My Lavender Candle Burns Perfectly but Vanilla Using the Same Wax, Jar, Wick and 8% Fragrance Mushrooms Badly — Why?

The setup is the same but the fuel is not

You have done the sensible thing. You approved a wick in one jar with one wax, you poured lavender at 8% and it burned cleanly, so you reused every setting for vanilla. Then the vanilla wick grew a black cap, the flame started to flicker, and perhaps a faint grey line appeared around the rim of the jar.

It is tempting to blame the wick batch or the wax delivery. Before you do, notice what the comparison is really telling you. Wax, jar, wick and load were held constant. The oil is the only thing that moved. That is actually good news, because a problem with one clear variable is much easier to solve than a problem where three things changed at once.

The phrase same 8% hides the important part. In a 200 g candle, 8% of total weight is 16 g of oil and 184 g of wax. Both candles contain 16 g of fragrance, but those 16 g are made of very different materials. Weight tells you how much oil is present. It tells you nothing about how that oil behaves when it is pulled up a cotton wick and burned.

Why vanilla loads a wick differently from lavender

CSI Lavender is described on its product page as a single-note lavender: fresh lavender at the top, herbal green in the middle and a soft musk base. That is a relatively light, volatile structure. A good share of it evaporates readily from the melt pool, which is exactly why lavender throws well without asking much of the flame.

CSI Dark Vanilla is built very differently. Its page lists rich vanilla and spice at the top, caramel and cream in the heart, and dark amber and musk at the base, and it states that vanillin is present. Vanillin and the heavy sweet materials that support a vanilla accord are less volatile. They stay in the liquid fuel longer instead of lifting off the melt pool, so more of them travel up the wick and reach the flame.

As a general principle, heavier and more complex molecules burn less completely than light ones. When the flame cannot fully combust everything the wick delivers, the unburnt residue deposits as carbon on the wick tip. That deposit is the mushroom. Over a long burn it grows, drops sparks or crumbs into the pool, and can make the flame flicker and smoke.

The wick is not wrong, it is mismatched

The wick you approved on lavender delivers a certain amount of fuel. With lavender, the flame burns that fuel cleanly. With vanilla at the same load, the same fuel delivery contains more slow-burning material, so the wick is effectively running hotter than the oil can tolerate. In practice, a vanilla candle often behaves as if it is slightly over-wicked, even though nothing about the wick itself changed.

The split test that separates load from wick

You have two sensible levers: reduce the vanilla load or reduce the wick size. Pulling both at once will fix the mushroom but will not tell you which change mattered, and that knowledge is what lets you set up the next vanilla product correctly. So test them separately, with lavender left untouched as the control.

Read the sheet honestly. If variant C cleans up and B does not, the load was the main driver. If B cleans up and C does not, the wick was oversized for this oil. If both help, choose the one that gives better hot throw, then confirm it over a few more sessions before you commit to production.

What each fix costs per candle

Cost matters here because vanilla is one of the more expensive oils on the shelf. CSI Dark Vanilla is ₹7,646.00 per kg, which is ₹7.646 per gram. CSI Lavender is ₹5,221.00 per kg, or ₹5.221 per gram.

If variant C works, dropping vanilla from 8% to 6% saves ₹122.34 − ₹91.75 = ₹30.59 in oil per 200 g candle. The wick option is cheaper to implement but saves less: on CSI Eco Wicks, stepping from a Thick C2 at ₹9.44 to a Thin C1 at ₹5.90 changes wick cost by ₹3.54 per candle. That is a real difference at volume, but the load change is where the money is, provided the scent still performs.

Keep, re-wick, reduce or accept

Once the log is filled, one of four outcomes usually applies.

Reduce the load if 6% stops the mushroom and your room test still reads as clearly vanilla. You gain a cleaner burn and a lower cost per candle, and the product page's stated soy range of 6–10% supports it.

Re-wick for vanilla only if 8% is needed for throw and the smaller wick keeps the flame clean. Record it as a vanilla-specific wick, not a jar-wide change. Your lavender stays on its approved wick.

Accept a light cap if the carbon only forms after an unusually long session and trims away cleanly. Put a trimming line on the label and move on. Do not chase perfection that customers will never see.

Change the oil if every variant still mushrooms heavily or soots. That is the point at which the formulation, not your settings, is unsuitable for your wax and jar. Test a different vanilla side by side under the same protocol before buying in bulk.

Whatever you choose, write the result into a wick record that names the oil, not just the jar and wax. A line such as 200 g jar, CSI 464, Dark Vanilla 6%, approved wick is far more useful six months from now than 200 g jar, approved wick. The next time you add a caramel-heavy or cream-heavy oil to the range, that record tells you to start testing at the vanilla settings rather than the lavender ones.

One separate note: Dark Vanilla's page also warns that it will yellow white or light-coloured wax and states this does not affect performance. Colour change is a cosmetic question, and it has nothing to do with the mushroom. Keep the two issues apart when you record results.

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