My Lavender Candle Is Stable at 10% but Vanilla Sweats in the Exact Same Wax — Can Fragrance Chemistry Change How Much Oil the Wax Can Retain?

Why the same wax can hold one oil and push out another

When two candles are made the same way and only one of them sweats, the difference is almost always in the oil. A fragrance oil is not one substance but a blend — fresh top materials, a heart, heavier base notes, sometimes solvents or carriers. How readily a soy wax takes those materials into its structure as it sets depends on what they are and how much of each is there. A wax can be comfortable with 20 g of one blend and crowded by 20 g of another.

That is why the useful unit is the wax-and-oil pair. Your lavender in that wax has a ceiling somewhere at or above 10%. Your vanilla in that same wax has a ceiling somewhere below it. Neither number transfers to the other.

What CSI's pages say about the two families

Look at what the product pages actually state, and only that. The Lavender page describes a single-note lavender with a soft musk base, a stated candle load of 6–10%, and the page's own rating of 'Strong'; flashpoint and vanillin are not published. The Dark Vanilla page describes rich vanilla and spice over caramel, cream, amber and musk, a flashpoint of approximately 191–193°C, and says vanillin is present without giving a percentage. It also says the oil will yellow white or light-coloured wax and that this does not affect performance. Vanilla Bean publishes 2.5% vanillin and a flashpoint of 203°C.

So the two oils you are comparing are, on paper, different kinds of blend. Makers commonly find that rich vanilla, gourmand and resin-heavy oils want a lower load in soy than light herbal or floral ones, but that is a pattern, not a law, and no page puts a number on it. Treat the vanillin note as a reason to test vanilla separately — and to expect colour change — not as a diagnosis of the sweat.

A five-candle test that separates oil from wax

The aim is to reproduce the sweat, then find the step where it stops, with a control in the same batch of wax that should stay dry. All loads below are % of total candle weight, and all are suggested starting points.

Totals: vanilla 20 + 18 + 16 + 14 = 68 g; lavender 20 g; wax 180 + 182 + 184 + 186 + 180 = 912 g. A 100 g pack of each oil and a 1 kg pack of wax cover the whole test with a little left over for a repeat of the winning step.

Reading the five candles honestly

Four outcomes are likely, and each points somewhere different. Decide in advance what each one means, so you are not tempted to read a borderline result kindly.

Lavender dry, vanilla dry from 9% or 8% down

The most common pattern. The vanilla simply has a lower ceiling in this wax. Approve the highest dry step, then repeat that one step in two more candles to confirm it was not luck. Your wax purchase does not change.

Lavender dry, vanilla sweats at every step

Now the pair is the problem, not the percentage. A load that still weeps at 7% is unlikely to be fixed by going to 6%, and a vanilla candle at very low load may also disappoint lit. This is when a second wax or a second vanilla oil is justified.

Lavender sweats this time too

Stop and look at the day itself before blaming either oil. A cooler addition temperature, a shorter stir, a different wax lot or a warm curing shelf can all push oil out of a candle that used to be dry. Repeat the control with your recorded process before you read anything into the vanilla candles.

Vanilla dry at 10% this time

Then the original sweat probably came from process or storage, not the oil. Compare your batch sheets for the failed candles with this test: temperatures, stir time, cooling spot and where the candles were kept. The difference is your cause.

What the lower vanilla load does to your cost

Vanilla is the dearer oil here, which makes the honest answer easier to swallow. Using the 1 kg prices as the production rate:

Vanilla at 10%: 20 × ₹7.646 = ₹152.92 of oil per 200 g candle. Vanilla at 8%: 16 × ₹7.646 = ₹122.34. Lavender at 10%: 20 × ₹5.221 = ₹104.42.

If vanilla passes at 8%, each candle carries ₹152.92 − ₹122.34 = ₹30.58 less oil than your current formula, and it stops sweating. These are oil costs only — wax, jar, wick, packaging, labour, shipping and GST come on top. If vanilla needs to drop to 7%, that is 14 × ₹7.646 = ₹107.04 of oil, still less than the failed recipe.

The test purchase and what comes after

For the comparison: 100 g Lavender ₹540.00, 100 g Dark Vanilla ₹790.00 and 1 kg Luxury Soy Wax CSI 464 ₹507.40. Total: ₹540.00 + ₹790.00 + ₹507.40 = ₹1,837.40, plus whatever jars and wicks you already use. If you want two candles per step — better evidence — add a 500 g pack of CSI 464 at ₹260.00 and a second 100 g of vanilla.

If your vanilla is from another source, keep the same plan and swap it in for Dark Vanilla. The lavender control still tells you whether the wax batch and process are sound.

If vanilla sweats even at 7% while lavender stays dry at 10%, you have learned something different: this vanilla and this wax are a poor pair. Then it is worth trying the same vanilla in a second wax — Chunks or CSI 468, for example — or trying a different vanilla oil in your current wax, before you approve anything. Only once a vanilla load is dry through your full observation window should you order its 500 g or 1 kg pack for production.

Record the outcome on your formula sheet as two separate lines: lavender, approved at 10% in this wax; vanilla, approved at its own tested load. Production staff should never be able to apply one oil's number to the other.

Where this page fits

CSI already has guides that cover parts of this question. This page answers one sweating decision; these cover the neighbouring questions:

Three things to buy for this fix

Frequently asked questions