My 8 cm Jar Takes Nearly Four Hours to Reach the Edges but Otherwise Burns Cleanly — Is the Wick Necessarily Too Small?

A slow edge is not the fault; a short session is the test

"Takes nearly four hours to reach the edge" is a measurement taken in a test that most customers never run. In a four-hour burn, the pool eventually touches the glass. In the two-hour or three-hour burn a customer gives it, the same candle stops short every time. The pool you measure at 3 h 50 min is not the pool your customer sees.

Here is why that matters. Hot throw comes from fragrance evaporating off the liquid pool, so a pool that covers 87% of the surface is releasing scent from 87% of it. And a soy candle that ends session after session with a ring of solid wax commonly builds that ring into a wall. The pool then tends to follow its own previous edge, which is how tunnelling starts. A clean flame, no soot and a calm pool are all good news, but they only show the wick is not too big. They say nothing about whether it is big enough.

So the question is not "is four hours too slow?". It is "how much of the surface is melted by the hour people usually put it out?". That is a measurement, and it takes a ruler and a watch. How fast the pool grows every 30 minutes in a 7 cm jar is the timed-pool-curve post. The continue-or-change gate for a 10 cm jar that takes four hours is the 10 cm four-hour post. This page is about the customer session in an 8 cm jar.

The session your customers actually burn

We don't know how long your customers burn their candles, and CSI has no data on it. What we do know is that the length has to be set before the test, because the same candle can pass at 3 hours and fail at 2. Pick the session you are prepared to print on the care label and stand behind. In our judgement, two to three hours is the realistic band for an 8 cm candle bought as a home or gift candle. An evening, not an afternoon.

If you settle on 3 hours, the 2-hour reading is still worth taking. It tells you what the shorter-session customer sees. In our worked example the C2 candle is at 63% coverage at 2 hours on burn 3. A customer who always stops there is looking at a lot of solid wax. What session length to print, and how much to trust people to follow it, is the policy question the soot-and-label post owns. Whether a wick survives a customer who stops the first burn at an hour is the short-first-burn stress test.

Coverage: turning a pool width into a share of the surface

Coverage is the melted share of the wax surface: (pool width ÷ inside width)², times 100. It is geometry, π r² on both, not a burn prediction. It turns "the pool is nearly to the edge" into a number you can compare between burns, wicks and jars. First, which 8 cm? The title does not say inside or outside. CSI lists no 8 cm jar (its nearest glass listings are at 7 cm and 10 cm), so this is your own jar. Measure the inside at the rim, the middle and about 1 cm above the base; the method is in the external-versus-internal diameter post. We read 8 cm as the outside with a 0.35 cm wall, which is our assumption: 7.3 cm inside, a 41.85 cm² surface. Read as 8 cm inside, it would be 50.27 cm².

Look at the 3 mm and 5 mm rows. A 5 mm ring is a thin line in a photo, but it is a quarter of the surface. That is why "it nearly reaches the edge at 3 hours" isn't a pass. In our worked example the C2 candle's burn-3 pool at 3 hours is 68 mm, a 2.5 mm ring, 87% coverage, with 5.54 cm² of the 41.85 cm² still solid.

Measuring coverage at 2 and 3 hours

The C2 candle is the one in the question. In a four-hour test it reached the glass at about 3 h 50 min, 124% of the 3.1-hour rule of thumb, 0.7 h over it. Clean flame throughout. Stopped at 3 hours, it climbs from 72% to 87% over three burns and then stalls, with a 2.5 mm ring. At 2 hours, about the length of a weekday evening, it is at 63%: a 26.42 cm² pool on a 41.85 cm² surface, 15.43 cm² still solid. That candle will tunnel in some customers' homes. It will not tunnel on a bench where it gets four hours.

The coverage rule, and why a clean flame does not rescue it

Why test one rung up when the flame is clean? Because "clean" and "big enough" measure different things. A flame too small for its jar is often a very tidy flame. What a small wick costs is not soot but surface: less pool, so less hot throw and a ring that can grow. The C2 candle above is exactly that case. Nothing about its flame would tell you it is underpowered for a 3-hour session.

The oil is part of it. A dense, vanillin-rich gourmand like Cinnamon Vanilla is the kind of oil commonly linked to needing more wick than a light herbal or citrus in the same jar (a tendency, not CSI data). If the same C2 reaches 90% at 3 hours with your lavender and not with this oil, the oil is the reason, and the step-up belongs to this fragrance only. Splitting oil from wick is covered elsewhere in the cluster, for example the 8 cm lavender-v-sandalwood bracket.

One rung up: C2 against Cotton Braided

Hold everything but the wick: the same wax bag, Cinnamon Vanilla at 8% of total weight, jar batch, 200 g fill (16 g of oil + 184 g of wax), pour temperature, cure and room. Three candles on your current wick (C2 in the worked example) and three on the next rung up. At CSI that is Cotton Braided Wicks Ready Made, whose page says it is "engineered for large-diameter candles exceeding 8 cm". There is no sized, tabbed wick between C2 and Cotton Braided, so this is a full step. That is the honest limit of our range. If Cotton Braided proves too hot for the lower third and C2 too small for the session, the in-between is not something CSI lists.

Per candle, before the jar, that is ₹180.21 on C2 and ₹182.77 on Cotton Braided. The step up adds ₹2.56 a candle in wick, ₹1,280.00 over a 500-candle run. That is real money at volume, and one more reason not to step up unless coverage says so. From nothing, the cash outlay is ₹1,465.60. If you already own the wax, oil, jars and C2, it is ₹120.00 for the Cotton Braided 10-pack. The 100 g bottle leaves only 4 g after six candles, too little for a seventh, so a retest needs a second bottle (₹967.60 for two, if you buy both now). Costs are worked examples and exclude jars, labour, freight, packaging and GST.

Protocol and decision

Pour all six on one day. CSI 464's page states "Pour Temperature 80°C – 90°C"; if your jar's page states no heat rating, pour at the low end and pre-warm clean jars (our judgement). Trim the Cotton Braided at setting; its page gives the wick as 10.16 cm tall. Cure identically, then run burns 1–3 at your printed session and log coverage at 2 and 3 hours exactly as above.

Cotton Braided passing coverage is only half a pass. A bigger wick that fixes the top of an 8 cm candle can overheat the bottom. That full-life comparison is its own post, with its lower-third checks, and we don't re-run it here. What to look for when the flame grows near the bottom is in the lower-third post.

What the care label can and cannot fix

Because coverage is judged at the session you print, the label is part of the result. Two things follow. If your candle passes at 3 hours but not at 2, say 3 hours on the label. And say it on the first burn especially, since a first session that stops early sets the edge the next burns follow. A Candle Care Stickers Roll is one way to carry that line. In the text we captured, the page doesn't show what is printed on the stickers, so ask us for a photo and print your own if the design has no session length.

What the label cannot do is rescue a candle that fails at the session you would honestly print. Printing "burn for 4 hours" to cover a wick that needs four hours to finish is asking every customer to keep a rule most people won't, in our judgement. When a candle only works if the customer follows the label to the letter, the wick is the fix.

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 counts are at 16 g (a 200 g fill) and 12 g (a 150 g candle).

Frequently asked questions