Ask any candlemaker what the most common return reads and you will hear the same line: “I still have half the wax left and it will not burn.” That wax is tunneling. The flame burned a narrow channel down the middle of the jar and never melted the wax against the glass. It is a melt-pool geometry failure, not a customer error, and it shows up in burn logs as a normal-looking candle.
What Tunneling Looks Like, Measured
A 12-ounce (355 mL) fill in a 75 mm glass jar gives you roughly 80 mm of wax to burn. A healthy first burn melts a pool that spans the full 75 mm and runs 10–15 mm deep, about 45–65 grams of liquid wax under the flame. A tunneled burn melts about the same number of grams, but into a 35–40 mm wide channel running 25–30 mm deep. The flame does not see the glass wall. It burns a cylinder the size of its own reach, and the wax frozen 20 mm outside that radius is never paid for.
In our factory runs, tunneled 12-ounce jars come back with 30–50% of the wax still against the glass. The customer burned it for an hour, twice a day, for two weeks. The burn log looks normal. The glass says otherwise.
The Three Causes, in Order of Frequency
Number one is a wick sized for a smaller jar. Our wick chart puts a 75 mm container in the CD-10 to CD-12 range (LX equivalent for soy). If the pool does not reach the glass by the two-hour mark, the wick is under-powered, and every burn after that digs the channel deeper. The failure is visible in session one: a pool stalling at 50–60 mm on a 75 mm jar means the wick is one size too small. Change the wick before you sell the candle.
Number two is melt point. Straight soy resists lateral spread, and a 10% fragrance load pushes the effective melt point up another 5–8 °C, so the pool runs deeper and narrower while the 3 mm of glass wins the heat battle. Blends with 10–20% coconut wax run a wider, shallower pool on the same wick. That is half the reason the 70/30 split works the way it does.
Number three is the customer’s burn habit, and it is the one you cannot fix on the wick. A 45-minute burn melts a pool that starts to cross the jar and then freezes mid-migration. The solidified ring acts as a thermal break, so the next burn restarts from a narrower pool. Two short burns and the candle has learned its tunnel.
The Fix, in Spec Order
Start with the label. “Burn at least 2 hours on first use” is the line that kills most tunneling, because it forces a full-pool burn before the wax sets its own shape. We print it on the bottom of every 3-inch jar. It costs nothing and answers the number-one return.
Then size the wick for the pool, not the wax. Our acceptance test: light a sample and start the clock. We log two numbers: time to the glass wall, and depth at the edge. A 75 mm jar that crosses in 1 hour 40 minutes to 2 hours 30 minutes, with an edge depth of 8–12 mm, passes. A pool that crosses at 3 hours is a wick too small. A pool at the wall in 45 minutes with a mushroomed center is a wick too big. Both are rework, not a customer problem.
Finally, run the 5-burn tunnel test in QC: five consecutive 2-hour sessions, pool diameter measured with a dial gauge after each burn. If the pool holds 70 mm or better across all five, the batch ships. If it drifts below 60 mm by burn three, the blend or the wick is wrong, and we fix the spec, not the packaging copy. The wax left behind is always a design number we chose. It is just the only one the customer gets to see.