Most pour-temperature guides hand you a single number. 135F for soy, 175F for paraffin, done. That number is a midpoint, and the midpoint is where your adhesion and your surface finish stop agreeing with each other. We have been pouring container soy at scale since 2019, and the 15-degree band between 125F and 140F is where the candle is actually decided.
The Adhesion Floor and the Frosting Ceiling
Soy 464 melts in the 113-119F range. Below about 125F at the pour, the wax is already part-solidifying at the glass wall. The surface you see in 24 hours is a shrinkage layer, not a bonded face, and cold-throw on the first few burns comes back weak because the top centimeter never fully wets the container. Push past 140F and the wax still bonds, but the surface is now fighting itself. Soy crystallizes on cooling, and a hotter pour gives the crystalline phase more time and more energy to form before the surface sets. That is the frosting line. It is not a defect in the wax; it is the pour temperature talking.
In our line, 130-135F is the band that clears both tests: clean adhesion under the 70F room-cool and a matte, frost-free surface within the first 12 hours. Preheating the glass to 100-120F is not optional. It closes the thermal shock at the wall and is the single cheapest fix for pull-away on the lower third of the vessel.
We have watched this fail in a specific shape. A batch poured at 138F into cold glass came out with a clean top and a 2mm pull-away ring at the bottom quarter. The top passed visual inspection; the bottom failed on the first burn, when the ring released and the wick migrated. The pour temperature looked fine on paper because the pot read 138F. The glass was sitting at 62F. The 76-degree delta at the wall is what tore the bond, not the pour number.
Why the Number Moves With the Wax
Paraffin 4630 pours at 170-180F. Coconut C-3 pours at 120-140F but wants the lower end. A coconut-soy blend shifts the window by 5-10F depending on the coconut share. The fragrance load does too. A 9% load adds a measurable amount of liquid to the melt and lowers the effective surface tension at the pour, which is why a blend that poured cleanly at 135F at 6% FO starts showing surface texture at the same temperature at 9%. Stir the fragrance in for a full two minutes, not the 30-second stir most recipes call for. Incorporation is a surface-energy problem, and a short stir leaves air and oil on the surface, not in the bulk.
If you pour a coconut-soy at 140F and get a rough top, do not reach for the dye or the wick. Drop the pour by 5F and preheat the glass. The wax is telling you the window moved; the wick chart cannot fix a surface.
One number worth pinning to the melt station: paraffin pours about 40F hotter than soy for the same vessel. If a supplier sends you one universal pour sheet for a soy line and a paraffin line, the soy is being over-poured or the paraffin under-poured. Both show up as surface texture, and both get blamed on the wax.
The Room Is the Third Number
A pour at 135F in a 75F room cools to set in a different shape than the same pour in a 60F room. Cold ambient pulls the top down first, and the bottom is still liquid under it. That is the crack that survives the pour and shows up six hours later. Keep the room in the 70-75F band, keep the candles off cold shelves, and do not draft on them. The candle is a small structure solidifying in air, and the air has a vote in the outcome. Record the pour, the room, and the preheat for every batch. The 5F that looks like noise in one batch is the whole window in the next.