My Double-Wick Candle Performs Perfectly at First but Becomes Dangerously Hot Near the Bottom — Should I Retest with Smaller Wicks?

Two flames, one pool: what changes near the base

A wide jar is double-wicked because one small flame cannot melt the whole surface in a reasonable first burn. In the worked example the inside is 9.0 cm across, a pool area of 63.62 cm², 65% more than a 7.0 cm pool (38.48 cm²). At the top, two flames sit in open air above a lot of cool wax; each melts its own circle, the circles meet in the middle, and the pool reaches the wall. The very rough rule of about an hour per 2.5 cm of width puts that at about 3.6 hours for this jar (a commonly cited rule of thumb; your log decides).

Near the base, three things tend to change at once. The two flames now burn inside the jar, sheltered by glass that has absorbed hours of heat, so they often stand taller. The merged pool is deeper, because there is more heat going into less remaining wax. And each wick stands in a deeper liquid pool, where it can lean or be nudged by convection; a wick that leans outward shortens the distance between its flame and the wall. With a single centred wick, the whole radius separates flame and glass. With two, each flame has only its own gap, which at the quarter-point placement is about a quarter of the diameter.

Why two wicks tend to heat glass more than one is the container series' question (two wicks, much hotter glass), and whether the jar's shape adds to it is the double-wick-too-hot-near-the-end post. This page takes it from there: you have a double-wick candle that failed late, and you need a retest that tells you which wick configuration, if any, this jar can carry to the bottom.

Double-wick patterns, ranked, and the gap that separates them

The order below is editorial judgement, most common first; CSI publishes no failure statistics and no measured burn data. The last two columns are what you can observe, cold, with a ruler: each cause has an observation that rules it in and one that rules it out.

Patterns 4, 5 and 7 are burn conditions; fix them before the bracket or every arm is judged under conditions nobody should burn in. Pattern 6 is the jar question (shape and heat; the narrow-bottom profile if your jar tapers). Patterns 1, 2 and 3 are what the bracket and the spacing log separate: 1 is a size problem, 2 and 3 are placement problems, and only size problems are fixed by smaller wicks.

The three-arm bracket for a wide jar

The bracket changes one thing, the wick configuration, and holds everything else: wax, oil, load, fill, jar, pour temperature, room and trim. The worked example uses the Apothecary Glass Candle Jar with Dome Lid (700 gm), CSI 464 and Eucalyptus Spearmint at 8% of total weight. We assume a 500 g fill because the listing's 700 g looks high for its stated outline; that gives 40 g of oil and 460 g of wax per candle. Weigh your own fill and scale the grams. CSI 464 states "Fragrance Load Up to 10%" and the oil page "up to 10% for soy wax", so 8% sits inside both ceilings (the ceilings' basis isn't stated: 50 g per candle on a total-weight reading, about 45 g on the more conservative of-wax reading, and more oil is not the fix for heat). The jar page's "Safe for hot-pour up to 85°C" and CSI 464's "Pour Temperature 80°C – 90°C" mean pouring at 80–85°C; the lower limit governs.

The placement for B is a commonly used starting point for two wicks (each roughly halfway between the centre and the wall, so the wicks sit 4.5 cm apart in the example), not a CSI figure; your own jar and your top-third result may move it. If you still have two unburned candles from the failing batch, at least as old as the new pour, use them as A1 and A2 and pour only four. Their cure is at least as long as the new candles', so the comparison is fair.

Costs are worked examples and exclude labour, freight, packaging and GST. Starting from nothing, the cash outlay is ₹4,593.60 for six candles (nine wicks, two sets of bamboo holders at ₹118.00 a set) or ₹3,019.80 for four (five wicks, one set), so every wick is held at once. The six-candle bracket uses 3 kg of candle; that is the price of testing a big jar honestly. CSI 464 is flat per gram from 1 kg to 10 kg, so three 1 kg bags cost the same per gram as the 5 kg and tie up less cash.

The spacing log: gap to the glass, burn by burn

The spacing log is what makes this a double-wick test rather than two single-wick tests in one jar. It costs nothing and takes a minute per candle. Measure only when the candle is cold and set: a ruler laid across the rim, reading from the glass to the centre of each wick, straight down from above. Never reach into a liquid pool, and never judge heat or position by touching hot glass.

Read the log in two directions. Down each gap column: a gap that shrinks steadily while the other holds is drift, and drift that reaches your limit is a fail on its own, even before the flame touches. Across a row: if both gaps hold and both flames still grow, the pair is too big for that depth. In the worked example, arm A shows drift on wick 1 on top of a pair that was already running hot: the drift made it fail sooner, but arm B's steadier gaps are what the comparison is for.

Two further numbers help. The gap between the wicks grows when both drift outward; if it grows and the flames merge into one tall flame in the middle of the pool, note it, because merged flames tend to run hotter than two separate ones (a tendency to observe, not a rule). The wax height at each failure, read from the outside marks, tells you whether the trouble starts well inside the final third or only below the stop line (pattern 5), and gives the stop line on your care card a reason.

Reading the bracket: which pair, or no pair

Two limits decide it, both written before the pour: the top-third pool limit and the final-third sheet, with the gap limit as one of its lines. The order of the rows below follows the question in the title: smaller wicks first, then the single wick, then nothing.

When more than one arm passes, prefer the one with the widest margin in the spacing log, not the one with the strongest throw; a pair that ended 0.1 cm inside your limit passed by luck. Throw is chosen among arms that are safe to the stop line. Worked into running costs, the wicks for 100 candles are ₹1,888.00 on 2 × C2, ₹1,180.00 on 2 × C1 and ₹1,200.00 on one Cotton Braided; moving from C2 to C1 saves ₹708.00 per 100 candles. That saving is a side effect, never the reason.

If you choose a slower pair because the faster one fails the bottom, how much first-burn tunnelling your brand can accept is a business decision, weighed in the slow-first-pool decision; how far down each test candle must go before a combination is accepted is the burn-depth gates.

What not to buy for a double wick that runs hot

A candle that frightened you near the bottom invites a big purchase. We would rather you ran the bracket first, even though several of these are ours.

What each outcome needs: adopting 2 × C1 needs ten C1 and a spec card with the measured positions. A drift outcome needs better holding while the wax sets, and nothing else. A single-wick outcome needs ten Cotton Braided. The all-fail outcome needs the container side of the question before any purchase at all.

Jar, wicks and holders for the bracket

Five items: the wide jar, the smaller pair, the single-wick alternative, the holders that set the spacing, and the worked-example oil. Chosen on page statements, per-unit price and stock on 26 September 2026; CSI has no measured burn or temperature data, so nothing is ranked on performance. The oil table shows every size, with 150 g candles at 12 g and 500 g candles at 40 g.

Frequently asked questions