My Candle Fill Weight Is Correct While Hot but Differs After Cooling — Which Measurement Should My Production QC Standardise?

What cooling changes, and what it does not

Melted wax is less dense than set wax. As a filled jar cools, the wax contracts, the surface drops, and in many soy blends a small dip or sinkhole forms around the wick. What you see is a lower fill line. What the scale should see, if nothing else has happened, is almost exactly the same mass it saw when the jar was full of liquid.

That is the key to this question. A candle poured to 200 g net should weigh very close to 200 g net the next morning. A tiny loss is possible, because a small fraction of volatile fragrance material leaves the surface while the wax is hot, but it is not the kind of difference an operator notices on a production scale. If your team is seeing a clear gap, several grams per candle, the cause is almost always in the measurement or the method.

As a rough guide, 200 g of melted soy wax occupies about 222 ml at roughly 0.9 g per ml. The set candle takes less room. That is why filling to a line while the wax is hot gives a lower line once cold, and why the line is a poor production standard on its own.

Find where the difference is coming from

Treat the gap as a small investigation. Four causes cover almost every case, and each can be tested in one production session.

1. The scale is drifting under a hot jar

A jar of wax near its pour temperature warms the scale platform and sets up rising air currents around it. Many digital scales creep for a while under those conditions. The operator reads 200 g at the moment of pouring; the cooled candle reads 197 g on the same scale the next day, and the scale was simply wrong while hot.

Test it directly. Pour one jar, leave it on the scale and note the reading at the moment of fill, then every minute for five minutes, a suggested starting point. If the display creeps, place a heat-resistant insulating mat on the platform, tare it, and repeat. A drift that disappears with the mat was the scale, not the wax.

2. Top-ups were added after the first weighing

Many soy candles get a second pour to fill a sinkhole. If the operator weighed 200 g on the first pour and then added 8 g of top-up without weighing, the finished candle is 208 g. That shows up as a cold weight higher than the hot one, and it also means your fragrance maths and costing assume less wax than was used.

3. Wax was removed after weighing

Drips wiped from the rim, a spill tipped out, or a crooked wick straightened with wax scraped away all reduce the finished weight. Each is small, but a team working fast can remove a few grams from many candles.

4. The tare was not the same

If one person tares the scale with the empty jar and another subtracts a nominal jar weight later, the two figures are measuring different things. Settle on one method: tare the actual jar with its wick fitted, every time.

Worked example: 200 g hot, 197 g cold

Suppose a maker pours CSI 468 at the wax maker's stated pour temperature of 58–60°C into 200 g jars. Operators weigh each jar hot and stop at 200 g net. The next morning QC weighs ten candles and finds an average of 197 g net. The following readings are illustrative, to show how the test separates causes.

Nothing about the wax changed in this example. The formula was right; the measurement was not. That is the usual outcome, which is why the scale and method come first in any hot-versus-cold investigation.

What to do with a batch already weighed the wrong way

If a batch was released to the shelf on hot readings and you now suspect drift, the candles are still there to be measured. Do not guess at the correction; weigh the evidence you already have.

Take a sample of cooled candles from the start, middle and end of the pour, ten from each as a suggested starting point, and weigh them net using each jar's own tare where you recorded it, or an average of ten empty jars from the same carton where you did not. Compare the cold average against the target. If it sits a few grams low across the whole batch, the scale drifted consistently and the fix is method, not rework. If only the early or late candles are low, the drift followed pour temperature or scale warm-up, and the batch sheet should show which candles are affected.

Release candles that meet your cold acceptance band. Hold any below a stated label weight and either remelt them into the next batch of the same fragrance or sell them where the claim allows. Record the decision on the batch sheet so the next reconciliation of yield and cost uses the corrected figures rather than the hot ones.

The QC standard to write into your SOP

Once you know why the numbers differed, write down which measurement does which job. Ambiguity here is what lets two people argue about whether a candle passed.

For costing and yield, use the cooled net weight. It includes every top-up and excludes every gram that was wiped away, so it is the true amount of scented wax that left your batch in that candle. If the cold average consistently sits above target because of top-ups, your planned batch weight is too small for the number of candles you are promising.

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