I Have 12 Fragrances Going into the Same 8 cm Jar — How Can I Identify Which Scents May Need a Different Wick Without Testing Every Possible Wick Size?

What a product page can tell you about a wick

CSI publishes no per-oil wick data, so nothing on our pages will name the wick an oil needs in your jar. Pages do state things that commonly travel with wick trouble: the family, the notes (especially heavy base notes), a vanillin percentage on some pages, and a stated candle load range. Heavier, vanillin-rich or resinous oils commonly leave more carbon on a wick and can make a flame smaller or sootier; light florals, herbals, citrus and aquatics often burn more easily. These are tendencies with plenty of exceptions, which is why the output of a paper screen is a list of suspects, not verdicts.

A few things pages say that look useful and aren't, for this purpose. Throw wording ("Strong throw", "Moderate to strong throw") is marketing about scent, not burn behaviour. Wick lines such as Sage and Cedarwood's "Cotton or wooden wicks for a steady burn. Test wick size to suit your container and wax type" and Palo Santo's "Eco wicks or wooden wicks" are generic advice that appears on many pages; they name no size and don't separate one oil from another. And a "0% vanillin" line is the page's claim about one discolouring component, not a promise about carbon on the wick.

The jar matters too. On a 0.35 cm wall (our assumption; measure yours) an 8 cm outside is 7.3 cm inside, a 41.85 cm² pool. At that width a single house wick is usually working near its limit, which makes it more sensitive to the oil. CSI lists no 8 cm jar; the nearest listings are 7 cm and 10 cm, so the jar in this test is yours.

Four flags and a flag line

Each flag is one point, scored only from what the page states in the text we captured. If a page is silent on something, that flag scores 0 and the cell says "not captured": silence is not evidence either way. The choice of flags and the line are our judgement.

The flag line is 2. One point alone is too common to be useful: plenty of easy-burning oils have an amber note or a woody family. Two independent reasons to worry is where it is worth pouring first. Your line can move after two or three collections, once your hit-rate record shows whether 2 over-flags or under-flags on your wick. Don't move it mid-collection.

Twelve CSI oils scored from their pages

Our worked example is a twelve-scent range in one 8 cm jar at 8%, using twelve in-stock CSI oils. Swap in your own; the method is the same. Each cell quotes the page and shows the point it scored in brackets. The family is our grouping from the stated notes; the vanillin and load cells are the pages' own figures where captured.

Three things stand out. First, the screen splits cleanly on this set: nothing scores 1, so no oil sits on the line. That is a property of these twelve, not of the method; expect borderline oils in yours. Second, four of the five flagged oils carry vanilla in their stated notes, which is why a vanilla-heavy range tends to need more wick work than a floral one (a tendency, not a rule). Third, two oils sit just under the radar. Gardenia states "Vanillin content 0.2%" and "Rich, creamy, and velvety" notes, below our 1% vanillin threshold; Sage and Cedarwood is woody but names no heavy base note. Neither is flagged, and the second is a control precisely to test whether "woody" should have counted.

The confirm-only test

"Confirm-only" means every candle burns on your house wick: the wick already bracketed and full-life tested on your lead oil (C2 in our worked example). No wick changes, no load changes. Each flagged oil either confirms (shows a trigger) or clears; each control should clear. Only confirmed oils go on to a bracket, and the fault they show tells you which direction.

That is ₹199.55 a candle on average; costs are worked examples and exclude labour, freight, packaging, jars and GST. The seven 50 g bottles are ₹2,116.30 together; each leaves 18 g, enough for a third candle if twins split, not enough for a bracket. Three 1 kg bags of CSI 464 (₹1,522.20) leave 424 g; CSI 464 is flat at ₹0.51 a gram from 1 kg to 10 kg, so a bigger bag saves nothing per gram.

How the four-oil version of this idea works as a straight same-wick screen in a 7 cm jar is the four-oil screen; how to fill a grid adaptively after a first pass is the adaptive grid. This page's difference is that the paper decides who burns first.

Recording the hit rate honestly

After the confirm test, fill in two numbers. The hit rate is flagged oils that confirmed, out of flagged oils tested. The miss check is controls that triggered, out of controls tested. In our worked example (illustrative, not CSI data) three of the five flagged oils confirmed, a hit rate of 60%, and neither control triggered.

Why record false alarms at all, when they cost nothing but a few candles? Because two false alarms out of five is information: on this wick, "gourmand with amber" is not enough on its own. After two or three collections the sheet becomes your own evidence about your wick, which is worth more than any general table. Whether the oils that did confirm form groups that can share a wick is the question the behaviour-grouping post owns.

What the screen saves, and what it never saves

Testing every possible wick size for twelve oils on CSI's three rungs (C1, C2, Cotton Braided), two candles each, is 72 candles. The pre-screen route in our worked example is 14 confirm candles, 10 safety burn-down candles for the five unflagged oils that weren't controls, and 12 bracket candles for the three confirmed oils at four each: 36 candles, 36 fewer. At the confirm test's average of ₹199.55 a candle, that is roughly ₹7,183.68 against ₹14,367.37 of materials (an illustration; each oil's price differs).

Notice what did not shrink: every one of the twelve oils still burns at least two candles to about 1 cm left on the house wick before it goes on sale. That floor is the policy in the every-fragrance burn-down post, and the screen works on top of it. What the screen changes is the order (suspects first, so their brackets start weeks earlier) and the number of brackets (only the confirmed).

Diwali is 8 November 2026, 40 days from 29 September 2026. A confirm test poured this week and cured about a week can show most triggers in the first few sessions, early enough to start brackets for the confirmed oils; their full lives will run into late October at best. Launch what clears; don't rush a confirmed oil onto the shelf to fill a slot.

What to buy for the confirm test

Four items: the two controls, the top-scoring flagged oil, and the house wick, 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 16 g (200 g at 8%) and 12 g (150 g at 8%).

Frequently asked questions