Every outdoor LED candle that fails at a customer’s venue did not fail because of the LED. It failed at the seam. Rain walked in through the battery cover, or through the gap between the flame lens and the housing, and three months later the battery contacts are green with corrosion. If you buy led candles for a covered porch, you will never see this. If you deploy them in a garden or along a pathway, you will.
The rating that decides whether that happens is the IP code, and most retail listings treat it like a marketing sticker. The two digits mean different things, and the second digit is the one your installation cares about. Here is what we actually test on the bench, and what we ask suppliers to prove before a batch ships.
What IP65 Actually Measures
IP codes have two digits. The first rates solid-particle protection, from 0 (nothing) to 6 (dust-tight). The second rates water protection, from 0 (no protection) to 9 (high-pressure, high-temperature jets). For an outdoor candle, the first digit is basically a footnote. Dust on a wax-coated flame or a glass housing is cosmetic. Water is what kills the electronics.
Here is the mapping that matters for placement: IP44 handles splashing water from any angle and works fine under a covered eave. IP65 means the enclosure is dust-tight and protected against water jets up to 6.3 mm nozzle from any direction, roughly 12.5 litres per minute. That is the difference between a candle that survives a covered pergola and one that survives an open rain event. IP67 adds temporary immersion for up to 30 minutes at one metre, and IP68 is continuous submersion. We sell IP65 as the standard for garden and pathway use, and we move to IP67 only when a customer is placing candles in water features.
The honest limit of IP65: it does not protect against standing water. If the candle sits in a saucer that holds rain, the seal is not what is protecting it anymore. That is an installation problem, but it lands on the product’s reputation.
The Battery Cover Is Where the Rating Lives or Dies
IP certification is done on a sealed unit, and the seal on a battery-candle is almost always the battery cover. This is the single most common point of failure we see in returns. The cover is hinged or removable so the customer can swap AA or AAA cells, and that opening is the weakest point in the whole housing. A supplier that presses a thin silicone gasket into the cover and calls the unit IP65 is describing a wish, not a measurement.
When we audit a factory, this is the first part we ask to see. We want a double-lip gasket, or a gasket with a lip that presses against a ribbed mating surface so the compression is mechanical, not just elastic. We check the screw torque on the cover fasteners. We look at whether the cover threads or clips. A hinged cover with a compression seal is easier to get right than a screw-off cap with a single O-ring.
One number that tells you a supplier has actually thought about this: the runtime of the unit after a controlled water exposure. We ask for a spray test at 12.5 L/min for 3 minutes from each direction, then a 24-hour hold, then the unit goes on the bench and we measure the actual runtime against the spec. If runtime drops more than 5% after the cycle, the seal is passing the letter of the test and failing its purpose. A battery-candle that has water behind the LED driver will still light, but it will run warm, drain faster, and the contacts will corrode on a timeline your customer will notice.
There is a second failure path that is less obvious. The flame lens. A lot of cheap flameless candles use a thin acrylic or resin flame insert, and that insert is held by friction, not a seal. Rain that gets behind the flame looks exactly like condensation in the lens, and the customer concludes the whole candle is wet. We specify a flame insert that seats in a chamfered pocket with a thin silicone ring, or we move to a sealed glass dome where the application allows it.
How to Verify the Rating Before You Ship
The practical test for a buyer who is not running an IP-certification lab is a two-step bench check. First, ask for the supplier’s IP test report, and verify it names the specific model and the second-digit rating. A report that says “IP65” without the standard reference (IEC 60529) is not a report, it is a label. Second, do the spray test on two or three units from the production lot, not the pre-production sample. We time the spray, measure the runtime after, and photograph the battery compartment for corrosion or moisture. It takes twenty minutes per unit and catches almost every supplier that is over-claiming the rating.
For solar garden candles specifically, add one more check. The photovoltaic cell is mounted on the top face and exposed to direct rain, so the IP rating of the solar unit is set by the cell’s seal, not the housing. We inspect the cell bezel under magnification before we test it, because a hairline crack there will pass a 3-minute spray and fail on the first hard downpour.
Our rule for wholesale buyers: do not specify IP65 on a product you have not spray-tested, and do not accept a supplier’s IP65 claim on faith. The rating is the product’s outdoor life, and it is either there or it is not. The twenty-minute bench test is the cheapest insurance you will buy on this line.