My Wooden Wick Shipment Contains Pieces With Different Grain Patterns and Slight Thickness Variation — How Should I Sample Them for Testing?

Why grain and thickness make wooden wicks a sampling problem

Cotton braid is made on a machine that repeats the same construction thousands of times. A wooden wick is cut from a sheet of real timber veneer, and no two trees, or even two parts of one log, lay down fibre identically. Tight, straight grain tends to draw molten wax steadily up the face of the wick. Open, wavy or knotty grain can wick unevenly, crackle louder, or let the flame wander to one side. Thickness matters as much: a sliver more wood per piece can mean a larger flame and a hotter melt pool, while a thinner piece may struggle to stay lit once the pool deepens.

None of that means your shipment is faulty. Natural variation is built into the material. The goods-in question is narrower and more useful: does the range inside this delivery sit within the range you have already proven in your candles? If the thickest and thinnest pieces in the box both burn acceptably in your jar, wax and fragrance, the middle almost certainly will too. If you only test one average-looking wick, you learn nothing about the ends of the spread, and the ends are where customer complaints come from.

So the job at goods-in is not to test many wicks. It is to test the right wicks: deliberately chosen extremes, drawn from across the shipment, compared against retained pieces from the lot you originally approved.

Sort before you sample: build the spread first

Random sampling is the right tool when you cannot see differences between units. With wooden wicks you can see and measure them, so a smarter approach is to sort first and then pick from the sorted groups. Statisticians call this stratified sampling; in a candle workshop it simply means laying the wicks out in piles before choosing which ones to burn.

A test plan that represents the whole shipment

Once the pieces are sorted and measured, the test plan writes itself. Each candidate is burned in your production jar with your production wax and fragrance, at your normal load, poured and cured exactly as you would for customer candles. Change nothing except the wick. If the new lot's thick pieces give a flame that climbs noticeably higher than the retained control, that is a signal. If the thin pieces tunnel where the control reaches the edge, that is also a signal.

Five candles is a suggested starting point, not a rule. If the sorting step showed a very narrow spread, you might drop to three. If it showed a wide spread or two clearly different veneer sources mixed in one box, you might test two pieces from each end. Decide the number from what the sorting told you, and write the reasoning down.

Worked example: cost of a representative wooden-wick test

Say you buy a pack of 100 Performance Wooden Wicks (Imported) from CSI at ₹4,130.00. That works out at ₹4,130.00 ÷ 100 = ₹41.30 per wick. You pour into the Clear Glass Jar with Golden Lid (150 gram) at ₹944.00 for ten, which is ₹944.00 ÷ 10 = ₹94.40 per jar. Your wax is Premium Coconut Soy Wax CSI 468, bought as a 5 kg pack at ₹3,185.00, so ₹3,185.00 ÷ 5 = ₹637.00 per kg. Your fragrance is Dark Vanilla Oud at ₹5,691.00 per kg.

A 150 g candle at 8% load, calculated on total candle weight, needs 150 × 0.08 = 12 g of oil and 150 − 12 = 138 g of wax. The wax costs 0.138 × ₹637.00 = ₹87.91. The oil costs 0.012 × ₹5,691.00 = ₹68.29. Add the jar at ₹94.40 and the wick at ₹41.30, and one test candle comes to ₹87.91 + ₹68.29 + ₹94.40 + ₹41.30 = ₹291.90.

Four new-lot candles plus one retained control is 5 × ₹291.90 = ₹1,459.50. The four wicks sacrificed from the new lot cost 4 × ₹41.30 = ₹165.20, which is four per cent of the pack price. Set that against the value of the other 96 wicks going into customer candles: 96 × ₹291.90 = ₹28,022.40 of finished material. A test costing about one-twentieth of what it protects is not an overhead. It is the cheapest insurance in your workshop.

One caution on Dark Vanilla Oud: its product page states a candle load range of 6–10% but does not publish a vanillin percentage, so if your jars are clear and your wax is light, watch for colour change during the cure as part of the same test. It will not change the wick decision, but it belongs on the same record sheet.

Reading the results and making the goods-in call

Burn every candle in the set in the same room, in the same session, away from draughts, and never leave any of them unattended. Record flame height by eye against a marked card behind the jars, melt pool reach at each hour, any soot on the glass, and how the wick head looks after extinguishing. Stop any candle that smokes heavily or flares, and treat that as a result, not an inconvenience. Continue at least until the pool should have reached the edge on your approved control; for a first delivery from a new wood source, take one candle from each end of the spread all the way down.

A conditional release is often the right answer for wooden wicks, because the variation is visible and can be screened. If thick pieces cause the trouble and your caliper can find them in seconds, sorting them out at the tabbing step is cheaper than rejecting the whole box. Write the screening rule on the quarantine label so whoever tabs the wicks next week applies it without guessing.

If the median piece behaves differently from your retained control, the variation is no longer at the fringes. That is a question for the supplier. For CSI wicks, send the photos, your caliper readings and the burn notes on WhatsApp to +91 7397976926 and ask what may have changed; CSI does not publish wooden wick widths or thicknesses, so record whatever answer you receive alongside your own measurements. Honest advice applies whoever supplied the box, including us: a lot that fails against your retained control stays in quarantine until you understand why.

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