I Have Already Poured 200 Candles From a Questionable Batch — Should I Hold the Remaining Production Until Samples Have Been Evaluated?

Stop first, investigate second

Something made you uneasy: the thermometer read far above your validated window, the oil went in later than planned, the wax looked streaky, a pour felt thinner than usual. Whatever the trigger, the batch is now questionable, and the question on this page is simple: keep going or pause?

Pause. Two hundred candles already contain whatever went wrong. The remaining wax may be affected too, and pouring it simply multiplies the number of candles you may need to rework or scrap. Holding costs time; continuing costs material, jars, wicks and labour on top of an unknown.

Stopping also preserves evidence. Once the last of the wax is in jars, you lose the chance to sample the melter contents directly, and the story of what happened is spread across 250 containers instead of sitting in one vessel.

Containment in the first hour

Sample the poured candles by position

A fault that appeared part-way through the run should, in principle, affect only candles poured after it. Sampling by position tests that idea. Take candles from the start of the run, from both sides of the point where doubt arose, and from the end.

Cure all samples together, burn them in the same room on the same day, and code them blind so the judge does not know which is which. Smell only. The sample counts above are a suggested starting point; add more near the trigger point if you need a sharper boundary.

Worked example: a 50 kg coffee batch

Suppose the batch was 50 kg of scented wax at 8% of total weight: 46 kg of Natural Soy Pearl Wax and 4 kg of Roasted Coffee, planned for 250 candles of 200 g. Two hundred had been poured when the thermometer showed the melter had been running well above the validated window for the last stretch of the run.

Material value. Pearl Wax in the 50 kg pack is ₹21,240.00, or ₹424.80 per kg, so 46 × ₹424.80 = ₹19,540.80. Roasted Coffee at ₹3,738.00 per kg gives 4 × ₹3,738.00 = ₹14,952.00. The batch holds ₹34,492.80 of wax and oil. The 200 poured candles are 80% of it: ₹34,492.80 × 0.8 = ₹27,594.24, plus 200 thick eco wicks at ₹9.44 each, ₹1,888.00. The unpoured fifth is worth ₹34,492.80 × 0.2 = ₹6,898.56.

Roasted Coffee's page states a flashpoint of approximately 191–193°C, well above melter range, so the question is not safety but volatile loss. Its page also notes vanillin in the base and slight yellowing in white wax, described as cosmetic — so a colour shift in later candles may be heat-related and is worth recording alongside throw.

Now suppose the samples show S1 to S3 matching the reference, S4 and S5 weaker, and the evidence jars weaker still. The boundary sits near the trigger point. Candles before it can be released once burn tests pass; the later range and the held wax stay on hold for a rework or write-off decision.

Set that against the cost of carrying on. Pouring the last 10 kg would have added 50 jars, 50 more wicks at ₹9.44 each — ₹472.00 — and the labour of filling, labelling and packing, all onto wax that the samples later showed was the weakest part of the batch. Stopping protected that ₹6,898.56 of wax from becoming finished candles you could not sell.

Looking after the quarantined candles

Held candles still need care, because the decision about them depends on tests that take days. Leave them in the same conditions as your normal curing stock — out of direct sun, away from heat sources, lidded if that is your usual practice — so that when you do release them, they have cured the way your approved candles do.

Hold off on the final steps that are expensive to undo. If labels, sleeves or gift boxes have not gone on yet, keep it that way until the decision is made; stripping packaging from rejected candles doubles the loss. If some candles are already packed, mark the outer cartons HOLD in large letters and log how many sit in each.

Keep the cure clock honest. The day the batch was poured is day zero for every sample, including those from before the trigger point. Burn-test at the product page's stated cure minimum and again later, so that a candle judged weak at day 3 is not condemned before it had the same chance as the approved reference.

Decide: release, part-release, rework or write off

Rework of poured candles is possible but slow: remove the wax, clean the jars, remelt and re-validate. It rarely pays for a large range. Write-off is a legitimate decision when the affected value is below the cost and risk of rework.

Whatever you decide, close the hold on paper. Write the release decision, the sample results behind it, the candle numbers released and the numbers rejected, and who signed it off. Keep one or two candles from each side of the boundary as retained samples; if a complaint arrives later, they show exactly where the line was drawn and why.

To prevent a repeat, give the pour a trigger list — conditions that mean stop and call someone: wax outside the validated window on the separate thermometer, oil added outside its planned time, a visible change in the wax. Write the list on the batch sheet so the decision is made in advance, not under pressure at candle 200.

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