Before I Reject a Fragrance Oil for Weak Hot Throw, What Variables Should I Test in Wax, Wick, Dosage, Cure and Room Conditions?

The eleven-variable checklist

A single weak candle can be weak for a dozen reasons, and only one of them is the oil. Before you cross an oil off your list, work down the checklist below. The first eight items cost little or nothing to check because they concern how the candle was made and judged. The last three cost a few candles and are best tested together in a small matrix. The mechanisms in the right-hand column are general candle-making practice, not CSI measurements.

The order matters. Variables 1 to 9 can each explain a weak candle on their own, and they are cheap to rule out. If you skip them and go straight to buying a new oil, you may carry the same fault into the next test and reject that oil too. Makers who reject oil after oil without improvement usually have a fault in this upper half of the table.

Clear the free variables first

Load convention. CSI's convention, used throughout this blog, is load as a share of the finished weight: a 200 g candle at 8% holds 16 g of oil and 184 g of wax. If you instead add 8% of the wax weight (8% of 184 g is 14.72 g), the true load is lower. On a 1 kg wax batch, adding 80 g gives 7.41% of the 1,080 g total; to reach a true 8% on 1 kg of wax you need 86.96 g, which is wax × load ÷ (1 − load). If your weak candle was made the second way, it was never an 8% candle.

Addition and mixing. Add fragrance at the temperature the wax maker specifies; CSI's own guidance for CSI 464 is an 80–90°C melt and pour, to be confirmed on the product page. For other waxes, follow the maker's instructions. Stir thoroughly; two minutes is a common working practice at small scale, not a rule. What happens at the wrong temperature explains why a normal-looking candle can still be affected.

Cure, first burn, room and nose. These four account for a large share of "weak" verdicts. A candle judged at two days, before its pool reached the glass, in a large room with a fan, by someone who had been pouring all afternoon, has been set up to fail. Re-judge any rejected candle at a fair checkpoint, after a full first burn, in the target room, by walk-in scoring. If it passes, nothing was wrong with the oil.

Oil lot. If an oil used to perform and now does not, compare a retained sample of the old lot with the new one in two candles poured the same way. CSI does not publish batch-consistency or reformulation data for any oil, so ask on WhatsApp about your specific lot, and keep a retained sample of every lot you buy.

The 2 × 2 × 2 matrix

Once the free variables are clear, three remain that interact with each other: wax, load and wick. A higher load can need a different wick; a different wax can prefer a different load. Testing them one at a time can miss the combination that works. A small matrix tests every combination of two levels of each, which is eight candles. The loads here are 7% and 10%: both within the 13% maximum stated for Luxury Soy Wax Chunks and the 10% stated for CSI 464; for CSI 464, 10% is at its stated maximum. The wicks are Eco Candle Wicks Thin C1 and Thick C2. CSI does not publish wick sizes by jar diameter, which is exactly why both are in the matrix.

If eight candles is more than you want to commit, a one-at-a-time plan tests a baseline and then changes one variable per candle: wick, load, wax. It uses 68 g of oil and costs ₹1,784.12 in materials against ₹2,608.54 for the matrix. The saving is real; the cost is that it can miss a combination. For example, the higher load may only work with the thicker wick, and a one-at-a-time plan never pours that candle.

Reading the matrix and the verdict

Set the pass mark before any candle is burned, for example a mean of 3 on the anchored scale at the room centre at 60 minutes, in a room of the target size. Then read the matrix in this order.

Note the wording of the rejection verdict: rejected for this candle system. An oil that fails in your jar, wax and room may still be right for a smaller room, a wider vessel or a wax melt. Melts versus candles and room ratings explain why. And a rejection is worth recording in full, with the matrix results, so that you do not retest the same oil in the same way next season.

What to buy

The matrix needs one oil in some quantity, two waxes and two wick sizes. Peach Melon Berry is the worked example because it is a value-tier fruit oil in stock at every size, which keeps a thorough trial affordable; we make no claim about its throw, and CSI publishes no throw data for any oil. Prices are CSI live pricing, pulled 24 September 2026, with wicks from the 25 September pull.

This checklist closes the bottle-versus-wax series. For the oils that pass, finding the lowest effective load and a repeatable mixing and curing protocol take you into production; for a large shortlist, the 30-sample elimination sequence applies the same discipline at scale.

Frequently asked questions