Multi-Cavity Profile Dies: Balancing Flow Without Losing Output

Why multi-cavity is a volume decision

Running two or four profiles through one die is the standard way to raise output without buying a bigger extruder. The arithmetic is attractive: a 75 mm extruder producing 250 kg/h can feed four cavities of a small profile at 62 kg/h each, or one cavity at 250 kg/h — but the single cavity would need a line speed four times higher, and most cooling sections cannot support that.

So multi-cavity is often not about output at all. It is about line speed. Four cavities let you run at a comfortable 3–4 m/min instead of 12–16 m/min, which the cooling tank can actually handle.

The manifold: where balance is won or lost

A multi-cavity die takes one melt stream and splits it. The splitting geometry is the manifold, and its job is to deliver identical pressure and identical thermal history to every cavity.

Table 1 — Manifold types for multi-cavity dies. The T-manifold is cheap and produces cavities that run at different pressures, which is exactly the problem multi-cavity is trying to avoid.

The balanced bridge is the reference design: the runner path from the inlet to each cavity is made exactly the same length, so each cavity sees the same pressure drop. It costs more to machine and it takes up more space in the die body, but it removes the single largest source of cavity-to-cavity variation.

Land length: the fine adjustment

Manifold balance gets you close. Land length gets you exact. The land is the final parallel section of the flow channel just before the die exit, and its length controls the pressure drop for that cavity.

Table 2 — Land length as the balancing tool. A die designed with 20% extra land on each cavity can be tuned by machining, which is much cheaper than re-making the manifold.

This is why a well-designed multi-cavity die is machined with the land deliberately long, then shortened cavity by cavity during commissioning until the outputs match. Once material is removed it cannot be put back, so the die starts rich and is tuned lean.

Some dies use replaceable restriction inserts in each land. They cost more upfront and they make the die serviceable — a worn or damaged cavity can be re-balanced without re-machining the body.

Thermal balance: the invisible asymmetry

Flow balance is visible in the drawing. Thermal balance is not, and it causes as many problems.

Table 3 — Thermal sources of cavity imbalance. The outer-cavity problem is the most common: cavities near the die edge lose heat to the surroundings and run cooler.

A 5 °C difference in melt temperature between cavities changes the viscosity by roughly 3–5%, which changes the output by a comparable amount. On a profile with a ±0.2 mm tolerance that is enough to put one cavity out of spec while the others are fine.

Downstream: four profiles, one haul-off

Multi-cavity dies create a downstream problem that is often underestimated. Four profiles leave the die at slightly different speeds because the melt velocity differs across the die face, and they must all be pulled by the same haul-off.

Table 4 — Downstream problems in multi-cavity lines. The cut length spread is the one that shows up in the customer's inspection, because it is measurable on every piece.

Cut length is where imbalance becomes commercially visible. If one cavity produces profile at 3.0 m/min and another at 3.05 m/min, and the saw cuts on a length signal from the puller, the two cavities produce pieces that differ in length by 1.7%. On a 3 m length that is 50 mm — far outside most tolerances.

The standard solution is to cut on a signal from the fastest cavity, or to use a saw that cuts each cavity separately. Neither is free. The better solution is to balance the die well enough that the spread is under 0.3%.

Cost and when to choose multi-cavity

Table 5 — Die cost against cavity count. The cost rises faster than linearly because the manifold, the thermal zoning and the balancing effort all scale up.

Above six cavities the balancing problem starts to dominate. The manifold becomes long, residence time increases, and the thermal zoning needs more zones than most controllers provide. Most profile extrusion stays at two or four cavities, with six used only for small, simple sections.

Commissioning a multi-cavity die

Table 6 — Commissioning sequence. The thermal check before any material runs is the step most often skipped, and it is the one that explains most of the initial imbalance.

The 4-hour stability run matters because a die that balances in the first 20 minutes can drift as the manifold heats unevenly. Sampling hourly for four hours catches the drift while the die is still on the bench and the die maker is still available.