My 7 cm Candle Reaches the Glass Only After Three Hours — Is That Too Slow or Can It Be Acceptable for a Properly Wicked Soy Candle?

Where three hours sits for a 7 cm jar

Start with the jar. The White Matte Glass Jar page gives "Height 8 cm width 7 cm", and like every one of CSI's 72 container listings it states no internal diameter. "Width" on a jar page is usually the outside. On a 0.35 cm wall, our labelled assumption, the inside is 6.3 cm and the pool can grow to 31.17 cm² before it meets the glass. If your ruler says 7 cm inside, the pool is 38.48 cm² and the band moves out a little. Measure yours at the rim, the middle and about 1 cm above the base; the method is in the 6–7 cm hub, and why the outside figure misleads is argued in the external-versus-internal diameter post.

Then the two rules of thumb. Neither is CSI data and neither is a standard; they are what many makers use as a first check.

So a three-hour edge is not "slow" by either yardstick. It is at the unhurried end of normal. The reason people worry is that they compare it with a candle that pooled in two hours, and in a small jar a faster candle is often a hotter one. Speed is not the goal; a pool that reaches the glass while the flame, depth and glass stay calm is. If yours is getting there very fast instead, the step-down post for fast 7 cm pools is the right page.

One more reason to keep the band wide: soy commonly needs a cure before it behaves as it will in a customer's home. The Gardenia page in our worked example states a cure time of 48–72 hours for soy; 25 of 102 CSI oil pages state a cure time in the text we captured, and many makers test at day 3, 7 and 14. A first burn on day 2 and a first burn on day 7 can draw different curves. Keep the cure identical across the candles you compare.

Why the last few millimetres take the longest

Watch a first burn with a clock and a pattern appears. The pool opens fast around the wick, then slows as it approaches the glass. In our worked-example burn 1 the diameter grows by about 0.33 mm a minute between one hour and ninety minutes, and by about 0.17 mm a minute in the last half hour. The last 5 mm, from 58 mm to the glass, takes 30 minutes on its own.

Three things commonly cause that slowdown, and none of them is about the wick being wrong. The heat has further to travel from the flame to the edge. The glass, cold at the start of a first burn, draws heat out of the wax next to it. And the wax at the wall is the wax that was poured against a cooler surface. These are tendencies, not measurements, but they are why a correctly wicked candle can look "stuck" at 58 mm for a while. What matters is that the last few millimetres also carry a lot of surface: the last half hour of burn 1 melts 15% of the pool's area, nearly as much as the first half hour (10%).

The two-hour row matters commercially. At 2:00 the worked-example candle has melted 66% of its surface. If your customers typically burn for two hours, they are living on that row, not on the three-hour one. Whether a jar should be judged on the sessions customers actually run is owned by the 8 cm near-four-hours post; for 7 cm, it is the reason the curve records every half hour rather than only the end.

How to draw the timed pool curve

The curve is a measurement, so the set-up matters more than the equipment. Use three candles from the same pour: same wax, oil, load, fill weight, wick and cure. Three, because one candle can have a leaning wick or an air pocket and you would never know which result was the candle and which was the wick. If you only have the one candle that worried you, draw its curve anyway; just don't make a production decision from it.

Never judge the glass by touching it, during or just after a session. If you want a heat reading, a non-contact thermometer is the tool; CSI does not list one. For this post the glass is an observation: does the flame lean towards it, does soot form on it, does the pool reach it evenly. Burn within sight throughout.

The acceptance rule and the five other outcomes

Read each candle's curve, then the three together. A cell "passes" when at least two of the three candles meet the rule; a 2–1 split is a pass with a note, a 1–2 split is a fail. The thresholds below are our judgement, chosen because each can be read off a photograph or a skewer. Set your own if you prefer, then keep them fixed.

Look at what the table does not do. It never turns a three-hour edge into a wick-up order on its own. Only two rows send you to C2, and both need evidence beyond the clock: a second burn that is no faster, or no edge at all by four hours. The second row also asks you to clear the room first, because a cold bench and a slow burn 2 look the same as a small wick.

Clearing the room before you blame the wick

If burn 2 was no faster than burn 1, check four things before you pour a single C2 candle. Room temperature: a bench under an air-conditioner or a cold monsoon evening can add a lot of time to the last few millimetres; the Indian winter, roughly December to February and colder in the North, does the same. Session gap: burn 2 must start from a fully cooled candle, but not one left in a colder room than burn 1. Trim: a wick trimmed shorter for burn 2 makes a smaller flame. Centring: a wick that drifted during burn 1 leaves a crescent that burn 2 inherits. If any of these differed, burn 2 did not test the wick. Repeat it.

Reading a 2–1 split

Three candles rarely agree perfectly. If two meet the rule and one reaches the glass at 3:45 with a 1.1 cm pool, look at the odd one before you average: a wick 2–3 mm off centre in a 63 mm jar is enough to slow one side. If the odd candle's wick was central and the room was the same, the wick is marginal for this system. Keep it in the full-life test, but pour a fourth candle so the full-life result is not resting on two.

If the curve fails: the C2 stage, costed

Stage 1 is the curve itself, on the rung you already use. In the worked example that is C1: three 150 g candles at 8% on total weight, 12 g of Gardenia and 138 g of CSI 464 each, in White Matte jars. Stage 2 happens only if the rule sends you one rung up: three more candles, identical except for a C2 wick. If your current wick is C2, stage 2 would be CSI's next rung, Cotton Braided Wicks Ready Made, whose page describes it as "engineered for large-diameter candles exceeding 8 cm". A 7 cm jar is outside the range its page describes, so treat it as a separate, labelled arm, watch the glass and the lower third closely, and ask us first.

One stage-1 candle costs ₹201.98 in materials: oil, wax, jar and wick. A C2 wick adds ₹3.54 per candle over a C1. All costs are worked examples and exclude labour, freight, packaging and GST.

Which way to buy? If you think stage 2 is unlikely, buy stage 1 small: ₹955.02. If you then need stage 2 and top up with the same small sizes (a second 50 g bottle, a second 500 g bag, a second pack of 5 jars and the C2 10-pack), you will have spent ₹1,886.44 in all, ₹25.64 more than buying both stages up front for ₹1,860.80. That is the honest trade: a small saving now against a slightly larger bill later. For most makers whose three-hour edge came with a calm flame and a shallow pool, stage 1 is all they need.

Timing, honestly: the oil page's 48–72 hour cure is the minimum; many makers wait about a week before the first burn. Burns 1 and 2 take two days with a full cool-down. So a curve poured this week can be read by early October. A failed curve adds the C2 stage on the same timetable. The full-life burns that follow take longer; plan them with the full-life calendar.

What to buy, and what to skip

Five items, in the order the test uses them, chosen on page statements, price and stock on 29 September 2026. CSI publishes no wick-performance data, so nothing here is ranked on burn results. Oil candles are counted at 12 g per 150 g candle.

Frequently asked questions