My Jar Narrows Toward the Bottom and the Flame Becomes Too Aggressive — Should I Choose Wick Size Based on the Narrowest Part of the Container?

Why the flame turns aggressive where the jar narrows

A wick draws melted wax at a rate set mostly by its own size and the fuel around it; the jar decides how much wax there is to melt and how close the glass is. In a straight jar, the pool and the wick stay in proportion from top to bottom, and the changes low down come from heat building up as the level falls (the late-flame hub maps them). In a jar that narrows, a second change is layered on top: the pool gets smaller while the wick stays the same. A wick that was just right for the rim is, commonly, more than enough for the base. The pool can reach the glass sooner in each burn, deepen faster, and the flame can grow and lean as the glass wall comes closer.

That is why the question "which diameter should I wick for?" comes back after a failure like yours. Before production it is a design question, and the jar-narrows sizing post answers it: usually wick for the widest part the pool must reach in early burns, then check late burns. Burn testing through a taper in general is wider at the top than the bottom. This page starts later: you already chose, the bottom failed, and you need to decide whether the fix is a smaller wick.

Aggressive-flame patterns in a narrowing jar

"Too aggressive" covers a flame that is tall, flickers, leans towards the glass, smokes or leaves a hot rim. Not all of it comes from the narrowing. The table ranks the patterns we would suspect, most common first; the ranking is editorial judgement, and CSI publishes no failure statistics or burn data.

Only pattern 1 is the narrowest-first question. Pattern 2 is a whole-jar over-wick, and the candidate in the retest will probably pass easily, which is fine, but the lesson is different. Patterns 3 and 4 are what the two-hour versus longer-burn post and the trimmed-versus-untrimmed twin separate. Pattern 5 is centring (the base audit). Pattern 6 belongs to the formulation, and the fix there is a lower load inside the lower of the oil page's and wax page's ceilings, not a wick.

Set your tunnelling limit before you light anything

The smaller wick's risk is at the top, where the jar is widest. It may take longer to melt out to the glass, and a first burn that stops short can leave a ring of hard wax that later burns follow down, a memory ring. So the retest needs a written limit for how much tunnelling you will accept, set before burn 1. It is your commercial judgement, not a CSI figure. Ways to write it:

For scale, one very rough rule of thumb is about an hour of burning per 2.5 cm of jar width, about 2.8 hours for a 7 cm rim, and a first full pool within about 3–4 hours is commonly cited. Both are rules of thumb, not CSI data, and neither is a limit on its own; your limit should describe what your customer will see. Whether a slightly slower first pool is a better business risk than a hot final third is the slow-first-pool decision; the short version, which we share, is that a final-third safety failure outranks a first-burn tunnel.

The retest: current wick against one step smaller

The retest changes only the wick. Wax, oil, load, jar, fill line, pour temperature, cure age, room and trimming stay the same. A1 and A2 carry the wick that failed; if you still have two unburned twins from the failing batch, use them, as they are at least as old as anything you pour now, so cure cannot favour the new pair. B1 and B2 carry the one-step-smaller candidate, poured now. Two of each, because one candle that behaves oddly settles nothing.

Pouring the candidate pair

Worked example in CSI 464 and the 150 g jar: 150 g candles at 8% of total weight, 12 g of oil and 138 g of wax each (8% of total is 8.70% of wax weight). That sits inside Japanese Cherry Blossom's stated "6-13% for candles" and CSI 464's "up to 10%", even if that ceiling is read on wax weight (9.09% of total); of the two ceilings, the lower one, the wax's, governs. The pair takes 24 g of oil and 276 g of wax; all four take 48 g and 552 g. Pour at 80–85°C, since CSI 464 states 80–90°C and the jar page "Safe for hot pour up to 85°C", and the lower limit governs. Centre each wick and hold it upright while the wax sets. Japanese Cherry Blossom's page states no cure time in the text we captured; many makers start burn tests at day 7 (common practice), and A1/A2 will be older than that.

Costs are worked examples of what the candles consume and exclude labour, freight, packaging and GST. Starting from nothing, the cash outlay for the candidate pair is ₹1,780.40 (a whole bottle, bag and pack each). If your failing candle used a different oil or wax, use yours: the retest must match the candle that failed.

Burning and logging

The decision rule, outcome by outcome

The first row is the one this page exists for, and the diagram's worked example lands there: the candidate leaves 3 mm at burn 1 and nothing by burn 3, while in the final third its flame reaches 27 mm, 9 mm above its 18 mm at burn 1, against 48 mm for the current wick, 26 mm above its 22 mm at burn 1, and touching the glass. A flame touching glass fails a wick however good the rest of the log looks.

The second row is the honest ending for jars that narrow a lot. A wick that is right for the base cannot always melt the rim, and a wick that melts the rim can run hot at the base. Then there is no size to choose, and the options move sideways: a different wick type, a double wick, a lower fill line that keeps the pool out of the narrowest section, or a straighter jar. Each has its costs and risks, laid out in the no-perfect-wick options table. The jar change is the one sizing post 2724 anticipates.

If the smaller wick wins, the saving follows you into production: C1 costs ₹3.54 less per candle than C2, ₹3,540.00 over 1,000 candles at the flat per-wick price. That is a side benefit, not a reason. A wick is chosen on the burn, and a cheaper wick that fails the top is the most expensive mistake on this page.

Wicks, oil and jar for the retest

Four items: the two wick rungs, the wick that is often the "current" in an aggressive narrowing jar, the worked-example oil and the worked-example jar. Chosen on page statements, price per unit and stock on 26 September 2026; CSI has no measured burn data, so nothing is ranked on performance. The oil table shows every size, 150 g candles at 12 g, and 24 g candidate pairs.

Frequently asked questions