My Candle Soots More After I Increased Fragrance from 8% to 10% — Could Fragrance Load Be Contributing to Incomplete Combustion?

First, is 10% inside your ceilings?

Restate the situation exactly. The same jar, the same wax, the same wick and the same oil. The only change: fragrance up from 8% to 10%. Since then the candle leaves more soot, on the rim, on a white card held above the flame, or both. You want to know whether the extra two percentage points caused it and what to do about it. Two limits sit over every load decision, and both come before any soot test. The wax maker's stated maximum and the oil page's stated candle maximum: the lower of the two governs. These are formulation limits, not regulatory ones, and the maximum is not the best-performing load.

CSI's own convention, and the one this series uses, is load as a percentage of the total candle: 10% of a 200 g candle is 20 g of oil and 180 g of wax. On that reading the CSI 464 page's "Up to 10%" allows your 10%, with nothing to spare. The page doesn't say which basis it uses, and the conservative reading is the stricter one: 10% of the wax weight is only 9.09% of the total. So a 10% candle on CSI 464 is at the edge either way. That doesn't prove the load causes your soot. It does mean you are formulating at the limit, and the test below includes 9% because it is the highest rung inside both readings. If you want the basis confirmed, ask us on WhatsApp. Load maths in depth, including the wax-weight against total-weight trap, is the load-reduction post in the formulation series.

How more fragrance can mean more soot

A container candle's wick draws liquid from the pool up to the flame, where it vaporises and burns. At 8% that liquid is mostly wax with some fragrance; at 10% there is 25% more fragrance per candle. Fragrance oils are blends of many materials with different weights and volatilities, and they don't all burn like wax. The tendencies makers commonly report, none of them CSI measurements:

More carbon on the wick. Heavier and darker components, common in gourmands, vanilla-rich and resinous oils, tend to leave more residue at the wick tip, which builds a mushroom. A capped wick behaves like a bigger one, so the flame grows and smokes later in the session. A less tidy flame. Some makers see a higher load make the flame smaller or lazier; others see it flicker and smoke. Either can mean less complete combustion. A wick that was just right becomes marginal. A wick chosen at 8% was chosen for that fuel. At 10% the same wick may be slightly over-powered or slightly clogged, and either way the margin you had has gone. Oil the wax doesn't fully hold. At the top of the range, some oil may not stay fully bound in the wax, and a load the wax struggles with commonly shows as sweating or a wet surface; some of it may reach the flame unevenly.

Notice that these can push the wick either way. That is why this page tests both directions: 10% on a smaller wick (arm C) in case the load has made your wick too strong, and 9% or 8% on the same wick (arms D and A) in case the load itself is the problem. If your issue is a mushroom rather than soot on the glass, the same logic applies with a carbon focus in the 10%-against-7% mushroom post (6948). For the reverse case, a bigger flame after lowering the load, see 6915. And for the general version of this question, why a higher load can change wick performance covers it for a 7 cm jar.

More soot at 10%: the causes, ranked

Ranked by how often each explains "more soot since the load went up" (our judgement). Three are how the candle is being burned or measured; the rest are how it was made.

Cause 3 is common and costs nothing to remove: if your 8% candles were poured last month and your 10% candles last week, you are comparing two batches, not two loads. Cause 5 is the burning side. A 10% candle that only soots when it has burned five hours untrimmed under a fan has a different problem; the two-hour smoke post (6922) and trimmed but still sooting (6924) cover those. Cause 2 is a tendency: dark gourmands like coffee and chocolate, vanilla-rich and resinous oils commonly carry load effects more visibly than light florals or citrus. If one scent is the problem and the rest are fine, vanilla against lavender and oud against citrus take that up scent by scent. When increasing the load also made the hot throw worse, not better, the combustion-interference post explains why.

The four-arm test: 8%, 9%, 10%, and 10% one rung down

Default test candle, labelled: 200 g in your jar. At 8% that is 16 g oil + 184 g wax; at 9%, 18 g + 182 g; at 10%, 20 g + 180 g (load as a share of total weight). Luxury Soy Wax CSI 464 is the worked-example wax ("Pour Temperature 80°C – 90°C"). Your current wick is assumed to be C2; adjust the rung names if it isn't.

Load or wick: choosing between the arms that pass

When both C and D pass, the decision is no longer about soot. Pool: C carries 20 g of oil on a smaller wick; if its pool is late reaching the glass, D wins. Throw: your blind sniff decides whether 10% is noticeably stronger than 9%. Money: in our worked example with Roasted Coffee at its 1 kg price (₹59.81 of oil at 8%, ₹67.28 at 9%, ₹74.76 at 10%), 10% costs ₹7.48 more a candle than 9% and ₹14.95 more than 8%. At 200 candles a month, 10% against 8% is ₹2,990.40 a month in oil alone. A smaller wick saves ₹3.54 a candle going from C2 to C1, which is less than a tenth of that.

Argued against our own interest: we sell fragrance oil, and the cheapest way out of this is often to use less of it. If arm D or A passes and the blind panel can't tell it from 10%, stop buying the extra 2%. The two percentage points cost money every month and bought you the soot you are now fixing. Change control applies when you settle on a new formula: record the load, wick, wax and oil lot (redesigning around a lower effective load).

The buy list, oil grams and the production step

CSI's sized, tabbed wicks are three rungs: C1 → C2 → Cotton Braided. One rung down is Cotton Braided → C2, C2 → C1, or a C2 pair → a C1 pair. There is no sized, tabbed wick between C1 and C2 or between C2 and Cotton Braided, no smaller sized, tabbed wick than C1, and one wooden-wick listing with no widths. That is exactly why arm D exists: moving the load by one percentage point is a smaller step than any wick change we can offer. If C2 soots at 10% and C1 tunnels, ask us on WhatsApp; we will say plainly that we have nothing in between.

On oil packs: three 100 g bottles (₹1,170.00) would leave more spare; the 500 g (₹1,890.00) is cheaper per gram (₹3.78 against ₹3.90) but leaves 278 g you don't need until a formula passes. With your own oil, holders and C2, the test costs ₹1,392.80.

The step to production quantity

Only after an arm passes on three full-life candles. Worked example, labelled: 200 candles a month, a 2% wick allowance, so 204 wicks. If D (9% on C2) wins: 210 C2 wicks (2 × 100 + 10) for ₹1,982.40, and 3,600 g of oil a month, bought as 3 × 1 kg + 500 g + 100 g for ₹13,494.00 (four 1 kg packs would be ₹14,952.00). If C (10% on C1) wins: 220 C1 wicks (2 × 100 + 20) for ₹1,298.00, and 4,000 g of oil, 4 × 1 kg for ₹14,952.00. If you return to 8% on C2: 3,200 g of oil, 3 × 1 kg + 2 × 100 g for ₹11,994.00. Check per-gram prices for your own oil: on some oils a 1 kg pack costs more than two 500 g. Reorder point = daily use × your lead time + safety stock; CSI publishes no lead times, so ask on WhatsApp.

What to buy, and what to skip

Five items, chosen on page statements, price and stock on 7 October 2026. CSI publishes no soot, flame or load-performance data, so nothing is ranked on performance, and each card says when not to buy.

Skip for now: a second wax while the load question is open; any 1 kg oil pack until an arm passes; a new oil "that burns cleaner" mid-test; any 500- or 1,000-wick pack; and Vybar or other additives, which add another variable to a test that should change only the load.

Frequently asked questions