How Many Wax Types Should I Test Before Deciding That a Fragrance Oil Is Unsuitable for Candle Production?

Why testing every wax is the wrong instinct

When an oil you like keeps sweating, it is natural to keep looking for the wax that will finally hold it. One more soy, then another, then a blend from somewhere else. Each test takes a pour, a cure, an observation window and a burn, and each new wax that passes brings its own wick tests and stock line with it. The search can run for months and end with a product nobody else in your range shares a wax with.

The better approach is to design a small panel before you start, choosing waxes that differ in ways likely to matter, and to fix in advance the result that ends the search. Most of the information comes from the first contrast. Beyond that, returns drop quickly.

Choose the panel for contrast

A useful panel spans the kinds of container wax you could realistically sell. For most soy-focused makers that means one 100% soy and one blend with a different base. If your product could be described simply as a scented candle without a soy claim, paraffin is the third, because it differs most from soy in how it sets and holds oil.

When you compare published loads, put them on one basis. CSI wax pages state loads as a share of wax weight; this series uses a share of the finished candle. The conversion is the wax figure divided by 100 plus that figure, times 100.

From that list, CSI 464 and CSI 468 make a natural pair: one pure soy, one coconut-soy, with stated ceilings of 9.09% and 11.50% of the candle respectively. Paraffin pellets add the widest contrast. CSI 400 and Soy Chunks are both soy, so adding them to a panel that already contains CSI 464 tells you less than adding a different type of wax. CSI also lists a Coconut Wax, a separate Premium Coconut Soy and a Soy Butter Wax without captured specifications; if you include any of them, establish every figure yourself.

Test at your floor, with a control in every wax

Before you pour, decide your floor: the smallest percentage at which this scent would still be worth selling. For many makers that is around 6% of total weight, which in a 200 g candle is 12 g of oil and 188 g of wax. Test the oil at your floor and one step above it, for example 7% (14 g of oil, 186 g of wax). Testing only at 10% proves little, because many oils sweat at high loads in many waxes.

Pour a control alongside in each wax: an oil you already know holds at that load. The control protects the verdict. If the control sweats in a wax, the problem lies with that wax in your process or storage, and that wax's result for the difficult oil does not count.

Rule out the obvious first

Before the panel, check two things that waste more test time than any wax. Confirm the beads are oil: a blot that stays oily on the paper and carries the fragrance is oil; a blot that dries clean is not. And confirm your load arithmetic: if the original candles were made at 10% of total weight, they held 20 g of oil in 180 g of wax, which is 11.1% of wax weight and already above CSI 464's stated figure. An oil that sweats there may be perfectly happy at your floor, in which case it never needed a panel at all.

Make sure the oil sample itself is representative. If you are testing a bottle that has been opened, warmed and cooled many times, or one from a different lot than you would buy, note it on the log. A verdict reached on an unrepresentative sample may not hold for the oil you would actually order.

Reaching a verdict

Notice the phrase unsuitable for your production. You are not ruling on the oil for every maker in the country. Another maker with a different wax and a different product might use it happily. Your job is to decide whether it fits your system at a load you would sell, and two good tests are enough to say no.

What the panel costs

Buying 1 kg of each of the three contrasting waxes costs ₹507.40 + ₹650.00 + ₹354.00 = ₹1,511.40. At 7%, each 200 g candle needs 186 g of wax, so 1 kg fills 1,000 ÷ 186 = 5 candles, rounding down. That is enough in each wax for the difficult oil at two loads, a control and two spare candles for repeats or a burn test.

The difficult oil itself needs 12 g + 14 g = 26 g per wax for one candle at each load, or 78 g across three waxes. A single 100 g bottle covers it. The controls use whatever oil you already trust.

Compare that with an open-ended search. Each extra wax adds another kilogram, another observation window and, if it passes, another full wick test across every jar you would use it in. Five waxes cost far more than two in time as well as money, and they rarely overturn what the first contrast showed.

Keep the panel results on file. If you later change your production wax, or a supplier reformulates an oil, the earlier panel tells you which oils were borderline and need re-testing first, and which were rejected for reasons that may no longer apply.

What to buy after the verdict

If the oil passed in one panel wax and that wax suits your product, move to a 5 kg pack for a pilot batch; CSI 468 is cheaper per kg from 5 kg, while CSI 464 costs the same per kg in every pack. If the oil failed, buy nothing more of it. Spend the next 100 g on a replacement oil, tested in the wax you already use. Using less oil, or none of a stubborn one, is often the cheapest result of a well-run panel.

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