
- 1. What makes cable extrusion different from pipe extrusion
- 2. The crosshead: where concentricity is made
- 3. Conductor preheating: the step that is always underestimated
- 4. Cooling: the trough is not just a trough
- 5. Materials and what each one demands
- 6. Faults and where they come from
- 7. What gets tested, and what it tells you
- 8. Specifying an insulated wire
What makes cable extrusion different from pipe extrusion
At first glance a wire insulation line looks like a small tube line: an extruder, a crosshead, a cooling trough, a capstan and a winder. The difference is that the "tube" is being formed around a solid core that is moving at the same speed, and the wall that matters is the one between the conductor and the outside world.
That single fact changes the whole process:
Wall thickness is measured as a minimum, not a nominal. A cable standard does not say "0.7 mm insulation"; it says "minimum 0.6 mm at any point". The nominal is set higher than the minimum so that the eccentricity of the process does not breach the floor.
Concentricity is a product of tooling and preheat, not of the screw. The conductor must be centred in the die by the tooling, and the polymer must wet it before it has a chance to sag.
The line speed is set by the conductor, not by the extruder. Output is what it is; the capstan pulls the wire at whatever speed consumes that output at the required wall.
The number that decides everything: eccentricity ratio, usually expressed as the minimum wall divided by the maximum wall. A good line runs at 90–95%. A line at 80% has to run 25% more insulation to guarantee the minimum, which is a direct material cost on every metre.
The crosshead: where concentricity is made
The crosshead turns the melt through 90° and around the conductor. There are three common designs and the differences are not cosmetic.
| Crosshead type | How the melt flows | Concentricity | Changeover |
|---|---|---|---|
| Pressure (offset) crosshead | Melt enters from one side and flows around a torpedo | Good; the flow is balanced by the torpedo shape | Moderato |
| Tube-on (jacketing) | Melt forms a tube and is drawn onto the conductor by vacuum | Good, but needs vacuum control | Veloce |
| Pressure-tight / self-centring | Melt completely surrounds the conductor in a sealed chamber | Best; centring is mechanical | Lento |
| Semi-tube (semi-pressure) | Melt tube is drawn onto the conductor without full pressure | Moderate; relies on draw-down | Veloce |
Table 1 — Crosshead types. Pressure-tight gives the best concentricity but the slowest changeover, which is why it dominates high-volume single-product lines and loses to semi-tube in job shops.
The die and tip set the wall and the centring. The tip is the internal mandrel that the conductor passes through; the die is the outer ring. The gap between them is the annular opening, and the conductor must sit exactly in the middle of it. Centring screws adjust the tip position in two axes, and setting them is the single most important setup task on the line.
Centring is done by running a short length, cutting a cross-section, and measuring the wall at four points. Adjust, run again, measure again. A good operator gets to 95% eccentricity ratio in three iterations. A line that will not centre usually has a bent tip, a worn die, or a crosshead that is not square to the line axis.
Conductor preheating: the step that is always underestimated
Cold copper entering a hot die does two things: it chills the melt at the interface, and it condenses any moisture on its surface. Both cause adhesion problems that show up later as a loose insulation that slides off the conductor.
| Preheat method | Temperature reached | Effect |
|---|---|---|
| None | Ambient | Poor adhesion, chill marks, moisture condensation |
| Resistance (current through conductor) | 80 – 150 °C | Good, uniform, needs slip rings; standard on fine wire |
| Induction | 100 – 200 °C | Fast, contactless, good for larger conductors |
| Hot air oven | 60 – 120 °C | Slow, bulky, but simple and works for large cables |
Table 2 — Conductor preheating options. The target is usually 80–120 °C for the conductor surface, which is enough to drive off moisture and promote wetting without overheating the polymer.
Preheat also removes the moisture film. Copper wire stored in a humid plant carries a microscopic water layer, and that water turns to steam at the die interface. The steam expands, creates a bubble, and the bubble becomes a void in the insulation. Voids are a dielectric failure waiting to happen and they are invisible from outside.
Cooling: the trough is not just a trough
Cable cooling is usually a series of troughs with decreasing water temperature — hot at the front, cold at the back. The reason is stress: quenching hot polymer in cold water freezes the outside while the inside is still shrinking, and the result is a insulation that is tight on the conductor and stressed through the wall.
| Zona | Water temperature | Scopo |
|---|---|---|
| First trough | 60 – 80 °C | Gradual set, low stress, good adhesion |
| Second trough | 40 – 60 °C | Main cooling |
| Third trough | 20 – 30 °C | Final set |
| Air wipe | — | Removes water before the capstan; wet cable slips |
Table 3 — Typical three-zone cooling. A single cold trough is faster and produces insulation that shrinks back off the conductor over weeks.
The distance from the die to the first water contact matters too. Too close and the polymer sets before it has wet the conductor; too far and it sags under its own weight on small cables. The usual guideline is that the insulation should enter the water while it is still slightly tacky, which on a small line is 150–300 mm from the die.
Materials and what each one demands
| Compound | Melt temp (°C) | Typical use | Characteristic issue |
|---|---|---|---|
| PVC (plasticised) | 160 – 185 | Building wire, flex, low voltage | Plasticiser migration; HCl corrosion of the die |
| PE (LDPE / HDPE) | 170 – 210 | Power cable, telecom | Low adhesion without preheat; shrinkage |
| XLPE (crosslinked PE) | 170 – 200 (before curing) | MV/HV power cable | Must be cured after extrusion; scorch risk in the extruder |
| PP | 200 – 230 | Automotive wire, thin wall | Brittle at low temperature; high shrinkage |
| TPU / TPE | 180 – 210 | Flexible cable, robotic, charging | Soft, easily stretched at the capstan; adhesion control |
| FEP / PFA | 340 – 380 | Plenum, high temperature, data | Very high temperature; corrosive decomposition |
| LSZH compounds | 150 – 180 | Building, transit, marine | High filler loading; poor flow, brittle, moisture sensitive |
Table 4 — Cable compounds. LSZH is the one that most often surprises a line set up for PVC, because the 60% filler loading changes flow, die swell and cooling behaviour completely.
XLPE is a special case. The compound contains a peroxide that crosslinks the polymer after extrusion, and the crosslinking happens in a continuous vulcanisation line — either steam, nitrogen, or infrared. The extrusion step is therefore only half the process, and the extruder must be run cool enough that the peroxide does not decompose prematurely in the barrel. That is why XLPE runs at 170–200 °C when the finished material would happily process at 230.
Faults and where they come from
| Fault | Aspetto | Causa |
|---|---|---|
| Eccentric wall | Thin on one side, thick on the other | Tip not centred, bent tip, worn die, crosshead misaligned |
| Voids / bubbles | Internal bubbles, sometimes visible as surface blisters | Moisture on conductor, wet compound, air entrapment in the crosshead |
| Poor adhesion | Insulation slides off the conductor | No preheat, contaminated conductor, die too cold |
| Die lines | Longitudinal grooves | Die damage, die build-up, or cold compound |
| Surface roughness | Sharkskin or matte patches | Melt fracture; die too cold or output too high |
| Diameter variation | Periodic thick/thin along the length | Capstan speed variation, screw pulsation, haul-off slip |
| Shrinkage after winding | Insulation pulls back from cut ends | Excessive quench, high draw-down, insufficient annealing |
| Scorch (XLPE) | Gels and discoloured lumps | Barrel temperature too high, long residence time, dead spots |
Table 5 — Cable extrusion faults. Voids and adhesion failures both trace back to preheat more often than to the compound.
The diameter variation fault is worth a note because it is often blamed on the extruder. A capstan driven by a DC motor with a worn tacho will vary by 1–2%, which on a 1 mm insulation is 10–20 µm — enough to fail a minimum wall check. Modern lines use servo capstans with encoder feedback, and the difference in diameter consistency is immediate.
What gets tested, and what it tells you
| Test | What it measures | Typical frequency |
|---|---|---|
| Wall thickness and eccentricity | Cross-section at four or six points | Start of run, then hourly |
| Outer diameter | Continuous, by laser gauge | Continuo |
| Adhesion / strip force | Force to pull insulation off the conductor | Start of run, then per shift |
| Spark test | Holidays and thin spots in the insulation | 100% inline |
| Tensile and elongation | Compound integrity after processing | Per lot |
| Hot set (XLPE) | Degree of crosslinking | Per lot |
| Volume resistivity | Dielectric quality | Per lot, or per cable design |
Table 6 — Cable insulation testing. The inline spark test is the one that catches the failures the offline tests miss, because it inspects every metre rather than a sample.
Adhesion is the test that most often fails on a new setup and the one most often waived. A loose insulation is not just a handling problem — on a flexible cable it allows the conductor to move inside the insulation, which eventually breaks strands at the point of movement. If the strip force is below the standard, the fix is preheat and die temperature, not more cooling.
Specifying an insulated wire
A wire specification should state: conductor size and stranding, insulation compound and grade, nominal and minimum wall thickness, outer diameter with tolerance, adhesion requirement, and the standard it is being tested to. Add the operating environment — temperature rating, flex requirement, flame rating, oil or chemical exposure.
The minimum wall is the number that drives cost. A 0.6 mm minimum with a 0.7 mm nominal requires a process that holds 85% eccentricity. A 0.6 mm minimum with a 0.62 mm nominal requires 97%, which is a different class of line and tooling. It is worth knowing which one you are asking for before you compare prices.
What is eccentricity and why does it matter?
Eccentricity is how far off-centre the conductor sits inside the insulation, expressed as minimum wall divided by maximum wall. A 95% ratio means the thinnest point is 95% of the thickest. Lower ratios force you to run a thicker nominal wall to guarantee the minimum, which costs material on every metre.
Why does my insulation slide off the conductor?
Poor adhesion, almost always caused by insufficient conductor preheat or a cold die. Cold copper chills the melt at the interface before it can wet the surface, and any moisture film on the wire becomes steam that prevents contact. Preheat to 80–120 °C and check the die temperature.
What causes bubbles inside the insulation?
Moisture. Either on the conductor surface, in the compound, or entrained in the crosshead. Preheat drives off surface moisture; the compound needs proper drying if it is hygroscopic. Voids are a dielectric weakness and are invisible from outside, which is why the spark test matters.
Why does XLPE run cooler than PE?
Because the compound contains a peroxide crosslinking agent that decomposes if the melt gets too hot in the barrel. The crosslinking is supposed to happen in the vulcanisation line, not in the extruder. Running XLPE at 230 °C would scorch it in the barrel and produce gels.
Do I need a pressure crosshead or a tube-on crosshead?
Pressure-tight gives the best concentricity and is right for high-volume single-product runs. Tube-on and semi-tube are faster to change over and are what job shops use. If your eccentricity requirement is tight and the volume is high, pressure-tight pays for itself in material saving.
How is outer diameter controlled?
By a laser diameter gauge feeding back to the capstan speed or the extruder output. The gauge measures continuously and the controller adjusts one of the two. Most lines control on capstan speed because it responds faster, and adjust extruder screw speed only for large corrections.
What is a spark test?
An inline high-voltage test that passes the insulated wire through an electrode. Any holiday, thin spot or pinhole in the insulation causes a breakdown that is detected and marked. It is a 100% inspection, not a sample, and it is the only way to catch random defects.
Can I extrude insulation onto a cable that already has a jacket?
Yes — that is jacketing, and it uses a tube-on or pressure crosshead with a larger tip. The considerations are the same: preheat the core, centre the die, and control adhesion. Jacketing over a stranded core needs more vacuum to pull the melt into the interstices.
Sourcing insulated wire or cable compounds?
Send us the conductor specification, the insulation compound, the wall and diameter requirements and the standard you are testing to. We will come back with a process route and the eccentricity ratio we can hold.
Continue reading
- Plastic Sheet Extrusion: Roll Stack, Polishing and Thickness Control
- Plastic Pipe Extrusion: Materials, Standards and Sourcing
- Melt Pump and Screen Changer: Why Pressure Stability Beats Temperature
- Extruder Screw Design: Zones, Compression Ratio and Wear
- Extrusion Output Calculation: Throughput, Screw Speed and Line Speed











