Extrusion Screw Design: Compression Ratio, Metering Depth and L/D

Three zones and what each one does

The feed zone conveys solid material from the hopper and compresses it into a solid bed. Its depth is set by the bulk density of the material: a fluffy material needs a deeper channel to convey the same mass, which is why a screw for a low bulk density powder looks different from one for pellets.

The compression zone, sometimes called the transition zone, is where the material melts. The channel depth reduces along its length, which squeezes the solid bed against the barrel wall where the heat is, and generates the shear that does most of the melting. Its length is the single most important variable for melt quality.

The metering zone pumps the melt at a constant rate and builds the pressure the die needs. Its depth sets the output per revolution and its length sets how stable that output is. A short metering zone gives a surging output; a very long one gives a stable output with more shear heating.

The ratios between these three depths are the compression ratio, and it is quoted as feed depth divided by metering depth. That single number summarises most of the screw's behaviour, which is why it is the first thing asked for when a line is not performing.

Metering depth is the output knob

The chart on the left plots output against metering channel depth at three screw speeds. At 100 rpm the output rises from 150 kg/h at a 2.0 mm channel to 302 kg/h at 4.5 mm. The relationship is close to linear, because the output of a drag flow pump is proportional to the channel depth.

That looks like a free lunch, and it is not. A deeper metering channel reduces the compression ratio and reduces the shear in the metering zone, so the melt arrives cooler and less uniformly mixed. The output rises and the melt quality falls, and the two effects meet somewhere around a depth that depends on the material.

The speed curves in the chart also show where the limit is. At 140 rpm the output reaches 388 kg/h, but the gap between the 100 and 140 rpm curves narrows at the deeper depths, which is the signature of a melting limit rather than a pumping limit. Past that point, more depth and more speed both stop helping.

In practice the metering depth is chosen from a rule of thumb, 0.05 to 0.07 times the square root of the screw diameter for a polyolefin, and then adjusted on the first trial. A 90 mm screw would therefore run a metering depth of 2.4 to 2.9 mm, which is in the middle of the range on the chart.

Compression ratio, by polymer

The chart on the right gives the compression ratio each polymer is normally run at. Rigid PVC is the lowest at 1.8, because it is shear sensitive and degrades if it is worked hard. LDPE is 2.2, PC and ABS around 2.4 to 2.6, PP and HDPE 2.8 to 3.0, and PA 66 the highest at 3.4.

The pattern follows crystallinity and melting behaviour. Semicrystalline materials need a high compression ratio because their melting is abrupt and the solid bed has to be broken up mechanically. Amorphous materials melt gradually and need less. And PVC needs the least of all because the penalty for overworking it is degradation rather than poor melt quality.

A screw with the wrong ratio for the material produces a specific set of symptoms. Too low and the output surges, the melt is unmixed and the pressure fluctuates. Too high and the melt overheats, the drive current rises and the output falls as backflow increases.

Changing the ratio means changing the screw, not the setting. That is why the ratio should be chosen from the material at the specification stage, and why a screw bought for one product family rarely performs well on a completely different one.

L/D: what length actually buys

The length to diameter ratio is the other headline number. A 20 D screw is short and cheap; a 30 D screw melts better and mixes better; a 36 D screw is used for difficult materials and for high output. Each step adds barrel length, heater zones and floor space.

The gain from L/D is mostly in the melting and metering length. A longer screw has more area for heat transfer and more residence time in the compression zone, which is what a high output line needs. It does not increase the output per revolution, which is set by the metering depth.

There is a cost that is rarely mentioned. A longer screw has more residence time overall, which is good for melting and bad for heat sensitive materials. A 36 D screw running PVC will degrade it, because the material sits in the barrel for longer at temperature. That is why PVC lines are usually 20 to 25 D.

Table 1 — Zone lengths and what each one controls

When a barrier screw is worth it

A barrier screw separates the solid bed from the melt with a second flight, so that the solid cannot be carried forward unmelted. It is more expensive and it needs a matched barrel, and on the right material it delivers 20 to 30 percent more output at the same melt quality.

It is worth it when the line is melting limited and the material is a semicrystalline polyolefin at high output. It is not worth it on PVC, on small machines, or on a line where the limit is the die or the cooling, because it does nothing for those.

The decision test is the same one used for output: plot output against speed and melt temperature against speed. If the melt temperature falls as speed rises, the line is melting limited and a barrier screw will help. If the melt temperature rises, the line is already past its efficient point and a better screw will only let it run further past it.

Specifying a screw in six numbers

A screw specification needs the diameter, the L/D, the compression ratio, the three zone lengths, the metering depth and the material it is designed for. Everything else, flight width, helix angle, root radius, follows from those.

The material is the most important of the six and the one most often left off a purchase order. A screw is designed around a material's bulk density, melting behaviour and viscosity, and a screw that is excellent for PP may be poor for PET. Writing the material on the drawing costs nothing and prevents most of the disappointment.

The last number to agree is the wear protection. A glass filled material will wear a standard nitrided screw out in a year, and a bimetallic barrel with a hardened screw flight is the difference between a maintenance item and a replacement. That is a cost decision, and it belongs at the specification stage rather than at the first failure.

Table 2 — The six numbers that define a screw