I’m Using a 10 cm Jar with 10% Vanilla Fragrance and Both Wicks Mushroom Heavily — Should I Reduce Wick Size, Fragrance Load or Both in Separate Tests?

Why both wicks mushroom at 10% vanilla

Mushrooming is the cap of carbon that builds on a wick tip as it burns: a small bead is common; a cap wider than the wick that drops soot or debris into the pool is a sign the wick is getting more fuel than it burns cleanly. It is commonly linked to too much wick for the candle, to heavy or vanillin-rich oils, to high loads and to long sessions. Those are tendencies, not CSI data, and in your candle three of them are present at once: a pair in a large pool, a vanilla oil, and a load at the top of its range.

When both wicks mushroom, the cause is unlikely to be one faulty wick or one off-centre tab; those make one flame different from the other, which is a different question. Both wicks mushrooming points at something they share: the wick size you chose for both, the oil they both burn, and the load in the pool they share. That's why the two levers here are the pair and the load, not position or trim.

Dark Vanilla's page describes it as "rich vanilla and spice opening into deep caramel and cream, settling into dark amber and warm musk", states "Vanillin present - note for light-coloured wax", and gives "candles (6-10%)". Rich, vanillin-bearing gourmand oils are the ones many makers expect to need a wick adjustment (a tendency we repeat across this series, not a measurement). At 10% you are at the top of the page's range, and if your wax is CSI 464, whose page says "Fragrance Load Up to 10%" without saying of what, 10% of total is 11.11% of the wax weight. On the conservative reading the ceiling is 9.09% of total. That alone makes the load a fair suspect, just not the first one to test.

Why never both in one candle

The tempting fix is one new candle with a smaller pair and 8%. If it stops mushrooming you feel finished, but you can't tell which change did it, and both possible answers cost you something. If the smaller pair alone would have fixed it, you have given up 20% of your oil, and some of the throw your customers liked, for no reason. If the lower load alone would have fixed it, you have narrowed both wicks unnecessarily, and a smaller pair in a 10 cm pool may leave a ring at the glass or struggle lower in the jar. Either way, the next time something changes (a new oil lot, summer heat, a new jar batch) you won't know which lever to reach for.

A full factorial, both pairs at both loads, all at once, does separate the two effects, and it also shows whether they interact. That is the right tool when you need the interaction, and the 10% load-vs-wick post owns it. This post owns the cheaper path for a large candle: change the lever most likely to help and least likely to cost you the scent, see whether that alone is enough, and only then touch the second one.

Round 1: the pair steps down at 10%

Pour four candles the same day from one melt, using the cluster's starting build from the 10 cm hub: the Apothecary jar's 500 gm size at a 500 g fill, Dark Vanilla at 10% of total weight (50 g oil and 450 g wax per candle), same jar batch, same spacing (the cluster's starting point is the inside radius, 4.65 cm, about 2.33 cm from the glass along the wick line, in a jar about 9.3 cm inside on our 0.35 cm wall assumption), same pour temperature (80–85°C for CSI 464 in a jar rated "Safe for hot-pour up to 85°C"), same cure. Two candles get your current pair, the control; two get one rung down. On CSI's rungs, a 2 × C2 control steps to 2 × C1. If your control is already 2 × C1, CSI lists nothing smaller and tabbed, so skip to round 2.

The first row is the one you hope for, and in our judgement it is a common outcome for a vanilla candle that started on a generous pair. The pair carries the whole candle's heat; stepping it down usually reduces the fuel each flame handles, which is the side of the mushrooming problem a wick change can reach. The table's other rows send you to round 2, and each names which pair round 2 should use, so the second round starts from evidence rather than a pre-drawn grid.

Round 2: the load steps down on the chosen pair

Round 2 runs only if round 1 sent you here. The single-wick version of a load step-down, scored for what the lower load costs in scent, is the 8 cm coffee post; here it is the second round of a plan. Pour four fresh candles: two at 10% (a fresh control with the same cure date as the test cell) and two at 8% (40 g oil and 460 g wax per candle), all on the pair round 1 named. Everything else stays as it was. Log and read exactly as in round 1.

Score throw blind: someone else codes the candles, two or three people sniff from the doorway of the same closed room at the same point in the session, with fresh air between candles. A drop of less than about one point on a five-point scale is often hard to tell from day-to-day noise (our judgement). The fuller objective method is a sibling series' job; the load-first lavender post runs the same kind of blind comparison in the other direction.

Why fresh controls rather than reusing round 1's candles? Because round 1's candles are older, partly burned and have been trimmed and relit; comparing a new 8% candle with an old 10% one mixes the load with cure age and burn history. Two extra candles cost money, which is the honest weak point of the sequential plan, and the next section counts it.

Why sequencing saves money at 500 g

Per candle, at the 500 g oil price and CSI 464's flat per-gram price, here is what each arm costs in materials.

Two things stand out. The wick difference is trivial; the oil, the wax and the jar are almost everything. And stepping 10% to 8% saves ₹71.73 per candle (10 g less oil, 10 g more wax), which is real money in production but irrelevant to how many test candles you should pour. The number of candles is what drives the bill.

If round 1 answers, you have spent ₹3,360.28 instead of ₹6,433.66: a saving of ₹3,073.38. If both rounds are needed, the series costs ₹129.29 more than the grid, because round 2 pours two fresh 10% controls the grid would not need, and it takes about 22 days to both decisions against about 11 for the grid (worked example: about a week's cure plus four daily sessions per round). That is the trade, stated both ways: you bet four candles on round 1 being enough, and you pay a little more and wait longer if it isn't. In our judgement the bet is worth it for a vanilla candle that mushrooms on a generous pair, and not worth it if you already suspect the two effects interact, in which case pour the grid.

What to buy for round 1, and for both

Oil. Round 1 needs 200 g. Two 100 g bottles (₹1,580.00) are exactly that, with nothing to spare, so weigh carefully or add a 15 g bottle. Both rounds need 380 g: four 100 g bottles (₹3,160.00) leave 20 g; the 500 g (₹3,840.00) leaves 120 g and costs ₹680.00 more cash. Wax. Two 1 kg bags of CSI 464 (₹1,014.80) for round 1, four (₹2,029.60) for both; the price per gram is the same from 1 kg to 10 kg. Jars. Four per round at ₹212.40. Wicks. A C1 20-pack covers round 1 and a C1 round 2 with 8 spare; a C2 round 2 needs 2 more C2 than the 10-pack leaves. From nothing, round 1's cash outlay is ₹3,656.80.

What to buy, and what to skip

Five items, in the order the plan uses them, chosen on page statements, price and stock on 29 September 2026. CSI publishes no wick-performance data, so nothing is ranked on burn results. Oil candles are counted at 50 g (500 g at 10%), 40 g (500 g at 8%) and 12 g (the standard 150 g candle).

Frequently asked questions