My Vanilla Candle Soots but Lavender Does Not Using the Same Jar and Wick — Should Vanilla Have Its Own Wick Size?

Same jar, same wick, different oil: what changed

Your situation, restated. You make two candles in one jar with one wax and one wick. The lavender burns cleanly. The vanilla leaves soot: black on the rim after a burn, a smoky trail from the flame, or a grey card when you hold one over it. You want to know whether vanilla needs a different wick from lavender, or whether something else is going on. In our worked example both are at 8% in a 200 g candle on C2 in Luxury Soy Wax CSI 464; your figures may differ, and the logic is the same.

A wick burns whatever the pool feeds it. Lavender and vanilla oils are both blends, and they differ. Vanilla-rich oils commonly carry heavier, darker, sweeter materials; vanillin itself is a solid that some oils carry dissolved, and many vanilla blends also include caramel, cream, amber and resin notes. Makers commonly find these oils build more carbon at the wick tip, so the wick caps, the flame grows and smokes, and soot lands on the glass. Light herbal oils such as lavender often burn more easily on the same wick. These are tendencies reported by makers, and they don't hold for every vanilla. Some vanillas burn cleanly and some lavenders don't. That is why the answer has to come from your candles, not from the label on the bottle.

It can go the other way too: in some jars vanilla needs a larger wick than lavender, because a heavy oil can damp the flame rather than feed it (the vanilla-needs-a-bigger-wick post covers that case). Soot is the clue that, in your jar, vanilla is getting more fuel than its flame burns cleanly, so the ladder below tests in the smaller direction. If your vanilla problem is a huge flame rather than soot, that is 6914; if it is mainly a big carbon mushroom, that is 6946.

What CSI's vanilla pages say, and what they don't

If your vanilla came from us, here is what the product pages state in the text we captured on 7 October 2026. Read the vanillin lines carefully. On every page they are about discolouration of light wax, not about burning, and no CSI vanilla page says anything about soot.

Three practical points. First, "vanillin present" or "2.5%" doesn't predict soot; it is the pages' colour warning. Vanillin is one reason vanilla oils are treated as heavier, but your burn test, not a percentage, says whether your wick suits it. Second, Cinnamon Vanilla's "Up to 13%" doesn't let you go past your wax's ceiling: in CSI 464, "Up to 10%" governs. Third, a vanilla blend with oud, spice or tobacco is another fragrance with its own spec; the oud end of that question is oud against citrus (6931). CSI publishes no oil composition, and we can't tell you what is in a blend beyond the page. For lot details to record on your spec, ask us on WhatsApp.

Vanilla soots, lavender doesn't: the causes, ranked

Ranked by how often each explains this exact pattern (our judgement). The first two are about making the candle; the next three about how the two candles were made or compared; the last about how they are burned.

Cause 5 matters more for vanilla than for most oils, because many makers dye vanilla candles cream or brown to hide vanillin yellowing. Dye is another carbon source and can affect the wick (a tendency; no CSI dye page states a dose). If your vanilla is dyed and your lavender isn't, add an undyed vanilla candle, or you are testing the dye too. Causes 3 and 4 cost nothing and catch many false alarms. For the broader question of whether every fragrance in a range needs its own wick test, see the every-fragrance-tested post; for carbon build-up against an unscented control, vanilla carbon against unscented.

The vanilla soot ladder: three arms and a control

Default test candle, labelled: 200 g in your jar; at 8%, 16 g oil + 184 g wax; at 6%, 12 g oil + 188 g wax (load as a share of total weight). Your current wick is assumed to be C2. The ladder differs from the flame-focused vanilla ladder in 6914: it adds a lower-load arm, because soot is often the one symptom a gentler load can fix without a wick change.

Writing a per-fragrance wick spec

Once an arm passes, write it down as vanilla's own specification, separate from lavender's, and pin both where you pour. The aim is that anyone making a vanilla candle six months from now makes the one that passed, not the lavender one with a different bottle.

The last row is the one makers forget. A different vanilla, even one sold as a direct replacement, is a new candle for wick purposes, and so is a new wick supplier; new wick works with lavender but not vanilla covers qualifying replacements across fragrances. If your range is growing past two scents, grouping them by behaviour rather than testing every combination is the systematic soot method (6940)'s territory. Load decisions per scent, independent of soot, belong to the per-fragrance load post.

The buy list and production for two specs

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. If vanilla soots on C2 and tunnels on C1, there is no half-step in our range: arm V3 is the finer adjustment, or ask us on WhatsApp and we will say so plainly.

Argued against our own interest: with your own oils, jars and holders, the test costs ₹1,392.80, and if V3 passes, vanilla's spec uses less of our oil, not more: ₹91.75 of Dark Vanilla a candle at 6% against ₹122.34 at 8% at the 1 kg price, ₹30.58 saved per candle. Two 100 g bottles of vanilla (₹1,580.00) would give more spare than 100 g + 50 g, but you don't need it yet. And because C1 and C2 cost the same per wick at every pack, a second wick spec adds stock, not cost per candle.

The step to production quantity: two specs, two orders

Worked example, labelled: 150 lavender and 60 vanilla candles a month, a 2% wick allowance. Lavender (C2 at 8%): 153 wicks, bought as 100 + 50 + 10 = 160 C2 for ₹1,510.40; oil ₹83.54 a candle at the Lavender 1 kg price. Vanilla, if C1 at 8% wins: 62 wicks, bought as 50 + 20 = 70 C1 for ₹413.00, and 960 g of vanilla a month: one 1 kg pack (₹7,646.00) is cheaper than 500 g + 5 × 100 g (₹7,790.00). Vanilla, if C2 at 6% wins: 70 C2 (50 + 20) for ₹660.80, and 720 g of vanilla: 500 g + 2 × 100 g + 50 g (₹5,830.00). The lower load saves ₹1,835.04 a month in oil at these volumes. Keep the two wick stocks in labelled bins; a vanilla candle wicked from the lavender bin is the commonest way a fixed spec breaks. 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

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

Skip for now: a different vanilla "that burns cleaner" while you test (a new oil is a new spec); Cinnamon Vanilla at its page's 13% in CSI 464, whose ceiling is 10%; any wick pack beyond the C1 20 until an arm passes; dye in the test candles; and Cotton Braided, which goes the wrong way for soot in a jar where C2 is already too much for vanilla.

Frequently asked questions