What Enclosure Ratings Mean for a Luminaire

Why this matters

A specification says IP66 and everybody relaxes, because 66 is a big number and the fixture is going somewhere wet. Two years later the housings hold standing water, the output is down, and nobody can say which promise was broken. The rating was never a promise about the location you put it in. It is a pass on one defined laboratory test, run on a new sample under conditions written into the test method, and heat, corrosion and the customer's actual cleaning method are not in that test at all. This card is mostly about what the rating leaves out, because that is where the failures live.

The rating is a test result, not a grade

An ingress rating is pass or fail against a procedure: water or dust applied at a stated rate, from a stated distance, for a stated duration, then the sample is opened and inspected against a stated acceptance. Nothing about the number is a durability scale.

Two systems are in circulation and they do not convert.

IP, from IEC 60529. Two digits: the first, 0 to 6, is solid-object and dust ingress, the second, 0 to 9, is water. An IP number reaches you as a manufacturer declaration against that standard, and the National Electrical Code does not ask for one.

Enclosure type numbers, from NEMA 250 and the corresponding UL 50E. A type number bundles ingress with things IP does not test at all, including corrosion resistance and, on the outdoor types, external ice formation. This is the family your inspector works in, binding through the listing mark plus whatever the code as adopted in your jurisdiction requires for that location.

For a luminaire there is a third marking that outranks both in a code conversation: the location suitability on the listing, marked under the luminaire product standard UL 1598, binding through the listing mark and through 29 CFR 1910.303(b)(2) for general industry or 29 CFR 1926.403(b)(2) for construction, each of which requires listed or labeled equipment to be installed and used in accordance with the instructions in its listing. NFPA 70 (National Electrical Code) Article 410, in the edition your authority having jurisdiction has adopted, works in wet, damp and dry locations, not in IP digits. A submittal carrying IP66 and no wet-location marking has answered a question the inspector did not ask.

The water digits are not a ladder

This is the single most useful thing on the card. Above the jet tests, the IP water digits stop being cumulative.

A housing marked for immersion has not necessarily passed a jet test, because immersion is static pressure and a jet is momentum against a gasket line. If you need both, the marking has to state both. The close-range, high-pressure, heated wash a wash bay or a food plant actually uses is not any of the digits in the table above. It is the second-digit 9 test, IPX9, carried into IEC 60529 from IPX9K, and a housing either holds it or does not. A housing marked IP66 has not been tested to it, so ask for IPX9 by name. A crew aiming a pressure washer at an IP66 housing is outside the test that number came from.

What no enclosure rating tests

• Heat. No ingress test says anything about the temperature the electronics inside reach. On a sealed housing the two pull against each other.
• Your atmosphere. IP does not test corrosion at all. The enclosure type numbers cover defined corrosive conditions, not the chemistry of your space. Salt spray, pool room air and a plating line each need a materials answer from the manufacturer.
• The optical materials. Lens yellowing and hazing from ultraviolet exposure or chemical attack are permanent losses that cleaning does not reverse, and they sit in the luminaire surface depreciation term rather than in any rating.
• Time. The test is run on a new sample. Gasket compression set, fastener relaxation and a lens that has been off twice for relamping are all real and none are in the number.
• The installed orientation. A housing tested in its design position can pool water in another one. If you tilt it, that is a question for the manufacturer.

Sealing is a thermal decision, not only a water decision

A luminaire's electronics reject heat through the housing. Seal the housing harder and the internal air temperature over the driver and the light-emitting assembly rises. Two things follow and they run the same way.

Output falls as the temperature at the light-emitting assembly rises. That is the ambient temperature term in the light loss factor product, and the light loss factor card owns how that product is built and how a maintained figure re-bases to an initial one; what belongs here is where the number comes from, which is the manufacturer's derating curve for that exact ordering code, read at the temperature the housing actually reaches rather than at the bench condition the published lumens were measured at.

Life falls the same way, so the sealed choice pays twice: less light now, and a steeper lumen maintenance curve later.

The counter-argument is real. A vented housing breathes, and in a dirty space what it breathes is dirt, landing on the inside of the lens where no rag reaches without opening the unit. So the trade is a thermal term against a dirt term, and which one wins is a property of the space, not of the product.

The call: a vehicle wash bay, two candidates

A four-bay wash building. Luminaires pendant-mounted at 13 ft, specification 30 fc maintained, horizontal at 36 in above finished floor, roughly hood height. Twelve luminaires of candidate A cover the layout.

Candidate A. Fully sealed housing, no breather, marked for wet locations on its listing, 12,000 lm from an absolute photometric report of the finished luminaire at the 25 C bench ambient.

Candidate B. Same optic family, listed breather fitted, also marked for wet locations, 11,000 lm at the same 25 C bench ambient.

Put the thermal term in one place only. The manufacturer's derating curve for A, read at the in-housing condition its own thermal data gives for a 32 C bay ambient, gives 0.94. B's curve at the same bay ambient gives 0.98, because the breather lets the housing track the room instead of climbing above it. That factor is the ambient term of the light loss factor, and it does not also come off the published lumens. Taking it in both places is the double-count the light loss factor card names; it would run pessimistic on both candidates and change their ranking not at all.

Light loss factor, per candidate. A: ambient 0.94 x luminaire dirt depreciation 0.90 x lumen maintenance at interval 0.96 = 0.812. B: ambient 0.98 x luminaire dirt depreciation 0.82 x lumen maintenance 0.96 = 0.771. The dirt terms differ because B breathes wash-bay aerosol onto the inside of its lens on the same 12-month cleaning interval, and interior lens dirt is what a wash-down routine cannot recover.

Maintained flux per luminaire. A: 12,000 x 0.812 = 9,744 lm. B: 11,000 x 0.771 = 8,481 lm. A delivers 9,744 / 8,481 = 1.149 times B, so B needs 12 x 1.149 = 13.8, round up to 14 luminaires against A's 12.

Then the gate that actually decides it. The bay is cleaned with a pressure washer and the operators run the wand across the ceiling to knock down soap film. Neither candidate's marking covers that. Close-range high-pressure washing is a different exposure with its own test method, and neither unit carries it. The gate closes on both candidates, and the flux comparison above was answering the wrong question first.

What was done. Two changes, one of them not a product change. The aiming rule went into the customer's written cleaning instructions - wand kept off the luminaire plane, ceiling soap film knocked down with a low-pressure rinse from below and to the side, which removes the exposure rather than trying to survive it - and the specification was reissued asking for a housing marked for the high-pressure test rather than for a higher IP digit. Candidate A stayed the base, on 12 units, because with the exposure removed its thermal penalty is smaller than B's dirt penalty.

Both changes carried a hazard. Water under pressure directed at an energized luminaire is a water-plus-electricity exposure, so the cleaning instruction states the bay lighting circuit is opened at its own disconnect before wash-down and the disconnect stays under the wash operator's control rather than on a wall switch anyone can flip back. Work at the 13 ft mounting height is done from a lift rated for the height with the fall protection the equipment requires under 29 CFR 1910 Subpart D in general industry or 29 CFR 1926 Subpart M in construction. Opening a housing to fit or check a breather is electrical work: de-energize, lock and tag, and prove dead live-dead-live per NFPA 70E-2021, 120.5, in the edition your employer's electrical safety program adopts, under 29 CFR 1910.333(b)(2).

What would flip this. Take the pressure washer out entirely, as on a covered parking deck that is only ever rained on, and the dirt term for a breathered housing collapses toward the sealed one; B's higher ambient factor then wins. Move the same two into a coastal or pool-room atmosphere and neither light loss factor decides it, because corrosion is not in any ingress rating.

Checking your own figures

• The two light loss factors recompute. A: 0.94 x 0.90 x 0.96 = 0.81216, reported as 0.812. B: 0.98 x 0.82 x 0.96 = 0.771456, reported as 0.771.
• The ambient derate appears exactly once each. 0.94 sits inside A's light loss factor and the 12,000 lm stays at its published 25 C basis; 0.98 sits inside B's and the 11,000 lm stays at 25 C. Neither lumen figure was pre-derated.
• The count ratio is derived, not asserted. 9,744 / 8,481 = 1.1489, so 12 x 1.1489 = 13.79, rounded up to 14 because a partial luminaire does not exist and rounding down would land the design under its own maintained target.
• Every illuminance carries plane, height and basis. The only illuminance in this card is 30 fc maintained, horizontal at 36 in above finished floor, and it is never compared against an initial figure.
• The gate closed on both, and the text says so. Two candidates evaluated, two failed the exposure test; the flux comparison is reported as having answered the wrong question first, not as the decision.
• Five items in the does-not-test list, five distinct subjects. Heat, atmosphere and corrosion, optical materials, time, orientation.

References

• IEC 60529, degrees of protection provided by enclosures (the IP code), reaching a product as a manufacturer declaration rather than through the electrical code; NEMA 250 and UL 50E enclosure type designations, binding through the product listing mark and the code as adopted in your jurisdiction
• NFPA 70 (National Electrical Code) Article 410, in the edition your authority having jurisdiction has adopted, for luminaire location classification; UL 1598 wet and damp location marking, binding through the listing
• 29 CFR 1910.303(b)(2) and 29 CFR 1926.403(b)(2), installation and use of listed equipment per its listing; 29 CFR 1910.333(b)(2) with NFPA 70E-2021, 120.5, for de-energized work on the luminaire; 29 CFR 1910 Subpart D and 29 CFR 1926 Subpart M for the work at height
• Manufacturer thermal derating curve and materials data for the exact ordering code, which own the ambient factor and every corrosion question
• See related: What a Light Loss Factor Is Accounting For; How to Read a Luminaire Cut Sheet; What Lumen Depreciation and Lifetime Ratings Mean