Measure the enclosed rating against the air in the top third of the box: hang the sensor 2 in [50 mm] below the top wall, keep it 4 in [100 mm] clear of the nearest coil, and log that reading as your ambient. Enclosed ambient limits are limits on air temperature, and the hottest air in any loaded box sits at the top. The worked example below takes a 24 x 20 x 10 in [610 x 508 x 254 mm] stainless box at 35 °C room ambient carrying 60 W of losses: the average air reaches 49 °C and the top third reads higher, 9 K past the example contactor's enclosed limit. Here is what the two datasheet lines grade, where the sensor goes, how to take a reading a test lab can stand behind, and what the box can carry.
What the two datasheet lines grade
The contactor in this article's example is a small IEC motor starter. Its datasheet prints -40 to +60 °C ambient, open, and -40 to +40 °C ambient, enclosed, with the data graded to IEC 60947-4-1, EN 60947-4-1 and VDE 0660. The enclosed figure sits 20 K lower, and that gap is the manufacturer's allowance for the air warming inside a typical enclosure.
IEC 60947-1 defines ambient air temperature as the air surrounding the equipment when it stands in free air, and the air surrounding the enclosure when the equipment is mounted in one. Read literally, that grades an enclosed device against the room, and the datasheet leaves the assumed internal rise unpublished, so the room-air reading alone leaves you designing blind. The way out costs nothing: hold the air inside the box at or below the enclosed limit. Air inside a loaded box always runs above the air around it, so one inside measurement keeps both readings of the definition satisfied, and it is the number a test lab will not argue with because it bounds the room-side reading too.
What is tested and what your box adds
Switchgear temperature-rise data comes out of a free-air test: rated current through the device until temperatures settle, with the rises graded against the standard's reference ambient. The test never sees your enclosure. The enclosed line is a blanket allowance for the device living inside some box, with some layout, beside some other heat sources; the 20 K between the two datasheet lines is the whole of what the manufacturer tells you.
Your panel is its own heat problem. The same contactor runs cooler beside one small relay than under a switching power supply, and the air at the top of the box carries every loss below it. The enclosure heat calculation guide works that rise end to end; the input it cannot supply is the one this article is about — the temperature your components actually live in, which only a measurement gives you.
Where the sensor goes
Grade the hottest bulk air: the top third of the interior, 2 in [50 mm] below the top wall, 4 in [100 mm] or more sideways from the nearest coil, terminal block or power supply. Hang the probe on a thin leash or stand it off a small spacer so air moves past it. Keep it away from the door gasket line and any vent, where air slides along the box skin instead of through the box.
Layout moves the reading. A box with its supply mounted low reads several kelvin cooler at the sensor than the same box with the supply at the top, so fit the big losses low and leave the top third clear; the component layout guide shows how that gradient forms and how to flatten it.
| Sensor position | What it reads | Verdict |
|---|---|---|
| Top third, 2 in [50 mm] below the top wall | Hottest bulk air — the conservative ambient | Use this one |
| On a coil, terminal or supply body | The component's own skin heat | That is the device, not its air |
| Outer wall or door skin | Close to room temperature | Reads far low |
| Bottom third of the box | Coolest bulk air | Flatters the design |
| 4 in [100 mm] off the wall, shaded, mid-height | Room-side ambient reference | Log it alongside |
How to take a reading the file can stand on
Load the panel to full continuous load, close the door, fit every blank, and give the box 2 h before the first logged value; a stainless box of this size reaches steady state in tens of minutes, and 2 h covers it with margin. Call the reading steady when three values 15 min apart stay within 1 K. Log the room ambient at the same timestamps, from a sensor in the shade and away from the box skin — air hugging a hot or sunlit wall reads several kelvin high, and why an outdoor temperature sensor reads high traces that error to its source.
Then write the method into the file: sensor type, position with a sketch or photo, load conditions, room ambient, time to steady state, final values. The lab grading the product works from that page, so the declared position and the measured position must match on paper.
Isolate and lock off the panel before you open it to place a probe.
What a 24 x 20 x 10 in box can take at 35 C ambient
- Enclosed limit: 40 °C, from the example contactor's datasheet, graded to IEC 60947-4-1.
- Room ambient: 35 °C — an assumption for the example; substitute your own worst summer figure.
- Losses: 60 W — an assumption; sum the dissipation lines of the component datasheets, not the load ratings.
- Box: 24 x 20 x 10 in [610 x 508 x 254 mm] stainless, inside the wall-mount range.
- Wall heat transfer: k = 3.7 W/m²K for stainless, the value this site's guides use.
A sealed metal box sheds its heat through the outer skin, so the average air rise above room temperature is the heat load divided by k times the skin area: ΔT = Q ÷ (k x A).
Skin area: A = 2 x (24 x 20 + 24 x 10 + 20 x 10) = 2 x 920 = 1840 in², which is 1.19 m². Rise: ΔT = 60 ÷ (3.7 x 1.19) = 60 ÷ 4.4 = 13.6 K. Average air: 35 + 13.6 = 48.6 °C, and the top third reads above the average. The model is linear: 30 W of losses gives half the rise, 120 W gives double. Painted steel, at k = 5.5 W/m²K, lifts the same 60 W by 9.2 K — 35 + 9.2 = 44.2 °C, still 4 K over the limit at the average.
Decision: the limit binds, so the average alone is 9 K over before stratification. The same line gives the bounds in watts. At 22 W (5 K x 4.4 W/K) the average air reaches the limit exactly, so the load that passes sits below that. Budget 2 K for the climb from average to top-third sensor until your own readings replace it, and the average may rise only 3 K: 3 x 4.4 = 13 W. A bigger box barely moves the verdict at 60 W: the largest in the range, 60 x 36 x 16 in [1524 x 914 x 406 mm], has 4.77 m² of skin, so the rise is 60 ÷ (3.7 x 4.77) = 3.4 K and the average sits at 38.4 °C with 1.6 K of margin — which the climb to the top third spends. At 35 °C ambient, 60 W and a 40 °C enclosed limit, the answer is a cooler, not a box.
When the top air runs past the limit
Work down this list; each step keeps the limit figure intact.
- Cut the losses. At this box's 4.4 W/K, every 10 W you remove drops the average air 2.3 K; the example passes only below 13 W, so most of the load has to go.
- Move the losses down and apart. Heat sources low, top third clear: the sensor sees the gain directly.
- Stir the air. An internal fan evens the gradient so the top-third reading closes on the average, but the walls still carry every watt — see what internal fans do in a sealed box.
- Push air through where the room air is clean. Through-flow holds the box near room temperature, so the component lives in near-room air. A filtered fan drops the assembly to the filter's own rating, so this route belongs indoors, where the washdown seal is not the point.
- Seal it and cool it. Where washdown, dust or fumes set the rating, a closed-loop cooler holds the limit inside with 35 °C outside once its capacity at the working 5 K lift covers the 60 W. Pick between fan, heat exchanger and air conditioner with the cooling comparison, and use closed-loop cooling where the air itself is the problem.
- Keep the box off the sun. A south-facing wall stacks solar load on top of the 60 W; the solar heat load guide puts a number on that rise.
Ordering a box you can measure
Fit the monitoring at order time and the rating survives it. We cut the gland hole where your drawing puts it — low on a side wall, so hose-down water never runs at the fitting — and the cable runs up inside to the probe tied off 2 in [50 mm] below the top wall. Size the entry with the cutouts, glands and windows guide, and rate the gland for the box: a gland below the box's own figure drags the whole assembly down to the gland's rating, the same weakest-part rule that governs drains, vents and windows.
The box itself is manufacturer-rated NEMA 4X / IP66; not UL listed. Choose 304 or 316L stainless and any size from 6 x 4 x 4 in to 60 x 36 x 16 in [152 x 102 x 102 mm to 1524 x 914 x 406 mm]: the stainless size finder filters the range by size and grade, and custom stainless enclosures take the cutout drawing straight from you. Delivered in 10 business days, counted from your approval of the drawing.
Catalog dimensions to compare
Compare these catalog dimensions with your component layout. Open each enclosure for its rating, material and usable space, then check those specifications against your application.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 24 × 20 × 10 in | 304 stainless | 24 x 20 x 10 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 16 × 12 × 8 in | 304 stainless | 16 x 12 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 24 × 20 × 8 in | 304 stainless | 24 x 20 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 30 × 24 × 10 in | 304 stainless | 30 x 24 x 10 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Why measure 2 in below the top wall rather than right at it?
The top layer of air sits against the ceiling metal, which is shedding heat outward, so a probe in the top 1 in [25 mm] reads slightly below the bulk. At 2 in [50 mm] down, convection moves air past the probe and the reading steadies. Keep the same 4 in [100 mm] sideways clearance from coils and supplies: a probe facing a hot body picks up that body's radiant skin instead of the air you are grading.
What sensor should I use for the reading?
A fine Type K bead thermocouple or a small PT100, read to about 1 K — the same tolerance as the steady-state test and finer than the 5 K of margin the limit leaves at a 35 °C room. Small mass matters more than calibration grade, because a heavy probe carries its own heat and lags the air behind a 13.6 K rise by minutes.
Which limit governs when the components disagree?
The lowest enclosed figure in the panel. If the contactor stops at 40 °C enclosed and a neighboring relay allows 50 °C enclosed, the shared air is designed to 40 °C. Most drives and electronic gear carry a 40 °C ambient limit unless their datasheets print more, so that is the usual number to hold, and the worked example is the normal case, not a worst case.
Does opening the door ruin the reading?
Yes. An open door dumps the stratified top air you are grading, and a box running a 13.6 K rise needs the better part of an hour to rebuild it. Place the probe before power-up, keep the door shut for the full 2 h, and log only once three readings 15 min apart agree within 1 K.
Where does the room-side ambient sensor go?
About 4 in [100 mm] off the nearest wall at mid-height, on the shaded side. Air hugging a sunlit or heat-soaked skin reads several kelvin high, and at the limit with a 35 °C room, 2 K of sensor error burns 40 percent of the 5 K margin. An outdoor box adds solar load of its own, so grade it from shade.
