Why Is My Candle Sweating Oil?

What sweating actually is

Pick the candle up and touch the surface. If your fingertip comes away oily and the residue smells of the fragrance you poured, you are looking at fragrance oil that came out of solution. That is the entire diagnosis in one sentence, and everything else on this page follows from it. The industry word for it is sweating, and the word is unhelpful because it invites the wrong mental model — moisture, humidity, a monsoon problem. It is none of those things.

The useful model is dissolution. Solid wax is a crystalline matrix and fragrance oil sits in the spaces between the crystals, of which there is a finite amount. Every wax has a quantity of oil it has been formulated to hold, stated by the manufacturer as a percentage. Below it the oil is bound — locked into the structure, invisible, released only as the wax melts under a flame. Above it the oil is free, and free oil is mobile: it moves through the setting wax toward the path of least resistance, which is the top surface and the gap between wax and glass, and pools there.

Sweating also sits in a family of related faults that people routinely confuse with each other, and the confusion leads to the wrong fix. Beading on top is oil. A rough, crystalline white bloom on the top or sides is frosting, which is the soy crystals themselves and is cosmetic. A dip or cavity under the surface is a sinkhole, which is a contraction problem, not a solubility one. Wet-looking patches against the glass with no smell are usually adhesion. If you are not certain which one you have, work through the full diagnostic for candle wax problems before you change a single variable — and note that fragrance load pushes on more than one of them at once, which is why oil load also changes frosting behaviour and why an over-loaded candle can tunnel as well.

The four causes, ranked

Ranked by how often each one turns out to be the answer when a maker sends us a photograph of a beaded candle. Work down the list; do not skip to the interesting one at the bottom. The first two account for most cases, the third explains why a recipe that seemed fine suddenly is not, and the fourth is real but rare.

The ceilings, stated correctly

There is no universal soy fragrance load, and the number circulating in maker groups is usually somebody else's wax. The figure that governs your candle is the maximum load stated by the manufacturer of the wax you actually bought. It describes how much oil that specific formulation will hold in solution. Two waxes from the same supplier can carry different numbers, and they are not interchangeable.

Above the ceiling, nothing gradual happens. The surplus does not partially bind, or bind more slowly, or find room during a longer cure. It is sitting in the candle from the moment you pour, waiting to come out, and cure time is simply how long it takes to reach the surface. This is why a longer cure does not fix sweating and sometimes makes it look worse: you have given the free oil more time to travel. It is also why a candle can pass a two-day inspection and fail at day twenty-one, which is precisely when it is being packed for dispatch.

The cause people miss: temperature at addition

If your load is inside the stated ceiling and the candle still beads, this is almost certainly where the fault is. It is missed because nothing about it is visible at the time: a pitcher of wax with oil stirred in at 65°C looks exactly like one stirred in at 85°C, and the jars look the same when poured and the same the next morning. The difference declares itself a week or two later, on the surface.

Eco Soy CSI 400 specifies fragrance addition at 80–90°C with two minutes of stirring, and CSI 464 melts and pours in the same band. Two mechanisms are at work in that specification. The first is energy: the wax needs to be hot enough that the oil goes into solution rather than sitting between wax molecules as suspended droplets. The second is time: two minutes is a long stir, far longer than it feels, and most makers stop at thirty or forty seconds because the mixture already looks homogeneous. It looks homogeneous long before it is dissolved.

Going too hot is a different mistake. Above the band you flash off the lighter, more volatile materials before the wax is even in the jar, which costs you cold throw and the opening of the scent. Over-hot addition does not usually give you sweating; it gives you a candle that smells thinner than the one you designed. The band exists because both edges are real.

Two practical points that matter more than they sound. First, a pen thermometer at ₹354 is the cheapest instrument in the workflow and there is no substitute for it — pouring "when it looks right" is how a batch drifts out of band halfway through. Second, the melt cools while you work, so on a large pour the last jars are being filled from wax that may be twenty degrees below the first. Either work in smaller batches, hold the melt in a mini electric wax melter at ₹2,360 that maintains temperature, or add fragrance to each pouring vessel rather than to the master melt. For the stir itself, bamboo stirring sticks at ₹118 for five are fine for small batches and a CSI Pro Mixer at ₹349 gives a genuinely even two-minute incorporation on larger ones.

The confirming test: four jars, fourteen days

Two causes produce one symptom, so guessing has a fifty per cent hit rate and costs a production run either way. The test separates them. It holds the wax, the oil, the vessel, the wick and the room constant and varies exactly two things — load and add temperature — across four jars. After fourteen days, the pattern of which jars beaded names your cause; if all four are clean, the fault is storage heat or a pairing issue and you can go to cause three or four with confidence.

What over-loading actually costs

Take a 200 g candle in CSI 464 with Sandalwood at 1 kg pricing — ₹5,026.80 per kilogram, or ₹5.03 per gram. At the stated 10% ceiling that is 20 g of oil, ₹100.54. At 14% it is 28 g, ₹140.75. The surplus alone is ₹40.21 per candle, and it is the eight grams that will end up on the surface.

Read the bottom row twice. Over-loading a sixty-jar run costs ₹2,169.00 more in materials than pouring it correctly, of which ₹2,412.60 is surplus Sandalwood that will not stay in the wax. That is the cheap half of the loss. The expensive half is that the ₹17,017.20 of finished stock is not sellable: a beaded candle either fails your own inspection, in which case you have destroyed the batch, or it passes and reaches a customer, in which case you have a refund, a return, a replacement pour and a public review. There is no version of this where the extra oil earns anything.

We should be straight about our position here, because it is a conflict. CSI sells fragrance oil. We have 92 oils in stock and we would make more money if every reader of this page poured at 14% instead of 10% — a customer at 14% buys forty per cent more oil than a customer at 10%, on every batch, forever. We are telling you to buy less of it and to spend an extra two minutes stirring what you do buy, because a maker whose candles arrive sweating stops being a maker, and a supplier whose customers quit has no business. Emma Bechara's review of British Rose at the top of this page makes the same point from the other side: "…you can go for the minimum percentage in candle still it will give amazing smell…". That is a customer telling other customers to buy less of our product, and we have left it on the site because it is true.

So the correct purchase, if your candles are sweating, is not more oil and not a different oil — it is a ₹354 scale and a ₹354 thermometer, ₹708 the pair, which is less than the surplus oil in one sixty-jar run. Then buy oil against a number you can prove. Once the recipe is stable, premium versus cheap fragrance oils and the 2026 oil round-up are the next reads, and how to verify wax quality before you buy covers the incoming side.

Frequently asked questions