I Have 6 cm, 7 cm, 8 cm, 9 cm and 10 cm Candle Jars — How Do I Build a Wick-Testing Matrix Without Randomly Buying Every Wick Size?

Why every wick for every jar is the wrong matrix

The obvious matrix puts every wick against every jar. With CSI's range that means five arms (Eco C1, Eco C2, Cotton Braided, two C1, two C2) against five jars: 25 cells, 50 candles at two per cell, 12,000 g of wax and oil, 30 C1, 30 C2 and 10 Cotton Braided wicks. Most of it is wasted. A pair of C2 wicks in a 6 cm jar is two flames sharing 22.06 cm² of wax; a single C1 in a 10 cm jar has to reach glass 4.65 cm away. Neither needs a candle to predict.

The wasted cells are also the dear ones. At the fills this series uses (100 g for 6 cm, 150 g for 7 cm, 200 g for 8 cm, 250 g for 9 cm, 500 g for 10 cm, all worked examples), a 10 cm cell weighs five times a 6 cm cell. Pour every arm in the 10 cm jar and you have poured 5,000 g of candle there alone, including singles that cannot reach the glass.

The banded matrix keeps only the cells where the answer is in doubt: for each diameter, the two or three arms that could plausibly be right, plus one held back in case the bracket misses. That is 11 cells and 22 candles: 28 fewer candles and 52% less wax and oil than the full grid. And because each arm sits in two neighbouring bands, the cells you skip are covered by the logic of the matrix: if C2 is already too hot at 7 cm, you have learned something about whether 2 × C2 belongs in an 8 cm jar.

Can one wick serve several of these jars? Sometimes, if their inside diameters are close. Five jars a centimetre apart are not close: each step adds 41%, 34%, 29% and 26% of pool area going up from 6 to 10 cm. How close is close enough is the post on standardising one wick across jars; the matrix is what you build when the answer is "not this close".

Five jars, measured from the inside

A matrix is only as good as the diameters on its column headings. The title says 6, 7, 8, 9 and 10 cm but not inside or outside, and CSI's own listings don't say either: none of CSI's 72 container listings states an internal diameter. This page reads every figure as the outside and assumes a 0.35 cm wall, so the five jars become 5.3, 6.3, 7.3, 8.3 and 9.3 cm inside. That is an assumption, and the matrix must be built on your measured figures, not ours (why the external figure misleads).

Three of your five sizes have CSI listings near them; two do not. The 6 cm row has eight CSI glass listings that give a 6 cm figure (seven 120 gm shiny jars and the Shot Glass Jar); the 7 cm row has the Clear Glass Jar with Golden Lid (150 gram), whose page gives an outer diameter of approx. 7 cm and "Safe for hot pour up to 85°C", plus four coloured 150 g siblings (some of their titles still carry an old delivery note, so confirm dispatch on WhatsApp); the 10 cm row has the Apothecary Glass Candle Jar with Dome Lid, "Diameter (outer rim): Approx. 10 cm", also "Safe for hot-pour up to 85°C". CSI lists no jar at 8 cm or 9 cm. Those two columns are your own jars, from your own supplier, and we have no figures for them. The 85°C wording is a pour limit, not a burn rating: with CSI 464 (pour 80–90°C) pour those jars at 80–85°C.

The starting-rung table: one bracket per band

The table below is where each band starts. It is built from three things: the pool area each jar asks a wick to melt, the rungs CSI actually sells, and the one diameter any CSI wick page names (Cotton Braided, "exceeding 8 cm"). It is not a wick chart: every band can move once your wax, oil, load and measured inside diameter go into the candle.

Why these brackets. At 6 and 7 cm, C1 v C2 is the whole of CSI's tabbed small range; nothing smaller is tabbed (the untabbed Eco Candle Wicks Loose is "designed specifically for 2 inch diameters") and the next rung up is described for candles over 8 cm. At 8 cm, C2 meets Cotton Braided at the border its page draws: about 7.3 cm inside on our reading, just under "exceeding 8 cm", so the two belong in one bracket. At 9 cm each wick of a pair serves 27.05 cm², between the whole pool of a CSI 6 cm listing (22.06 cm²) and a 7 cm one (31.17 cm²), which is why 2 × C1 is the pair arm there. At 10 cm each wick of a pair serves 33.96 cm², a little more than a 7 cm listing's pool, so the pair arms are 2 × C1 and 2 × C2 against the large single. Pairs start at a centre-to-centre spacing equal to the inside radius (4.15 cm at 9 cm, 4.65 cm at 10 cm), which leaves each wick about 2.1 cm and 2.3 cm from the wall along the line: geometry and a labelled starting point, not a burn prediction.

Why the overlap matters. Read the table down the "Round-1 bracket" column and every arm appears twice: C1 at 6 and 7 cm, C2 at 6, 7 and 8 cm, Cotton Braided at 8, 9 and 10 cm, 2 × C1 at 9 and 10 cm. That is deliberate. When the 8 cm band fails high, the arm you would step down to (C2) has already been watched in the 7 cm jar; when the 9 cm band fails low, the arm above (2 × C2) is being poured at 10 cm. The neighbour's cell is not a result for your jar, but it tells you how that arm behaves before you pour it.

Round 1: eleven cells, two candles each

Round 1 holds the fragrance still with one light reference oil, so that what changes between columns is the jar and what changes between rows is the wick. Our worked example uses Lavender Fragrance Oil at 8% (its page states "Recommended fragrance load 6-10% for candles"; CSI 464's page states "Fragrance Load Up to 10%"), poured into each jar at the fills in the table: 8 g + 92 g at 6 cm up to 40 g + 460 g at 10 cm.

Materials consumed: 464 g of Lavender at ₹5.24 a gram (₹2,431.36), 5,336 g of CSI 464 at the 10 kg price (₹2,707.49), and ₹237.20 of wicks (₹70.80 C1, ₹94.40 C2, ₹72.00 Cotton Braided): ₹5,376.05. Your jars are not in that figure because you already own them. The 10 cm row alone is ₹2,743.38 of it. All costs are worked examples and exclude labour, freight, packaging and GST.

Then run round 2: the same 11 cells with one heavy oil. Heavier, vanillin-rich or resinous oils are commonly linked to more carbon and a need for a wick change, so a matrix built only on Lavender can sit a rung low for them. Our worked example uses Woody Oud Fragrance Oil (page: "candles (6-10%)"; notes of "citrus and spice top notes opening into rich oud and patchouli heart"), priced at ₹4.53 a gram from 100 g to 1 kg: 464 g from the 500 g costs ₹2,102.48. How to sort twenty oils into groups that share a matrix is the grouping post; how to keep two groups as a house standard is the behaviour-groups post.

Reading the matrix band by band

Each band is read on its own, with its two or three arms side by side, then against its neighbours. The thresholds below are our judgement, chosen because they can be read off a photograph and a ruler; the flame range many makers use is about 2.5–5 cm and the common pool-depth target about 0.5–1 cm. "Passes" means both candles in the cell pass at burn 3 with no heat signal.

A neighbour's result is evidence about an arm, not a verdict for your jar: C2 running calm at 7 cm does not mean C2 passes at 8 cm, where the pool is 34% larger. At 10 cm, if all three arms pass, take the one with the most margin at the bottom of the jar (the 10 cm pair-or-single hub).

The first wick shopping list, in packs and rupees

The per-wick price is flat across every pack size: ₹5.90 for C1, ₹9.44 for C2 and ₹12.00 for Cotton Braided, whether you buy 10 or 1,000. A bigger pack only saves a reorder, and commits you to a rung before the matrix has chosen it.

Our recommendation: buy round 1 (₹426.80), run it, and buy the 50s of C1 and C2 (₹767.00) when you start round 2, because by then both are clearly in use across four bands. Hold the Cotton Braided at 10-packs until a 9 or 10 cm candle has passed full life on it. Argued against our own interest: the three 50-packs together are ₹1,367.00; for the matrix, the lean list does the same job for ₹1,007.00.

What to buy, and what to skip

Five items, in the order the matrix uses them, chosen on page statements, price and stock on 29 September 2026. CSI publishes no wick-performance data, so nothing here is ranked on burn results; the brackets are starting points and the matrix is how you replace them with your own evidence.

Map of this cluster: 4281–4300

This page owns the method: one bracket per diameter band, the starting-rung table, round 1 and the first shopping list. Each other post in the cluster owns one commercial question about wick data: grouping, launch counts, test design, standards, stock, change control and the internal guide.

For many fragrances at once, the tiered burn plan for 20 fragrances shows how to cut obvious failures early without skipping final safety testing; the order to test raw materials in is the test-order post.

Frequently asked questions