I’m Using a 6.5 cm Glass Jar with Soy Wax and 8% Oud Fragrance — Why Does the Wick Struggle After the Candle Has Been Burning for Several Hours?

What changes inside one long session

"The wick struggles" usually means one of four things you can see: the flame gets shorter, it starts to flicker or sputter in still air, a black cap grows on the tip, or the flame sits low over a pool that has become deep. Each has common session-level causes. All are tendencies, not certainties, which is why the clock measures rather than guesses.

Carbon builds up. Over a long burn, some of what the wick draws up doesn't burn off at the tip and collects as a carbon cap (a mushroom). A growing cap commonly chokes the flame: it gets shorter and duller. Heavier, resinous oils (oud, patchouli, amber, sandalwood, dense gourmands) are often reported to do this. How to score a cap in detail is the mushroom-scoring post; here you only note none, small or large each hour.

The pool deepens. A pool keeps deepening after it reaches the glass. A common rule of thumb targets about 0.5–1 cm; in a small jar, three or four hours can take it well past that. A deep pool means more liquid fuel around the wick and less wick standing above it, and a flame that struggles can then sit low and sputter. Reading depth against that rule of thumb, and combining it with flame height and session length, is the 7.5 cm pool-depth post.

The small jar heats up. A 5.8 cm inside (6.5 cm outside on our 0.35 cm wall assumption) is a pool of 26.42 cm². Glass above the pool warms through a session, and the whole candle is warmer at hour 4 than at hour 1. That can speed the pool and the carbon both. Judge it by what you see, flame height, flame position and any darkening on the glass, never by touching the jar; CSI lists no non-contact thermometer.

The session clock, hour by hour

Use two twins from the same pour of your current build (Woody Oud at 8% on your house wick), cured as the page says ("Cure soy wax candles for 48-72 hours before burn testing"; many makers wait about a week). If you want the oil-swap check at the same time, pour two White Royal Oud twins on the same day, same wick, same wax and fill. Our worked example uses 100 g fills (8 g of oil + 92 g of CSI 464) because CSI lists no 6.5 cm jar; use your own jar and fill.

In the example, Woody Oud A falls 47% from hour 1 to hour 4 (3.8 to 2.0 cm; two-thirds of 3.8 cm is 2.5 cm, so hour 4 crosses both lines), while White Royal Oud falls 14%. Both oils deepen the pool at almost the same rate, so the difference is at the wick tip, not in the pool. That is the pattern the swap is designed to show.

Reading the onset hour

Compare the onset with the labelled session length: the longest single burn your label tells customers to do. If you don't have a label yet, the jar sets a floor. By the very rough one-hour-per-2.5 cm rule of thumb, a 5.8–6.5 cm inside needs about 2.3–2.6 hours to pool to the glass, so a 6.5 cm candle can't sensibly be labelled for less than about 2.5 hours; many makers print "burn up to 3–4 hours". Our worked example uses 3 hours.

The third row is the one to handle with care at 6.5 cm, because CSI's rungs are short. One rung down from C2 is C1; one rung up from C1 is C2; there is no sized, tabbed wick between them. Above C2, Cotton Braided's page describes it for candles "exceeding 8 cm", outside a 6.5 cm jar. If the answer lies between C1 and C2, we would rather tell you than sell you the wrong pack: ask us, or try another supplier's in-between size.

Woody Oud against White Royal Oud

Why White Royal Oud as the check? Because it is the closest CSI oil in family and stated load: both pages state 6–10% for candles, and both are in stock in every size on 29 September 2026. The structures differ: Woody Oud is "citrus and spice top notes opening into rich oud and patchouli heart, settling into warm woody notes, amber, and musk"; White Royal Oud opens on "bergamot and saffron", with rose and labdanum in the heart and "creamy sandalwood, ambergris, and soft musk" below. The White Royal Oud page also says "No heavy smoke"; that is the page's claim, and the clock is how you check it on your wick.

Money, honestly: switching oils to fix a late-session struggle costs ₹23.78 more per candle at 50 g prices and ₹28.57 more at 1 kg prices, which is ₹2,857.44 more per 100 candles. A label that says "burn up to 3 hours" costs a sticker. That is why the swap is a check in this post, not a recommendation: it tells you whether the oil matters, and the onset rule tells you whether you need to act on that.

Two cautions. First, don't raise or lower the load in the same test; the confound between oil and load in a wider jar is the 7.5 cm oud-at-10% post. Second, both are ouds: if you need to know whether the family is the issue, add a light floral twin on the same wick as the 6.5 cm vanilla post does with Lavender. A first-level version of the question is why a wick struggles after fragrance oil is added.

When the fix is a sentence on the label

When the onset lands after your labelled session on both twins, in both sessions, the build is doing what you designed it to do. The fix is to tell the customer: burn for up to the hours you tested, extinguish, let it cool, trim before relighting. One CSI trimmer page words it as "Trim your wick to 1/4 inch before every burn". A small jar also runs out fast, so sessions matter: in a worked example where a 100 g candle loses 15 g in a 3-hour session (about 5.0 g an hour; weigh your own on a scale, ₹354.00), it gives about 6 full sessions before the last centimetre.

People do burn longer than labels say. That is a reason to run your full-life test with at least one longer session, and to watch the final third, where heat commonly builds in a small jar. It is not a reason to re-wick a candle that passes the session you promise: a smaller wick that survives five-hour burns may never pool the jar in three.

What to buy, and what to skip

Four items, chosen on page statements, price and stock on 29 September 2026. CSI publishes no wick-performance data, so nothing is ranked on burn results.

Frequently asked questions