My New Glass-Jar Shipment Has Slightly Different Internal Diameters — How Much Variation Is Enough to Affect My Previously Approved Wick?

What a few millimetres of internal diameter actually changes

A wick is not approved in the abstract. It was approved for a particular wax, a particular fragrance at a particular load, and a particular circle of wax surface that it has to melt from wall to wall within a reasonable burn. The internal diameter of the jar sets that circle. When a new jar shipment arrives and the inside measures a little wider or a little narrower than the jars you tested, the wick is being asked to do a different job, even though the label on the carton reads exactly as before.

The surface the flame must melt grows with the square of the diameter, not in a straight line. That is why small differences matter more than they look. A jar that is 2 mm wider across the inside does not ask for 2 mm more melting; it asks for a noticeably larger area of wax, and the extra area sits right at the wall, which is exactly where a marginal wick leaves a ring of unmelted wax.

The reverse also applies. A narrower inside gives the same wick less wax to melt, so a wick that was already on the hot side of acceptable can start to throw a taller flame, heat the glass more at the rim, and soot. Neither direction is automatically a problem. The job at goods-in is to find out how far the new shipment has actually moved and whether that movement crosses the edge of what your approved wick can handle.

Measure the shipment before you argue about the wick

Before anyone changes a wick, get numbers. Guesswork from looking into a jar is unreliable because glass thickness, colour and a rolled rim all distort the eye. A vernier caliper with inside jaws is the right tool, and it is cheap enough to be part of every goods-in bench. CSI does not sell calipers, so buy a generic one from any tool supplier.

Twenty jars and forty readings take a practised person well under an hour. Compared with pouring hundreds of candles into containers you have not checked, that hour is the cheapest insurance on the bench.

How much variation is enough to matter

There is no universal number, and nobody can honestly give you one without knowing your wick's margin. A wick that left a clean full pool with room to spare in the approved jar can absorb more variation than one that only just reached the wall. What you can do is set bands for your own product line and treat them as a suggested starting point to validate against real burns.

The ±1 mm and 3 mm figures above are not an industry standard and are not CSI policy. They are a sensible place to begin for jars around the 6–7 cm range. After a few shipments, your own burn records will tell you whether your wick tolerates more or needs tighter limits, and you should adjust the bands to match what the burns show.

Worked example: a 200 ml jar shipment with a wider spread

Suppose your approved reference jar measured 63.0 mm across the inside. These are your own illustrative readings, not a CSI figure. From 20 jars in the new shipment, the narrowest measured 61.8 mm and the widest 65.2 mm, averaging close to the reference.

Work out the melt surface change using the diameter squared. For the widest jar, 65.2 ÷ 63.0 = 1.035, and 1.035 × 1.035 = 1.071, so the wick has about 7.1% more wax surface to melt. For the narrowest, 61.8 ÷ 63.0 = 0.981, and 0.981 × 0.981 = 0.962, so it has about 3.8% less. The widest jar is 2.2 mm over the reference, which lands in the conditional band.

The test itself is inexpensive. Four candles (two in the narrowest jars, two in the widest) at 200 g each with 8% fragrance of total candle weight need 184 g wax and 16 g oil apiece, so 736 g wax and 64 g oil in total.

Against a production run in the hundreds, spending ₹731.19 and a few evenings of supervised burning to learn whether the widest jars still reach full melt pool is an easy decision. If they pass, release the shipment and note the measured range. If the widest jars tunnel, you have a choice: sort the shipment by measurement and hold the wide jars for a different wick, or return them as not matching your approved sample.

Running the extreme-jar burn and making the call

Burn the four test candles side by side with your retained reference candle from the approved jar if you still have one, in the same room, on the same day, after the same cure. Your cure period should at least match the stated minimum on the fragrance product page; for soy with most CSI oils that is 48–72 hours. Change nothing except the jar.

Watch for three things at each burn: how long the pool takes to reach the wall, whether the flame is steady or tall and flickering, and whether the rim glass becomes noticeably hotter than on the reference. Burn the candles down to the end, never leave a lit test unattended, and extinguish any candle that smokes heavily or flares, recording it as a failure.

If both extremes pass, the shipment is accepted and the measured range becomes part of your record for the next order. If one extreme fails, quarantine the affected jars rather than the whole shipment where you can sort them by measurement. And be honest with yourself about the result: a shipment that fails your comparison goes back in quarantine, whoever supplied it, including us. Send the readings and photographs to the supplier so the next order can be specified more closely.

Once a pattern emerges, say three shipments that all sit within 1 mm, you can reasonably reduce the dimensional sample. Until then, measure every shipment, because the one you skip is the one that moves.

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