How Do I Create a Candle Production SOP That Prevents Fragrance Separation When Scaling From 1 Kg Development Batches to 25–50 Kg Production?

Why a recipe is not enough at 25–50 kg

A development batch carries a lot of knowledge that never gets written down: how you stirred, when you checked the temperature, how long the jug sat before you poured, where the candles cooled. At 25–50 kg, other people, larger vessels and longer pours are involved, and each unwritten habit becomes a place where fragrance can separate. The SOP's job is to make the habits explicit and checkable.

This page lays out an SOP structure built for one aim: carry a stable 1 kg formula to production without introducing sweating. It uses a 7% load as the worked example; replace it with your own validated figure.

Section 1 — the approved formula, printed for every size

Begin with the formula specification, and print the arithmetic so it never happens on the production floor. Fragrance load is a percentage of total candle weight: 200 g at 7% is 186 g wax plus 14 g oil.

The 25 kg oil purchase works out as ₹3,068.00 + ₹1,652.00 + 2 × ₹354.00 + ₹212.40 = ₹5,640.40, and the 50 kg purchase as 3 × ₹3,068.00 + ₹1,652.00 = ₹10,856.00. Wax for the 50 kg run is 46,500 × ₹0.5074 = ₹23,594.10 at the CSI 464 10 kg rate; CSI lists no 25 kg or 50 kg pack, so 25 kg is bought as 2 × ₹5,074.00 + ₹2,537.00 = ₹12,685.00 and larger quantities are quoted on WhatsApp. All figures exclude jars, wicks, packaging, labour, shipping and GST.

Write the load basis in words on the sheet as well: "7% of total candle weight". CSI 464's page says "up to 10%" on a wax-weight basis, which is 9.09% of total. Your approved 7% is a choice you validated, not the page maximum, and the sheet should say that nobody may raise it for a "stronger" batch.

Keep the 1 kg development record attached to the formula sheet. When the production record for a 25 kg sub-batch is compared with it, every difference is visible at a glance, and the SOP can be corrected where production has drifted from development.

Note what the SOP deliberately does not contain: a range of loads for staff to choose from. One approved value per formula, printed in grams, is easier to follow and much harder to get wrong under pressure.

Section 2 — vessels and sub-batches

Few workshops melt 50 kg in one vessel. CSI's Electric Wax Melter is listed in 10 kg and 20 kg capacities, so a 50 kg production day is usually several vessel loads. The SOP should treat each vessel load as a sub-batch with its own weights, temperatures, times and sub-batch number on the record. Validate the largest single vessel load you will use; a formula proven in a 1 kg jug and assumed for a 20 kg vessel has not been validated at that size.

The melter page states a ±10°C deviation on its display, so the SOP must require a separate thermometer reading in the wax at every temperature step.

Add two rules that stop sub-batches blurring into each other. Never top up a vessel that is part-way through pouring with freshly melted, freshly scented wax; finish the sub-batch, record it and start the next one. And never combine the remains of two sub-batches in one pitcher unless the SOP explicitly allows it and both were made inside tolerance. Mixed pitchers are where traceability is usually lost.

Raw-material receipt belongs in this section too. Record the wax and oil lot numbers against each sub-batch, and keep oil at room temperature before use rather than bringing it straight from a cool store, so the addition temperature you measure is the one the batch actually experiences.

Sections 3 to 8 — the process, step by step

Mixing is where scale most often defeats a development recipe. A method validated in a 1 kg jug does not prove itself in a 20 kg vessel. During validation, draw a small sample from near the surface and another from near the base once mixing ends, pour each into a marked test jar, and compare those candles through observation. If the two behave differently, the mixing method does not reach the whole vessel yet.

Section 9 — QC, retained samples and release

Fragrance separation often appears days after pouring, so the SOP cannot release a batch the same afternoon. Set a release rule, such as: retain a set number of candles from each sub-batch (our suggested starting point is three, from first, middle and last pour), blot-test them at day 1, 3 and 7, and burn-test one after the stated cure on your oil's product page (48–72 h in soy for many CSI oils). Release only when all retained candles pass.

Section 10 — validating the SOP itself

An SOP is a hypothesis until it has produced clean candles at full scale. Run the first 25 kg or 50 kg day as a validation run: follow the document exactly, retain more candles than usual, and observe them through your full stability schedule — 14 days at minimum, with outside-room and warm-room samples if the candles will ship in summer. If anything sweats, compare the record with the 1 kg development batch, find the difference and amend the SOP before the next run.

If the validation run sweats even with every value inside its tolerance, the formula itself has too little margin at scale. Step the load down — 7% to 6% is 188 g wax plus 12 g oil per 200 g candle — before changing wax or oil. Less of our oil is a perfectly good outcome if the candle still throws well and stays clean.

Train against the document, not around it. Walk each person who will run production through one full sub-batch using the SOP, with the record sheet filled in as they go, and have them sign that they have done so. When someone suggests a better way of doing a step, test it as a change, validate it and then update the document; do not let improvements live only in one person's routine.

Once the validation run passes, sign and date the SOP, file it with the approved-formula sheet, and require that any change to wax, oil, load, vessel or batch size triggers a fresh validation.

Where this page fits

CSI already has guides that cover parts of this question. This page answers one sweating decision; these cover the neighbouring questions:

Three things to buy for this fix

Frequently asked questions