I Changed from Soy to a Soy-Paraffin Blend to Improve Hot Throw — Do I Need New Wick, Fragrance-Load and Stability Tests?

Why a throw-driven change needs more testing, not less

A supplier swap asks "is the new wax the same?". A move from soy to a soy-paraffin blend asks the opposite: you want a difference, in hot throw. That changes the whole test. It is not enough to show the blend is no worse than soy on every line; you also have to show that it is actually better on the line you changed it for, and that the improvement did not come from somewhere you would not accept. In general candle-making practice, paraffin and paraffin blends are commonly chosen for projection. The same practice notes that a bigger or hotter burn can also raise throw. A test that only asks "does it smell stronger?" cannot tell those two apart.

So the short answer to the three things in the title is yes, yes and yes. The wick has to be re-established, because the melt pool is a property of the wax system and you have changed the wax system. The fragrance load becomes a new question, because the ceiling you relied on — CSI 464's 10% — belongs to CSI 464 alone, and no ceiling is quoted for the paraffin or the blend. And stability — tops, adhesion, sweating in warm storage — has to be repeated, because the solid candle is a different material. The table below adds the lines makers usually forget: the benefit test itself, the blend ratio, the mixing order and the label.

The three gates, and why the order matters

The diagram above sets out the logic. Gate 1 — safety. Every blend candle is scored beside a soy control on flame height, soot, mushrooming, jar heat and tunnelling. If the blend is worse on any of them with Wick A, try Wick B; if it is worse with both, the blend fails, however well it throws. Gate 2 — benefit. Only the candles that passed gate 1 go into the blind throw sessions. If the blend does not beat the control, stop: you have a candle that is no better at doing what you changed it for, and the soy candle is already validated. Gate 3 — stability. The winner is stored and burnt to the end to prove it holds up through the warm season and the last third of the jar.

The order saves work as well as risk. If you run the throw panel first and fall in love with the result, the safety data arrives as an inconvenience rather than a gate. Running safety first means the panel only ever smells candles you could actually sell. The hotter-jar post covers the same trade-off when a new wax throws better and runs hotter, and the wick-driven version shows it from the wick side.

Test protocol with candle counts

The protocol is a working practice, not a standard: three candles per variant, 3–4 hour burn cycles with full cool-down, cure checkpoints at 48 hours, 7 days and 14 days as labelled examples. The soy is still available, so pour a fresh control the same day. The candle is 200 g at 8%: 16 g of Bergamot Amber and 184 g of wax, which in the 70:30 blend is 128.8 g of CSI 464 and 55.2 g of paraffin. Weigh both — ratios estimated by volume are not formulas.

The safety sheet is the same one used across this series — melt-pool timing and depth, flame height, soot, mushrooming, jar heat, consumption in grams per hour, appearance and sweating — with one rule on every safety line: no worse than the soy control, and within your jar supplier's guidance and the fire-safety standard you work to (ASTM F2417 is commonly cited in the US). CSI publishes no numeric flame-height or jar-temperature limit. The benefit sheet above is the addition this change needs.

The priced test kit and what the blend does to cost

Everything in the kit is a listed, in-stock CSI size. Oil per candle follows the CSI convention (load = share of the finished candle), with a 5% pour-loss allowance as a working assumption. The whole test needs 523.5 g of paraffin — just over one 500 g pack — so the 1 kg pack at the same per-kilo price is the buy, and the rest is the start of your first production batch if the blend passes.

Now the part that often makes the blend attractive regardless of throw. Paraffin pellets cost ₹354.00 a kilo against CSI 464's ₹507.40, so every point of paraffin in the blend lowers the wax cost. The table below prices the wax in one 200 g candle at 8% (184 g of wax) at different soy : paraffin ratios. The ratios are examples; only the one you test is a formula.

At 70:30 the wax in each candle falls from ₹93.36 to ₹84.89, a saving of ₹8.47. With Bergamot Amber at its 1 kg rate, the full wax-plus-oil cost moves from ₹180.35 (soy, 8%) to ₹171.89 (blend, 8%) — and to ₹161.93 if stage 2 shows the blend holds its throw at 7%. That last number is why stage 2 is worth three candles. The total-economics post sets out how to compare waxes on the finished candle rather than the kilo price, and the payback post shows how many candles it takes for a saving like this to cover the testing.

Decision rule, release and the new formula card

When the blend is approved, the formula card gets a new version with the ratio as a weighed line ("128.8 g CSI 464 + 55.2 g Fully Refined Paraffin Wax Pellets per 200 g candle"), the melting order, the temperatures and their source, the load and its source, the wick, the test-candle numbers, the panel sheets and the sign-off. Then the label: a soy-paraffin candle should be described as one. Release in a small first batch with two candles retained as the new controls, and keep one burning to the end while the rest ship. The formula-record post shows the layout.

Where soy-paraffin switches go wrong

The first error is scoring throw unblinded. Once you know which candle is the blend, you will smell what you expect. The second is testing the blend at a higher load "to really see the difference". That changes two variables — wax and load — and any throw gain is now unexplained. Keep the load where it was until the wick and the throw are settled, which is the one-variable rule explained here.

The third is treating paraffin as an additive rather than a wax. At 30% it is nearly a third of the candle's wax, and everything that depends on the wax — melt pool, adhesion, sweating — is in play. The fourth is keeping the soy wording on the label. And the fifth is skipping the full-life burn because the first three cycles looked good: the harder-wax post explains why the last third is where blend changes most often show their real effect. If the move to a blend was driven by a stock problem rather than throw, the emergency-order post covers staged release under time pressure, and the master checklist lists every line for any wax change.

Frequently asked questions