I Bought 25 Kg of Soy Wax and This Batch Looks More Yellow Than the Previous One — Should I Test It Before Making 500 Candles?

Yes, test it — but test the right thing first

Soy wax is a natural material. It is made from a crop, refined and hydrogenated, and no two production runs are chemically identical. A shade of difference between one delivery and the next is common, and on its own it does not mean the wax will burn, throw or set differently. What it does mean is that you have lost the right to assume. You are about to commit 25 kg of wax, the jars, the oil and a week of labour to 500 candles, and the only thing you know about this lot so far is that it looks different.

So the answer to the title is yes, test before the 500-candle run. The better answer is that the test does not need to be a full revalidation. At goods-in you are trying to answer one narrow question: is this yellow a cosmetic shade of a normal natural product, or is it a signal of oxidation, age or heat damage that will show up in finished candles? A retained reference, a fair light source and a small melt answer most of that in under half an hour.

Until you have that answer, the bag sits in quarantine. Label it, move it away from the production shelf, and carry on making candles from your previous lot if you still have some. Nothing about a colour change forces you to stop working. It only forces you to stop using the new bag.

Compare against a retained reference, in the same light

The single most common goods-in mistake with colour is comparing a fresh bag against your memory. Memory is a poor instrument. Warm tube lights, a sunny window and the grey light of a monsoon afternoon all shift how white a flake looks, and the bag you remember was probably seen under different light, in a different room, weeks ago.

If you kept a retained sample from the previous lot — a sealed bag of a few hundred grams, labelled with product, invoice number and date received — this is the moment it pays for itself. If you did not, start now: the check below still works against the best reference you have, which might be a finished unscented candle from the old lot or the last kilogram in the old bag.

Melt a small sample — cosmetic or signal

Take roughly 50 g from each lot. Melt them side by side in identical clean containers over the same gentle heat, and keep the temperature moderate — overheating a sample is one of the easiest ways to create the very yellowing you are trying to judge. Check the melt with a thermometer rather than trusting a dial.

Now compare three things. First, the liquid colour: melted soy is usually clear to pale straw, and a new lot that is noticeably darker in the liquid state deserves more attention than one that only looked warmer as flakes. Second, the odour of the warm melt: a normal soy smell is faint and slightly fatty; a sharp, rancid or burnt note alongside the yellow suggests oxidation or heat history rather than natural variation. Third, clarity: haze, specks or sediment in the liquid are separate findings and belong on the record even if the colour is fine.

Then pour each into an identical jar or test cup, unscented and undyed, and let them set in the same place overnight. Compare the set tops and the view through the glass the next day in your chosen light. This is the version that matters, because it is what the finished candle will look like before fragrance is added.

Separate the wax colour from the fragrance colour

A lot of "yellow wax" complaints are really fragrance complaints. Several oils contain vanillin, which warms or yellows light wax over time on its own. CSI's Dark Vanilla product page says plainly that it will yellow white or light-coloured wax and that this does not affect performance; the Chocolate Fudge page says the wax will warm in tone over time. If you make vanilla, coffee or gourmand candles, their finished colour already drifts, and a slightly warmer wax lot may be invisible in them.

That is why the goods-in melt is done unscented. You are judging the wax alone. Only once the unscented comparison passes do you pour a scented pair — old lot and new lot with the same oil at the same load — to see whether the combined colour still meets your standard. If your 500 candles are a pale floral or a white clean-linen style, this second pair is worth doing, because a low-vanillin line is where wax colour shows most.

If the unscented pair matches and the scented pair does not, the wax is not the problem. You have learnt something about how the oil and wax interact, which is useful, but it is not a reason to reject the bag.

The worked example: what a check costs against 500 candles

Take your own figures. 25 kg of wax for 500 candles works out at 25,000 g ÷ 500 = 50 g of wax per candle, so these are small candles — tins, votives or minis. If the wax is CSI 464, the 25 kg was bought as two 10 kg bags and one 5 kg bag: 2 × ₹5,074.00 + ₹2,537.00 = ₹12,685.00.

The colour check uses about 50 g from each lot for the melt, and for the unscented pour you can reuse that same melt. Call it three small pours from the new lot as a suggested starting point: 150 g of wax. At the 10 kg rate that is ₹5,074.00 ÷ 10 = ₹507.40 per kg, so 0.15 × ₹507.40 = ₹76.11 of wax. Add a few test cups and half an hour of your time.

Against that sits the whole run: ₹12,685.00 of wax before you have added jars, fragrance, wicks, labels and the hours to pour and finish 500 candles. If a yellow lot turns out to be heat-damaged and you only discover it when customers see cream-coloured candles in a white range, the loss is the full run, not ₹76.11. The maths is not close. What would be a waste is a full wick and burn revalidation for a colour question that a melt and a pour can answer.

Honesty cuts both ways here. If the new lot fails the comparison, it stays in quarantine and goes back to the supplier as a question — whoever supplied it, including us. Send the side-by-side photos, the invoice number and the date received on WhatsApp (+91 7397976926 for CSI) and record the answer you get. CSI does not publish colour specifications or batch numbers for its wax, so your photos and your retained reference are the evidence.

Decide, record, then release

End the check with a written decision, not a feeling. A single line in a goods-in notebook is enough: date received, product, invoice number, bag weight, the light you used, the colour grade, melt observations, the decision and who made it. Staple or file the photo with it. Keep a sealed 200 to 300 g portion of this lot as the reference for the next delivery — a suggested starting quantity that you can adjust once you know how many comparisons you actually run.

There are four honest outcomes. Accept when the set, unscented wax matches. Conditional release when it is marginally warmer but otherwise normal — use it on coloured, dark or vanilla lines first and keep it away from your whitest products. Quarantine while you wait for a supplier answer on anything with odour, haze or dark liquid colour. Reject when it is clearly degraded. Whichever you choose, do not quietly blend the yellow bag into the old one to average the colour out. That hides the lot, makes the next problem harder to trace, and teaches you nothing.

Where this page fits

CSI already has guides that cover parts of this question. This page covers the goods-in check for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions