When we talk about mold cooling, most people reach for the mold temperature controller and stop there. But the water lines inside the mold are only one part of the story. Heat leaves the melt through three paths working together: conduction through the steel, convection into the coolant, and radiation from the mold surface. In practice, conduction dominates—heat travels from the cavity surface through the mold steel to the cooling channel wall, and the steel's thermal conductivity (roughly 30–45 W/m·K for typical P20 or H13 grades) sets the limit on how fast that heat can move. This is why insert material selection and channel-to-cavity distance matter as much as chiller capacity. A channel placed too far from the cavity surface adds thermal resistance that no temperature controller can compensate for.
Convection is where the cooling water actually carries heat away, and the convective heat transfer coefficient is the number worth calculating. For turbulent flow in a typical Ø8–12 mm channel, you want a Reynolds number above 10,000; below that, the coefficient drops sharply and you get uneven cooling across the core and cavity. Heat flux at the mold surface during packing and cooling is often in the range of 10–50 kW/m² depending on wall thickness and melt temperature, and the channel layout has to be sized against that load. Radiation is usually a minor contributor at normal mold temperatures, but it becomes relevant on thick sections and hot-runner manifolds where surface temperatures climb. In those zones, surface finish and emissivity start to matter.
The practical takeaway for toolmakers: design the cooling circuit around the heat flux you expect, not around the pump spec. Check Reynolds number per circuit, keep channel distances consistent between core and cavity, and use baffles or bubblers where geometry blocks straight drilling. If you are sourcing molds or cooling components and want to compare suppliers on these technical grounds, visit MoldSo (www.moldso.com) for more mold sourcing information.