My New Fragrance Gives Better Hot Throw but Causes a Less Stable Flame — How Do I Balance Fragrance Performance Against Clean Burning?

Why stronger throw and a calmer flame can pull in opposite directions

You swapped one fragrance for another in a candle that already burned well. Same wax, same jar, same wick, same 8% load. The new candle fills the room noticeably better, but the flame now flickers, sputters now and then, and occasionally throws a thin thread of smoke. Nothing else changed, so the fragrance is the obvious suspect. The question is not whether to blame the oil. It is how much stability you are willing to trade for that extra throw, and whether you actually have to trade anything at all.

Hot throw depends on how readily fragrance leaves a warm melt pool. An oil built with a bright, eager top and a rich base can release beautifully while also changing the liquid that climbs the wick. The wick is a capillary: it pulls melted wax and dissolved fragrance up to the flame, where both burn as fuel. When the composition of that fuel shifts, the flame responds. More volatile material near the flame can make it lively and restless; heavier, slower material can load the wick tip and leave it smouldering unevenly. Either way, the flame is telling you that fuel supply and wick capacity are no longer matched as neatly as before.

That is why better throw and a less stable flame often arrive together. They are two readings of the same event: the new oil moves differently through heat. Your job is to separate the part of the change you want to keep, the scent in the room, from the part you do not want, the instability, without swinging several variables at once.

Define unstable before you try to fix it

Makers use the word unstable for very different things. A flame that dances because a ceiling fan is running is not the same fault as a flame that pulses on its own in still air. Before you change a wick or a load, write down exactly what you see, when you see it, and under what conditions. Otherwise you will fix a draft and call it a formulation problem.

Run the new candle and your previous, well-behaved candle side by side in the same still spot, lit at the same time. That pairing is your control. If both flicker, the room is the cause. If only the new fragrance flickers, you have isolated the oil as the variable. Observe at fixed checkpoints; hourly for a four-hour burn is a sensible starting point to validate, not a rule.

Use your own scoring scale and keep it the same across every candle. A 1–5 scale is a practical starting point; what matters is that you score both candles with the same nose, in the same room, after the same length of burn. Give your nose fresh air between checks, because olfactory fatigue will flatten the difference you are trying to measure.

A three-candle test that separates wick from load

Once you have confirmed the instability belongs to the new oil, you have two main levers: fragrance load and wick size. Change one at a time. The test below keeps the wax, jar, pour temperature and cure identical, and makes a single change per candle. Pour everything from the same batch of wax on the same day so the cure clock is shared.

Add fragrance below the oil's published flashpoint and at the temperature your wax supplier states, and never leave any of the test candles burning unattended. If any candle smokes steadily or the flame flares, extinguish it and log it as a fail rather than watching to see whether it recovers.

What each fix costs, using a real oil

Consider a concrete case. Suppose you moved to CSI Bergamot Amber in CSI 464 soy flakes, in a 200 g jar. The product page describes a bergamot and citrus top, an amber heart with soft floral, and a woody musk and vanilla base, with a stated load range on the product page of 6–10% for soy. It also gives an intensity guide: at 6% bergamot leads, at 8% the profile is balanced, and at 10% amber comes forward and the vanilla and musk base becomes more prominent. That guide matters here, because reducing load changes the character as well as the flame.

At the 1 kg price of ₹5,437.00, the oil cost per candle works out as follows. At 8%: 16 g × ₹5,437.00 ÷ 1000 = ₹86.99. At 7%: 14 g × ₹5,437.00 ÷ 1000 = ₹76.12. At 6%: 12 g × ₹5,437.00 ÷ 1000 = ₹65.24. For reference, CSI Lavender at 8% is 16 g × ₹5,221.00 ÷ 1000 = ₹83.54.

The difference between 8% and 7% is ₹86.99 − ₹76.12 = ₹10.87 per candle. That is a saving, not a cost, so if 7% delivers both a calm flame and the throw you liked, you have improved the candle and lowered its price in one step. If the flame only settles at 6%, weigh whether the bergamot-forward character still fits the product you are selling.

Clean burning first, then tune the throw

Treat clean burning as the fixed requirement and hot throw as the adjustable one. A candle that fills a room but smokes, sputters or leaves soot on the rim is not a better candle; it is a candle with a complaint waiting to happen. The reverse is not true: a calm candle with slightly softer throw can often be improved through cure time, melt-pool behaviour or a small wick change.

Reading the three-candle result

If Candle B wins, lower the load and keep your wick; you have fewer variables to manage in production. If Candle C wins, re-wick this fragrance only and record it as an exception in your wick sheet, because the wick that suits your other oils is still correct for them. If both B and C steady the flame, choose the one that preserves the scent profile you want. If neither steadies it, test B and C combined, and if that still fails, this oil may simply demand a different system, such as a different wax, before it becomes a product.

Finally, repeat the winning candle across a second pour before you approve it. One good burn is an encouraging sign; two matching burns from separate pours are a decision. Write the approved wick, load, wax and cure into your formula card so the next batch starts from evidence rather than memory.

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