Should I Scale From 1 Kg to 5 Kg to 10 Kg to 25 Kg or Can I Safely Jump Directly to a Much Larger Batch Once Percentages Are Established?

Why correct percentages do not make a jump safe

Once you know the formula is, say, 92% wax and 8% oil by total weight, multiplying it to any size is simple arithmetic. That part rarely fails. What fails is everything the recipe card does not describe: how quickly a deep tank of wax loses heat, how long it sits hot before and during pouring, whether a bamboo stick can actually move oil from the top of a full vessel down to the bottom, and how many minutes separate the first candle from the last.

A 1 kg batch melts quickly, is scented and stirred in moments, and is poured before its temperature moves much. A 25 kg batch may need several melts, loses heat slowly in the middle and fast at the walls, and can take long enough to pour that the last jars receive wax at a different temperature and after a longer time at heat than the first. The percentage is identical across all of it. The candle may not be.

That is why the real question is not ‘1–5–10–25 or jump?’ but ‘at which sizes does my process change, and have I proven each change?’

Build a ladder from process changes, not from numbers

List every batch size you intend to make and, for each, what is physically different about how it is made. Where something is different, you need a proven step. Where nothing is different, a step is a cost with no information in it.

Take a real example. A maker uses a double boiler for 1 kg tests and owns the 10 kg-capacity CSI melter, whose product page states 9.5 litres per melt — roughly 8.5 kg of melted soy wax at about 0.9 g per ml. Now look at the proposed ladder. At 5 kg, the melter is used for the first time: keep that step. At 10 kg, the batch no longer fits one melt and must be split: that is a real change, so a step is justified. At 25 kg, three melts are needed: if the 10 kg step already proved how two melts are matched, the 25 kg batch adds only more of the same, and can be run as a held first batch.

Putting rupees on each step

Price every step so the decision is made with the real cost in view. Take CSI 468 with Dark Vanilla at 8% of total weight.

Two things stand out. First, oil is more than half the material cost here, so a batch lost to a process fault throws away the most expensive ingredient. Second, a step that passes is not wasted: its candles can be sold once their retains match the control. The honest cost of a step is therefore the risk of failure times the batch value, plus the time — not the full batch value.

Compare the two routes. Jumping straight from 1 kg to 25 kg puts ₹29,943.00 on a single untested process. Taking a justified 10 kg step first puts ₹11,977.20 at risk while the process is unknown, and the 25 kg batch that follows then carries a much smaller unknown.

When a direct jump is defensible — and when it is not

A direct jump from 1 kg to a full commercial batch is reasonable when three conditions hold together. Your 25 kg process is already running reliably for other products. The same melter, pitchers, jars and crew will be used. And the new formula differs from a proven one in a small, understood way — typically the oil, at a load inside the stated range on the product page.

Even then, the first large batch is a trial, not a release. Number the candles in pour order, keep retains from the start, middle and end, and hold the batch until those retains match a 1 kg control poured the same day and cured the same length of time.

A jump is risky when any of the following apply:

Red flags for skipping steps

New equipment. The first run on a new melter is always a step. A new wax. Different waxes hold heat and accept oil differently. A load near the stated maximum. Less margin for uneven distribution. A low-flashpoint oil. CSI product pages publish flashpoints as low as 85°C for Neroli and 98°C for Gardenia, which narrows the safe addition window in a hot tank. A vanillin oil in light wax. Colour drift between first and last candles is easier to create and to see. A new crew. Hand mixing and pouring speed vary a lot between people.

Whatever the oil, add it below its published flashpoint and never with the melter at its maximum setting. Large volumes of molten wax cause deep burns; keep pouring areas uncluttered and cords out of the walkway.

How to prove each step you keep

A step is only proven when it is compared against something. Pour a fresh 1 kg control from the same wax and oil lots on the same day as the step batch. Cure both for the same period — the product pages state 48–72 hours minimum for soy — and, as a suggested starting point, recheck at around one and two weeks.

For each step, record the same short set of figures: wax temperature at oil addition from a separate thermometer, time spent mixing, time from oil addition to first pour, time from first pour to last pour, and the number of candles filled against the arithmetic. When a later step fails, those records tell you which figure moved.

Compare beginning, middle and end candles with the control for throw, melt pool, tops and colour. If they match, the step is proven and the next size inherits its settings. If they do not, the step has done exactly what it was for: shown you a process problem at a size where it costs thousands rather than tens of thousands.

If a step fails, keep its candles apart and labelled rather than selling the ones that look fine. Diagnose at that size by changing one variable at a time against the control — mixing time, oil addition temperature or hold time — and repeat the step once the fix is written into the batch sheet. Only a repeat that passes counts as proof; a fix that was never re-run is still a guess. The next size up should wait until then, however tight the order deadline looks. Never leave a test burn unattended, and extinguish any candle that smokes or flares.

Where this page fits

CSI already has guides that cover parts of this question. This page covers the failure investigation for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions