My 9 cm Double-Wick Candle Has Excellent Hot Throw but the Jar Becomes Extremely Warm — Should I Reduce Both Wick Sizes?

Throw and heat come from the same flames

Hot throw comes from fragrance evaporating off a heated pool. Two good flames in a 9 cm jar make a big, hot pool, which is why your throw is excellent. The same flames heat the glass: directly where a flame sits close to the wall, and indirectly through a deep pool and hot air trapped in the jar above it. So "great throw, warm jar" is not a contradiction. It is one candle doing a lot of one thing.

The jar is your own; CSI lists no jar at 9 cm. Measured 9 cm across the outside, on an assumed 0.35 cm wall, it is 8.3 cm inside with a 54.11 cm² pool, 27.05 cm² per wick. With the pair 4.0 cm apart on a line through the centre, each wick sits 2.15 cm from the glass along that line. That is the worked example's current candle, on a pair of Eco C2.

"Extremely warm" is a real warning, but it is not a measurement. Candle jars are commonly soda-lime glass, and cracking is commonly linked to thermal shock, flames touching the glass and existing flaws. No CSI jar page gives a burning-temperature figure; the "Safe for hot-pour up to 85°C" wording on some CSI jars is a pour limit, not a burn rating, and your 9 cm jar is not one of them anyway. So this page does not give you a temperature to aim for. It gives you observations to compare against your own control. For the jar side of this question, including whether to change the jar rather than the wicks, read the two-wick jar that runs hotter than the single.

Four levers and a control

L1, both wicks down. Less fuel at each flame, so less heat everywhere. It is the lever the question names, and commonly the one that costs the most throw, particularly in the first hour while the pool is still forming. It saves ₹7.08 a candle in wicks.

L2, one wick down. Half the cut. A C2 and a C1 side by side can melt unevenly, the C2 side faster, so watch the pool shape and the C2 flame's lean. If it works, it is a production nuisance: two wick types in every candle. Mixed-size pairs are unusual in production; this test tells you whether yours earns the extra handling.

L3, spacing inward. Moving each wick 0.5 cm inward takes the flames further from the glass (2.65 cm instead of 2.15 cm) without changing their size, so the throw often holds. The price is a hotter, deeper middle: the midpoint is now 1.5 cm from each flame instead of 2.0 cm. If the centre gets very hot and deep, that is the too-close spacing ladder territory; the wall-gap trade itself is wall gap v edge melt at 8.5, 9 and 10 cm.

L4, shorter session. Same candle, burned 3 hours instead of 4, 25% less time for heat to build. It usually keeps throw well and often clears the flags. But it changes the burner, not the candle. A customer who burns it for 4 hours gets your control candle. Keep it in the test, because it tells you how much of your heat is time rather than build, but read it with that in mind. The rough rule of thumb of an hour per 2.5 cm of width gives about 3.3 hours for 8.3 cm, so a 3-hour cap also risks a pool that stops short of the glass.

Logging heat without touching the glass

Every heat observation below is made by looking or measuring, never by touching the jar. Make them at the same time in every session (2 hours after lighting) so the candles are compared at the same point.

The flags are deliberately about the causes of a hot jar rather than its temperature: flames close to the glass, heat piling up in a deep pool, flames too big for the pool. A lever that removes them has removed the reasons your jar got warm, which is more useful than a number with no rating to compare it to. The Pen Thermometer CSI lists (₹354.00) is described on its page for monitoring "the temperature of your candle wax": it is for the melting pot, not for reading a burning jar from a distance.

Logging throw without fooling yourself

You want to know whether a lever cost you throw. That comparison is easy to fool: you know which candle is which, your nose tires after the first few, and a room that has had a candle in it for an hour smells of it regardless. So keep it blind, short and coarse.

Why only three words? Because a finer score invites false precision. "Same" and "weaker" are what a customer would notice; a 7 versus an 8 is not. And why two times? Because the levers tend to cost throw at different moments. A smaller wick often loses most in the first hour and catches up as the pool fills; a shorter session loses nothing at 1 hour and cannot be judged at 4. Olfactory fatigue is real: step outside between candles and air the room.

Choosing the lever

The rule (our judgement). First, drop any lever that still raises a heat flag. Of what remains, keep the lever whose blind throw is "same" at both 1 hour and 3 hours. If more than one qualifies, prefer the one that changes your production least: spacing (no new wick) before one down (two wick types) before both down (a new wick, but one type). If none keeps the throw, keep the lever that clears the flags with the smallest throw loss. If nothing clears the flags, the pair is too much for this jar at this fill: reconsider the fill, the jar, or a very different build, and see the deep-pool-at-three-hours post for the depth side.

The session lever. L4 can only win if the control, burned for your full session, is also flag-free, which in this situation it isn't. So in practice L4 is a label, not a fix: combine it with a build lever that already passes, and state a burn time on your label because it is good practice, not because it makes an overheating candle safe. Argued against our own interest: our Candle Care Stickers roll will not rescue a candle that fails at 4 hours, and the page does not state what the stickers say.

In the worked example, L1 cleared every flag with throw "weaker at 1 h" and "same at 3 h"; L2 and L3 kept the throw but each kept one flag (the C2 side still leaning in L2, a pool over 1 cm deep in the middle in L3); L4 looked perfect but is the session lever. By the rule, L1 wins: the candle's first hour is slightly softer and it no longer overheats. It also saves ₹7.08 a candle in wicks, ₹708.00 per 100 candles, which is a pleasant side effect rather than a reason.

Then the full life. Double-wick jars commonly warm most in their lower half, when the glass above the pool is exposed and heat builds. A lever that passes burns 1–3 must pass to about 1 cm left before it goes into production. If it fails late, the double-wick candle that gets too hot near the bottom is the next test; the cumulative-heat version of this question in a 12 cm jar is the 12 cm jar that gets hotter every burn.

Grams, wicks and rupees

From nothing: Eucalyptus Spearmint two 100 g and a 50 g (₹920.40), three 1 kg bags of CSI 464 (₹1,522.20), a C2 20 (₹188.80), a C1 20 (₹118.00) and a pack of 10 Steel Wick Holders (₹500.00): ₹3,249.40. Left over: 10 g of oil, 240 g of wax, 6 C2 and 14 C1.

Per candle, oil, wax and wicks only: ₹243.88 on a C2 pair (control, L3, L4), ₹236.80 on a C1 pair (L1), ₹240.34 on C2 plus C1 (L2). The levers move the cost by a few rupees at most; the oil itself is ₹84.96 a candle at the 100 g price and ₹79.20 at the 500 g price. None of that should decide this test. All figures exclude jar, labour, freight, packaging and GST.

What to buy, and what to skip

Skip: a bigger jar as a first move. CSI's widest listing, the Apothecary Glass Candle Jar with Dome Lid (₹212.40 and ₹250.00), is about 10 cm at the rim and base by its page and states "Safe for hot-pour up to 85°C", a pour limit, not a burn rating. A wider jar changes the pool, the wick distances and the fill, so it is a new candle to test, not a fix for this one. Skip a higher fragrance load to "make up" throw after L1: the oil's page and CSI 464 both state up to 10%, and more oil is not more throw in a straight line.

Frequently asked questions