Cable sheathing · Root-cause framework
A dull, rough, lumpy, bubbly or dimensionally unstable jacket is evidence of a process-system mismatch—not proof that one material property is wrong. The shortest path to a useful answer is a controlled comparison, not a temperature guess.
Flexible PVC jacket extrusion couples the compound, screw and barrel condition, screen pack, crosshead and die geometry, melt residence history, puller, cooling and cable construction. Similar surface symptoms can arise from different combinations of those factors. Build the trial around the actual formulation, machine and tooling. Record the evidence before changing a sample or process variable.
Record when the defect begins, whether it is continuous or periodic, which cable size and colour are affected, and whether it follows a material lot, a tooling change, a restart, a speed change or a shift. Preserve an untrimmed sample and take photos with a scale. Record the actual—not target—barrel/head readings, screw speed, line speed, puller stability, screen-pack history, cooling conditions and time at temperature. A single description such as “rough surface” is too broad to distinguish incomplete fusion, flow instability, contamination/gels, die condition, cooling or draw-down stress.
First establish material identity and handling: grade, colour, lot, packaging condition, feed continuity and any regrind or concentrate. Next separate melt-generation evidence (motor load trend, pressure trend, purge appearance and residence events) from downstream evidence (die condition, centring, puller consistency and cooling). Finally measure the product: thickness/eccentricity, surface appearance and any agreed mechanical or electrical checks. Do not change compound, temperature profile, screw speed and draw simultaneously; the result would not identify what changed the outcome.
| Observed issue | Evidence first | Influences to compare—not assume | Escalation point |
|---|---|---|---|
| Rough, dull or “orange-peel” surface | Retained sample, die/purge appearance, pressure/load trend, speed and temperature history | Melt condition, die condition, line speed, cooling and material dispersion | Repeatable defect remains after one controlled line comparison |
| Lumps, pimples or gels | Location/frequency, screen-pack history, hopper/feed record and a cut sample | Foreign material, dispersion, filtration, residence history or tooling deposits | Lot comparison or laboratory examination is needed |
| Bubbles or voids | Cross-section images, when the issue begins, feed and storage history | Entrapped gas, moisture/contamination, unstable feed or melt history | Safety- or electrical-critical construction; suspend acceptance pending agreed testing |
| Unstable thickness or eccentricity | Cross-section measurements, centring record, puller and line-speed trend | Tool alignment, draw-down, conductor movement, puller variation and cooling | Geometry affects the named standard or electrical test specimen |
Industry PVC extrusion guides from Teknor Apex and AlphaGary both publish troubleshooting categories such as surface irregularity, surging, bubbles and shrinkage. They are useful as prompts for what to inspect, but their recommended windows and remedies are compound- and equipment-specific. Use them to build a line record, then confirm a change through the customer’s actual construction and acceptance method. They do not prove that a different grade will run at the same settings.
Research on PVC cable ageing adds a second reason for careful records: the structure surrounding a PVC layer can affect environmental exposure and measured dielectric behaviour. Consequently, a pellet result, a free film and a completed cable should not be substituted for one another when the issue concerns cable performance.
Send the cable drawing, target standard/clause, jacket thickness and tolerance, colour, current material/TDS, photos of the defect, actual line record and any failed-test report. This lets the discussion begin with evidence rather than with a speculative reformulation claim.
Start an RFQ