IP code
IP code — An IP code is the result of a set of type tests defined in IEC 60529, where the first digit rates protection against solid ingress and the second rates protection against water ingress under a specific prescribed test.
CIP and SIP
CIP and SIP — Clean-in-place circulates cleaning solution through the plant without dismantling it, and steam-in-place raises the whole wetted path to sterilising temperature, so instrumentation sees hot detergent, hot steam and rapid cooling in a repeating cycle.
| Rating | Prescribed test | Failure mode it detects | Failure mode it misses |
|---|---|---|---|
| IPX7 | Immersion in calm water at 1 m for 30 minutes | Leakage driven by a small static head with the part cold | Anything caused by pressure, temperature or thermal cycling |
| IPX9K | Hot water at high pressure swept over the part from four nozzle angles while it rotates | Seal lift and gasket extrusion under jet pressure and heat | Slow ingress through a capillary cable path over days |
| IP66 / IP67 combined | Jet test followed by immersion on the same specimen | Both a jet leak and an immersion leak | Chemical attack, which neither test contains |
| IP69K | Steam-clean jet test, and the enclosure also carries dust-tight rating | Washdown ingress on a sealed enclosure | Crevice corrosion at the joint between two materials |
| EPDM, hot water and steam | good |
|---|---|
| EPDM, mineral oil and grease | poor |
| FKM, caustic at elevated temperature | variable |
| FKM, steam cycles | compression set risk |
| PTFE, chemical range | very wide |
| PTFE, as a static seal | needs spring energy |
- Illustrative share of washdown leaks found at the cable entry, not the housing
- 2out of 3 examples
- Source: Illustrative sample dataset, not a measured value.
If a sensor is rated IP69K, is it automatically IP67 as well?
No, and this is the single most common reading error on a datasheet. The 9K test sprays hot water for a few tens of seconds from outside and never submerges the part, while the 7 test relies on a static head with the assembly cold; a seal designed to resist jet pressure and heat can still pass water along a cold capillary path, so a genuinely washdown-and-submersion application needs both digits quoted, on the same test record.
Does a CIP cycle age a seal by its peak temperature alone?
No. Peak temperature sets the chemical reaction rate in the elastomer, but the damage that shows up as a leak comes from the number of thermal cycles: each heat-and-cool sequence makes the housing and the flange expand by different amounts, so the gasket is alternately compressed and relieved, and that pumping is what drives compression set and lets the seating load decay long before the material is chemically exhausted.
Why does a stainless housing still corrode inside a CIP plant?
Because the cleaning chemistry attacks where the design left a crevice or a low-alloy inclusion. Chloride-bearing detergents pit a passive surface wherever oxygen is starved, which is under a gasket, inside a press-fit joint and in a thread root, and the resulting attack is narrow and deep rather than uniform, so the first visible sign is usually a rust streak at a clamp rather than general dulling.
The parameters on this page, including the illustrative example in the stat block and the elastomer tendencies in the table, are illustrative sample values for a demonstration site. Replace them with your own test data before publishing, and never quote them as a specification.
What the digits are and what they are not
An IP code records which test a specimen passed, not how long the assembly will last. That distinction decides hardware selection, because the digits are neither cumulative nor interchangeable in the way a scale would suggest. Immersion at 1 m and a 100 bar jet are different physics: the first applies a small, steady, cold pressure head over half an hour, and it is a test of whether gaps are long and narrow enough to resist that head; the second applies a jet with enough momentum to force water past a seal lip that would hold under static pressure, at a temperature that softens the seal and opens clearances by differential expansion.
Two further properties of IEC 60529 matter when you read a datasheet. The second digit is not a superset: a higher number does not guarantee the lower tests were run, so an IP69K part is not automatically IPX7 unless the 7 test was also performed. And the water test is performed on a new, clean, dry assembly with a manufacturer-supplied cable or a test gland, which is not the state of a sensor that has been through two years of thermal cycling with a field-terminated cable in it.
The commonly quoted 9K conditions - hot water around 80 degrees Celsius, several tens of bar at the nozzle, a short dwell at each of four jet angles while the specimen turns - describe an aggressive but brief exposure. They do not describe a plant where a sensor sits wet with detergent overnight. Verify those figures against the current edition of the standard before quoting them; treat anything on this page as a prompt to check, not as a citation.
Where washdown water really gets in
Three paths account for almost all washdown ingress, and only one of them is the housing seam.
The cable entry is the first. A moulded cable has an interface between the moulding and the sheath, and a field-entry gland has an interface between the seal and a cable jacket that was cut on site. Water does not need a hole: a capillary gap of a few tens of micrometres will draw water along it, and the same gap opens and closes with every thermal cycle, pumping the cable. This is why a sensor can pass a washdown test with its factory cable and leak in service with a field-terminated one, and why the correct fix is a seal that grips an uncut jacket plus a cable that is not under tension at the entry.
The second path is the vent. A sealed enclosure that is heated and cooled breathes: expansion pushes humid air out and cooling draws it back. If the enclosure has no hydrophobic membrane, the returning air carries condensed water, and repeated cycles collect visible water inside a housing that has never been sprayed. The membrane is a filter, not a hole - it passes gas and blocks the liquid by surface tension - which is exactly why it must never be taped over to “make it more sealed”.
The third path is the process seal, and here the mechanism is load decay rather than material failure. A gasket seals because the clamp load presses it into the microscopic peaks of two faces. Every CIP cycle changes the gap between the flange and the housing body, since they are different parts and sometimes different materials with different expansion coefficients, so the gasket sees alternating over-compression and relief. Over enough cycles it stops recovering its thickness - compression set - and at that point the surface finish and the crevice geometry decide how fast water arrives at the electronics.
Hygienic design is a different requirement than IP
IP ratings ask whether water gets in from outside. Hygienic design asks whether product and cleaning solution can be completely removed from the surface, and it fails on features that improve the IP rating. A deeply pressed gasket pocket traps liquid that the rinse cannot reach. A thread exposed to product holds soil in its root. A crevice whose depth is large relative to its width dries slowly and becomes a biofilm substrate, which is why crevice geometry, not surface polish alone, drives the cleanability argument.
For a sensor this means the mounting decides the hygiene, not the body material. A flush diaphragm mounted in a rounded pocket drains; the same diaphragm stood proud of a flat surface leaves a ledge under which product bakes onto the steel during steam sterilisation. Slope matters more than polish: a face that drains to the process is cleanable, a face that holds a puddle is not, whatever its Ra value.
Choosing for the real cycle
Write down four things before you choose an elastomer: the peak temperature and how long it is held, the number of cycles per week, the chemistry of each step in the sequence, and whether the seal is static or sees movement. EPDM resists hot water, steam and caustic well and performs badly against mineral oils and greases, which makes it the default in dairy-type plants and a poor choice anywhere lubricants migrate. FKM covers a wider chemical range and handles hot oils but suffers in repeated steam and in strongly caustic service, and its long-term behaviour is compression set at the top of its temperature range rather than sudden attack. PTFE is almost inert but has no elastic recovery, so a pure PTFE seal depends on continued clamp load and needs spring energy to survive cooling - a fact that decides whether you can use it at all in a cycling joint.