Can Repeated Heating and Cooling During Ecommerce Delivery Make Glass Adhesion Progressively Worse?

Does each cycle make it worse? The short answer and the first test

A wet spot is commonly described as a thin air gap where wax has separated from the glass, which looks dark and wet through clear glass without any liquid being present. Ecommerce delivery in India commonly means several days in cartons that warm by day and cool by night, and makers commonly report that a separated patch, once started, can spread and does not reattach without remelting. So it is reasonable to suspect that every warm/cool swing makes adhesion a little worse. But it is not the only possible shape. A candle might lose contact at its weakest points on the first swing and then stop changing. Or it might hold for several cycles and then give way.

Those three shapes need different fixes. If damage grows with every cycle, the number of days in transit is a direct cost and shorter or better-buffered routes help. If one swing does all the damage, transit length hardly matters — the pour has weak spots and the fix is on pour day. If there is a threshold, you need to know your route's cycle count against it. We publish no cycle-test results for any CSI wax or jar, and there is no published curve to look up. The cheapest way to find yours is a cycle-count test: the same numbered jars, scored and traced after cycle 1, 2, 3, 5 and 8.

Start simply. Pour ten jars from one batch. Keep three on your shelf as the control, cycle three continuously, cycle three for three cycles and then rest them, and keep one as an untouched reference. After each cycle, trace every patch on a clear acetate sleeve with a fine marker. The traces stacked on top of each other show you, without any arithmetic, whether the patches grew.

Why repeated swings can keep doing damage

In general candle-making practice, wax is described as contracting as it cools and crystallises, and wax and glass are commonly described as expanding and contracting by different amounts with temperature. Every warm phase and every cool phase therefore puts a small shear on the line where wax meets glass. At the edge of an existing gap, the wax is no longer supported by the glass on one side, so it is plausible that the next swing peels it a little further — the way a sticker that has started to lift lifts more easily. That is the mechanism behind "progressively worse", and it is commonly described rather than measured: we do not publish contraction figures, cycle limits or failure rates for any wax or jar.

There are also reasons a patch might stop growing. Once the weakest contact has let go, the rest of the wall may be held firmly enough that the same swing does nothing more. And soy wax is commonly described as settling for days after pouring, so a candle cycled at three days old may behave differently from one cycled at three weeks. That is why the test records the age of the candle when cycling begins, and why the post on patches that appear after a week is worth reading alongside this one.

The example wax is Natural Soy Pearl Wax. Its page publishes no melt point, pour temperature, fragrance ceiling, cure time or adhesion statement, so pour it to the wax maker's instructions (ask us on WhatsApp if you need them), stay at or below the Clear 150 g jar's stated hot-pour limit of 85°C, and use 8% only as a labelled test load. That makes it a useful example for a different reason: with no page claim to lean on, the test is the only evidence you have — which is true of every wax in this series, whatever its page says.

The cycling ladder

The full series ladder runs from "is it really a wet spot?" through glass cleanliness, glass and pour temperature, cooling, storage and courier cycling, fragrance and dye, jar glass and the wax. This post sits on the cycling rung, but the rungs above it decide whether cycling has anything to work on. Climb in this order.

Rung 3 turns the curve into a business number. A parcel that spends five days in transit is not necessarily five cycles — a van parked overnight in a cool hub may swing twice in a day, and a parcel in an air-conditioned warehouse may barely swing at all. Log one real parcel. The courier-delivery hub shows how to send a logger on the route; the simple cycle-test post shows how to copy the log onto a bench.

The cycle-count test

The worked example is the series default candle: 150 g in CSI's Clear Glass Jar with Golden Lid (150 gram) at 8%, which is 12 g of oil and 138 g of wax per jar. The jar's page states an approximately 150 g fill and that it is safe for hot pour up to 85°C. The wax is Natural Soy Pearl and the oil Blue Ocean. The cycle settings below are labelled examples only. Replace them with your logged parcel readings as soon as you have them; until then, treat every result as provisional.

Three jars per arm screens for a large, consistent effect. Because the same jars are scored repeatedly, you get more information per jar than in a one-off test — each sleeve is a whole curve. If two of three arm B jars grow and one does not, add three more cycled jars rather than more checkpoints; in this example each extra jar costs about ₹274.24 in CSI materials. Testing across temperatures before pan-India approval shows how the cycle arm fits into a launch decision.

Turning the curve into a route decision

Count your routes in cycles, not days. If your logged parcel shows one warm afternoon and one cool night per day, a three-day delivery is three cycles; if it sits in an air-conditioned hub for two of those days, it may be one. Then read your arm B curve at that count. Shape 1 means every extra cycle costs you; shape 2 means the first cycle does the damage and route length barely matters; shape 3 means there is a count your routes must stay under. The table below turns that into money with labelled example figures.

In the example the far route adds ₹59.11 to every acceptable candle against ₹9.85 for the local route. Those spotted shares are placeholders: read yours off your own arm B curve at each route's cycle count. If your curve is shape 2, the three rows would be nearly the same, and the money is better spent on the pour. The discounted-stock post covers counting complaints and discounts in the same sum.

The scoring sheet

Score every jar on one house scale. The example grades the visible wall 0–4; the sleeve count gives a second, finer number. Both are examples of a house standard — there is no industry figure for an acceptable wet spot — so set your pass line before the first cycle and do not move it.

An example pass line for a pan-India SKU might read: "no arm B jar above grade 1 at the cycle count of our longest logged route". That is a house choice. How much variation is normal helps you set it; what share of spotted jars is acceptable in a pilot helps you defend it when the test grows.

The priced test kit

With your own stock, this test costs only time. From scratch, here is the ten-jar kit with listed, in-stock CSI packs, using oil = W × L and a 5% pour-loss working allowance: 10 × 138 g wax = 1,380 g, plus 5% = 1,449.0 g; 10 × 12 g oil = 120 g, plus 5% = 126.0 g.

That is ₹274.24 of CSI materials per test jar before the IPA spray. Two oddities are worth knowing. The oil line is two 50 g and two 15 g bottles (₹680.44) because that combination is a rupee cheaper than a 100 g plus two 15 g for the 126.0 g needed. And Natural Soy Pearl starts at 1 kg, so you keep 551 g for a repeat round. Acetate sheets, markers and the thermometer are non-CSI items at your own cost. The Eco Thin C1 line is the example wick; CSI publishes no wick size for this jar.

The decision rule

Apply it after the 30-day checkpoint, from the top, and stop at the first true line.

If Natural Soy Pearl is the wax you keep, it is the one soy wax in our range with a real bulk step: ₹450.00 a kilo at 1 kg against ₹424.80 a kilo at 5, 10 and 50 kg — a 5.6% saving. For this candle that is ₹62.10 of wax at the 1 kg rate against ₹58.62 at the 5 kg rate. The 50 kg bag is listed in limited stock; confirm on WhatsApp before planning around it.

The CSI products in this test

Frequently asked questions