My Candle Has Weak Cold Throw Initially but Becomes Noticeably Stronger After Several Days — What Is Changing Inside the Wax During Cure?

What is physically happening inside the crystal

At the moment you pour, the fragrance is fully dissolved in liquid wax, a clear, uniform solution. As the wax cools it passes through its melting range and crystals begin to form. That first crystallisation is fast and untidy: many small crystals grow at once, in a less stable arrangement, and the fragrance molecules are caught wherever the wax happened to solidify around them.

A solid-looking candle is therefore not a finished material. Over the next days the soy continues to rearrange. Less stable crystals give way to more stable ones, and as the ordered regions grow tighter they push fragrance out of their way. The oil gathers in the softer, amorphous zones between crystals, a network that runs right up to the surface. More of it can now reach the air at room temperature, and your cold-throw score rises.

That is the physical story behind the live post on whether fragrance gets stronger as a candle cures. This page is about the mechanism and about recording it properly, so that you can separate a genuinely weak oil from one that simply had not finished settling.

Why heavy woods and oud are the slowest to wake up

Not every part of a fragrance moves at the same pace. Light top notes are volatile and leave any surface quickly, so they show up early and then fade. Heavy base notes, the woods, resins, musks and vanillin, have low volatility. Even once they are sitting in the soft zones near the surface, they need more time and more concentration before your nose registers them strongly.

CSI's own product pages reflect this. The Royal Oud page recommends a 72–96 h cure because oud takes longer to express. The Vetiver page says the oil builds slowly and asks you to allow the full cure. Woody Oud and Sandalwood both carry the standard 48–72 h soy statement, but their woody bases often keep developing after that minimum. A citrus candle might be at its loudest early; an oud candle may be at its quietest early.

Room temperature matters as well. Crystal settling is a slow physical rearrangement, and a workshop that swings from a cool night to a hot afternoon can make it less even from candle to candle. Candles cured near a window or on top of a warm cupboard may look and smell different from their twins on a lower shelf. For a fair log, store the whole test set together, out of sunlight, somewhere the temperature is reasonably steady.

There is also a matter of balance inside the blend. On day one the lighter notes you can still detect may be sitting on their own, without the heavier base that gives the scent its body. As the base catches up, the candle does not just get louder, it starts to smell more like the oil in the bottle. Many makers describe this as the scent rounding out, and it is one more reason not to judge the character of a woody or gourmand oil too early.

A dated sniff log you can actually trust

Impressions drift. A number written against a date does not. The template below is built for one pour of several candles, all lidded after setting and stored together. Scores use a simple 1–5 scale, which is a suggested starting point: 1 is barely detectable with the lid off, 5 is clearly noticeable at arm's length. Change the scale if you like, but keep it identical across every checkpoint.

Two habits make the log reliable. First, sniff a neutral reference between candles, such as your own sleeve or a cup of coffee beans, so your nose resets. Second, when you can, let another person score the candles without seeing your earlier numbers. A single sniffer who remembers the last score tends to confirm it.

Worked example: Woody Oud at 8% and the price of impatience

Suppose you pour Woody Oud at 8% in 200 g candles of Luxury Soy Wax CSI 464, which is 184 g wax and 16 g oil. Oil per candle: 16 g × (₹4,531.20 ÷ 1000) = 16 × ₹4.5312 = ₹72.50. On day one the candles seem thin, so the temptation is to re-pour at 10%, which is 20 g of oil: 20 × ₹4.5312 = ₹90.62. That step adds ₹90.62 − ₹72.50 = ₹18.12 to every candle.

Across a 50-candle production run the difference becomes ₹18.12 × 50 = ₹906.00, spent on a problem that might have disappeared by day seven. There is a second cost too: a wick chosen for 8% may carbon up or flood at 10%, sending you into a wick test you did not need.

When waiting will not fix it

Cure explains a candle that climbs. It does not explain a candle that stays flat. If your dated log shows the same low score at day 3, day 7 and day 14, and a burn with a full melt pool is also weak, the crystal story is finished and you need to look elsewhere.

Waiting is free, and for soy it is part of the process. Give the crystals their time, write down what you smell and when, and let the log rather than the first evening's disappointment make the decision.

Where this page fits

CSI already has guides that cover parts of this question. This page answers the specific oil-in-wax question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions