My Candle Looks Stable for a Week and Then Suddenly Develops Oil Droplets — Should Stability Testing Run Longer Before Commercial Approval?

What one week actually proved

Your test did its job. It told you the candle can hold its oil for seven days in the conditions it sat in. Then it told you something more important: it cannot hold it for longer. A formula that sweats on day eight, ten or twelve is not a stable formula that had bad luck. It is a borderline formula that needed more time to show itself.

This is common enough that a one-week window is best treated as an early warning check rather than an approval. The processes that push oil out of a candle are slow. Soy wax keeps settling for days. Surplus oil travels slowly through solid wax. And a week of warm afternoons and cool nights is only seven small cycles; a month of storage and a hot courier route add many more.

CSI publishes no stability duration and no heat-resistance figure for any of its waxes. Nobody can tell you from a label how long your formula will stay dry. Your observation log is the only evidence, so its length matters.

An approval window built as gates

Instead of one long wait, set gates. Each gate has a pass condition and a consequence. The day counts below are suggested starting points; extend them if your stock sits for longer or ships in hot months.

Three candles per gate set is a reasonable starting point: one in a steady room, one in the place your finished stock will actually sit, and one for heat cycling. Keep the burn-test candles separate, because a burned candle can no longer tell you about storage.

Worked example: what early approval puts at risk

Suppose, as an example, that you had approved on the day-7 read and poured a 200-candle run: 200 g candles at 8% of total weight in CSI 464 with Roasted Coffee, in matte 200 ml jars with Eco C1 wicks. Each candle holds 16 g of oil and 184 g of wax, so the run uses 200 × 16 = 3,200 g of oil and 200 × 184 = 36,800 g of wax.

Materials at CSI's listed prices: wax 36,800 × ₹0.5074 = ₹18,672.32; oil 3,200 × ₹3.738 = ₹11,961.60 at the 1 kg rate; jars 200 × ₹70.80 = ₹14,160.00; wicks 200 × ₹5.90 = ₹1,180.00. Total ₹18,672.32 + ₹11,961.60 + ₹14,160.00 + ₹1,180.00 = ₹45,973.92, before packaging, labels, labour, shipping and GST. If the formula sweats on day ten, that whole run is now a rework-or-scrap decision.

Now the cost of waiting. One test candle of the same formula costs 184 × ₹0.5074 = ₹93.36 in wax, 16 × ₹3.738 = ₹59.81 in oil, ₹70.80 for the jar and ₹5.90 for the wick: ₹229.87 in all. Six extra test candles to carry the formula through days 14 and 28 and a heat cycle cost 6 × ₹229.87 = ₹1,379.22. That is the price of knowing.

Set the reminders before you pour. The most common reason a long-hold test fails to protect a launch is not the formula; it is that day 14 and day 28 arrive in a busy week and nobody checks. Put each gate date in your calendar with the candle codes, and decide in advance who blots, who photographs and who records the room conditions. A gate that is not checked is a gate that was not passed.

Some oils deserve the full window more than others

Every oil needs its own approval, but some give you more reason to be patient. Roasted Coffee's page notes vanillin in its vanilla and caramel base, without giving a percentage. Royal Oud's page suggests a 72–96 hour cure because oud takes longer to express. Vetiver's page says it builds slowly. Oils like these may change in character over the first weeks as well as in stability, so a long window serves both questions at once.

Strength ratings on the product pages, such as "Strong" or "Moderate to strong", are the pages' own descriptions, not measured data, and they say nothing about how long an oil will stay put in your wax. A strongly rated oil is not automatically a sweating risk, and a moderate one is not automatically safe. Only your gates tell you.

Do not assume an oil that passed in one wax will pass in another, or that an oil that passed at 7% will pass at 8%. Each change of oil, wax or load restarts the window. That sounds slow. It is much faster than recalling stock.

What to do with the candles that already failed

The candles that went dry for a week and then wept are still useful. They are your best evidence of where the formula breaks, so keep a few of them back from any rework as a reference, labelled with their pour date and the day oil first appeared.

For the rest, the choice is rework or scrap, and it depends on how much stock there is and what the retest shows. If a lower load turns out to pass every gate, the dilution maths covered elsewhere in this series tells you how much unscented wax a remelt needs to reach it, and how many extra jars and wicks that creates. A remelt resets cure and can lose some fragrance to heat, so the reworked candles go back through the window too. If the retest shows the oil itself does not sit well in this wax, rework at a lower load may still fail, and scrapping the wax and reclaiming clean jars can be the honest call.

Whatever you choose, none of the failed candles goes out in the meantime. Wiping the tops and lidding them hides the first sign of the fault, not the fault itself, and the customer meets the candle after its next warm week.

Sign off a specification, not a feeling

When a formula passes every gate, write it down as a specification and treat it as fixed. If anyone changes a line, the formula goes back to G1.

Only once the specification is signed do you place the bulk order. For CSI 464 beyond the listed 10 kg pack, ask on WhatsApp for a quote based on your approved grams per candle, not on the load you started with. And the candles from the batch that sweated after a week stay off sale while you work through the gates.

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