Why Is My Candle Sweating Fragrance Oil After Pouring?

What those beads actually are — and the two things they are not

Set soy wax is a crystalline solid with fragrance oil distributed through it. The oil is not chemically bonded to the wax; it sits inside the structure the crystals leave behind. That structure has a finite capacity, and the wax manufacturer states it as a maximum load — up to 13% of total wax weight on the Luxury Soy Wax Chunks page, up to 10% on the Luxury Soy Wax CSI 464 page, up to 13% on the Premium Coconut Soy CSI 468 page. Put in more oil than the set wax can hold, or fail to disperse what you did put in, and the surplus goes where surplus always goes: to the nearest surface.

That is the whole mechanism. Beads on a two-day-old candle are surplus oil that the wax structure could not accommodate, arriving at the top surface and at the glass wall, usually helped along by a few degrees of warmth. They feel greasy between finger and thumb, they smell strongly of the fragrance, and they come back after you wipe them off. Nothing is leaking out of the oil bottle and nothing is wrong with the oil.

Two things get called sweating that are not. The first is condensation: in a Mumbai or Kochi workshop in August, a jar that has been in an air-conditioned room and is carried into humid air collects water droplets on the glass, and sometimes on a cool wax surface. Water is thin, odourless and evaporates within an hour on a shelf; oil sits there and smells of itself. The second is a wet sheen in the first twenty minutes, while the surface is still setting. On a fresh pour that is the last of the melt levelling out, and it is meaningless. Judging either of those as sweat sends makers back to the supplier for a new oil when the batch was fine.

The honest version of the bad news: beads are almost never the fragrance oil’s fault, so replacing the oil almost never fixes them. CSI sells fragrance oil. It is still the wrong first purchase here. The right first purchase, if you do not already own one, is a scale that reads in grams — CSI lists a Candle Making Weighing Scale at ₹354.00, with only two units showing in September 2026 — and any kitchen scale with a 1 g resolution will do the same job today.

The first 48 hours, hour by hour

Soy does not finish setting when it stops being liquid. Crystallisation continues for hours after the jar is firm to the touch, and it is during that period that oil which was never properly dissolved gets pushed out of the structure. This is why a jar can look perfect at bedtime and be beaded by morning, and why a jar that looks slightly damp an hour after pouring is often completely dry by the next day.

Work to the clock below. The two readings that matter are the wipe at hour 24 and the look at hour 48. Everything before that is watching wax set.

The 48-hour mark is not arbitrary. The Chunks page asks for a minimum 48-hour cure and says ideally one to two weeks before burn testing, so 48 hours is the first moment the product page itself treats the candle as a candle. It is also long enough for the crystal structure to have done most of its rearranging. Judge the throw far later — the 14-day cure is about scent development, not about beads.

Photograph the jars on a plain background under the same light at hour 24 and hour 48. Phone photographs beat memory, and if you eventually message CSI on WhatsApp about the batch, a pair of dated photographs plus your recipe in grams is worth more than any description you can type.

Sweat, wet spots, frosting or a sink hole? Reading the surface correctly

Before you change a recipe, be sure what you are looking at. Soy container candles produce four surface complaints that get muddled constantly, and three of them have nothing to do with the fragrance load. Diagnosing a wet spot as sweating sends you chasing your oil percentage for weeks while the real cause — a cold jar or a fast pour — is never touched.

The test that separates the first from the last is the wipe-and-wait already described: oil comes back, water does not. The test that separates the first from frosting is touch — bloom is dry and powdery and does not mark a tissue, while oil marks it instantly. And the test that separates sweating from a re-poured top layer is a torch held at a low angle across the surface: a top-up seam shows as a faint ring or line where the second pour met the first, and beads that sit only inside that ring are a top-up problem, not a batch problem.

Six causes, ordered by how cheaply you can rule them out

This list is not a ranking of how often each cause is guilty — no supplier has data for that, and inventing it would be useless to you. It is ordered so that the cheapest checks come first. Five of the six cost nothing but an evening; the sixth costs a remelt.

Why volume measuring produces beads so reliably

Every load figure CSI publishes is a percentage of wax weight. Fragrance oils are not water: a given volume of oil does not weigh the same as the same volume of water, the relationship differs from oil to oil, and CSI does not publish a density for each oil — so there is no conversion you can do on paper. Pouring “two capfuls” or “20 ml” into 200 g of wax means you do not know your load, and the error runs in both directions. Weigh the oil into the melt on a scale that reads to 1 g, or weigh the bottle before and after. The Mini Electric Wax Melter at ₹2,360.00 makes this easier because the pot sits still on a bench, but a scale is the part that matters.

Why a few degrees of storage warmth finishes the job

A batch poured at a borderline load can look clean in a 24°C room in Bengaluru and bead within a day on a shelf above a shop counter in Nagpur. The surplus oil is present either way; warmth softens the surface layer and lets it travel. CSI does not publish a storage temperature for finished candles, so there is no number to quote you — what the wax pages do state is cool, dry storage away from sunlight and moisture for the raw wax, and that is the sensible instruction for cured jars too. Cure on an open shelf, out of direct sun, away from a window that catches afternoon heat, and do not stack sealed cartons of warm jars on top of each other on pour day.

The ceiling arithmetic: which number actually governs your jar

Every fragranced candle has two stated limits: the wax grade’s maximum load and, on most CSI oil pages, a candle usage range for that oil. When they differ, the lower figure governs — always, and without exception. A maker who reads “up to 13%” on the Chunks page and pours 13% of an oil whose page says soy 6–10% has exceeded the oil page by three points and will often see it on the surface within two days. Working inside the safe load is the first thing to confirm and the cheapest to correct.

Now the money. Take the series costing unit: 200 g of CSI 464 at ₹507.40/kg is ₹101.48 of wax. At 8% that is 16 g of British Rose at ₹49.09; at 10%, 20 g at ₹61.36; at 12%, 24 g at ₹73.63. Going from 10% to 12% adds ₹12.27 of oil to each jar — 4.9% of a ₹250.87 candle — and on that grade takes you two points past the stated maximum. Across 50 jars you have spent ₹613.60 to manufacture a defect.

This is the part that surprises people: overdosing is not only a quality problem, it is a margin problem. Fifty jars poured at 12% on CSI 464, in a ₹94.40 jar with a ₹5.90 wick, cost ₹13,770.60 in materials. The same fifty at 8% cost less, cure cleaner and — going by what the load guides and most makers’ test burns show — usually smell no weaker, because past a point the extra oil is not being carried by the flame at all. It is sitting on the glass, which is exactly where you found it.

Diagnose your batch in one evening and one shelf

The protocol below is deliberately small. It uses 600 g of wax in total and answers the only question that matters on pour day: is this batch beading because of the dose, or because of the way the oil went in? Run it on the same wax, the same oil and the same wick as the batch that failed, in the same room. Changing anything else makes the result unusable.

If jar A beads too, your problem is not in this list — it is upstream. Re-check the grade you are using against what its page states, re-check that the oil is the one you think it is, and message WhatsApp with the batch details. It is also worth burning jar A at day 14 anyway, because a batch that beads and throws well and a batch that beads and throws badly are different conversations (add temperature for soy, wick sizing by diameter).

Can you rescue a sweating batch? The honest arithmetic

If the batch beaded because the load was above the governing ceiling, the fix is dilution: melt the jars back down and add wax until the oil is inside the limit. Nothing is lost except time, glass and a fresh cure. If the batch beaded because the oil never dispersed properly, a remelt at the right temperature with a proper stir will often also fix it, because you are giving the oil the chance it never had. What a remelt cannot do is turn 50 jars into 50 jars.

Work through a real batch. Fifty jars, 200 g of CSI 464 each with 24 g of British Rose — 12% of wax weight, two points above that grade’s stated maximum. That is 10,000 g of wax and 1,200 g of oil in the batch. To bring the load to 10% you need the oil to be a tenth of the wax: 1,200 g of oil needs 12,000 g of wax, so you add 2,000 g — 2 kg of CSI 464 at ₹1,014.80.

So the materials cost of the rescue is ₹2,017.80 and you end up with 60 saleable jars instead of 50 — more product, at a lower load, in a batch that will not bead. That is a good outcome. The bad outcome is the calendar: you have spent a pour day and you are 14 days from knowing what the throw is like. In the third week of October, with Diwali on 8 November 2026, that is survivable. In the first week of November it is not, and the honest advice then is to sell the batch as it is with the jars wiped and a card that says the candle may show oil on the surface, or hold it back for the next season.

The pour-day card that stops it happening again

Sweating is a process defect, and process defects respond to written procedure far better than to better ingredients. This is the card. It assumes a container soy grade and a CSI oil whose page states a candle range; adapt the temperatures to the grade you actually use, because the published figures differ from grade to grade.

A last honest note on grades. If your workshop runs hot for months at a time and you keep meeting borderline loads, the answer is usually to drop the load by a point or two and fix the wick, not to buy a different wax. But the published ceilings do differ: Luxury Soy Chunks at ₹599.00/kg and CSI 468 at ₹650.00/kg both state up to 13%, while CSI 464 states up to 10% at ₹507.40/kg. If you genuinely need to run a heavy gourmand at 12%, a grade whose page supports 13% is the correct place to do it — and it still has to sit inside the oil page’s own range (load by fragrance family).

Everything else on the shelf is secondary: a melter at ₹2,360.00 for steady temperature, a thermometer at ₹354.00 so that “80 to 90” is a measurement rather than a hope, and Eco Candle Wicks from ₹118.00 for a 20-pack so that you can wick-test without buying a hundred. The scale and the timer are the two that stop the beads.

Frequently asked questions