid, or from galvanic action between dissimilar materials. Then the specification work is material compatibility, and the surface treatment question is the one that decides service life, as set out in plunger material and surface treatment.

The requirement is written into a customer drawing as a bare IP65. That is the most common case and the least useful one. The productive move is to ask what the unit will actually be exposed to, and to write the answer down as fluid, concentration, temperature, pressure and frequency of wash. An IP number is a test result. The list of exposures is a requirement, and only the second one can be designed against.