My Bestselling Candle Suddenly Started Tunnelling Even Though I Didn’t Intentionally Change the Formula — Could a New Wax or Wick Batch Be Responsible?

First, confirm the tunnel is new, and yours

"Tunnelling" here means a ring of hard wax left standing against the glass after a burn long enough to melt to the edge. Commonly cited rules of thumb expect a container candle's first burn to reach the edge in about 3–4 hours, and a very rough guide is an hour per 2.5 cm of width, about 2.8 hours for the 7 cm-wide 150 g jar. A ring after a one-hour burn is not a tunnel; it is a short burn.

Before you suspect a lot, reproduce the fault. Take one candle from the batch that tunnels and one retained candle from the last batch that didn't. Burn each for four hours on separate days, in the same room, with no fan or AC blowing across it, and measure the ring at its widest point with a ruler after the flame is out. If only the customer's candles tunnel and yours don't, the likely story is burn habits, which is the customer-report post's job, not a lot change.

Write down, for each candle: batch, pour date, age at the burn, wax bag and wick carton (from your batch sheet or invoices), ring width at four hours, flame height, and whether the ring is even or lopsided. That last column matters: a lopsided tunnel points to the wick's position more often than to its lot.

Five things that make a bestseller tunnel

The table ranks the causes behind a candle that used to burn clean and now tunnels, most common first. The ranking is editorial judgement, not data; CSI publishes no failure statistics. Two rows are free to check before any test pour.

Row 1 is first because it is free and common: a wick that drifted while the wax set produces a lopsided ring that looks like an under-wicked candle on one side. Row 4 is in the table because a wider jar spreads the same wick's heat over more wax; within-delivery spread is the few-millimetres-wider post and a supplier change is the jar measurement-sheet post. Rows 2 and 3 are what the title asks about, and the 2×2 answers them together.

The swap-back 2×2, step by step

The 2×2 works only if you still hold some of the old wax and the old wicks: 552 g of each wax lot and four wicks from each carton. If you don't, go to the fallback below. Everything that isn't wax lot or wick lot is held identical: one oil bottle, one jar delivery, one scale, one pot per lot, one pour temperature, one cure shelf, one burn room.

Material value only, excluding labour, freight, packaging and GST. The 2×2 is 13.3% of one 60-candle batch, and it decides what the next sixty are made from. British Rose's page states up to 10% in soy; the Chunks page states up to 13%, so the lower figure governs and 8% sits well inside both. Keep the load identical in every cell; this is not the place to change it.

Which lot the tunnelling cells convict

Two cautions keep the verdict honest. First, a convicted lot is a lot that behaves differently in your system; it does not show that the material is out of specification, and CSI publishes no lot data to compare it with. Send us the result with order numbers and photos, and we will look at it with you. Second, "use the old stock while it lasts" is a bridge, not a fix. When the old stock runs out, the fix test below decides the wick for the new lots.

For a scent that tunnels while others in the same wax and jar don't, the first-level answer is one wick, one fragrance tunnels. For a jar change that brought the tunnel, see narrow jar fine, wider jar tunnels.

No old stock left: the retained-candle fallback

If you used up the old wax or the old wicks, the full 2×2 is gone, but three weaker tests remain, in order of what they can prove.

Half a 2×2

If one old material survives, say four old wicks but no old wax, pour two candles of new wax on old wicks and two of new wax on new wicks. If only the new-wick pair tunnels, the wick lot is implicated; if both pairs tunnel, the wick lot is cleared and the wax, jar or process remains. It cannot convict the wax, because there is no old wax to compare.

Retained candle against a new pour

Burn a retained candle from the last good batch against a candle from the tunnelling batch, four hours each, same room. The retained candle is older, at least as cured as the new one, so if it reaches the edge and the new one doesn't, the difference is in the lots or the process, not in cure. It cannot say which lot, and the hub's records-first audit is how you narrow it from there.

Test the fix, not the cause

Pour four candles from the new lots: two on your production C1 and two on C2, the next step up. That uses 48 g of oil and 552 g of wax, ₹892.35 of materials, and the only new cash is a pack of ten C2 at ₹94.40 if you have none. If C2 reaches the edge where C1 tunnels, C2 is the candidate, but only a candidate: it must pass the full jar to about 1 cm of wax left, with the flame calm and no soot in the final third (the acceptance gates at depth). Never judge glass heat by touching it; a non-contact thermometer is the tool, and CSI does not list one.

What not to buy for a tunnel

Everything the 2×2 needs, you already own: the two lots, one oil bottle, jars and a ruler. Our picks below are for the few cases that need something: a C2 step for the fix test, holders if your wicks drift, and wax or oil only if you run short for production.

Products for the 2×2 and the fix

Four items: the worked example's wax and its wick in both steps, a centring tool for the process suspect, and the bestseller's oil. Chosen on page statements, price per unit and stock on 26 September 2026; CSI has no measured burn data, so nothing is ranked on performance. The oil table lists every size.

Frequently asked questions