Why Does My Wick Drown in Highly Fragranced Soy Candles?

What drowning is, and what it is not

A candle flame does not burn solid wax. The heat melts a pool; the wick draws that liquid up by capillary action; the flame vaporises and burns what arrives. When the wick is sized well for the jar and the load, supply and burn stay roughly in balance: the pool widens to the glass over the first few hours and stays shallow, and the wick keeps a few millimetres standing clear of the liquid. Drowning is what happens when that balance tips the other way. The flame cannot use the liquid as fast as the pool offers it, the liquid level rises up the exposed wick, the flame gets smaller, and a smaller flame melts and burns even less. Sometimes it recovers. Often it goes out.

"Highly fragranced" enters the story in three ways, and it is worth keeping them separate because each has a different fix. First, a higher load means more of the pool is fragrance oil rather than wax; oil and wax do not necessarily burn alike, and at loads near or over what the wax holds some oil tends to sit less well bound in the wax. Second, heavy, sweet or resinous oils are often thicker, and makers commonly find they need a wick adjustment (that tendency is the subject of Do Heavy Fragrance Oils Need a Larger Candle Wick?). Third, a wick that coped at 6% may be borderline at 10%. None of these makes a well-matched wick drown at a sensible load. All of them push a borderline wick over the edge.

Two symptoms often travel together. A candle that sweats on the shelf and drowns when lit is telling you about load before it tells you about the wick. A candle that burns clean at 6% and drowns at 10% in the same jar and wick is telling you the wick has no headroom. A candle that drowns at every load you try is telling you the wick is simply too small for the jar.

Is it actually the fragrance oil? Four checks first

Makers who see a drowning candle tend to reach for one of two purchases: a different oil or a bigger wick. Before either, run these four checks on the candles you already have. They rule out causes that neither purchase would fix.

Check 1 — the wick length before lighting

A wick trimmed very short gives a small flame from the first second. The Candle Wick Trimmer page advises "Trim your wick to 1/4 inch before every burn", which is about 6 mm; take that as a page's guidance, not a rule for every wick. If you trimmed to a stub, re-test before drawing conclusions. If the wick was left long and mushroomed, that points the other way: too much wick or too much load, not too little.

Check 2 — the wick is upright and centred

An off-centre wick melts one side to the glass while the other stays solid, and the deep side can flood the flame. Look straight down on the unlit candle. If the wick leans, the fix is a holder while the wax sets: the Wick Centering Device is ₹118.00 for five and suits openings under 4 inches by its page. That is a production fix, not a fragrance fix.

Check 3 — the first burn and the room

A first burn stopped at an hour leaves a narrow pool that later burns deepen into a tunnel, and a tunnel is where wicks drown most easily. A commonly cited rule of thumb is that a container candle's first burn should reach a full pool to the edge in about 3–4 hours, or roughly an hour per 2.5 cm of jar width. A fan or open window can also flatten a flame and flood a wick that would be fine in still air. Test in a still room, and let the first burn run long enough.

Check 4 — the load, in grams, on total weight

Load is a percentage of the finished candle. A 150 g candle at 10% is 15 g of oil and 135 g of wax. The common slip is to take the percentage of the wax: 15 g against 135 g reads as 11.1% if you do it that way round, which is why makers think they are at 11% when they are at 10%. A target set above the ceiling is the dearer slip. Worked example: a maker aiming for "12%" melts 150 g of wax and adds 18 g of oil, a 168 g candle at 10.7%; that is already over CSI 464's stated 10%. Weigh on a scale; oils differ in density, so a spoon is not a gram.

"Highly fragranced" against two ceilings

Two figures cap your load, and the lower one governs. The wax maker's page sets one; the oil page, where it states a candle load, sets the other. These are formulation limits, not regulatory limits, and the maximum is not the best-performing load, but going past either is, in our judgement, one of the commonest reasons a heavily scented soy candle drowns. Past what the wax holds, extra oil tends to sweat, pool or burn poorly rather than add throw.

Read the last row carefully. Of the 102 oils in CSI's range, 50 pages state a candle load or range in some form (two of them, Lavender Oud and Amber Oud Royal, only as "Candles (Soy/Paraffin/Blends) 10%" in a perfume usage table) and 52 do not. For those 52 you are working inside the wax's ceiling alone, which is permission, not a recommendation. We would start such an oil at 6% or 8%, not at the wax's maximum. Waxes other than the three above (Natural Soy Pearl Wax, Coconut Wax, Soy Butter Wax, Eco Soy CSI 400) state no load ceiling in the page text we hold, so ask the maker before you push them.

What over-loading costs, besides the burn

Worked example: Fresh Strawberries at 13% in Luxury Soy Wax Chunks is 19.5 g of oil in a 150 g candle, against 13.5 g at the page's 9%. At the 1 kg price that is ₹21.24 of oil per candle the page does not ask for, ₹2,124.00 per 100 candles, and a wick that now has to cope with it. Cinnamon Vanilla at 13% in CSI 464 is 4.5 g over the wax's 10%, ₹20.50 per candle at the 1 kg price. None of that money buys throw if the pool is drowning the flame.

Sizing the wick for a heavily scented pool

Once the load is inside both ceilings, a candle that still drowns is usually asking for more wick. CSI's container wicks form a short ladder: Eco Candle Wicks Thin (C1), then Thick (C2), then Cotton Braided Wicks Ready Made, whose page says they are "engineered for large-diameter candles exceeding 8 cm". The Eco wick page publishes no diameter chart, and jar diameters vary even between jars sold for the same fill weight, so we will not give you a wick size for your jar. What we can give you is the measurement that decides it.

The melt pool is the verdict

On a cured candle, burn for up to four hours and log the pool's width and depth every hour. Commonly cited targets, which are rules of thumb and not CSI data: a full pool to the glass in about 3–4 hours on the first burn, and a pool roughly 0.5–1 cm deep. A drowning candle usually shows the opposite pattern: a pool that is deep near the wick and narrow, with solid wax left at the glass. Worked example: a 4 cm pool in a jar 6.5 cm across inside melts only 38% of the surface, because area goes with the square of the width. That candle is both drowning and wasting fragrance in the ring it never melts.

Some oil pages point you the same way. The Vanilla Bean page says "Wick testing is recommended based on your container size and wax type"; the Palo Santo page recommends "Eco wicks or wooden wicks for an even, clean burn". Neither gives a size, and neither can, because your jar is not on their page. How load changes the wick you need, across a full load × wick grid, is in Does Increasing Fragrance Load Mean I Need a Bigger Wick?; the general question of whether oil weakens a wick is the hub, Can Fragrance Oil Make a Candle Wick Too Weak?.

Cure: judge the wick on the candle you will sell

Cure is not usually the cause of drowning, but it decides whether your test tells the truth. Several CSI oil pages state a soy cure before burn testing: Apple Cinnamon's load line reads "cure 48-72 hours", the Woody Oud and Citrus Lemon pages say "Cure soy wax candles for 48-72 hours before burn testing", and Roasted Coffee's says "at least 48-72 hours". A wick judged on a candle poured last night is judged on a different candle from the one your customer lights next month. Many makers test at day 3, 7 and 14; that is common practice, not a rule. The discipline that matters is this: never compare two candles of different ages. If candle A is 20 days old and candle B is 3 days old, you are testing age, not wick.

The drown test: three candles, one variable each

This test assumes checks 1 to 3 passed and your load is already inside both ceilings. If it was over, re-pour at the governing ceiling first and burn that; in our judgement that alone settles many drowning candles. Worked example throughout: Apple Cinnamon at 10% in CSI 464, in the Clear Glass Jar with Golden Lid (150 gram) (page: outer diameter approx. 7 cm), on a C1 wick, drowning by hour 2.

Oil: up to 42 g (15 + 12 + 15 if B wins), which a 50 g bottle covers with 8 g to spare. Wax: 408 g. Wicks: one C1 you already own, and up to two C2 from one pack. A must be burned alongside at a comparable age; if it is much older than B and C, pour a fresh A as well, which adds one candle and 15 g of oil (and then you need a second 50 g or your own stock).

All figures exclude labour, freight, packaging and GST. The jar is the dearest line in each candle; wash and reuse test jars.

What to buy, in order

Two non-oil fixes first, because in our judgement they fix more drowning candles than any bottle does; then three oils for makers who want a strong candle without living at the ceiling. The oils are chosen on stated basis: a page candle load that sits at or below CSI's soy wax ceilings, stated notes, price per gram and stock on 25 September 2026. CSI has no measured throw or burn data for any oil, so none is ranked on performance. Candle counts assume 150 g at 8%.

We sell the 1 kg of every oil above. For a drowning problem, do not buy it: 50 g covers the whole drown test, and a kilogram of an oil you then pour at 2 points less than planned lasts longer than you expected anyway. At 8%, 1 kg makes 83 candles of 150 g; at 10%, 66.

Frequently asked questions