My Candle Jar Cracked Near the End of the Burn — Should I Investigate Wick Size, Glass Quality and End-of-Life Heat Together?

The crack, made precise

Your situation, stated exactly: a container candle burned normally for most of its life, and near the end the glass cracked; perhaps a hairline you noticed after it cooled, perhaps a crack during the burn, perhaps a piece came away. You want to know whether to blame the wick, the jar or the way it was burned, and you suspect, correctly, that it may be all three.

The worked example in this post, labelled: 150 g at 8% = 12 g fragrance + 138 g wax (load as a percentage of total weight) in CSI's Clear Glass Jar with Golden Lid (150 gram), the series default for 7–7.5 cm jars, about 25 g per cm of height. The jar cracked with about 12 g left, 0.48 cm of wax: below a 1 cm floor (25 g). The last session was long, and the wick's tab in the spent jar sits visibly towards one side. Each of those details is a suspect; none is a verdict.

Stop, keep, record

Do these before anything else, in this order:

If it was a customer's candle, the questions to ask them and the batch records to pull are the complaint post's subject; this post takes over once you have the jar and the facts.

Reading the crack

Glass failure analysis is a specialist field, and we won't pretend a photo settles it. But some patterns are commonly associated with particular causes, and they tell you which arm of the test to watch most closely:

The off-centre case is worth a number, because it's easy to underestimate. In the worked example's jar (7 cm outer, 6.3 cm inside with 0.35 cm walls assumed) a centred wick is 3.15 cm from every wall. Five millimetres off centre, the near wall is 2.65 cm away (16% closer); a centimetre off centre, 2.15 cm (32% closer). That's geometry, not a burn prediction, but it shows why a tab that wandered during setting can matter more at the end, when the pool is fully liquid and the flame is low. For coated or frosted jars, the published sibling testing whether a frosted coating tolerates repeated heating sets out the extra checks.

Causes ranked: heat, glass, shock

The ranking is our judgement for a crack near the end of a container candle. Signs are common observations; CSI publishes no breakage, heat or glass-strength data for any jar.

The interaction is why you test together. A long last session (1) on a centred wick in a sound jar may only make the glass hot; the same session with a tab drifted to one side (3) in a jar with a small chip (4) can crack it. Change only the wick and you may never see the chip; blame only the glass and the next lot cracks too. Why heat builds low in a jar at all is why a correct wick becomes too aggressive lower down.

What CSI's jar pages do and don't tell you

Of the 72 container listings in CSI's catalogue data, seven say "Safe for hot pour up to 85°C" (or "hot-pour"): the 150 g lidded jar family and the Apothecary Glass Candle Jar with Dome Lid. That is a pouring figure. It tells you not to pour wax hotter than 85°C into the jar (with Luxury Soy Wax CSI 464, whose page states 80°C–90°C, pour at 80–85°C in these jars), and nothing about how the glass behaves at the end of a burn. Other listings use general wording ("withstand high heat", "heat-resistant") or nothing. No listing states an inside diameter, a wall thickness or a glass type you could test against, and the listing titled "Borosilicate Glass Jar 600 gram" has a description that only says clear glass. The 150 g jar's page calls it thick-walled; heavier isn't automatically safer, because thick and uneven glass can hold bigger temperature differences.

Argued against our own interest: we can't tell you from our listings that any of our jars won't crack at the end of a burn, and we won't. If a CSI jar cracked, send us the photo and the order date on WhatsApp; we'll check what we can about that dispatch, and we'd rather you test another jar than keep selling one that fails. What to ask any supplier about glass for burning, and why thick glass isn't automatically suitable, is in the published sibling whether candle glass needs specific heat resistance.

The joint test: nine candles, three arms

Test candle, labelled: 150 g at 8% = 12 g fragrance + 138 g wax, one wax and oil lot, every wick held centred while setting. Arm A: current wick, jars from the cracked jar's lot. Arm B: one CSI rung smaller (C2 → C1, or Cotton Braided → C2), same jar lot. Arm C: current wick, jars from a different lot (or the jar you're considering instead). 9 candles: 108 g of oil and 1,242 g of wax.

We used Luxury Soy Wax Chunks here (its page states "up to 13% fragrance load capacity"; 8% is inside it); test in whatever wax the cracked candle used, since a wax change is another variable. About 9 days with daily sessions (worked example). If your wick is C1 already, arm B has no smaller CSI tabbed wick; replace it with your current wick and a stop designed higher, as in designing the candle to stop before the flame nears the base.

When to restart, and how to stop it happening again

Whatever the result, three habits prevent repeats: inspect every jar on arrival (chips, bubbles, rocking bases) and measure 5 per lot; centre and hold every wick and look at the tab in your own spent test jars; print the end-of-life rules you tested, about 1 cm floor, a maximum session, a heat-resistant surface, away from draughts, and never water on a hot jar. A jar that's fine on burn 1 can still fail at the bottom, which is why a new container is approved on full-life tests (the late-burn heat hub sets out the options when the bottom runs hot).

The CSI buy list and the step to production

Argued against our own interest: a crack is the one problem where we'd rather you test a different jar than buy more of ours on trust, and where the answer may be a label and a habit rather than anything from our store. CSI's wick range is also short: no sized, tabbed wick between C1 and C2, nothing tabbed smaller than C1, one wooden wick width, no stated neck heights.

The step to production quantity

Only after the restart rule is met. Worked example, labelled: 160 candles a month plus 2% is 163; if C1 replaced C2, 100 + 50 + 20 = 170 C1 for ₹1,003.00 (if C2 stays, ₹1,604.80). Jars: 8 × 20 + 10 = 170 of the 150 g jar for ₹16,048.00 (₹94.40 a jar, flat across packs), inspected on arrival, 5 per lot measured. Holders: four packs (₹472.00) hold a 20-candle pour and are reused. Reorder point = daily use × your lead time + safety stock; CSI publishes no lead times, so ask on WhatsApp.

What to buy, and what to skip

Four items for the joint test and a possible jar change, chosen on page statements, price and stock on 7 October 2026. CSI publishes no breakage, heat or glass-strength data, so nothing is ranked on performance.

Skip for now: a bigger or "thicker" jar bought on looks alone; wooden wicks as a crack fix (one width, a different system to test from the top); the Pen Thermometer (₹354.00) as a glass gauge (it's for wax; CSI lists no non-contact thermometer); and any 500-pack of wicks or 20-packs of jars before the restart rule is met.

Frequently asked questions