My Candle Flame Gets Larger as the Wax Level Falls — Is Heat Building Up Inside the Container and Changing the Burn?

Two causes that look the same from across the room

The question in the title assumes one cause, heat. Heat is real and it does build: the flame drops below the rim, the glass wall above it cuts off moving air so the flame stands straighter, and the glass absorbs radiant heat and runs warmer across the candle's life. Warmer glass and warmer wax underneath commonly give a deeper pool, and a deeper pool gives the wick more liquid fuel within reach. Any of those can lift the flame. That is the "container" story, and it is a property of the candle: jar, wick size, wax and oil together.

The second story is a property of how the candle is burned. In many candles a session consumes wax faster than it consumes the wick, so an untrimmed wick tends to stand a little taller at every lighting, and on heavier or sweeter oils carbon can collect on the tip. A longer wick, with or without a carbon cap, carries a longer flame at any depth. If your first test burns were done carefully, trimmed and timed, and your later burns were done like a customer would do them, you have compared two different wicks and blamed the jar.

The two stories need opposite fixes. The container story usually ends in one wick step down, confirmed through the whole jar. The wick-length story ends in a trimming routine and a care line on the jar, and a smaller wick would make it worse at the top, where it would struggle to reach the edge. That is why this test comes before any purchase.

What a growing flame looks like in each case

Before running anything, read the candle you already have against the table. The order is editorial judgement, most common first; CSI publishes no failure statistics or measured burn data. The two right-hand columns are what matters: what would rule each cause in, and what would rule it out.

Patterns 1 and 2 are the ones this post separates. Pattern 3 is the depth story in its own right, and measuring it burn by burn is the depth chart post. Pattern 4 is the base and centring audit in the uneven-near-the-bottom post. Pattern 5 is not a depth problem: go back to a first-three-wicks bracket. If what you mainly see is black on the glass rather than a taller flame, read the soot-band post first.

The trimmed-versus-untrimmed twin test

The design is a controlled pair, duplicated. T1 and T2 are trimmed to the same length before every burn. U1 and U2 are never trimmed; any carbon is left. All four come from one pour, so the wax, oil, load, wick, jar and age are identical. The worked example uses the 150 g jar, CSI 464 and your production wick (Eco C2 in the example), at 8% of total weight: 12 g of Apple Cinnamon and 138 g of wax per candle, poured at 80–85°C (the jar page's 85°C governs). Four twins take 48 g of oil and 552 g of wax.

Costs are worked examples and exclude labour, freight, packaging and GST. If you already own the wax, jars and wicks of the problem candle, the new cash is ₹355.18 (the oil and the centring device); starting from nothing, it is ₹1,900.98. Using your own production oil instead of Apple Cinnamon is better still: the test is about your candle. If you only have budget for two candles, run T1 and U1, and treat the result as provisional until a second pair agrees.

Reading the twins: the decision rule

For each twin, take its top-third flame at 1 h and 3 h as the baseline. At each later stage, is the flame at the same clock time taller than that baseline by at least the margin you wrote down before the test? The diagram's worked example uses a third: a 3 cm baseline flame counts as grown at 4.0 cm. That margin is our illustration, not a standard; what matters is that you set it first and apply it to all four twins.

Two cautions keep the reading honest. First, the U twins show what customers do, not what you should. If U grows dangerously while T holds, the candle is fine and the instructions are the product; a care line on every jar is part of the fix. Second, a growing T twin is a result about this jar, this wick and this oil, not about the wick in general. The same C2 may hold steady in a straighter or wider jar, or with a lighter oil.

When the trimmed twin grows too

The dark bottom-right cell answers the title's question with a yes: heat and depth are changing the burn. Two readings tell you how far. Pool depth, by skewer at the end of each burn near the wall, against the rule of thumb of about 0.5–1 cm: a pool that stays near that while the flame grows points to heat in the glass; a pool that runs well past it points to a wick at its hot edge for the lower section (the full chart belongs to the depth post). Glass heat, read with a non-contact thermometer at the same spot and times on T and U, lets you see whether the flame's growth tracks a warming jar.

Glass heat is never judged by touch. A hand on a jar in its final third can be burned, and "too hot to hold" says nothing repeatable. CSI does not list a non-contact thermometer; our Pen Thermometer is a probe for the wax pot, and its page states no range. Set your acceptance limit before the test and compare wicks against it; the safety criteria that fail a wick in the final third, whatever its throw, belong to the hot-jar post. When the question becomes the jar itself rather than the wick, read the very-hot final burn post and jars that survive early burns but crack late.

What usually follows a growing T twin is a one-step-down test: the current wick against the next smaller rung, in duplicate, to the stop line. For an Eco C2, that rung is C1; two confirmation candles cost ₹454.49 in materials in the worked example. If the smaller wick then fails to reach the edge at the top, you have the "no perfect wick" problem, and the options beyond wick size are the options post.

What not to buy before the twins have spoken

The twin test exists because the two causes need opposite purchases. Buying before the result is a coin toss, and several of the wrong options are ours.

What each outcome does need is short. Top-right: a trimming routine and a care line; the Candle Care Stickers Roll (₹413.00 for 500) carries it to customers, though whether care instructions are enough is a business decision. Bottom-right: ten wicks one step down and two candles. Top-left: a look at the room. Bottom-left: better twins.

The trimmer, the centring device and the rest of the set

Five items: the two tools that make the twins fair, the wick in the worked example and its step down, the oil and the jar. Chosen on page statements, per-unit price and stock on 26 September 2026; CSI has no measured burn data, so nothing is ranked on performance. The oil table shows every size, candles at 12 g, and four-twin sets at 48 g.

Frequently asked questions