How Do I Establish an Acceptable Temperature Window Rather Than Requiring Employees to Hit One Exact Number During Commercial Pouring?

Why a single target number fails on a production floor

A single number sounds precise and behaves badly. Wax in a pitcher is cooling from the moment it leaves the melter, so no one can pour a whole pitcher at one temperature. Thermometers differ slightly from each other. And people under time pressure round towards the number they were told. The honest alternative is a window: a range with edges you have proved, a working zone inside those edges, and clear instructions for what to do when a reading falls outside.

A window is not a looser standard. Done properly it is tighter, because every candle in it has been shown to pass, and every reading outside it triggers a defined action instead of a shrug.

The three layers of a working window

Think of the window as three nested layers, each with a job.

The guard band protects you from the two errors you cannot remove: the thermometer's own error, which is not published for many pen thermometers, and the fall in temperature between reading a pitcher and emptying it. The size of the guard band is a suggested starting point to validate; one or two degrees each side is a common place to begin.

Find your edges with a bracket study

Worked example. A maker plans 20 kg production batches of Premium Coconut Soy Wax CSI 468 with Blue Ocean at 8% of total weight. CSI 468's product page states a pour temperature of 58–60°C and a melt point of 50–53°C. That published range is where the study starts; the study tells the maker how far either side of it their jars and wicks still produce a good candle.

Suggested starting points to validate: pour three candles of 200 g at each of four temperatures, 55°C, 58°C, 61°C and 64°C, read on a separate thermometer at the moment of pouring, all from one small batch so everything else is shared. Twelve candles need 2.4 kg of scented wax: 2.208 kg wax + 0.192 kg oil.

What the study costs, with the working: wax 2.208 kg × ₹637.00 per kg (the 5 kg pack at ₹3,185.00 ÷ 5) = about ₹1,406.50; Blue Ocean 192 g from two 100 g packs at ₹495.00 each = ₹990.00. About ₹2,396.50 in all. The 20 kg production batch it protects: wax 18.4 kg × ₹637.00 = ₹11,720.80, oil 1.6 kg as 1 kg ₹4,956.00 + 500 g ₹2,478.00 + 100 g ₹495.00 = ₹7,929.00, so ₹19,649.80. The study costs roughly an eighth of one batch.

From these illustrative scores, the validated range is 58°C to 61°C with confidence, and 64°C is a marginal edge worth confirming before relying on it. Remember these are one maker's example results; your jars and wicks may shift the edges either way.

Turning the result into operator instructions

With a validated range of 58–61°C and a one-degree guard band each side, the operating window becomes 59–60°C, which matches the wax maker's stated 58–60°C closely. That is narrow for a pitcher that cools while pouring. The maker has two honest choices: confirm 64°C as a valid upper edge with a second study, which would widen the operating window to roughly 59–63°C, or keep the window narrow and pour fewer jars per pitcher so the fall between reading and last jar stays small.

Either way, the instruction sheet says the same four things. Start a pitcher only inside the operating window. Read again before the last jar. At a warning reading, pause and tell the batch owner. At a stop reading, the pitcher goes back to the batch and the reheat rule applies. Write the numbers on a card at each station, not only in a file.

Where the window is read matters as much as its numbers

A window only means something if everyone reads it at the same place. The wax that matters is the wax entering the jar, so the operating window applies to the pitcher, read with the thermometer in the wax just before pouring, after a gentle stir with a bamboo stick. A melter reading inside the window says nothing about a pitcher that has been standing for five minutes. Write the reading point on the station card: pitcher, stirred, immediately before first jar and again before last jar.

One window per wax is also rarely enough. A heavier jar, a different wick series or a fragrance family that sets differently can each move the edges. Keep a small window table with one row per wax, jar and oil family combination you produce often, and mark each row with the date of the bracket study that set it. Rows without a study date are guesses and should be treated as such until one is run.

Candles poured before the window existed

For stock poured under the old one-number instruction, the bracket study doubles as a reference set. Compare a few of those candles with the study candles at each temperature. If they look and burn like the 58–61°C set, release them. If they resemble the 55°C set, sort by appearance, repair tops where the defect is cosmetic and hold any that burn differently. Do not remelt stock just because the window is new; remelt only on evidence.

Revisit the window when you change jar supplier, wick or fragrance family, or when a season changes your room conditions. The melter's display has a stated ±10°C deviation, so every edge on this page is read on a separate thermometer. If you want a second view on a bracket result, CSI will look at it on WhatsApp (+91 7397976926).

Where this page fits

CSI already has guides that cover parts of this question. This page covers the failure investigation for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions