You hold it up to the light. It wavers. It bends. From three metres it looks like a real candle. Then six months into a hotel deployment, or one hot July afternoon on a windowsill, you find the flame bead warped, the base melted into a flat pancake, and a thin trail of wax that looks like it had a medical incident.

Every flameless candle I have opened in the last ten years contains the same component: a small LED flame bead, and a bead of plastic, and a pair of AA batteries. What almost nobody tells you is that the bead is not a decoration. It is a heat source. And heat, aimed at a wax column from the inside, is a slow demolition crew that does not even need a match.

Let me walk you through the actual numbers, because the numbers are where the cheap ones lose.

The Problem: the fake flame is a heater

A typical flame-bead LED draws 20-30 milliamps at 3 volts from two AAs. That is 60-90 mW of total power. The chip converts roughly 60% of that into light; the rest is heat, radiated from a bead that is 15-20 mm across, sitting two centimetres below the top of the wax column.

Individually that is almost nothing. But two facts change everything:

  • Paraffin begins to soften in the low 40s Celsius. A wax body held near the top at 35-40 degrees for long periods creeps, sags, and eventually slumps onto the bead.
  • Duty cycle. Retail demos run these candles 8-20 hours a day. A 24-hour cycling timer means the electronics never fully cool before the next window. That is not a candle burning for an evening. That is a 3-watt-equivalent heater parked inside a wax column, 365 days a year.

I have pulled open a flameless pillar after a summer on display. The bead had fused into a mushroom. The wax column had pulled a ring around it. The base was flattened where the heat concentrated through the stem. None of it took fire. It took a Tuesday in July.

The Insight: real wicks are hot too, briefly; fake ones are lukewarm, forever

Here is the part most reviews skip. A real candle wick sits at 40-50 degrees during a burn, sometimes higher at the tip. So why does the real candle not self-demolish? Two reasons: time, and airflow. A burn lasts a few hours, the flame convects heat upward out of the pool, and between burns the wax is stone cold. A flameless bead gives up less peak heat but never stops, and it radiates in every direction, including downward into the plastic base, where soft polystyrene and ABS start moving in their 70s and 80s Celsius range, and where the glue, the battery contact, and the wire solder start failing years early.

That is the entire failure map of a cheap flameless candle: thermal creep on the wax, creep on the base, creep on the electronics. Nobody tests for it, because in a retail demo it takes months to show up. By then the batch is sold.

The Execution: four checks that separate a candle from a fire hazard in disguise

1. The 60-second bead test

Buy the set. Light the flame mode. Count to sixty. Back of the hand, two centimetres below the top of the wax column, touching the wax near the base. If you feel a definite warm spot inside 60 seconds, the LED is dissipating into the column and that unit is a clock, not a candle. Premium units I have tested stay within a degree or two of room temperature at that point because the bead sits in an open cavity, not buried in the column.

2. Standoff distance

Between the top of the flame bead and the top of the wax column there should be a visible gap, roughly 10-15 mm of air. Units where the bead is pressed into or almost touching the wax are transferring heat by conduction, which is an order of magnitude worse than radiation. This one is visible in product photos, before you ever buy. Zoom into the flame cavity. See a gap, keep shopping. See the bead winking through the wax, walk away.

3. The base is a composite

Flip it over. Good units are machined or thick-moulded one-piece plastic, heavy in the hand, with the battery lid on a screw or a proper clip. Cheap units are a thin shell glued to a separate base. Glue lives about one decade, and it starts dying at 60 degrees. If you can feel the seam, that seam is where your candle will open its back like a clam in a hot van.

4. Battery drain is a spec, not a vibe

Two AAs should give you 200-500 hours depending on the driver and the flicker algorithm. The cheap ones publish a number; the honest ones publish a duty cycle. A 24/7 cycling timer at 4 hours on, 20 off is roughly a 1/6 duty cycle, and that is when the 500-hour claim becomes 3000. If a listing gives you an hour figure with no duty cycle, assume the worst duty cycle: always on.

Where the honest specs land

The LED itself is rated for 50,000 hours. That number is real. The battery window, 150-500 hours for an AA pair, is real too. What is fake is the assumption that 50,000 hours of LED life means 50,000 hours of the candle. The candle is a thermal system, and thermal systems die on the component with the lowest temperature margin, which in this design is never the LED.

Rule of thumb from the production side: the cheaper the unit, the more of that 60-90 mW of dissipation ends up in the wax and the base. The more expensive the unit, the more of it ends up radiating into open air from a cavity, where it is nothing. Same chip. Same wattage. Different geometry. That is the whole difference between a product and a problem.

What we do with ours

Our pillar and taper lines are built around the cavity: open air between bead and column, one-piece base, and a driver that cycles with a duty cycle you can see in the manual. If you are outfitting a venue and want the flame look without the thermal clock, look at our LED candle range. And if the candles are only part of the install, the same geometry logic applies to everything else you string around the venue - our Christmas lights lineup includes outdoor-rated drivers that were not designed to sit in a wax column, which is, for once, exactly what you want.

Frequently asked questions

Do flameless candles get hot enough to damage anything?
A real candle wick runs at 40-50 degrees for a few hours. A flameless LED bead gives off 60-90 mW continuously. Peak temperature is lower, but duration is infinite. Anything flammable, soft, or glued near the bead is exposed to a low-grade, unlimited-duration thermal load. The 60-second touch test above tells you in one minute whether your unit radiates or buries.

Why did my plastic-based flameless candle deform after a summer?
Polystyrene and ABS bases begin softening well before their rated temperature once held warm for hours. A heat-conducting bead pressed against the column, plus a sealed base with no airflow, is the classic setup. The deformation you find in August usually started in June.

How long should a flameless candle run between battery changes?
200-500 hours per AA pair is the honest range, and it scales with duty cycle. Always-on use halves or thirds the published figure. If a listing does not state the duty cycle behind its hour count, assume always-on.

Is a flameless candle safer than a real one?
For open flame, yes, unambiguously. For thermal creep, no: a flameless unit on a 24-hour timer in a warm room outlasts a real burn by weeks, and heat that never stops accumulates in ways heat that does stop never does. Safe means no fire; it does not mean no damage.