My 1 Kg Soy Candle Test Batch Has Excellent Hot Throw but the Same Formula Is Weaker When I Make 10 Kg — What Could Be Changing During Scale-Up?

The formula did not change. The process did.

When a 1 kg test candle fills a room and the 10 kg run is noticeably weaker, the instinct is to blame the fragrance oil or a bad carton of wax. Occasionally that is right. Far more often the percentages on paper are identical but the wax went through a different journey: it took longer to heat, sat hot for longer, received its oil at a different temperature, was stirred across a larger mass and was poured over a longer session. Each of those can shave something off hot throw, and together they add up to a candle that smells like a paler copy of the test.

The useful way to approach this is as a list of suspects with a way to clear each one, cheapest check first. None of it needs new equipment. You need your batch sheet, a thermometer, a few labelled test candles and some patience while they cure.

Suspect one: the load was calculated differently

Start here because it takes two minutes. CSI calculates fragrance load as a percentage of the total candle weight. Your 1 kg test at 8% was 920 g wax plus 80 g oil. The correct 10 kg version is 9.2 kg wax plus 0.8 kg oil.

The common slip at scale is weighing out a round 10 kg of wax, because that is one full pack of CSI 464, and then adding 800 g of oil. That batch weighs 10.8 kg and runs at roughly 7.4% (800 ÷ 10,800 = 0.074). The difference looks like a rounding error on paper. It is often noticeable from across a room once the candle is lit.

If your sheet shows the round-pack error, you have probably found most of the gap. If the maths was right, keep going.

Suspect two: the wax lived hotter for longer

A 1 kg jug heats in minutes and cools almost as quickly. Ten kilograms holds far more heat, so it reaches temperature more slowly and takes much longer to come back down. That creates two risks. You may have added oil hotter than in the test because you were waiting for the batch to cool and ran out of patience. And even at the right addition temperature, the scented wax almost certainly stayed hot for longer before the last candle was poured.

Fragrance oils are blends of many materials, and the lighter ones leave more readily when warm. A long hot hold gives them more time to go, and the loss usually shows up as a flatter throw rather than a changed smell. That is a general principle, not a measured figure for any particular oil, so treat it as a hypothesis for your test candles to confirm or rule out.

If you used a large electric melter, note that CSI's Electric Wax Melter (Imported) lists a display range of 60°C–100°C with a stated ±10°C deviation. A display reading is not a wax temperature. Measure the wax with a separate thermometer and add oil at the wax maker's stated temperature, not at whatever the display shows.

Suspect three: the oil never reached every candle equally

In a 1 kg jug a spoon moves the whole mass. In 10 kg the same effort reaches a smaller share of the wax, and oil that is not fully worked in can sit unevenly. The symptom is not uniform weakness but a spread: some candles strong, some weak. If you only burned one or two candles from the run, you may simply have picked weak ones.

The way to know is to burn candles from the start, middle and end of the pour, each labelled with its position. If throw varies across that set, mixing or settling is involved. If all three are equally weak, look back at the maths and the heat history.

Settling is worth a separate thought. A batch that sits for a while between pitcher fills is not being stirred, and if the oil was only partly incorporated it has time to drift. Re-stirring the whole melter before every fill costs a few seconds and removes that question from the investigation entirely.

Look hard at the cure row. It is very common to compare a test candle that has rested for a couple of weeks with a production candle burned a few days after pouring. Match the cure before concluding anything.

A controlled rerun that isolates the cause

What to decide after the rerun

If the 1 kg control and the 10 kg candles now throw alike, the process fix worked; write the new controls into your production sheet and stop there. If the fresh control is also weaker than your original test, the difference sits in materials, such as a new oil lot or wax pack, and the investigation moves to comparing batches rather than process. If the 10 kg candles vary by position, focus on mixing and hot hold time.

One more habit makes the next scale-up easier: keep the rerun sheet. Weights, the three temperatures, the stirring routine, pour start and finish times, cure dates and the blind burn notes together form the first real production record for this candle. The next time a batch disappoints, you compare it line by line against a sheet that worked instead of relying on memory.

What you should not do is raise fragrance load to cover the gap. Soy wax has a ceiling, CSI 464 states up to 10%, and pushing oil up to hide a process fault costs money on every candle and can bring sweating or a poor burn. Fix the process first. You may find the formula you tested was right all along.

Where this page fits

CSI already has guides that cover parts of this question. This page takes the production-floor angle; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions