My 10 cm Jar with Two Wicks Still Leaves Wax Around the Perimeter — Should I Increase Wick Size or Change Their Spacing?

Why a pair leaves wax where it does

A single centred wick is the same distance from every point of the glass, so whatever it fails to melt is left as an even ring. Two wicks on a line through the centre are not. In CSI's 10 cm listing, the Apothecary Glass Candle Jar, read as the outside with a 0.35 cm wall (our assumption), the inside is 9.3 cm and the surface 67.93 cm². Put two wicks 4.65 cm apart on a diameter, the halfway starting point from the 10–12 cm hub, and each is 2.33 cm from the glass at its own end of the line. The two points on the glass at right angles to the line are 5.20 cm from both wicks: 2.2 times as far, and further than the 4.65 cm a single centred wick would be.

That is not a flaw in your wicks. It is what a pair on a line looks like, and the diagram's lower chart shows it for three spacings: every curve has its peak at 90°. The formula is √(r² + (s/2)²), and it is smallest when the spacing s is zero, when it equals the radius. So no pair on a line, at any spacing, gets nearer to those two points than one centred wick does. What the pair does instead is put heat close to the glass at the line ends and let the merged pool spread sideways from there. Whether it spreads far enough depends on power, wax, oil and session length, which is why the shape of what is left behind is informative.

Three shapes are worth knowing. An even ring means the merged pool is behaving like one wide source that stops short everywhere: power. A ring heaviest at right angles means the pool reaches the glass where the wicks are close and stops where they are far: geometry. A ring heaviest at the line ends means the two flames are crowded into the middle, heating each other and the strip between them while the ends of the line lag: spacing. A one-sided ring is usually not about the wicks at all.

The oil plays a smaller part in this question. The worked example uses Fresh Strawberries at 8%, inside its page's "Recommended fragrance load 6-9% for candles" and below CSI 464's "up to 10%"; a light, fruity top on a vanilla-sugar base is not the kind of oil commonly linked to stubborn rings (a tendency, not CSI data). If your oil is heavy and the wicks mushroom, read the ring anyway, then see the vanilla double-wick post.

Reading the ring at eight points

Use the candle you already poured. A twin from the same pour is better still: if both show the same shape, it is the build, not the candle. Change nothing between burns.

Inside diameter matters here because every distance is measured from it. The Apothecary page gives outer figures only and describes a "Curved silhouette", so measure the inside at the height your ring sits, not just at the rim; the full method is in the external-versus-internal diameter post. If your own jar measures 10 cm inside, the halfway spacing is 5 cm and the distances become 2.50 cm at the line ends and 5.59 cm at right angles: the same shape, slightly larger.

Candle A's even 4 mm ring is 11.18 cm², 16% of the surface. Candle B's ring looks smaller, but if its pool is an ellipse touching the glass on the wick line and 5 mm short at right angles, the solid wax is 7.30 cm², 11% of the surface; if that share stayed solid all the way down a 500 g candle, it would be about ₹25.10 of CSI 464 by weight. Both are worth fixing; they are not fixed the same way.

The ring-shape map: five shapes, five next steps

Classify after burn 3, on the readings from all three burns. The thresholds are our judgement, chosen because they can be read off a photo with a ruler; set your own if you prefer, and keep them fixed.

Two shapes can combine. A candle with an even ring and a deep midpoint between the wicks is short of power and crowded at once; widen first, because spacing costs nothing, then reread. A ring that is heaviest at right angles on one side only is a one-sided ring first: fix centring or the draught and read again. If you need to separate spacing from size in a single designed test because the shape will not settle, that is the eight-candle 2×2, and the strip of solid wax that sometimes stays between two wicks is the between-the-wicks post.

The right-angle ring: what spacing can and can't do

Here is the whole distance map for three spacings in a 9.3 cm-inside jar. It is geometry, every figure declared, and it is the reason the right-angle ring is called geometry on this page.

Move a 4.65 cm pair 1 cm closer and the right-angle points come only 2.0 mm nearer, while the line ends move 5.0 mm further away. Move it 1 cm apart and the line ends come 5.0 mm nearer while the right-angle points move 2.4 mm further away. In other words, for a right-angle ring, widening is the wrong direction and narrowing buys a couple of millimetres at the price of a cooler line end and a hotter middle. If your ring at 90° is 5 mm, spacing alone will not close it.

First, check the spacing you actually got

Measure centre to centre at the wax surface. If it is more than about 5 mm wider than you planned, the right-angle ring is partly self-inflicted and the first move is a re-pour at the planned spacing, with both tabs stuck down and the tops held until the wax sets. A pair whose midpoint is off the jar's centre also shows up here: shift a 4.65 cm pair 5 mm along its line and one wick ends up 55% further from its wall than the other, which reads as a one-sided ring, not a right-angle one.

Then, decide what the ring is worth

The option that genuinely changes the geometry is a third wick, because three wicks on a triangle bring a wick nearer to more of the glass. In a 10 cm jar that is often more total heat than the surface needs (the hub's configuration starter shows each of three would serve about the whole pool of a CSI 6 cm listing), so treat it as a planned test, not a fix. Spacing three wicks is the triangle geometry post; for 11 cm jars, where three wicks become a more common choice, the 11 cm two-or-three post has the coverage arithmetic.

The even ring: a four-candle size re-pour, costed

An even ring all round, the same width at every point and still there after burn 3, is the one shape that asks for more power. Hold the spacing at what you measured in the candle, not what you planned, so the only change is the size. Two candles per arm is the minimum that shows an odd candle. The worked example uses CSI 464 and Fresh Strawberries at 8%: 500 g = 40 g of oil + 460 g of wax in the Apothecary 500 gm.

Pour: all four the same day. CSI 464's page states "Pour Temperature 80°C – 90°C" and the Apothecary page states "Safe for hot-pour up to 85°C", so pour at 80–85°C into clean, pre-warmed jars. Cure: the Fresh Strawberries page states no cure time in the text we captured; the 25 of 102 CSI oil pages that do say 48–72 hours for soy, and many makers wait about a week; keep it the same for all four. Burn and read: exactly as above, eight points after burns 1–3, then to about 1 cm left for any arm that clears the glass. Judge glass heat with a non-contact thermometer if you own one (CSI lists none), never by touch.

Per candle, with stickers, that is ₹604.20 on a C1 pair and ₹611.28 on a C2 pair (₹141.60 of oil, ₹233.40 of wax, ₹212.40 of jar). The pair size changes the cost by ₹7.08 a candle; the jar and the wax are most of it. Costs are worked examples and exclude labour, freight, packaging and GST. From nothing the cash outlay is ₹3,946.80; the 340 g of oil left covers 8 more candles if a cell splits.

A pair that reaches the glass far too quickly once you size up is the opposite problem, and the one-hour step-down post owns it. A pair that takes about four hours with clean flames is the continue-or-change gate.

What to buy, and what to skip

Five items, in the order the reading and the re-pour need them, chosen on page statements, price and stock on 29 September 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