My Candle Has a Very Tall Flame but the Melt Pool Still Does Not Reach the Glass — Why Might Simply Wicking Down Create Another Problem?

Height and width: two different jobs

Your symptom, stated exactly: the flame is very tall (well above the 2.5–5 cm range many makers use as a practical rule of thumb), yet the melt pool still stops short of the glass, so there is a ring of solid wax around the edge after a burn. That pairing is the whole diagnosis, because the two things measure different jobs.

Flame height mostly reflects how much fuel the wick lifts and how much wick stands above the pool. A long wick, a carbon cap or a fast-drawing construction all make it taller. Pool width reflects how far heat travels sideways from the flame into the wax, over time. That depends on where the wick sits, how the heat is distributed, how wide the jar is, and how long you burn. A tall, narrow flame sends much of its heat upwards. A wick a centimetre off-centre leaves the far wall short. And a single wick has only so much reach: in a wide jar, turning it up raises the flame far more than it widens the pool. These are common observations about combustion, not CSI test data.

So "too big" and "too small" can both be true of one candle, about different things. That is why the question in your title is the right one. The obvious fix answers the height and makes the width worse. The closest published post in our wick series, tall flame, pool short of the edge: wrong wick type?, makes the same point from the purchase side. This post works through the causes, the four free checks and the 2×2 test. (If your pool does reach the glass and the flame grows as it deepens, you want the pool-depth post instead.)

Five reasons for a tall flame over a short pool

Ranked by likelihood for exactly this pattern and by cost to test (our judgement). Signs are common observations, not CSI burn data.

Cause 4 needs a word of care. When people say a heavy oil "starves" a wick, they usually mean that a viscous, resin- or vanilla-rich oil clogs the wick or builds carbon on it. The flame then burns tall and dirty from a capped tip while drawing fuel unevenly. That is a common observation, not a measured property of any CSI oil, and some heavy oils behave perfectly well. The unscented control settles it for your oil: if the same wick reaches the glass unscented, the oil is in the way, and the next step is the oil or load, not the wick (lavender fine, vanilla huge has the per-fragrance logic). Mushrooming as the main event belongs to the mushrooming hub; a wick that bends to one side is the leaning-wick post.

What a smaller wick would actually do

Here is the honest prediction for each cause if you simply drop one CSI rung (Cotton Braided → C2, or C2 → C1), based on common observations rather than CSI test data.

Tunnelling is not cosmetic for a seller. The ring is wax the customer paid for and never burns, and it tends to deepen with every short burn as the pool "remembers" its narrow shape (a common observation). In a deep, narrow tunnel the flame can also struggle for air and drown in its own pool, or burn close to a high glass wall. If you have already wicked down once, got a tunnel, and wicked back up into a tall flame, you are in the loop that the wick-down-then-tunnel post breaks; don't start it.

Four free checks before any purchase

Write each result down before you move on. The sheet matters later: when the 2×2 picks a winner, you will want to know it was not an off-centre wick or a cap all along.

The 2×2 test: wick system × oil

The worked example is labelled: an 8.5 cm (inside) jar, currently on a single C2, with a 250 g fill at 8% = 20 g fragrance + 230 g wax (load as a percentage of total weight). Use your own jar and fill. The pool area is 56.75 cm²; a centred single wick sits 4.25 cm from the wall. A pair spaced 4.0 cm apart on the diameter (a starting point for testing, not a rule) sits 2.25 cm from the wall, each wick serving about 28.37 cm². That is geometry, not a burn prediction; one larger wick or two smaller is the full method for choosing between them, and this test is its short form for your symptom.

How to read it: compare across a row to see the oil effect, and down a column to see the system effect. If both systems reach the glass unscented but not in your oil, the oil is the lever, and a wick change alone won't fix it. If only the pair reaches the glass, in both columns, the jar wants two wicks. If Cotton Braided reaches the glass with a flame inside your band, and the pair runs hot or deep, the single thick wick wins. If everything passes, prefer the system with the calmer flame and the pool that reaches the glass by about hour three to four from burn two. A pair that reaches the glass but runs two large flames and a hot jar is a different question: reduce both wicks before removing one.

Worked example for the calendar: at about 8 g an hour (your scale decides), a 250 g candle is about 31.2 hours, roughly 8 four-hour sessions, with all 15 candles burning side by side on separate heat-safe spots. Keep pair candles further apart from their neighbours, since two flames heat the air around them more.

Already own holders and wax? Then the purchase is the wicks: ₹238.00. Wax is the big line here because the grid has 15 candles; it is still cheaper than pouring a production batch on a guess. If you want a scented comparison instead of unscented, Eucalyptus Spearmint 100 g + 50 g (₹566.40) covers six 250 g candles with 30 g spare.

CSI buy list, honest limits, and what to skip

CSI's sized, tabbed wicks are three rungs: C1 → C2 → Cotton Braided, plus pairs of C1 or C2 and one wooden-wick listing with no width options. There is no sized, tabbed wick between C1 and C2 or between C2 and Cotton Braided. In CSI terms, "a different construction" means switching between Eco (C1/C2), Cotton Braided, a braided roll cut to length (the Braided Wick Roll with 100 wick tabs, ₹706.82, which you "cut … to the desired length"), the loose Eco wick, or the wooden wick. We do not sell other named wick series. If your jar needs something we don't stock, we would rather tell you than sell you the nearest pack: ask on WhatsApp.

Argued against our own interest: most candles with this symptom fail check 1, 2 or 3, and the fix costs nothing. Buy the 2×2 wicks only after all four checks pass. And don't choose between Cotton Braided and a C1 pair on price: a pair is ₹11.80 a candle against ₹12.00, a difference of ₹0.20, which the test result will outweigh many times over.

After approval, buy by the month. Worked example, labelled: 100 candles a month plus 2%. On C1 pairs that is 204 wicks, so 2 × 100 + 20 = 220 for ₹1,298.00. On Cotton Braided it is 102, so 100 + 10 = 110 for ₹1,320.00. Same per wick in every pack. The reorder logic is in the hub, and two-wick spacing in production is worth a holder per candle. A later change of wick lot or supplier is a formula change (what to revalidate after moving to two wicks).

Five items, chosen on page statements, price and stock on 7 October 2026. CSI publishes no burn, flame or pool data, so nothing here is ranked on performance, and each card says when not to buy.

Skip for now: a smaller single wick (the whole point of this page); any 100-plus pack before approval; a new oil while the wick question is open (unless check 4 points at the oil); and Cotton Braided in a jar of 8 cm or less unless the checks and a C1 pair have already failed.

Frequently asked questions