Injection Mold Cooling Design: Channel Layout and Cycle Time

In most injection molds, cooling takes more than half of the cycle. It also decides whether the part comes out flat or warped. Yet cooling channels are often drilled last, wherever the ejector pins and screws leave room. This article gives the rules for sizing, placing and connecting cooling channels, and a short calculation to check that the circuit can remove the heat.

How much heat, and how long

Two numbers frame the design: the cooling time the part needs and the heat the mold must remove per second.

Cooling time depends mainly on wall thickness. The plastic is a poor heat conductor, so heat from the middle of the wall must travel through the plastic itself before the steel can take it away. The time rises with the square of the wall thickness.

These values assume the cavity surface stays at the set mold temperature. A badly cooled mold surface runs hotter than the water, and the real cooling time is longer. That is the part the mold designer controls.

Heat load is the mass of plastic per shot multiplied by the heat each kilogram gives up between melt temperature and ejection temperature, divided by the cycle time. Semi-crystalline materials give up more heat than amorphous ones.

Channel diameter, depth and pitch

The logic behind the table:

• Channels too close to the surface or too far apart give alternating cold and warm stripes on the cavity. This shows as gloss differences and uneven shrinkage.
• Channels too deep give an even surface temperature but respond slowly and remove less heat.
• Keep the same depth along the whole cavity. A channel that follows the part contour at constant distance is worth more than a larger straight channel that is near in one place and far in another.
• Keep at least 3–4 mm of steel between a channel and any ejector pin hole, screw or insert pocket, more in hardened steel under high injection pressure.

Turbulent flow is not optional

Water in slow, layered (laminar) flow takes up heat several times less effectively than water in turbulent flow. The measure is the Reynolds number: above about 4,000 the flow begins to be turbulent, and a value above 10,000 is the usual design target.

For water at 20 °C, the flow needed for a Reynolds number of 10,000 is roughly 0.47 litres per minute for each millimetre of channel diameter. Warm water is thinner and needs less.

The second check is the temperature rise of the water between inlet and outlet. Keep it within 2–5 °C, and within 2–3 °C for precision parts. A larger rise means one end of the cavity is cooled by warmer water than the other.

Worked example

A PP part and runner weigh 0.12 kg per shot. The cycle is 20 seconds. The mold has four circuits of 10 mm channels.

The result says: with chilled water, raise the flow to about 5 L/min per circuit or accept three circuits. The calculation takes five minutes and prevents a mold that can never reach its planned cycle.

Circuit layout

Practical rules:

1. Use several short series circuits rather than one long circuit or many unbalanced parallel branches.
2. Give each mold half its own circuits, and separate circuits for slides and large inserts, so their temperatures can be set independently.
3. Put the inlet near the hottest area, usually near the gate.
4. Avoid internal parallel branches. If you must use them, make the branches identical in length and diameter.
5. Do not let hoses and quick couplings be the bottleneck. A coupling with a 6 mm bore on a 12 mm channel cancels the benefit of the larger channel.
6. Mark every inlet and outlet on the mold and draw the circuit plan on the mold drawing.

Cooling the core

The core receives more heat per unit of surface than the cavity, because the part shrinks onto it and it has less steel to carry heat away. Cores that run hotter than the cavity pull the part into a warped shape. Options, from simple to advanced: