My Dark-Coloured Candle Soots More Than the Undyed Version — Could Heavy Dye Loading Be Affecting the Wick?

Dyed soots, undyed doesn't: the symptom, stated exactly

You make the same candle twice: same jar, wax, fragrance, load and wick. One is undyed, or pale. The other is dark: black, chocolate brown, deep burgundy or navy. The undyed one burns clean. The dark one soots more: a darker card held above the flame, a grey band on the rim, a carbon cap on the wick earlier in the session, sometimes a flame that looks lazier or flickers more. You want to know whether the dye load is affecting the wick, and what to change without losing the colour your customers chose.

Pin down four things first. Were the two candles made from one scented batch, split before dyeing, or on different days? How was the dye measured: drops, a "pinch" of chips, or weighed? Was it fully dissolved, with no specks on the base or sides of the set candle? When does the soot start, in which third of the jar, and after how many hours? Check the bottom of the set dark candle against the light: dark specks or streaks are undissolved dye, and they are a process fault before they are a formula one.

How a colourant can reach the flame

A container candle's wick draws liquid wax up to the flame, where it vaporises and burns. Fragrance oil is partly volatile, so some of it leaves the pool surface as scent; that is the hot throw. Candle dye is not meant to evaporate. It is dissolved in the wax, it travels up the wick with the wax, and at the flame it has to burn or be left behind. The tendencies makers commonly report, none of them CSI measurements:

More dye, more residue at the wick. A heavy dose means more non-volatile material arriving at the wick tip per hour. Some makers see the wick cap sooner, and a capped wick burns like a bigger one: taller, less steady, sooty. Undissolved dye is the worst case. Specks from chips that weren't fully melted in, or liquid dye that separated in a cool pot, are particles. Particles can lodge in the braid and choke the wick's draw. Pigments are not dyes. Mica and other insoluble pigments sit in the wax rather than dissolving, which is why many makers keep them out of container wax or on the surface only; that is the decorative-colour question in the decorative-colour post (6934). Dark candles often carry dark oils. Coffee, oud, chocolate and vanilla-rich oils are commonly chosen for dark candles, and they bring their own carbon tendency (oud against citrus (6931)). If your dark candle and your undyed candle use different oils, you have two variables.

The soy factor deserves its own line. CSI's black liquid page says black dye in soy "tends to produce lighter to medium shades of black" and that paraffin, being clearer, gives "more vibrant color payoff". In plain terms: a deep black in soy takes a lot of dye, and that is how many dark soy candles end up overdosed. The lever is not only "less dye"; it may be accepting a charcoal rather than a jet black, or choosing a shade that reads dark at a lower dose. And because no CSI dye page states a dosage or a maximum, there is no printed ceiling to tell you when you have gone too far; your burn test is the ceiling.

The causes, ranked: dose, dissolving, or something else

Ranked by how often each explains "the dyed candle soots, the undyed one doesn't" (our judgement, not CSI data). Two are about how the candles were compared or burned; five are about how they were made.

Causes 2, 3 and 6 cost nothing to remove and are common. Cause 2 is the one to rule out on day one: if you can see specks, no wick change will fix it. Cause 3 matters because without a weighed dose you can't halve it, repeat it or write it into a spec. Cause 5 explains why some dark candles soot and others don't on the same wick: if the undyed candle is already near the edge, a little dye is enough to tip it. If airflow is a suspect, the oversized-wick-against-draught post (6928) separates them. If your dark candle also went from 8% to 10% fragrance when you darkened it, read the 8%-to-10% post (6929) first, because that is a second change.

Measuring dye by weight: the concentrate and the colour ladder

The problem with dye is scale. Take an illustrative dose of 4 g of dye per kg of finished candle (0.4% of the candle's weight); this is a placeholder for your own figure, not a recommendation, because no CSI dye page states one. In a 200 g candle that is 0.8 g of dye. On a scale that reads to 1 g, a 1 g slip is 125% of that dose: you are guessing. The fix is a concentrate: 10 g of dye fully dissolved in 90 g of plain wax makes 100 g at 10% dye. Now one candle's dose is 8 g of concentrate, and the same 1 g slip is 12.5%. Halving the dose means weighing 4 g instead of 8 g. Liquid dye is sold by the millilitre on some CSI listings; weigh it anyway, because the pages don't state a density.

The burn test: undyed, current dose, lowest accepted dose

Default test candle, labelled: 200 g at 8% = 16 g oil + 184 g on the wax side, in your jar, on your current wick. Each arm is three candles, 600 g of mix: 48 g of oil, 528 g of plain wax, and 24 g of additions. Arm B's 24 g is all concentrate (the full illustrative dose, 0.8 g of dye per candle); arm C's is 12 g of concentrate + 12 g of plain wax (half dose); arm A's is 24 g of plain wax. That way every arm has exactly 8.0% fragrance and the same wax, and only the dye differs.

Cure all nine together for at least 3 days (many CSI soy oil pages say 48–72 hours before burn testing). Burn them side by side in still air, fans and AC off, trimmed to a measured 5–6 mm before each session and never during, in 4-hour sessions to about 1 cm from the bottom. At hours 2 and 4 hold a white card briefly well above each flame; at hour 4 photograph the wick beside a ruler; after each session photograph the cooled rim. Judge the top, middle and bottom thirds separately. Expect about 11 days from the colour ladder to a full-life result.

The buy list and the production dose

Argued against our own interest: the most likely outcome of this page is that you buy less dye than before, and no wicks. That is fine. If arm C passes, you keep your wick spec and halve a material. If C fails and A passes, the next step is a choice between a lighter colour and one rung down on the wick, and that choice is 6933's, not this page's. On wicks, be clear about what that would mean at CSI: our sized, tabbed wicks are three rungs, C1 → C2 → Cotton Braided, with no size between C1 and C2 or between C2 and Cotton Braided and one wooden-wick listing with no widths. If C2 soots at your lowest acceptable dose and C1 tunnels, tell us on WhatsApp; we will say plainly that we have nothing in between.

If you already hold dye, wicks and holders, the test costs ₹1,274.80 in wax. Of the 100 g concentrate, the test uses 46 g; keep the other 54 g, labelled with its strength and date, for the first production batch.

The step to production dose

Worked example, labelled: 200 dark candles a month. At the illustrative full dose that is 160 g of dye a month; at half, 80 g, made up as 800 g of concentrate. Make concentrate in batches you can weigh accurately and label each one. How many bottles that is depends on your dye's density, which the pages don't state, so weigh your first bottle and note its net grams. In money the dye is small: in our worked example, at the Blue 100 gm price of ₹9.00 a gram, 0.8 g is ₹7.20 a candle and 0.4 g is ₹3.60. The saving is not the point; the clean burn is. Wicks don't change unless the burn test says so, and if it does, buy for a month only after three full-life candles pass, not a 1,000-pack.

What to buy, and what to skip

Five items, chosen on page statements, price and stock on 7 October 2026. No CSI dye page states a dosage, a maximum or burn data, so nothing here is ranked on burn performance, and each card says when not to buy.

Skip for now: new wicks before arm C is read; mica or other pigments mixed into container wax to deepen a colour; a second black "that burns cleaner" mid-test; and the out-of-stock "Black Candle Dye Liquid" listing, which is a different product from the in-stock Candle Dye Liquid | Black.

Frequently asked questions