Should Fragrance Oil Be Added Directly to a Commercial Wax Melter or Should Unscented Wax Be Transferred Into Smaller Controlled Batches First?

Two kinds of control, pulling in opposite directions

Both methods put the same oil into the same wax. What differs is where the risk sits. Adding fragrance directly to a commercial melter makes the weighing easy and the batch large, but leaves a big mass of scented wax sitting hot for as long as it takes to pour. Transferring unscented wax into smaller controlled batches keeps each batch's time in the heat short, but multiplies the number of weighings and makes each one more sensitive to error. The right choice depends on which of those risks your production actually runs into.

The weighing argument favours the tank

Any scale has some reading error. CSI does not publish the capacity or resolution of its own scale, so the figures here are purely an illustrative example: imagine a scale whose reading could be off by 1 g either way. On a 20 kg batch at 8% load, you weigh 1.6 kg of oil once. A 1 g error on 1,600 g is about 0.06% of the oil. Split the same 20 kg into ten 2 kg sub-batches and each needs 160 g of oil. A 1 g error on 160 g is about 0.6% of that portion's oil — ten times the relative error, repeated ten times.

Those percentages are small in absolute terms, and a steady hand can keep them smaller. The point is direction: every time you divide a batch, you make each weighing matter more. If you choose sub-batches, the scale and the discipline around it become the most important part of the method.

Reducing error when you do divide

If you choose sub-batches, a few habits claw back most of the lost accuracy. Pre-weigh the oil for each sub-batch into its own labelled container before any wax is portioned, so the oil weighing happens calmly at a clean bench rather than next to a hot melter. Weigh the wax portion into a tared vessel and calculate that portion's oil from the actual wax weight rather than a target. And weigh the oil onto the scale on its own, not poured into wax already sitting on the scale, where drips and heat make the reading harder to trust.

The heat argument favours smaller batches

Fragrance oils contain components with different volatilities, and the lighter ones leave faster the longer scented wax is held hot. A large batch poured slowly spends a long time at temperature after the oil has gone in. Suppose, as an assumption to test, that your team pours 60 candles an hour. A 20 kg direct-scented batch of 95 sellable 200 g candles is then held for well over an hour and a half between the first pour and the last. A 2 kg sub-batch of ten candles is poured in about ten minutes.

Whether that difference shows in the finished candle is something only your own testing can tell you. Pour the first, middle and last candle from a large direct-scented batch into labelled jars, cure them together, and compare their cold and hot throw side by side. If the last candle is noticeably weaker, the hold time matters for your oil and your wax, and smaller batches will help. If you cannot tell them apart, direct scenting is working.

Run that comparison more than once, and with more than one fragrance. Oils behave differently, and a result from a heavy woody scent does not prove anything about a light citrus one poured the same way.

Temperature control in each method

There is a second, quieter advantage to transferring. With direct scenting, the whole tank has to come down to the addition temperature before the oil goes in, and then stay there for the whole pour. With sub-batches, the tank can hold unscented wax at whatever temperature suits melting and holding, within the wax maker's guidance, while each portion is brought to the addition temperature on its own. For Eco Soy CSI 400, the wax maker states adding fragrance at 80–90°C and stirring for 2 minutes. For Premium Coconut Soy CSI 468, the stated pour temperature is 58–60°C. A small portion reaches those windows quickly; a full tank takes much longer.

The stirring guidance deserves care. A stated 2 minutes of stirring describes normal making quantities. Treat it as a minimum for a 20 kg tank, not a target, and validate your own mixing with the first-middle-last comparison above.

Choosing between them

There is also a middle route that suits many workshops: scent the tank directly, but only as much as you can pour within your validated window. If your first-middle-last test shows 50 candles hold up and 95 do not, fill the 20 kg melter with 10 kg of scented wax per batch. You keep the single-weighing advantage and cap the hold time.

What to write down in either method

The batch sheet looks slightly different depending on the route. For direct scenting, one line per tank batch: date, fragrance, oil lot, wax weight, oil weight, addition temperature on your thermometer, stirring time, and the time of the first and last pour. That last pair is what tells you, months later, how long the batch was held hot. For sub-batches, the same fields repeated per portion, plus which jars or trays that portion filled. The extra lines are what let you trace a complaint to one portion rather than a whole day.

If you move from one method to the other, note the date of the change on the batch sheet. When you compare candles across months, you will want to know which were scented in the tank and which in portions, rather than trying to reconstruct it from memory.

Keep the first-middle-last candles from any new batch pattern until the burn tests are done. They cost three candles, take a little shelf space while they cure, and are the only direct evidence you will have that your chosen method keeps the batch uniform.

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