My Candle Jar Gets Much Hotter Near the End of Its Life — Can the Wick Become Effectively Too Aggressive as the Candle Burns Deeper?

Why the wick gets "bigger" as the candle gets lower

On the first burns the flames sit near the rim with open air above and around them. By the final third they sit several centimetres down a glass tube whose walls have absorbed heat through every session above. Commonly reported consequences: the pool deepens, the flame lengthens, the exposed wick grows faster than it is trimmed, carbon builds, and the glass near the flames gets much hotter than it did at the top. In a thick-walled jar with a broad base, that stored heat can carry from one session into the next if the rests are short.

For an 8 cm jar, the mechanism and a rest-interval test that separates stored heat from wick size are in the lower-third aggression post; the safety-first acceptance criteria for a hot final third (flame touching glass, a wick drifting to the wall, sustained soot) are the hot-jar-near-the-bottom post. This page assumes you have seen the symptom and need to decide what to buy and how many jars to stake on the answer.

The jar's page lists the rim and the base both at about 10 cm, while also describing a "curved silhouette". If your jar measures narrower near the base, the late-heat effect is stronger, and whether to carry a reserve of the down-rung for a narrowing jar is the narrowing-jar post. Measure the inside at the rim, the middle and near the base before you trust any geometry on this page.

Down-rung set or configuration change

A late-heat problem has two plausible fixes, and they cost different things. Dropping one rung reduces the heat every flame makes, all the way down, and risks a slower start at the top. Setting the same wicks closer together moves each flame further from the glass in the final third, costs no wicks, and risks concentrating heat between the flames. Neither is known to work in your jar until three candles of each reach the bottom.

The spacing figures are geometry and starting points for testing (our judgement), not burn predictions: wick-to-wall = r − s/2 on an inside radius of 4.65 cm. Closer wicks can merge their heat in the centre, so the configuration arm has its own failure to watch for; whether spacing alone can fix a hot centre is the spacing-only post, and the fixture count for setting pairs in production is the two-wicks-leaning post. A single centred wick (4.65 cm from the glass) is a third possibility for this jar, but in a 9.3 cm pool it is a bigger change with its own early-tunnelling risk; pricing that retest against reducing both wicks is the double-wick retest post.

Why the down-rung is the only wick purchase

Two C1 per candle, three candles: 6 C1, plus two for one re-pour. The C1 10-pack is out of stock on 1 October 2026, so the 20 (₹118.00) is the smallest that covers it, leaving 12. The control and configuration arms use 12 C2 between them plus four for two re-pours; if your stock is short, the C2 20-pack (₹188.80) covers it with 4 over. Per candle, 2 × C1 saves ₹7.08 in wicks against 2 × C2: true, small, and not the reason to choose it. The reason is the final third.

The jar order and its breakage allowance

For most wick tests the jar is a constant you buy once. Here it is part of what is being tested, and a test that runs every candle to the end in the hottest stage can crack one, or end one early because a flame drifted to the glass and you stopped it. If that happens in an arm of three, you have two results and a decision on two candles. The allowance exists so that every arm finishes with three.

From nothing, the full test is ₹7,605.20; with a home-made two-hole card jig instead of bought bars, ₹7,033.20. The nine candles consume ₹5,307.88 of materials, jars included: ₹592.12 for a candle on 2 × C2 and ₹585.04 on 2 × C1. Costs exclude labour, freight, packaging and GST.

The allowance is our judgement: two jars on nine is 22%, which is generous for a sound jar and modest if your glass is suspect. It is not a prediction that two jars will crack. If none do, the two allowance jars go into production. If one does, it is evidence: note where the crack started and what the flame was doing, because a crack that begins near a drifted flame is a different finding from one at a chip in the rim. How to read a crack is the cracked-jar post, and why the final third is the stage to test glass in is the final-third glass post.

Judging heat without touching the glass

Readers often ask for a glass temperature that marks the line. We do not give one: no CSI jar page in the text we captured states a burn-temperature rating (the Apothecary jar's "Safe for hot-pour up to 85°C" is about pouring wax into it), and an invented figure would be worse than none. What you can do is judge the final third on things you can see, and keep the same observations for every arm.

The Pen Thermometer (₹354.00) is the tool CSI lists, and it is for wax: its page describes it for "the temperature of your candle wax". It is right for checking the pour at 80–85°C in this jar; it is the wrong tool for a burning jar, and nobody should press a probe or a hand against hot glass to find out. A relative reading between arms from a non-contact thermometer is useful if you own one; it still does not turn into a safe/unsafe line we can print. If the jar, not the wick, is your real question, the crack and final-third glass posts linked above are the place to start; this page stays with the wick purchase.

From a calm final third to production

Approval: all three candles of an arm reach the glass early, and stay calm through the final third to about 1 cm on every observation above. Then:

Reorder the approved wicks on daily use × your own lead time + safety stock (the formula in the cluster's pilot post); CSI dispatches from Pune and publishes no lead times or stock depth. Keep a jar allowance in mind for production too, for transit and handling (your own figure); that is a different allowance from the test's. When to time the production order against the burn calendar is the final-third order-timeline post, and which symptoms justify any wick purchase is the cluster hub.

Late-heat test items: what to buy and what to skip

Five items, in the order this test needs them, chosen on page statements, price and stock on 1 October 2026. CSI publishes no wick-performance or glass-heat data, so nothing here is ranked on burn results. Oil candle counts use 40 g per 500 g candle at 8%.

Skip for this decision: Cotton Braided Wicks (₹120.00 for 10) as a pair, a step up when the problem is too much heat; the Steel Wick Holder (₹250.00 for 5), whose page says it fits jars "up to 9 cm wide", narrower than this jar's 10 cm rim; and Performance Wooden Wicks (₹590.00 for 10), a different construction and a different test.

Frequently asked questions