In extrusion blow moulding, the parison is extruded downward from an annular die at melt temperatures between 180 °C and 230 °C for high-density polyethylene grades. Parison sag is defined as time-dependent elongation under gravitational load after the parison exits the die, while die swell is the radial expansion of the viscoelastic melt caused by recovery of shear-induced normal stresses. The die swell ratio B is expressed as D_parison/D_die, where D_parison is measured with a laser diameter gauge positioned 10–15 mm below the die face and D_die is the inner diameter of the die bushing. Capillary rheometry according to ISO 11443:2021 at an apparent shear rate of 200 s⁻¹ and 190 °C provides the entrance pressure loss and shear viscosity data used to estimate recoverable strain. On a production line with a 60 mm single-screw extruder having a 24:1 L/D ratio and an accumulator head, the parison hang time typically ranges from 0.5 s to 5 s, during which the parison may elongate by 5–30% depending on melt strength, die gap, and molecular weight distribution. Die gap programming through a servo-hydraulic mandrel actuator with position resolution of 0.01 mm and response time of 5–10 ms modifies the local parison wall thickness before the mold closes. The processing conflict is direct: reducing melt temperature suppresses sag but increases die swell and head pressure, while raising melt temperature reduces die swell but accelerates gravitational creep. Melt flow rate measured according to ISO 1133-1:2022 at 190 °C/2.16 kg is therefore an incomplete predictor because it does not capture elongational strain hardening that governs sag resistance.

How Does Die Swell Influence Initial Parison Diameter in HDPE Extrusion Blow Moulding?

Measured at an apparent shear rate of 200 s⁻¹ and 190 °C, the die swell ratio for a bimodal HDPE blow moulding grade typically falls between 1.5 and 2.0 when the annular die land length-to-gap ratio is 15:1. This ratio must be determined on the specific die because changes in land length, inlet angle, and mandrel surface roughness alter the recoverable shear strain. Die swell increases when molecular weight distribution broadens, particularly when the high-molecular-weight tail increases normal-stress relaxation time. On an accumulator-head machine fed by a 75 mm/30:1 grooved-feed extruder with a melt pump, volumetric output variation can produce parison diameter fluctuations of ±0.15–0.30 mm unless the melt pump suction pressure is controlled to ±0.2 MPa. Radial die centering with four or six adjustment screws compensates for die swell asymmetry caused by uneven temperature distribution around the head; temperature differences of 2–5 °C across the die circumference are sufficient to create measurable wall thickness eccentricity. Die swell also raises the effective parison diameter above the die bushing diameter before mold closure, which influences the blow ratio—defined as the ratio of the mold cavity diameter to the parison outside diameter—and therefore the hoop draw. For HDPE containers with a blow ratio of 2.5:1 to 3.5:1, final wall thickness is approximately the initial parison wall thickness divided by the blow ratio after accounting for local stretching and cooling. The initial parison diameter must be selected such that the fully inflated wall thickness falls within the specification, typically 0.8–1.5 mm for a 1 L container sidewall, while pinch-off zones require 2.5–4.0 mm after welding.

Axial wall thickness distribution in a 1 L high-density polyethylene container is controlled primarily by servo-hydraulic parison programming applied during the extrusion stroke. The controller divides the parison length into 100–200 discrete program points, each assigned a target die gap, and the hydraulic actuator moves the tapered mandrel relative to the die bushing in 5–20 ms per step. Near the top seal area, die gap may be set to 2.0–2.8 mm to provide enough material for the neck and shoulder, the body section may be programmed to 1.2–1.8 mm to save weight, and the bottom pinch-off region may be increased to 2.8–4.0 mm to maintain weld strength. Because the parison continues to sag during the extrusion stroke, the programmed profile must be corrected by a sag model that accounts for instantaneous parison length, local strain rate, and melt temperature; otherwise the bottom section arrives at mold closure with a wall thickness 10–20% below the programmed value. Ultrasonic and beta backscatter thickness gauges mounted on the container discharge conveyor measure the distribution non-destructively with a resolution of 0.01 mm per wall, following calibration against a certified step wedge whose thickness traceability follows ISO 16809:2017. Field experience on shuttle and continuous wheel machines with 12–20 stations shows that the largest thickness variation appears opposite the parting line, where mold pinch-off flashing and uneven cooling alter local stretching. Mold temperature differences of 5–10 °C between the parting line and the cavity centre shift the local wall thickness by 3–8% due to crystallization-induced shrinkage differences. For a 1 L bottle with a target body wall of 1.0 mm, a process capability study would therefore require at least 30 samples per cavity, measured at 8–12 defined points per container, to establish a C_pk above 1.33 under stable production.

Thermal and Rheological Boundaries for Sag Control in HDPE Accumulator-Head Lines

Sag kinetics during the 0.5–5 s parison hang time become the limiting factor for containers taller than 200 mm. The instantaneous Hencky strain is ε = ln(L/L₀), and the strain rate is ε̇ = (1/L)dL/dt; under constant gravitational stress, the apparent extensional viscosity η_E = σ/ε̇ falls with increasing strain rate for linear polyethylene and rises with strain rate for strain-hardening grades. Strain hardening in bimodal HDPE arises from the high-molecular-weight tail, and it can be characterized by the maximum force at break in a Rheotens test conducted after a capillary rheometer at 190 °C, with the force values quoted in centinewtons. A resin with a Rheotens force of 0.20 N at 190 °C may exhibit a sag length increase of 8–12% in a 3 s hang time, whereas a low melt strength grade at 0.08 N may sag 20–30% under identical conditions. On an accumulator-head machine with a shot capacity of 2.5 kg and a die bushing diameter of 25 mm, increasing the melt temperature from 190 °C to 220 °C reduces head pressure by 15–25% but raises sag rate by a factor of 2–4 because elongational viscosity has Arrhenius temperature dependence with an activation energy near 25–30 kJ/mol for high-density polyethylene. The practical processing window is therefore bounded at the lower end by excessive die swell and head pressure above 35 MPa and at the upper end by sag-induced thickness variation exceeding ±10% of target. Pre-drying of HDPE is not normally required when resin moisture is below 0.01% by weight, but humid regrind above 0.05% moisture can create steam bubbles and surface pitting during extrusion above 200 °C.

In injection blow moulding, the preform is injection moulded over a core rod inside a closed cavity at injection pressures of 80–140 MPa, and subsequent blow forming occurs either in the same station or after rotation to a blowing station. The wall thickness distribution is fixed almost entirely during injection, so core-to-cavity concentricity must be held within 0.03–0.05 mm to prevent radial wall thickness variation greater than 0.1 mm in a 2.0 mm preform wall. Core rods are commonly made from beryllium-copper or tool steel, with internal channels supplying oil at 120–150 °C, while the injection cavity may be cooled to 10–20 °C to solidify the surface without freezing the gate. The injection speed and holding pressure are programmed in multiple stages to minimize gate blush and sink marks; typical injection time is 0.5–2.0 s, followed by a holding pressure of 40–60 MPa for 2–5 s. Because the preform is transferred at elevated temperature, its surface temperature uniformity must be within ±3 °C at the blow station; otherwise local stretching during inflation creates thin spots near the shoulder and base. Wall thickness distribution is measured after sectioning by optical microscopy or by Hall-effect sensing on a ferromagnetic core rod analogue, with data acquisition at 12–24 points per bottle. For polyethylene and polypropylene pharmaceutical containers, the sidewall thickness is typically 1.0–2.5 mm, the shoulder is 1.2–2.8 mm, and the base is 1.5–3.5 mm, but published data for any single tooling configuration is limited and must be established by capability studies per ISO 9001:2015 or ISO 13485:2016 control plans. The process accommodates complex neck finishes but requires that the core rod temperature, injection velocity, and holding pressure all be stable within ±2% to maintain C_pk above 1.33.

When Parison Sag Supplements Radial Die Profiling in Complex Container Geometries

When the container neck and handle sections require wall thickness variations greater than 0.8 mm over an axial distance of 10–20 mm, sag-assisted preform shaping is combined with radial die programming. This technique intentionally allows the parison to elongate so that the lower region becomes thinner while the upper region remains thicker, reducing the required die gap excursion. Finite-element simulation using ANSYS Polyflow or Moldex3D with an integral K-BKZ constitutive model requires shear and extensional relaxation spectra extracted from dynamic shear measurements according to ISO 6721-10:2015 and capillary shear measurements according to ASTM D3835-16. The simulation meshes the annular die, the free parison surface, and the moving mold halves, with boundary conditions including no-slip at the die wall, free-surface tension, and contact with mold walls after clamping. For a 1 L HDPE bottle with a blow ratio of 3:1, a die gap programme of 1.8 mm at the neck tapering to 2.4 mm at the base, combined with 8% sag-induced bottom thinning, may yield a body wall thickness distribution of 0.9–1.2 mm. Published simulation and experimental data for handle pinch-off zones with complex parison weld lines is limited, so each tool requires synchronous wall-thickness scans across at least six angular positions and 20 axial stations before release. The acceptable process window is narrow: a deviation in die gap of ±0.05 mm at any programme point can shift local wall thickness by more than 5% in thin body sections. Proportional hydraulic valves controlling the mandrel must therefore be tuned with a dead time below 10 ms and a position repeatability of ±0.005 mm over a 4.0 mm stroke, while the extruder screw speed and melt pump suction pressure must be maintained within ±1.5% to avoid cyclic swell variation.

Ultrasonic Thickness Metrology and Parison Programming Standards

Non-destructive wall thickness measurement on blow moulded containers is conducted with ultrasonic pulse-echo gauges operating at 5–20 MHz, with calibration against certified reference blocks per ISO 16809:2017. The gauges measure time-of-flight between the outer and inner wall echoes; for HDPE with a longitudinal sound velocity near 2400–2500 m/s, a 1.0 mm wall gives a transit time of approximately 0.80–0.83 µs. Beta backscatter gauges may be used on moving lines but require radiation safety interlocks and periodic verification of the isotopic source per local regulations. For compliance with dimensional control requirements under ISO 9001:2015 clause 7.1.5, measurement systems are subjected to Gage R&R studies where repeatability and reproducibility should account for less than 10% of the total tolerance. In high-volume closure and container manufacturing, wall thickness distribution is monitored at 12–32 points per container, with sampling frequencies of 1 in 30 to 1 in 100 cavities depending on criticality. The measured data are compared to setup limits derived from minimum burst strength testing according to ASTM F2057-17 for household containers, and top-load testing according to ISO 12048:2011. Correlation between wall thickness and top-load force is not linear; thin sidewall sections below 0.5 mm can reduce top-load by 30–50% compared with 0.8 mm sections because buckling initiates at the thinnest effective ring. Therefore, thickness distribution specifications are usually stated as minimum values at each defined measurement location rather than as a single average.

Two-stage reheat stretch blow moulding of polyethylene terephthalate bottles imposes wall thickness distribution requirements that originate in the injection-moulded preform and are only partially modified during stretch blow. The preform is injection moulded from PET resin with an intrinsic viscosity of 0.80–0.84 dL/g measured according to ASTM D4603-18, dried to a residual moisture content below 50 ppm at 160–180 °C, and then reheated to 90–115 °C in a multi-zone infrared oven with 6–10 heating zones. Axial stretch ratios of 2.5–3.0 and hoop stretch ratios of 3.0–4.5 give a planar stretch ratio of 8–12; the final sidewall thickness therefore falls from a preform wall of 3.0–4.5 mm to 0.25–0.40 mm for a 0.5 L carbonated soft-drink bottle. Stretch rod speed of 1.0–2.0 m/s, pre-blow pressure of 0.8–1.2 MPa, and final blow pressure of 3.5–4.0 MPa determine how the wall thickness distributes across the shoulder, body, and base. Too aggressive stretch rod movement causes excessive material transport to the base and leaves thin sidewall bands, while delayed pre-blow creates a thicker shoulder and thin panel region. Wall thickness distribution is inspected by sectioning bottles at 12–16 circumferential positions and measuring with a Hall-effect sensor or optical comparator with 0.005 mm resolution. Burst strength verification per ASTM F2057-17 and carbonation retention tests at 24 °C and 4.0 volumes CO₂ provide end-use correlation. Residual moisture above 50 ppm reduces intrinsic viscosity during melting by hydrolytic degradation, increasing the variation in stretch behaviour and resulting in wall thickness fluctuations of ±15% across a cavity. For these reasons, preform drying, infrared lamp output stability within ±2%, and ambient relative humidity below 30% are essential process boundaries that define the operational window for stable wall thickness distribution.