Short answer: IP67 means dust-tight plus 1 m [3.3 ft] of immersion for 30 min under IEC 60529. That test stops liquid water and nothing else. It never looks at water vapor, hose jets, or a second dunk. A hermetically sealed IP67 box holds the humidity that was in the room when you closed the lid — about 18 mg in a 2 L box sealed at 21 C [70 F] and 50% RH — and cycles that water as condensation every cold night. A temperature probe on a cable is happy inside one. A humidity, gas, or pressure element is not, because it can only read the air in the box. Give anything that measures air a breather membrane that carries an IP66 or IP67 rating of its own.
What the two digits actually test
IEC 60529 splits the code into two separate laboratory tests. The first digit 6 means dust-tight: the enclosure sits in a circulating dust chamber and no dust enters at all. The second digit 7 means temporary immersion: the box spends 30 minutes with its lowest point a meter under clean water, and no harmful amount gets in. Both tests run once, at room temperature, on the maker's own sample. The rating is the manufacturer's declaration — our enclosures are manufacturer-rated NEMA 4X-type / IP66 and not UL listed, and NEMA itself tests nothing.
The jet grades are different tests, not harder versions of the same one. Second digit 6 fires a 12.5 mm nozzle at 100 L/min from 2.5-3 m [8-10 ft]; digit 5 fires 6.3 mm at 12.5 L/min. A box marked only IP67 has never met a hose. Because the tests are independent, washdown enclosures often carry both marks: IP66 for the jets, IP67 for the dunk. The full digit table is in the IP rating chart, and the two jet tests are compared in IP65 vs IP66 water jet tests.
What IP67 leaves untested
No vapor test exists anywhere in the standard; IEC 60529 moves liquid water at enclosures and stops there. There is no long-immersion test either — that is IP68, with depth and time fixed by agreement between maker and buyer. There is no corrosion test: NEMA 250 gives Type 4X an extra 200 h salt-spray exposure judged against a 304 stainless specimen, which is why road-salt and coastal jobs specify the NEMA type. And in US installs, NEC 110.28 is explicit that an IP code alone never satisfies a spec written in enclosure types, so order the type with the IP mark beside it. NEMA's own conversion table puts Type 4X at IP66, not IP67; the NEMA 4X vs IP66 comparison lays out that mapping.
Vapor goes where liquid cannot
Water vapor is a gas, and a barrier that holds back a water column does not have to stop single molecules. The mesh over a waterproof phone's speaker passes pressure waves and blocks raindrops; the same physics runs a breather vent. An expanded-PTFE membrane lets vapor through and holds liquid water out, so a vented box tracks outside humidity instead of its own history.
A sealed box forces the same exchange through the gasket, badly. Cool a 2 L box from 21 C to -5 C and the trapped air drops about 9 kPa [1.3 psi] in pressure: 101.3 kPa x 26/294. That presses on a 6 x 4 in [152 x 102 mm] door with roughly 140 N [31 lbf], the gasket flexes, and the box inhales at its weakest point — usually the cable gland. A membrane rated for the job turns that breathing into harmless vapor exchange, and the vent's own marking sets the assembly's rating, because the weakest part always does.
The water you seal in with the lid
Take the snowmelt-style job: a temperature probe on a cable in the sidewalk, radio and battery in a 2 L stainless box on a pedestal. You assemble indoors at 21 C [70 F] and 50% RH. Saturated air at 21 C holds about 18.3 g/m³ of water, so your 0.002 m³ of trapped air carries 0.5 x 18.3 x 0.002 = 18 mg. That air dew-points near 10 C, so from the first clear autumn night the coldest wall collects droplets. On a February night at -5 C the air holds only 3.4 g/m³, or 7 mg in your box; the other 18 - 7 = 11 mg condenses on the lid, re-evaporates when the sun loads the box, and condenses again the next night, forever.
Eleven milligrams is nothing to a temperature logger mounted off the lid; it is enough to fog a viewing window at dawn and to pin a humidity element at 100% RH.
The real filling route is the gland. A gland that lets the box change 5% of its volume a day brings in about 0.5 mg of water daily, with outside air at 5 C and 80% RH holding 5.4 g/m³. Every inhale arrives moist and every exhale leaves drier, because the cold box keeps the difference. Over a 90-day winter that is roughly 50 mg, on top of the 18 mg you sealed in. Fix the entry the way the IP67 project-box condensation article walks through, then drop in a 10 g silica sachet for the rest: it holds about 3 g of water, some 15 years' worth at that leak rate.
Which measurements work inside a sealed box
| Measurement | Sealed box works? | Why | What to do |
|---|---|---|---|
| Temperature, probe on a cable | Yes | The box only shelters radio and battery | Gland the cable, seal the lid dry |
| Temperature, element inside | Partly | Reads the box wall, tens of minutes behind the air | Put the element on the cable for air temperature |
| Relative humidity | No | Reads the box microclimate, 100% RH at a condensing wall | Breather rated IP66/IP67, element near the vent |
| Barometric pressure | No | About 9 kPa of thermal drift swamps the 2-4 kPa day-to-day weather signal | Same breather; it equalizes pressure |
| Gas (CO2, VOC, O2) | No | No airflow reaches the element | Vent through a membrane loop |
The humidity row is where most sensor projects stumble, and the humidity sensor in a sealed enclosure article shows the size of that error. When the element must live outdoors, the outdoor humidity sensor enclosure comparison weighs a membrane against a shielded sensor head. Gas sensing follows the same rule with a twist: you need a flow path, and the gas-sensor venting article shows how to build one that keeps the rating.
Seal it or breathe it: the decision
- Temperature only, outdoors all winter: seal it. Assemble in a dry room, gland the probe cable, add the 10 g sachet, and order 316L when de-icing salt can reach the box — 304 pits under chloride deposits even at room temperature.
- Humidity, pressure, or gas in the reading: breathe it. A membrane breather rated IP66 or IP67 in its own right, mounted high on the upper wall, with the sensing element beside it.
- Washdown line: a NEMA 4X-type / IP66 box with a vent rated for hose-down. A drilled drain hole belongs to the Type 3R world and would pull the whole assembly down to it.
- Flooded pit or ditch, deeper or longer than 1 m for 30 min: IP68 with depth and time in writing. The IP67 vs IP68 battery-project comparison and the IP68 depth and time limits article cover how those numbers get set.
- Batteries in the same box: mount cells off the outer wall. Typical datasheets stop lithium-ion discharge at -20 C, while lithium thionyl chloride primaries run to -55 C, the usual choice for an all-winter logger.
What to put on the order
- Material: 316L wherever salt, washdown chemicals, or standing chlorides reach the box; 304 for dry indoor panels.
- Size: the wall-mount line runs from 6 x 4 x 4 in to 60 x 36 x 16 in (H x W x D) outside, and the stainless size finder filters it by grade and inside dimensions. If you are still matching a box to the environment, start at the enclosure selector.
- Cutouts: order every gland and vent hole cut and sealed at the factory rather than field-drilled; the cutouts and glands guide sets the clearances.
- Vent: a breather carrying its own IP66 or IP67 rating, on the upper wall away from direct spray.
- Glands: one per cable, sized to the cable OD — the seal is made at the gland, not at the box wall.
We build to your drawing in 304 or 316L with every hole you mark, and delivery is 10 business days from your approval of it.
Enclosures for this job
Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 12 × 10 × 6 in | 304 stainless | 12 x 10 x 6 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 16 × 14 × 8 in | 304 stainless | 16 x 14 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 20 × 16 × 8 in | 304 stainless | 20 x 16 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 12 × 10 × 6 in | 316L stainless | 12 x 10 x 6 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Does an IP67 enclosure protect a sensor from rain and hose spray?
Rain, yes; a hose was never tested. Rain and splash are digit 3-4 territory, far gentler than the digit 7 immersion test, so weather alone rarely challenges an intact IP67 seal. Pressure washers and hose-down belong to digit 6: a 12.5 mm nozzle at 100 L/min. For those duties order a box marked IP66 as well as IP67, or a NEMA 4X-type / IP66 enclosure with an equally rated vent.
Why does my sealed IP67 box fog at night when it passed the immersion test?
Because the water was inside when you shut the lid. A 2 L box sealed at 21 C and 50% RH holds about 18 mg of vapor that condenses on any surface below its 10 C dew point. IP67 has no test for this cycle. Assemble in a warm, dry room — sealing at 30% RH halves the inventory to about 9 mg — and add a 10 g silica sachet or a rated breather membrane.
Can I drill a drain hole in the bottom to let condensation out?
Drilling it turns the box into a vented one and drops the assembly to that class. Drainage openings of 3-6 mm [1/8-1/4 in] are a Type 3R practice that NEC 314.15 permits only in boxes listed for damp or wet locations, and the opening sets the rating. Clear water as vapor instead, with a breather membrane rated IP66 or IP67.
Is IP67 the same rating as NEMA 4X?
The two overlap but say different things. NEMA's conversion table puts Type 4X at IP66 — hose-down plus a 200 h salt-spray test — and reserves the IP67 and IP68 pairings for the submersion types 6 and 6P. An IP code reports ingress tests only; a NEMA type adds corrosion checks. On a US spec, order the type and let the IP mark ride alongside.
How do I keep a humidity sensor accurate behind a breather membrane?
Keep the element near the vent and away from anything warm. A membrane exchanges vapor in minutes for a small box, so the sensor tracks outside air. Heat is the enemy: a 5 C warmer enclosure reads about 25 RH points low at 20 C, because warm air at the same moisture content carries a higher saturation value. If the electronics run hot, give the sensor its own vented pocket.
