My Soy Wax Manufacturer Says the Wax Holds 10% Fragrance but My Candles Still Sweat — What Else Determines Actual Fragrance Compatibility?

What "holds 10%" promises, and what it doesn't

Every wax maximum you read is a formulation limit: roughly, the most oil the maker expects the wax to carry in normal practice. On CSI's container soy, Luxury Soy Wax CSI 464, the page says "Fragrance Load Up to 10%". Premium Coconut Soy Wax CSI 468 and Luxury Soy Wax Chunks say up to 13%. Those are ceilings, not targets, and not regulatory limits. The supplier-claim question in general, and what a bigger number like "12%" does and doesn't mean, is the supplier-12% explainer.

Three things a ceiling cannot promise. It can't promise every oil. Oils are mixtures whose composition isn't published, and a wax holds some less readily than others; one oil can sweat where another sits dry at the same load (the same-load range map shows it across a range). It can't promise your method. How hot the wax was when the oil went in, how long you stirred, how cold the jar was and how fast the candle cooled all change the outcome, as tendencies. It can't promise your shelf. Candles that heat and cool repeatedly tend to sweat more readily than candles kept steady. And behind all three sits a fourth problem that isn't about chemistry at all: the ceiling usually doesn't say what it is a percentage of.

The basis audit: what your 10% really was

That's the whole audit. From those two numbers, work out oil ÷ (oil + wax) and oil ÷ wax. Then read the result against both possible meanings of the maker's 10%. The table shows sheets that commonly turn up, for a 150 g candle.

Read your row and you are in one of three places. Row 1: you are at the ceiling on CSI's reading and past it on the conservative one, so the load is a live suspect. The audit's answer is to make round 1 at 9% of total (13.5 g of oil + 136.5 g of wax), which is inside both readings. Whether 8% would then be better still is the 8-versus-9% tie-breaker; this page doesn't re-run it. Rows 2 to 4: your real load is under 10% of total and no more than 10% of wax, within the ceiling on both readings. A load within a ceiling can still sweat, so round 1 still runs at 9%, but "too much oil" is not over any stated figure, and everything below matters more. Row 5: nothing can be concluded until the oil is weighed. A ml of fragrance oil is not a gram, and oils differ in how much a ml weighs.

Two more checks belong in the audit because they change the grams without changing the sheet. The scale: weigh something of known weight on it, or the same object on a second scale; a scale that drifts by a gram matters at 13.5 g. The top-up: if you top up sinkholes with scented wax from a different pot or a later session, the top layer may not match the sheet; note it.

Sweat patterns that point away from the load

When the audit clears the load, the pattern of the sweating usually points at the next suspect. The ranking is editorial judgement, most likely first; CSI publishes no failure statistics.

Rows 3 and 5 have their own posts: the controlled hot-cool protocol is the cycling-protocol post, and dye on top of a high load is the dye × load grid. Row 1 is the one this cluster handles oil by oil in the per-oil practical-max card. The rounds below separate row 1 from rows 2 to 4 in one step.

The non-load checklist, ranked

The by-the-book column uses only figures CSI pages actually state. Where a page states nothing, the setting is labelled judgement.

A word on 2 and 3, because they are where makers most often improvise. Adding oil to cooler wax to protect the scent is common, and the tendency with too-cool addition is poorer incorporation: mottling, separation, oil that doesn't stay put. The corrective order for that is the too-cool addition post. Short stirring is the other half, and the 20-second stir test runs it properly. Here each is simply one add-back round.

One factor at a time: round 1, then add-backs

Changing everything at once can stop the sweat without telling you why, and the next batch drifts back. The rounds below change one thing at a time, all at the audited load of 9% of total, in the wax, jar and wick your sweating candles used. The worked example uses Roasted Coffee in CSI 464 and the 150 g jar; use your own oil.

Round 1 and one add-back round together are seven candles and ₹1,555.47 of materials, excluding labour, freight, packaging and GST. One 100 g bottle, one 1 kg bag and a pack of ten jars and wicks cover both; a second add-back round needs more oil (the bottle has 5.5 g left) and more wax. Starting from nothing, the outlay is ₹1,900.40. In practice you already own the oil and wax that sweated.

What not to buy when the load checks out

For the first-level version of this question (is my load simply too high?), see the sweating first-level post; for oil that won't mix in at all, the oil-or-temperature question. This page starts where those end: the load is inside the ceiling on paper and the candle still sweats.

What the audit and the rounds use

Four items: the wax whose ceiling is in question, the scale the audit depends on, the thermometer add-back A depends on, and the worked example's oil. Chosen on page statements, price and stock on 26 September 2026; CSI has no measured sweat data, so nothing is ranked on performance. The oil table shows every size.

Frequently asked questions