Solid Rod Profile Extrusion: Cooling, Die Swell and Internal Voids

Why solid rod is harder than it looks

A round rod has no corners, no thin walls and no difficult flow paths. On paper it should be the easiest thing an extruder makes. In practice it is one of the harder products to hold to tolerance, because every one of its problems is a volume problem rather than a shape problem.

A 40 mm rod has a surface area of 0.126 m² per metre of length and a volume of 1.26 litres. That ratio — ten times worse than a 4 mm wall tube — is the whole story. Heat has to leave through a small skin, and the core stays molten long after the outside is hard.

The line, step by step

Table 1 — The rod line. The two-tank arrangement is not optional above about 25 mm; one tank cannot remove the heat from the core in the distance available.

Small rod — under 12 mm — is often run without a sizing sleeve at all, straight from the die into a water bath with the haul-off setting the diameter by drawdown. That works because the surface-to-volume ratio is high enough for the rod to freeze quickly. Above 20 mm the free-extrusion approach loses control of diameter and the sleeve becomes necessary.

Cooling a cylinder is a different problem

Cooling time scales with the square of the characteristic dimension, exactly as it does for a flat sheet, but the constant is different because heat is converging on an axis rather than a plane. For a cylinder cooled from all sides, the practical rule is that the time to solidify the core is roughly proportional to the square of the diameter.

Table 2 — Bath length and throughput for PVC rod at 3 m/min line speed. The bath length grows faster than the diameter, which is why large rod lines are long and slow.

The practical consequence is that large rod cannot be run at the same line speed as small rod. Trying to do so produces a rod that is hard on the outside and liquid in the middle, and the internal shrinkage that follows shows up as a void, an oval section or a lengthwise crack discovered by the customer.

Spray cooling between tanks helps because it removes the insulating film of still water that builds up around a hot rod. A rod sitting in a tank develops a boundary layer of warm water that travels with it; spray breaks that layer and can double the heat transfer coefficient on the surface for very little capital cost.

Die swell and the sizing decision

Every melt expands as it leaves the die. For rod the swell is larger than for a thin profile because the die land is short relative to the diameter and there is no calibration surface to restrain it.

Table 3 — Die swell by material. The die diameter has to be sized for the swell at the running melt temperature, not at a nominal value from a table.

Because swell depends on melt temperature, shear rate and land length, the die has to be trimmed on the actual line at the actual running conditions. A die that produces a 20.0 mm rod at 180 °C will not produce 20.0 mm at 195 °C. This is why rod dies are usually supplied with a slightly oversized orifice and finished by the processor.

The drawdown ratio between die and finished rod is the second control. Pulling a rod down hard from a large die gives good concentricity but builds orientation along the axis, which shows up later as lengthwise splitting when the rod is machined or heated. A drawdown of 1.2 to 1.6 is a reasonable compromise; above 2.0 the orientation becomes a liability.

Straightness: where bend comes from

Rod bend is almost always a cooling asymmetry, not a die problem. The melt leaves the die at 170–190 °C and has to be supported until it is stiff enough not to sag. Any asymmetry in that support — gravity, uneven water flow, a haul-off belt that grips one side harder — produces a curve.

Table 4 — Sources of rod bend. The last one is the frustrating case: the rod passes inspection and fails a week later.

The air gap between die and tank is the single most effective lever. Every millimetre of unsupported melt hanging in the air adds sag, and sag on a 40 mm rod at 180 °C is not small. Getting the tank close to the die — 50–150 mm — and supporting the rod with a small roller in that gap fixes most bend complaints on large rod before anything else is touched.

Internal voids and the coring problem

A thick rod cools from the outside in. The skin freezes first and stops moving. The core continues to cool and shrink, but it is now enclosed by a rigid shell, so the shrinkage has to be accommodated by something. What happens is a void: a cavity at or near the centre, sometimes a continuous pipe running the length of the rod.

Table 5 — Voids in solid rod. Distinguishing shrinkage voids from moisture voids takes one cut with a saw: frothy is moisture, smooth-walled and central is shrinkage.

Where the void cannot be eliminated — very large diameter, fast cycle — the industry answer is deliberate coring. A small axial hole, 2–4 mm, is formed by a pin in the die and gives the shrinking core somewhere to go. The rod is then sold as cored rod, or the hole is plugged. It sounds like a defect and it is standard practice on rod above 50 mm.

Square, hex and special sections

Non-round solid sections add a corner problem on top of the volume problem. A square rod has four corners that are further from the cooling surface than the faces, so the corners cool last and shrink most. The result is a section that is slightly concave on the faces and rounded at the corners.

Table 6 — What is realistically achievable on solid sections. The tolerances are looser than for hollow profile because there is no internal calibration to hold the shape.

The corner radius in the die is the main design lever. A sharp corner in a solid die creates a low-flow region and a stress concentration in the product. Building in a radius of at least 0.5 mm — more for large sections — improves both flow and the finished corner. Buyers who specify a sharp corner on a square rod are usually specifying something that will be radiused in practice anyway.

Tolerances you can actually hold

Rod tolerance is a function of diameter and of whether the rod is sized or free-extruded. Free-extruded rod below 10 mm can hold ±0.10 mm on a good line; the same rod at 40 mm free-extruded will drift ±0.5 mm over a shift. Sizing brings large rod back to ±0.15–0.25 mm, at the cost of a sleeve that wears.

The measurement itself needs care. A soft rod measured with a micrometer immediately off the line reads small because the material is still warm and compressible. Measure at least 30 minutes after cutting, at 23 °C, and measure ovality at three positions along a metre rather than at one.