My Candle Produces Soot Even Though I Trim the Wick Before Every Burn — Should I Test a Smaller Wick?

What "soot even after trimming" is telling you

Your symptom has two parts. The wick is trimmed before every burn, so it starts at a sensible 5–6 mm (one CSI trimmer page says 1/4 inch). The candle still soots: black marks above the pool, a dark rim, or a thin black trail from the flame. So the starting length is not the variable driving it. Write down three more facts before you change anything. When does the soot start: from lighting, after an hour, or only late in a long burn? Is it all round the rim or on one side? And which fragrance, load and dye are in this candle? Those answers decide which of the checks below can be skipped.

Soot is unburned carbon. Commonly, a flame sheds it when it takes up more fuel than it burns cleanly (a wick too strong for its candle) or when moving air breaks its shape (a fan, AC vent or open window). A heavy oil, a high load or a lot of dye can tip a borderline wick into either. Trimming closes only one route: a long wick at lighting.

A wisp of smoke when the flame is put out is extinguishing smoke, normal and separate from soot during burning (how to tell the two apart). Marks that build over burns, or a dark trail while lit, are combustion soot: this page. If the marks appear only near the bottom of the jar, start with the soot hub (6921). If the smoke appears at a set hour of a burn and you need to reproduce it on demand, the reproduction test (6926) sets that up.

Trimmed and still sooting: the causes, ranked

The ranking is our judgement for your exact pattern and the cheapest order to test. Signs are common observations, not CSI test data.

Cause 1 sits at the top because of what you have already ruled out. If the wick were simply long at lighting, your trimming would have cured it. A wick that is the right length at lighting and the wrong length later is a wick being outrun by its own pool. This is commonly linked to over-wicking: the wick melts wax faster than its own tip burns away, so more wick stands proud as the wax level drops. That is a tendency observed by many makers, not a CSI measurement. The next section shows how to see it with a ruler.

Causes 2 and 3 can mimic cause 1, so they are checked before any money is spent. Cause 4 is a policy question: soot only after four hours owns it. Dye has its own post. Cause 7 takes a ruler: many makers find their "5–6 mm" was nearer a centimetre.

The drift check: does the wick grow back mid-burn?

This check owns the part of your question that trimming cannot reach. Light a candle trimmed to a measured 5–6 mm in still air and do not touch the wick again for four hours. At 0, 1, 2, 3 and 4 hours, record the exposed wick length (from the pool surface to the tip), the mushroom (cap) width, the flame height, the pool depth and whether you can see smoke during a 30-second watch. Photograph the wick beside a ruler at each reading. Repeat on two more candles from the same pour. Three candles separate a wick from one odd candle.

Read it in one line: the current wick grew 9 mm in a session and sooted; one rung down grew 3 mm and did not. Your numbers will differ. Set the limits before you light the first candle, so the results cannot move them. For example, write "exposed wick no more than 10 mm and cap no wider than 3 mm at hour four", and treat any visible smoke trail as a fail. These are labelled example limits for your own protocol, not industry figures. CSI publishes no wick-growth, mushroom or soot data.

Customers trim once, if at all, and never at hour two. Whether a care label can carry that burden is the trimming-versus-formulation question (6957); this page assumes you would rather fix the wick. If you want to see how much margin your candle has for a lazy trim, the real-world wick-length test sets that up, and the cost case for a smaller wick against a trimmer in the box is the smokes-when-long post.

Air and load: the two checks before any wick

Check 1: the still-air re-burn

Take a candle from the same pour, trim to a measured 5–6 mm, and burn it with every fan and the AC off, doors and windows closed, well away from vents. Log it exactly as above. If the soot that appears in your living room does not appear here, the candle is reacting to air, and a smaller wick would treat the wrong cause; ceiling fans and AC vents commonly make a correct wick flicker and soot. Placement is the airflow post (6927)'s territory. If you are not sure whether you are looking at a draught or an oversized wick, the two-arm test (6928) separates them.

Check 2: the load audit and a light-oil control

Work out your load the way CSI does: oil grams ÷ total candle grams. A common slip is adding oil "at 10%" to the wax weight. Worked example: 20 g of oil stirred into 200 g of wax is 10% of the wax weight but 9.09% of the finished candle. Then compare it with two ceilings, and the lower one governs. One is the wax page's stated maximum: Luxury Soy Wax Chunks "up to 13% fragrance load capacity", Luxury Soy Wax CSI 464 up to 10%, Premium Coconut Soy Wax CSI 468 up to 13%. The other is the oil page's stated candle load, where it states one. The wax pages do not all say which basis they use. Read cautiously as a share of wax weight, 10% is 9.09% of total weight and 13% is 11.50%. These are formulation limits, not regulatory ones. If you are near or over either line, bring the load inside both and re-test before you touch the wick. Whether more oil makes more soot is the 8% to 10% post (6929)'s question.

Then pour three candles of a light oil at 8% on your current wick and burn them beside your problem candle. Eucalyptus Spearmint is a reasonable control: its page states "up to 10% for soy wax" and "No vanillin" (the page's claim), and 50 g covers three 16 g candles with 2 g to spare. If the light oil candles also grow back and soot, the wick is too strong for the jar whatever the scent, and the ladder will tell you by how much. If they burn clean, your oil is the variable, and that oil needs its own wick on its own ladder. Per-fragrance wick specs are normal; heavier, vanilla-rich and resinous oils commonly build more carbon than light ones. That is a tendency, not CSI data.

The ladder test: current rung, one down, and a pair

If the drift log says the wick is too strong, and the air and load checks are clear, this is the test that decides the replacement. It uses the series' default test candle, labelled: 200 g at 8% = 16 g fragrance + 184 g wax (load as a share of total weight), poured in your jar with your own oil. Keep your own fill if it differs, and scale the grams. In a 7 cm jar a full pool commonly takes about 2.8 hours by the rough one-hour-per-2.5 cm rule, and in an 8 cm jar about 3.1, so 4-hour sessions give every candidate a fair chance to reach the glass.

Own a cutter and holders, and skipping the control? The ladder alone is ₹1,316.00 (C1 20-pack plus 2 × 1 kg Chunks for 1,656 g), or ₹118.00 if you have the wax; 2.5 kg of Luxury Soy Wax CSI 464 instead is ₹1,274.80. The Chunks page says "Clean burning with minimal soot": a claim about the wax, which no wax can make true over a wick too strong for its jar.

The CSI wick buy list and the step to production

CSI's sized, tabbed wicks are three rungs: C1 → C2 → Cotton Braided. "One size smaller" means 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, and none smaller than C1. There is one wooden-wick listing with no widths. The braided wicks by the metre (Eco Braided Wicks, the Braided Wick Roll with 100 tabs) and the Eco Candle Wicks Loose (page: "designed specifically for 2 inch diameters") are cut-to-length constructions, not half-sizes. If your ladder lands between two of our rungs, we would rather tell you than sell you the wrong one: ask on WhatsApp, test a pair, or make the fine adjustment on the oil or load side.

Argued against our own interest: you may not need new wicks at all. If the still-air burn is clean, your spend is nothing. If the load audit puts you over a ceiling, fixing the load costs less than any wick. And because C1 (₹5.90), C2 (₹9.44) and Cotton Braided (₹12.00) cost the same per wick at every pack, a bigger pack gives you no saving, only stock. Stepping C2 → C1 saves ₹3.54 a candle, while a C1 pair costs ₹2.36 more than one C2. Both are side effects, never the reason to choose.

The step to production quantity

Once a rung has passed on three full-life candles, production wicks = candles × wicks per candle × (1 + a defect allowance). Worked example, labelled: 250 single-wick candles a month with a 3% allowance needs 258 C1. Because C1 is flat per wick, buy the packs closest to the need: 2 × 100 + 3 × 20 = 260 for ₹1,534.00. The same month as 100 + 100 + 50 + 20 is 270 wicks for ₹1,593.00. A 500-pack (₹2,950.00) costs the same ₹5.90 a wick, ties up ₹1,416.00 more cash and leaves 242 on the shelf. If C2 passes instead, 260 C2 are ₹2,454.40. If the C1 pair passes, 515 wicks round up to 520 (one 500 + one 20) for ₹3,068.00. Reorder point = daily use × your lead time + safety stock. CSI publishes no lead times, so ask on WhatsApp before you set yours. A later change of wick lot or supplier is a formula change: retest it (why a good first burn after a size change is not enough).

What to buy for the ladder, and what to skip

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

Skip for now: any 500- or 1,000-pack before approval; a new "cleaner" oil while the wick is unresolved; Performance Wooden Wicks as a soot fix (one width, nothing narrower); and the dearer cutters if the ₹177.00 one reaches your wick. Centring bars (₹118.00 for 5, one 4 mm hole each) are worth having if any test candle's wick sits off centre.

Frequently asked questions