How Many Candidate Wicks Should I Test Side-by-Side When Troubleshooting Soot, Mushrooming and Excessive Flame?

Your question, made precise

Here is the question in this page's terms. Your candle shows one or more of three combustion symptoms after the free checks (a 5–6 mm trim, still air, 4-hour sessions): soot (a smoke trail, a grey card, a dark band at the rim), mushrooming (a carbon cap on the wick tip that grows through a session), or an excessive flame (taller than your band and growing). You want to test wick candidates together and need to know how many. The candle is the series default, labelled: 200 g at 8% = 16 g fragrance + 184 g wax in your own jar. Our costing uses Cinnamon Vanilla because spiced, vanilla-rich gourmands are where these symptoms commonly appear (a tendency, not CSI data); use your production oil.

"Side by side" is doing a lot of work in your question. It means every candidate is poured on the same day from the same wax and oil lot, cured for the same time, and burned in the same room on the same days, so the only difference between them is the wick. Two candidates burned a fortnight apart, in a different season or with a different oil lot, aren't side by side, and their comparison can mislead you. This page owns the count and the schedule; buying a sample of every construction is the sampling post, and the options when two sizes bracket your jar are the hub.

Why three to four, and why three candles each

The recommendation is judgement built on two needs that pull against each other. You need enough candidates to bracket the answer: one that is still too strong (your control, usually), one that may be right, and one that shows what happens if you've gone too far. And you need few enough that you can observe every candle at every hourly check, space them so their heat doesn't interact, and afford to burn each one to the bottom of the jar, because soot, mushrooming and flame growth commonly get worse in the lower third.

Why three candles each? One candle per candidate tells you about that candle: a slightly off-centre wick, a bubble under the tab, a draught on one side of the bench. Two candles that disagree leave you with a coin toss. Three let a single odd candle show itself, and under the approval rule below one failure still fails the candidate, which is what protects your customers. Replicating further once something has passed (more pours, other lots) is the how-many-repeats post's job; this page is about choosing between candidates.

Why "more candidates" makes results worse. Every extra candidate is another chance that one wins by luck on a good day, and every extra candle on the bench adds heat and air movement to its neighbours. Many makers find that the hourly log becomes the bottleneck long before the wax does: at a worked-example 2 minutes per candle per check, four checks a day for 12 candles is 96 minutes; for 18 it is 144.

Choosing the candidates from CSI's ladder

CSI's sized, tabbed wicks are three rungs, C1 → C2 → Cotton Braided, with no sized, tabbed wick between them and nothing smaller than C1. The other candidates are configurations or constructions: a pair of C1 or C2, the one Performance Wooden Wick (no widths), and the Braided Wick Roll cut to length. So from any starting wick there are only a handful of honest candidates, and three or four of them almost choose themselves once you read the symptom.

Two rules keep the list honest. Change one thing per candidate: a C1 pair is one change (configuration); a wooden wick at a lower load is two, and if it wins you won't know why. Don't fill a slot with a wick you have no reason to expect could work just because the bench has room. If you genuinely need a construction comparison rather than a size comparison, that is a different test, with its own basket (construction or formula first).

Before any of this: causes, ranked

A candidate test answers a formulation question. Make sure that's the question you have. The ranking is our judgement for soot, mushrooming or a big flame; signs are common observations, not CSI data.

What each extra candidate costs

Per test candle in the worked example: 184 g of Luxury Soy Wax CSI 464 at its flat per-gram price (₹93.36), 16 g of Cinnamon Vanilla at the 100 g price (₹77.41) and one C2 (₹9.44): ₹180.21, before your jar and your time. That is why each extra candidate, at three candles, adds about ₹540.63, and why the wick is never the reason to cut a candidate: a candidate's three wicks cost ₹28.32 at most among the Eco sizes.

Ceilings stay where they are whatever the count: Luxury Soy Wax CSI 464 states "Fragrance Load Up to 10%" (read conservatively on wax weight, 9.09% of total), Cinnamon Vanilla's page "Up to 13% load for candles"; the lower, the wax's, governs, and 8% is inside both. The jar is often the dearest line of all; if you are testing in a jar you sell, every candidate you add is three more jars you can't.

Scheduling the burns on a real bench

Count how many candles you can burn at once safely: spaced so no flame heats its neighbour, on a heat-safe surface, away from fans, vents and windows, all within sight. That number, not your ambition, sets the plan. In the worked example the bench holds 6, and each 200 g candle burns about 28.6 hours at 7 g an hour, about 7 four-hour sessions to the 1 cm floor (a common rule of thumb), one session a day so each candle cools and sets between burns.

Narrowing as you go, without fooling yourself

Full-life tests are slow, and it is tempting to cut them short. You can, within rules you set before the first burn. Drop a candidate early when two of its three candles fail the same limit in the top third: a tunnel collar after the second full session, or continuous smoke in a session. That saves wax and bench space for the others. Never crown a candidate early because it looks clean in the top third: soot, mushrooming and flame growth commonly appear lower in the jar, which is exactly where your current wick probably failed.

Fragrances multiply the count. Three candidates in one oil is 9 candles; the same three in four oils is 36. Before you multiply, ask whether your oils behave differently enough to need separate tests; testing in the production fragrance covers when similar oils can share a result, and the ten-fragrance launch post how many sizes to start each fragrance with. If a jar test is part of a wider matrix of jars and oils, the wick-matrix post keeps it from becoming a shopping list. Eliminating obvious failures early across many fragrances without skipping full-life tests is the 20-fragrance post.

The CSI buy list and the step to production

Argued against our own interest: if you have no reason to expect the fourth candidate to work, leave it out and save ₹120.00 on wicks and three candles of wax and oil. And if the free checks (trim, still air, four-hour sessions) clear the symptom, you need none of this.

The step to production quantity

Once a candidate passes, production wicks = candles × wicks per candle × (1 + a defect allowance). Worked example, labelled: 200 candles a month at 2%. C1 single: 204 wicks, bought as 2 × 100 + 20 = 220 for ₹1,298.00. C1 pair: 408 wicks, 4 × 100 + 20 = 420 for ₹2,478.00. The full formula with safety stock and lead time is the purchase-quantity post; CSI publishes no lead times, so ask on WhatsApp.

What to buy, and what to skip

Four items, chosen on page statements, price and stock on 7 October 2026. CSI publishes no burn, soot or flame data, so nothing is ranked on performance, and each card says when not to buy.

Skip for now: a fifth or sixth candidate "while the bench is set up"; the wooden wick and braided roll unless the symptom table points there; and any pack above the test's needs. Mushrooming as the main symptom has its own hub, the mushrooming hub (6941); soot as the main symptom, the soot hub (6921).

Frequently asked questions