My 10 cm Jar Takes Four Hours to Melt Close to the Edge but Has Clean Flames — Should I Increase the Wick Size or Continue the Full-Life Burn Test?

Is four hours slow for a 10 cm jar?

Against the commonly cited rules of thumb, four hours is borderline, not bad. The one-hour-per-2.5 cm rule gives 4.0 hours for a full 10 cm and 3.7 hours for a 9.3 cm inside; a full pool within about 3–4 hours of a first burn is the other common expectation. Neither is CSI data, both are rough, and both were built for a single centred wick. A pair in a 10 cm jar commonly reaches the glass along the wick line first and closes the gap at right angles last, so "close to the edge" at four hours usually means a thin ring at two points.

A thin ring is still worth taking seriously. On a 9.3 cm inside, a 2 mm ring all round is 8% of the surface and a 4 mm ring 16%. What matters is its direction of travel. A ring that narrows every burn is a candle catching up as the jar and wax warm through over successive burns. A ring that holds steady at the same width is a candle at its limit. Clean flames rule out the opposite problem: if the wicks were overloaded you would expect a deep pool, tall flames, carbon or soot, and you have none of them.

The oil in the worked example is Sandalwood Fragrance Oil at 8%, inside its page's "Recommended load up to 10% for soy" and CSI 464's "up to 10%". Woody, resinous oils are commonly linked to needing a little more wick than light florals (a tendency, not CSI data), which is one reason a woody candle can sit at the slow end of the clock with perfectly clean flames.

The continue-or-change gate

The gate turns the question into something you can read with a skewer, a ruler and a clock. The thresholds are our judgement, built on commonly cited rules of thumb (depth about 0.5–1 cm, flames about 2.5–5 cm); set your own if you prefer, then keep them fixed for every candle and every reading.

Read depth with a skewer beside each wick and at the midpoint between them, flame height against a ruler held behind the jar (not touching it), and the ring at four points round the glass from a photo taken from above. If the ring is heavier at the two points at right angles to the wick line than at its ends, the stall may be the geometry of a pair rather than power, and the ring-shape map reads that before you pour anything.

Two cases sit outside this page. A pair that is too fast, under an hour to the glass, is the one-hour step-down. A candle that is fine for half its life and hot near the bottom is the multi-wick life-cycle post; the gate's mid-life and final-third readings are how you would catch that early.

The parallel run: test the next rung without stopping

When depth and flame pass but the ring stalls, you have two reasonable hypotheses: the current wicks are adequate and the ring will close lower down, or they are a rung short. Stopping the current test to try the next rung throws away the first hypothesis's evidence. Running the next rung alongside keeps both, and because a 500 g candle burns for many sessions, it also saves calendar time.

The two arms then answer each other. If the current pair's ring closes by mid-life and the gate passes in the final third, it wins, and the C2 arm becomes a documented "too much" on the other side, useful for closing the bracket. If the current pair's ring persists and the C2 arm clears the glass with calm flames to the stop line, the C2 pair is your candidate. If the C2 arm runs deep and tall while the C1 pair rings, your answer sits between CSI's rungs: we list no sized wick between C1 and C2, and we would rather say so; ask us, or look at another supplier's in-between size.

Full life in grams, hours and days

A 500 g candle burns for a long time, which is why the parallel run pays off. Weigh each candle before and after each session (cooled, lid off) to get its burn rate. Then the arithmetic is simple. The figures below are a worked example: a 60 g remainder at the ~1 cm stop line and burn rates of 11 and 13 g an hour are illustrations, not CSI data; your own scale replaces them after two or three sessions.

With one four-hour session per candle per day, the current pair finishes on day 7 from the gate decision. The C2 arm, poured the same day and cured 7 days, runs 9 sessions and finishes on day 16. Done one after the other, the C2 arm would start only after day 7 and finish on day 23. Running in parallel saves 7 days, and the two C2 candles add 67.6 hours of burning you were going to do anyway. More than one session a day per candle shortens everything, as long as each candle cools fully between sessions and is never left unattended.

The burn rate does a second job. It is the number your care label's burn time and your customers' experience rest on, and it is the first thing that changes if a candle starts running hot in its lower third. A rising grams-per-hour figure over successive sessions, with the same session length, is worth noting at every gate reading; the weigh-per-burn approach is set out in the faster-wax-use post.

What the parallel arm costs

Per candle, with stickers, a 500 g Sandalwood candle costs ₹663.68 on a C1 pair and ₹670.76 on a C2 pair (₹201.07 of oil, ₹233.40 of wax, ₹212.40 of jar); the rung up adds ₹7.08 a candle. Costs are worked examples and exclude labour, freight, packaging and GST. From nothing, the arm's cash outlay is ₹3,640.00 with the 500 g Sandalwood, the smallest size in stock on 1 October 2026 (the 15 g, 50 g and 100 g are out of stock), which leaves 420 g for later candles. If you already own 80 g of the Sandalwood your current pair uses, the outlay is ₹1,126.60.

What to buy, and what to skip

Five items, in the order the gate and the parallel run need them, chosen on page statements, price and stock on 29 September 2026 (Sandalwood stock rechecked 1 October 2026). CSI publishes no wick-performance data, so nothing is ranked on burn results. Oil candles are counted at 40 g (500 g fill) and 12 g (150 g).

Frequently asked questions