An IP rating is a test result, not a service life. A waterproof cob led strip that passes IP65 survived a defined water jet for a defined number of minutes, in a laboratory, at room temperature, when the sample was new. Nothing in that test describes what the product does after two summers on a south-facing facade, or after a thousand cycles of warming and cooling inside a sealed silicone tube.
That gap is where most outdoor and wet-area installations fail: the specification was met and the run still failed, because the failure mode was not the one the test measures. This guide explains what IP65, IP67 and IP68 each certify, what sealing a COB LED Strip costs thermally and optically, and which questions decide whether a sealed run survives its location.
What an IP Number Actually Certifies
IP codes are defined by IEC 60529. The first digit is solid-particle ingress, the second is water ingress, and the two are independent — a 6 in the first position tells you nothing about the second. For lighting the second digit is the one under discussion, and each level is a specific test of specific duration:
| Rating | The test that was passed | What it does not cover |
|---|---|---|
| IP65 | Water jet from a 6.3 mm nozzle, all directions, roughly 12.5 litres per minute, about 3 minutes | Standing water, submersion, pressure |
| IP67 | Immersion in 1 m of water for 30 minutes | Continuous immersion, chlorinated or salt water, thermal cycling while wet |
| IP68 | Immersion deeper or longer than IP67, at conditions the manufacturer declares | Nothing standard — no fixed depth, so the declared condition is the spec |
Read the IP68 row again, because it is the most misused rating here. IEC 60529 does not fix a depth or duration for IP68; it requires the manufacturer to state them. An IP68 claim with neither declared is not stronger than IP67 — it is unspecified.
Two consequences follow. IP67’s 30 minutes certify survival of temporary submersion — a flooded planter, a washdown, a drain that backs up — not permanent underwater service. And the step from IP65 to IP67 is not a small upgrade of the same product; it is usually a different sealing method, which changes the thermal behaviour discussed next.
The Thermal Cost of Sealing a COB Strip
A COB strip is a continuous phosphor line running continuous current. Unlike an SMD strip, which has gaps between packages where the PCB sheds heat, a COB emitter generates heat along its whole length, so the circuit and its mounting surface are the entire heat path.
Encapsulation interrupts that path. Silicone has a thermal conductivity around 0.2 W/m·K against roughly 400 W/m·K for the copper in the circuit. A sealing layer over the emitting face means heat that used to leave through the top must now leave through the bottom, through the adhesive, into the mounting surface — the direction it was already travelling. It gets no better route, only a blanket over the alternative.
The result is a higher junction temperature at the same drive current, which accelerates lumen depreciation and shifts colour. IES LM-80 data only describes that rate at the case temperatures actually tested, so a sealed strip running hotter than the tested range has no published depreciation curve at all.
This is why the same circuit is normally derated when sealed, and why a sealed run belongs in an aluminium profile even when the sealing appears to make one unnecessary. Here the profile is not weather protection — it is the heat sink replacing the one you covered. In practice, if an open build is specified at 14 W/m, the sealed equivalent is either rated lower or rated the same with a maximum ambient temperature attached. Both are honest. A sealed strip quoted at the open wattage with no ambient limit and no profile requirement is the one to question.
How Sealed Runs Really Fail
Have an LED Lighting Project?
Share your project requirements — our engineering team will configure the complete LED strip system (strips + profiles + controllers + power supplies) and send you a detailed proposal within 24 hours.
Water rarely enters through the middle of a sealed strip. It enters slowly, at the three points the seal is interrupted.
The cut end
Every cut opens the sealing envelope. Field-sealing it with an end cap and adhesive is a manual operation whose quality varies with the installer, the temperature on the day, and the cure time the schedule allowed. Both ends of a site-cut run need the same treatment even if only one sits near visible water.
The cable entry
The wire must pass through the envelope, and the wire is stranded. Water reaching the strands travels inside the insulation by capillary action, so moisture can enter at a dry-looking junction box metres away and arrive inside the strip. Sealing the strip end while leaving the cable route open protects the wrong end of one path.
Condensation from inside
This is the failure that surprises people, because no water ever got in. A sealed strip that warms when switched on and cools when switched off breathes: the air inside expands and pushes out through microscopic gaps, then contracts and draws ambient air back in with its humidity. Over hundreds of cycles that moisture accumulates, condenses on the coolest inner surface and corrodes the solder pads — inside an envelope that passed its ingress test.
Nothing in IP65, IP67 or IP68 tests for this, because all three are short tests on a thermally stable sample. It is controlled by construction quality and by not oversizing the sealed air volume. Which is the conclusion for a project buyer: the rating filters candidates, the construction decides outcomes, and two IP67 products can have entirely different service lives in one location.
Matching the Rating to the Actual Location
Specify from the water that will reach the strip, not from the room’s label. A bathroom cove and a shower niche are one room and two requirements.
| Location | Water it actually sees | Minimum sensible rating |
|---|---|---|
| Kitchen under-cabinet, splash zone | Occasional splash, wiped surfaces, steam | IP65 |
| Bathroom cove, above the door head | Humidity and steam, no direct water | IP65 |
| Shower niche, wet room, spa | Direct spray, standing water on ledges, daily cycling | IP67 |
| Covered soffit or canopy | Wind-driven rain, dust, wide ambient swing | IP65, UV-stable sealing |
| Exposed facade, parapet, handrail | Driving rain, pooling in the channel, freeze-thaw | IP67 |
| Ground level, planter, pathway edge | Irrigation, temporary flooding, grit abrasion | IP67, plus drainage in the detail |
| Marine deck, pool surround, salt air | Chloride attack on solder and copper | IP67 up, plus a declared corrosion approach |
| Continuously submerged | Permanent immersion | Not a strip application — use a declared fixture |
Two rows deserve emphasis. Ground-level and channel installations fail from pooling, so if the extrusion holds water the rating must cover immersion however the location is described. And salt is a chemical problem, not an ingress one: an IP67 seal keeps water out and does not stop chloride corrosion at a compromised cut end. Marine work such as the planned yacht LED strip waterproof COB reference sits in that second category.
What Sealing Does to the Light
The reason to choose COB outdoors is the continuous line — no visible dots, no scalloping on a nearby surface. Encapsulation interacts with that in two ways worth knowing before approving a sample.
The sealing layer adds thickness above the emitter, diffusing the line slightly further — neutral or mildly helpful on a COB source, since the emitter was already continuous. The second effect is less welcome: silicone and comparable materials can yellow under sustained ultraviolet exposure, which shifts output toward the red end while reducing total lumens. On a 4000K run that shift is more visible than on a warm one, because the eye reads a warm drift against a neutral reference more readily than warm-on-warm.
So UV stability is a facade specification, not a general waterproofing one. A sealed strip for a shaded soffit and one for a south elevation face very different exposure; asking which sealing material is used, and whether it is specified as UV-stable, separates them. Colour consistency over life matters most where the run is long and continuous — the case in facade work such as the planned outdoor waterproof COB LED strip facade reference. Where the project needs high colour fidelity as well as sealing, the two requirements compound: check the high CRI COB LED strip guidance for what the CRI figure does and does not control, then apply the sealing decision on top.
Voltage and Safety in Wet Locations
Low-voltage DC COB strips run as SELV circuits under IEC 61140, which is much of why they are acceptable in wet areas. Two things follow outdoors.
The driver becomes the weather-exposed component. A strip may be IP67 and the driver enclosure IP20, in which case the assembly is IP20. The driver either goes indoors with a longer DC run to the strip, or outdoors in an enclosure rated for the location — and an outdoor-rated driver still has to breathe, or it accumulates the condensation described earlier.
That longer DC run brings voltage drop into the calculation. Current at a given power is inversely proportional to voltage: a 10 W/m load draws 0.83 A/m at 12V and 0.42 A/m at 24V, so the 24V build loses roughly a quarter of the voltage over the same conductor. Where the driver sits indoors and the cable is long, that decides whether the far end is visibly dimmer — the main reason exterior sealed runs are specified at 24V, as set out in the 24V COB LED strip guide. Where run length makes even 24V DC awkward, mains-voltage construction removes the drop problem and takes on different safety and dimming constraints; see 120V COB LED strip for where that trade-off makes sense.
Keep the driver’s continuous load near 80% of its rating, not at 100%. Outdoors that is not headroom for its own sake: ambient temperature is part of a driver’s rating, and one sized with no margin in a 25°C workshop is past its limit in a 40°C parapet cavity in July.
Sealed COB or Neon Flex Outdoors?
For exposed exterior work these are the two realistic constructions and they fail differently. A sealed COB strip is a flat circuit with a sealing layer applied, so its weak points are the interruptions — cut ends, cable entries, field-made joints. Neon flex is an extruded body with the circuit inside, so it has fewer interruptions by construction and the body itself resists impact and abrasion.
The choice follows from what the light is for. If the strip is hidden and the surface it lights is the visible element, sealed COB in a channel is right. If the line itself is visible — a facade outline, a sign, a handrail seen directly — neon flex is normally more robust; it is a different product class, not a sealed version of a flat COB circuit. The comparison is developed in the COB neon LED strip guide, and for facades combining several constructions, the commercial outdoor LED strip lighting overview covers how the mounting details are coordinated.
Detailing a Sealed Run So It Drains
A sealed strip in a channel that holds water is being asked to pass an immersion test permanently. The detail decides whether it has to.
- Give the channel a drainage path — weep holes at the low points, or a profile mounted so its opening faces down or outward rather than up.
- Plan cuts where they can be sealed and inspected, not at the low point of the detail.
- Put a drip loop below the cable entry, so gravity works against capillary travel.
- Mount to metal, not into a sealed void. A sealed strip needs the thermal path more than an open one.
- Allow for movement. Sealed strips and aluminium expand at different rates, and a long run fixed hard at both ends puts the difference into the seal.
- Feed long runs from both ends where the layout allows — this roughly quarters the worst-case drop.
In wet interiors the same logic applies with less exposure but more thermal cycling. Spa and steam-room ceilings cycle from ambient to warm and humid several times a day, the condition that drives internal condensation — the planned waterproof COB LED strip bathroom spa reference deals with a detail of that type, where the water is airborne rather than falling. Treat all six points as additions to the normal installation sequence.
What to Ask Before Approving a Sealed Sample
A supplier who can answer these has tested the product. One who answers them with the rating number has tested the rating.
- Which sealing method achieves the rating, and is the material specified as UV-stable?
- Are the IP68 depth and duration declared in writing? Without both, the code means less than IP67.
- What is the maximum ambient temperature when sealed, and is the wattage derated?
- Is an aluminium profile still specified once the strip is sealed?
- How are cut ends and cable entries sealed, and is a factory-terminated length available?
- Is LM-80 data available at a case temperature representative of a sealed installation?
- For salt or chlorine exposure, what protects the solder joints beyond the seal?
Then test the assembly, not the strip. Ask for a short sealed length terminated the way the project will terminate it, in the profile the project will use, at the specified CCT. Power it for several hours and check three things: whether the far end has dimmed, whether the profile is warm or hot to the touch, and whether fogging appears inside the sealing after it is switched off and left to cool. The third check predicts the condensation failure and costs only time.
Sealed variants of this range are being added to the catalogue rather than published as fixed SKUs, so sealing method, rating and derating are confirmed per project. Review the published COB LED strip product range for the base circuits and specify the sealed construction on top, stating the exposure rather than only the rating.
FAQ
Is IP68 always better than IP67?
Not necessarily. IEC 60529 fixes IP67 as 1 m for 30 minutes but leaves IP68’s depth and duration to the manufacturer to declare. An IP68 product declared at 2 m for 24 hours is stronger than IP67; an IP68 claim with nothing declared is weaker, because nothing has been specified.
Can a waterproof COB LED strip be submerged permanently?
No rating in normal strip construction covers permanent submersion, and IP67’s 30-minute test explicitly does not. Continuous underwater service needs a fixture declared for it, not a sealed strip.
Why did a sealed strip fog up inside when no water reached it?
Thermal cycling makes the sealed volume breathe. Air expands when the strip warms and is drawn back in with ambient humidity as it cools; over many cycles that moisture condenses inside. It is a construction issue, not an ingress failure, and no IP test measures it.
Should an outdoor sealed run be 12V or 24V?
24V in almost all cases, because outdoor layouts push the driver further from the strip and a 10 W/m load draws 0.42 A/m at 24V against 0.83 A/m at 12V.
Send Us the Exposure, Not Just the Rating
The most useful thing you can send is a description of where the strip goes: whether water falls on it or only reaches it as vapour, whether the channel drains, how far the driver sits from the run, the ambient range at the mounting position, and whether salt or chlorine is present. Those five facts set the rating, the voltage, the sealing material and the profile together, which a rating alone cannot. Contact Senfey with the location detail and we will specify against it.
