I Forgot to Stir My Fragrance Oil Properly and Have Already Poured the Candles — Can I Save Them by Remelting the Entire Batch?

Why this is a variance fault, not a quantity fault

Most of the expensive mistakes in candle making are arithmetic: you measured 100 g of oil where the recipe wanted 80 g, and every candle in the batch is wrong by the same predictable amount. Those faults are unpleasant but tractable, because you can calculate your way out of them — that is what the 100 g instead of 80 g correction does. An under-stirred pour is a different species of problem. Your batch average is exactly right. The oil-rich wax near the top of the pot went into the jars you poured first or last, depending on how you laddled, and the oil-poor wax went into the others. No single jar is the recipe.

That has one consequence which should decide your behaviour immediately: you cannot fix this by testing. Burning one jar tells you about that jar. Burning three tells you about three. There is no sample size that makes eighteen jars of unknown, unequal load into a product, because the thing you would be measuring is not a property of the batch — it is an accident of the pour. Nor can you sort them by smell: cold throw at day two is a poor guide to hot throw at day fourteen for reasons set out in how cure time changes the smell of different fragrance families, and a nose cannot resolve a two-point difference in load.

Why all eighteen jars go back into one pot

This is the single technical decision on the page, so it is worth doing the arithmetic out loud. You weighed in 2,000 g of wax and 200 g of fragrance oil: a batch of 2,200 g at a load of 10% of wax weight. You poured eighteen 120 g jars — 2,160 g — and about 40 g stayed in the pot and on the utensils. Whatever the distribution across those eighteen jars, the sum of the oil in them is still roughly 200 g minus whatever went into the 40 g residue, and the sum of the wax is still roughly 2,000 g minus the same. Pool them and the ratio reasserts itself.

Remelt them separately and nothing reasserts itself. Jar 4 melts back into a 120 g puddle at whatever load jar 4 received, and you pour it back into jar 4. You have spent eight minutes to produce the same defective candle with a fresh wick in it. This is the reason the title's word "entire" is load-bearing, and it is why the practical instruction is: do not begin until you have a vessel that will hold the whole batch. At 2,160 g of wax you need roughly 2.4 litres of headroom; a 1-litre Mini Electric Wax/Soap Melter will not do it in one pass, so either work in two halves that you then combine in a larger pot before pouring, or use a pot big enough for all of it.

One more reason the combined pot matters: it gives you a weight you can check. Weigh the pooled, molten batch. If it reads 2,100 g rather than 2,200 g you have lost about 100 g between the jars, the utensils and the original pot residue — and you now know your top-up. That is where the 500 g pack of Luxury Soy Wax Chunks at ₹299 earns its place: it lets you correct a recovery shortfall in wax rather than pouring fewer candles. If you top up with wax alone, remember that you have very slightly lowered the load — 200 g of oil in 2,100 g of wax is 9.52% of wax weight rather than 10% — which is a change small enough to accept and large enough to write down.

Getting the wax out without losing the glass

The wax is the cheap part of this operation. Eighteen Amber Shiny Glass Jars at ₹47.20 each are ₹849.60 of glass, which is 40% of the material value you are trying to rescue, and glass is the thing a careless decant destroys. Work at the slowest temperature that moves the plug.

The 18-jar recovery ledger

Here is the whole operation in rupees, with every assumption labelled. The worked batch is 2,000 g of Luxury Soy Wax Chunks at ₹599 a kilo plus 200 g of Apple Cinnamon Fragrance Oil, priced from the 500 g pack at ₹2,371.80, which is ₹4.74 a gram. Substitute your own oil; the shape of the answer does not move.

The ratio to remember is ₹117.10 recoverable against ₹28.40 to recover it — a little over four to one. That is why "remelt" wins on money at any home or small-studio scale, and it keeps winning as the jars get bigger, because the material in a jar scales with its fill weight while the labour per jar barely moves. A 200 g candle holds ₹195.17 of wax and oil at the same rates and still costs about ₹28.40 to recover. The route where remelting stops paying is the one where labour per unit rises instead: hundreds of candles, a paid team and a deadline, which is the subject of remelting 100 candles together or in smaller batches.

The cure calendar — the constraint that actually decides this

Every number above says rescue. The calendar is the only thing that can say no, and it says it absolutely: you cannot compress a cure. The Luxury Soy Wax Chunks page asks for a minimum of 48 hours and ideally one to two weeks before burn testing; the working standard across this corpus for soy containers you intend to sell is 14 to 21 days, and that is an assumption you should treat as a planning figure rather than a specification. Repouring restarts it from zero. There is no partial credit for the days the first pour already sat.

Read the last two rows together, because they are the actual decision. If your deadline is Diwali or later, remelt: you have the days, and ₹2,146.72 of material is worth 2.4 hours. If your deadline is inside two weeks, the remelt does not solve your problem even though it saves your material — so do both things separately. Remelt the batch for stock you will sell later, and tell the customer today that the order will be late or short. That is a better conversation on the 29th of September than on the 7th of October, and it is the only version of this that does not end with an untested candle going out of the door because the date arrived.

What you need to buy to repour

Four lines, and only one of them is wax. Each card lists every size CSI stocks with the live price, so you can size the order to eighteen jars rather than to a guess.

What cannot go back on sale

Three categories, and being firm about them is what makes the rest of the rescue worth doing.

The original eighteen, in any form. Not the strong ones, not the ones that look perfect, not the ones you would keep for yourself. They have unknown and unequal loads, and the strong ones are the specific jars most likely to be above the wax's stated capacity. Once pooled and repoured they become sellable stock again — after a burn test. Before that they are material.

Any jar the decant damaged. A chip, a star fracture on the base or a hairline crack rules the glass out of a sellable candle, full stop, whatever it does to your unit count. This is not a cosmetic judgement: a jar that holds a burning wick for forty hours is a pressure and heat vessel. The cracked-glass posts in this series treat it as a safety question and so should you.

The repoured batch, until one jar has burned. A corrected batch is an untested batch. Pour eighteen, set one aside as your test jar, cure it to at least fourteen days and burn it for three hours in a closed room: time to full melt pool, flame height, any soot or mushrooming, and throw scored at 30 and 120 minutes against your reference notes. The hot throw testing guide has a fuller sheet. If the pooled load came out different from your design — because you topped up with wax, or because recovery was poor — then the wick is facing a different fuel and the wick question is open again.

Frequently asked questions