I Changed Only My Soy Wax Supplier but Kept the Fragrance, Jar and Wick Identical — Do I Need to Repeat the Complete Burn Test?

Yes, the burn test — but not everything, and not from zero

Makers ask this question hoping for a no, and it is a fair hope. The fragrance is the same bottle, the jar is the same carton, the wick is the same pack. Surely most of the old testing still counts? Some of it does. But the wax is the only material in a container candle that is burned from the first minute to the last. The wick draws it up, the flame consumes it, and the melt pool carries the fragrance into the room. When the wax changes, every result that describes how the candle burns was measured on a fuel you no longer use.

So the honest answer is partly. You do not repeat the parts of your development that were about the jar or about the oil as a material: the jar's dimensions, the oil's safety documents, your label artwork. You do repeat — against a control candle, not from memory — everything the flame and the melt pool decide: wick performance, melt-pool diameter and depth, flame height, mushrooming, soot, jar heat at every stage, throw, and the full life of the candle. And you add a few checks that never existed before, because the new wax brings its own instructions, handling and lot record.

That is not a CSI rule. It is what a controlled change looks like in any small manufacturing process: sort the old results by whether the change could have touched them, keep the ones it could not, check the ones it might, repeat the ones it certainly did, and write down what is new. This page sets that out once, as the method the rest of this cluster uses. The siblings each take one symptom — a tunnel, a faster melt pool, a hotter jar, sweating, droplets, rough tops, a yellower candle — and go deeper.

The wax-change retest matrix: KEEP, CONFIRM, REPEAT, NEW

The matrix below is the backbone of every post in this cluster. It uses four words consistently: KEEP means the earlier result is still valid and you do nothing; CONFIRM means a short side-by-side check that you expect to pass; REPEAT means the full test again, against a control; NEW means a test or record you never needed before the change. The example is an approved 200 g candle in a White Matte Glass Jar, Lavender Fragrance Oil at 8%, an Eco Candle Wicks Thin C1 wick, moving from Luxury Soy Wax Chunks to Luxury Soy Wax CSI 464.

Two things are worth noticing. First, there is no row where the old burn data can be kept on the grounds that the new wax "looks the same" or "is also soy". Whether a wax that looks identical can keep your wick is its own post, because it is the most common shortcut. Second, the CONFIRM rows are genuinely short. Weighing one pour and reading one product page take ten minutes. The work is in the REPEAT rows, and the rest of this page is about doing them efficiently.

Why a wax change reaches the wick, the jar and the throw

A wick does not burn wax directly. The flame melts a pool, the molten wax climbs the wick by capillary action, vaporises near the flame and burns. How much fuel reaches the flame, and how fast the pool widens and deepens, depends on the wick's construction and on the wax: how it melts, how runny the melt pool is, and how the fragrance oil sits in it. These are wax properties that differ between products, and CSI does not publish melt points, viscosity or hardness for its waxes. That is exactly why the test is necessary — nobody can tell you from a data sheet how your wick will behave in a wax it has never burned in.

In general candle-making practice, a wax that melts more readily can widen the pool sooner, deepen it faster and let the wick draw more fuel, which commonly shows up as a taller flame, more mushrooming or a hotter jar later in the burn. A wax that holds its structure longer can leave a narrower pool and, in the worst case, a tunnel. Neither direction is a verdict on the wax. It is a mismatch between wax and wick that only the burn test reveals — and it can appear only halfway down the jar, where the flame sits closer to the glass. Why a correct wick can start tunnelling after a wax change and what a much faster full melt pool means take those two directions one at a time.

Throw follows the pool. Fragrance leaves the candle mostly from the hot liquid surface, so a different pool diameter, depth and temperature can change hot throw even at the same 8%. Cold throw depends on how the oil is held in the set wax and how the wax crystallises during cure — again, wax behaviour. Better throw with a hotter jar is the case where the change looks like an improvement and is not approved yet, and summer sweating at 8% is the case where the oil and the new wax do not stay together on the shelf.

The minimum test plan, candle by candle

The plan below is the smallest one that still answers the question properly. Every number in it is a labelled working practice you can change, not a standard. The core idea is that nothing is judged on its own: every new-wax candle burns beside a control made with the old approved wax, poured the same day, with the same oil lot, the same wick from the same pack and the same jar. If you have no old wax left, a retained candle from your last approved batch is a weaker but acceptable control — note its age, because a candle cured for three months will not throw like one cured for a week.

The scoring sheet: new wax against control, line by line

Fill in one sheet per burn cycle for every pair. The column that matters is the last: is the new wax no worse than the control on this line? Safety lines have no tolerance at all. Quality lines allow a tolerance that you set before you start, not after you see the result. CSI publishes no numeric flame-height or jar-temperature limit, and this page will not invent one. The acceptance standard is: no worse than the approved control on every safety line, and within your jar supplier's guidance and the recognised candle fire-safety standard you work to (ASTM F2417 is a commonly cited US standard; read it for its own requirements).

Put the jar reading into context before anyone gets excited about throw. If the new wax throws better and the glass reads hotter, the candle has not improved; it has become more aggressive. A hotter jar or a taller flame is never traded for more throw. That trade is the most tempting mistake in the whole cluster, and it gets a full post of its own.

The priced test kit from CSI

Here is Stage A priced with CSI's listed, in-stock packs. Grams come from the CSI load convention (oil = candle weight × load): each 200 g candle takes 16 g of oil and 184 g of wax. Four candles per wax need 736 g of each wax, and a 5% pour-loss allowance (a working assumption) makes that 772.8 g. The oil is shared: eight candles need 128 g, or 134.4 g with the same allowance.

Notice the control-wax line: 772.8 g of Chunks is cheaper as two 500 g bags (₹598.00) than as one 1 kg bag (₹599.00) — the listed price is ₹598.00 a kilo in 500 g bags. CSI 464 goes the other way: one 1 kg bag at ₹507.40 is cheaper than two 500 g bags at ₹520.00. Stage B, if you need it, adds 3 candles in the new wax with an example alternative wick (Eco Thick C2, 10-pack ₹94.40): the leftover 227.2 g of CSI 464 is not quite enough, so it needs another 1 × 500 g of wax, and the stage costs ₹978.40 in materials. Thick C2 is only an example of a different construction to bracket against — CSI does not publish wick-size charts, and no CSI wick is claimed as a size step from another.

What the retest costs against what the change saves

The test kit looks small next to a production order, and it should. Here is the per-candle material cost at the 5 kg wax rate and the 1 kg oil rate, with the ₹5.90 wick and ₹70.80 jar, excluding labour, packaging, freight and GST. On the old wax a candle costs ₹270.45 (₹110.22 of Chunks); on the new wax ₹253.60 (₹93.36 of CSI 464). The saving is ₹16.85 a candle — worth stating plainly because it is one of the real reasons makers switch.

Even with an example labour figure, the testing pays for itself inside roughly 248 candles. The more useful comparison is the other way round: a single 100-candle batch in the new wax carries ₹25,359.76 of materials. Skip the test and a late tunnel or a hot jar puts that whole amount at risk, plus customer returns. The ₹15-per-candle savings post builds this payback into a full decision, and the higher-price, lower-load comparison handles the case where the new wax costs more but might need less oil.

The decision rule: pass, adjust, or stop

Decide on the whole sheet, not on the best-looking line. The rule below uses three outcomes, and it is written so that two people applying it to the same sheet reach the same answer.

Two restrictions keep the rule honest. Change one variable at a time — if the new wax tunnels and you change both the wick and the load, you will not know which fixed it. And never move a failed safety line into the "adjust" row because the throw improved. Testing a "direct replacement" before 500 candles applies this rule to a supplier's claim, and the 1,000-candle emergency shows the true minimum when time is short.

Releasing the new wax and recording the change

A passing sheet is the start of the release, not the end. Make one pilot batch at your normal size and pull three candles from it — first, middle and last poured — to burn beside a control. Scale changes heat-up, stirring and cooling, so the pilot proves the process, not just the materials. Then update the recipe card and the batch sheet so the new wax, its lot, its temperatures and its fill weight are the only version anyone can use.

Keep a retained sample of every wax lot — 200 to 300 g in a sealed, labelled bag — and one finished candle from each approved batch. CSI does not publish lot numbers or lot-to-lot consistency data, so your own retained samples are the only way to compare a future bag against the one you approved. That is also what lets you answer a customer complaint months later with evidence rather than memory.

Which sibling answers your exact symptom

This page is the hub for the twenty wax-change questions. Every sibling uses the same KEEP / CONFIRM / REPEAT / NEW vocabulary and the same control-candle method, but each starts from a different symptom and goes deeper into the part of the matrix that symptom touches. The worked examples deliberately use different CSI waxes as the old and new wax — not because one CSI wax behaves differently from another as a fact (we publish no data that says so), but because any two waxes might, and the test is what shows it.

Outside this cluster, the same matrix appears with different materials: changing only the fragrance supplier keeps your wax data and repeats the oil-driven lines, and which results to keep when only the wax supplier changes turns this page into a reusable change-control template.

Frequently asked questions