PET sintering shrinkage behavior testing is an experimental procedure used to quantify the dimensional changes that polyethylene terephthalate (PET) powders undergo during thermal or laser sintering. By systematically measuring linear, volumetric, and anisotropic shrinkage under controlled process parameters, this test reveals the shrinkage rules required to predict final part dimensions, compensate CAD models, and stabilize product quality. This article explains the sample preparation, testing items, measurement methods, typical shrinkage patterns, and practical applications of PET sintering shrinkage characterization.
Technical Overview
Polyethylene terephthalate (PET) is a semicrystalline engineering thermoplastic with a glass transition temperature of approximately 76 to 80 degrees Celsius and a melting point of approximately 250 to 260 degrees Celsius. When PET powder is processed by laser sintering, powder-bed fusion, hot compaction, or conventional thermal sintering, individual particles soften, form necks, coalesce, and progressively eliminate internal porosity. This densification, combined with crystallization and thermal contraction during cooling, produces a systematic and measurable reduction in dimensions known as sintering shrinkage.
PET sintering shrinkage behavior testing is designed to quantify the relationship between process parameters and dimensional change. Unlike injection molding shrinkage, which is largely governed by cavity pressure and cooling, sintering shrinkage depends strongly on energy input, particle packing density, layer-by-layer construction, and thermal history. As a result, shrinkage in sintered PET is typically anisotropic, with the build direction often shrinking more than the in-plane directions.
The primary objectives of the test are to establish quantitative shrinkage rules, identify stable processing windows, derive dimensional compensation factors, and provide input data for warpage simulation and quality control. A well-executed shrinkage study converts an apparently random dimensional error into a predictable, correctable process characteristic.
Testing Samples
Powder Materials
- Virgin PET laser-sintering powder with a controlled particle size distribution, typically 20 to 100 micrometers.
- Recycled PET powder derived from post-consumer bottles or industrial feedstock, evaluated for batch consistency.
- Filled or modified PET powders, including glass bead filled, mineral filled, and nucleated grades that reduce or stabilize shrinkage.
- PET composite powders blended with flow agents to improve bed spreadability and packing uniformity.
Specimen Geometries
- Cubic specimens of 20 or 30 millimeters for measuring linear shrinkage in three orthogonal directions.
- Rectangular bars, for example 80 by 10 by 4 millimeters, for evaluating warpage and bowing.
- Thin-walled plates and stepped features for assessing wall-thickness sensitivity and fine-feature fidelity.
- Gauge blocks with internal and external features for verifying scale factors and fit tolerances.
Preparation and Conditioning
- Vacuum drying of the powder to a moisture content below 0.02 percent, preventing hydrolytic degradation during sintering.
- Building of specimens in three orientations, namely flat, on-edge, and upright, to expose anisotropic effects.
- At least five replicate specimens for every parameter combination to support statistical evaluation.
- Conditioning of finished specimens at 23 plus or minus 2 degrees Celsius and 50 plus or minus 5 percent relative humidity for at least 24 hours before measurement, following ISO 291.
Testing Items
The test program covers dimensional, process-dependent, and structural items that together describe the complete shrinkage behavior of the material.
Dimensional Shrinkage Items
- Linear shrinkage in the X, Y, and Z directions relative to the nominal CAD geometry.
- Volumetric shrinkage calculated from the three orthogonal linear values.
- Shrinkage anisotropy ratio, defined as the ratio of build-direction shrinkage to in-plane shrinkage.
- Local shrinkage at thin walls, edges, holes, and small features where dimensional deviation is often amplified.
Process-Dependent Shrinkage Items
- Shrinkage as a function of laser energy density, combining laser power, scan speed, hatch spacing, and layer thickness.
- Shrinkage as a function of powder-bed or furnace temperature and dwell time in isothermal sintering studies.
- Shrinkage variation with cooling rate and part-build height, which reflect thermal history effects.
- Shrinkage drift over repeated powder reuse cycles for recycled feedstock.
Structure and Quality Items
- Relative density and residual porosity, which correlate directly with the degree of densification.
- Warpage, curling, and flatness deviation of thin sections.
- Crystallinity determined by differential scanning calorimetry, since crystallization contributes additional contraction.
- Repeatability indicators, including standard deviation and process capability indices for each shrinkage value.
Testing Methods
Direct Dimensional Measurement
The core method compares the nominal dimension L0 from the CAD model with the measured dimension L of the sintered specimen. Linear shrinkage is calculated as S equals L0 minus L, divided by L0, multiplied by 100 percent. Measurements are taken at three to five positions per direction using calibrated digital calipers with 0.01 millimeter resolution, a coordinate measuring machine, or optical and laser 3D scanning for full-surface deviation mapping.
Thermomechanical Analysis
Thermomechanical analysis or dilatometry is applied to powder compacts or miniaturized specimens to record a continuous shrinkage-versus-temperature curve. This identifies the onset temperature of sintering, the maximum shrinkage rate region, and the plateau where densification saturates. Differential scanning calorimetry following ISO 11357 is performed in parallel to track melting, crystallization, and the crystallinity level associated with each shrinkage stage.
Density and Porosity Evaluation
Bulk density is measured by the Archimedes method according to ISO 1183, and geometric density is calculated from mass and measured dimensions. Comparing the measured density with the fully dense PET reference density yields relative density and residual porosity, providing an internal check on the densification interpretation of the shrinkage data.
Parameter Matrix Testing and Statistical Modeling
A designed experiment varies the governing parameters systematically. In laser sintering, energy density is calculated as laser power divided by the product of scan speed and hatch spacing, and it is varied across at least five levels while layer thickness and bed temperature are held constant. In furnace sintering, a temperature-time matrix is applied, for example sintering between 200 and 260 degrees Celsius for dwell times from 5 to 60 minutes. Regression analysis is then used to fit the shrinkage data, commonly yielding power-law or logarithmic relationships that constitute the empirical shrinkage rules for the material.
Microscopic Verification
Scanning electron microscopy of specimen fracture surfaces confirms the physical stages behind the measured shrinkage, including neck growth between particles, progressive pore closure, and the transition from porous packing to a dense sintered body. This step links dimensional data to microstructural mechanisms and validates the model interpretation.
Typical Shrinkage Rules and Data Interpretation
Although exact values depend on the powder grade and equipment, several consistent rules emerge from PET sintering shrinkage studies.
- Shrinkage increases with energy density, rising sharply at first and then approaching a saturation plateau once particle coalescence and pore elimination are essentially complete.
- Build-direction shrinkage in layer-wise laser sintering is generally larger than in-plane shrinkage because the loose powder bed packs at only about 50 to 60 percent density and each new layer partially re-melts the previous one.
- In isothermal furnace sintering, linear shrinkage follows classic sintering kinetics, increasing approximately with time raised to a fractional exponent and accelerating strongly as temperature approaches the melting point.
- Higher crystallinity produced by slower cooling or elevated bed temperature adds measurable contraction and must be separated from pure densification shrinkage.
- Excess moisture causes hydrolytic degradation, lowers melt viscosity, and produces irregular, excessive shrinkage together with embrittlement and yellowing, which is why strict drying is mandatory.
- Compensation factors are derived directly from the rules, for example enlarging the CAD model in each axis by the factor 1 divided by 1 minus the fractional shrinkage of that axis.
Applications
- CAD model scaling and per-axis dimensional compensation for PET parts produced by laser sintering and powder-bed fusion.
- Definition of stable process windows for laser power, scan speed, hatch spacing, and bed temperature in additive manufacturing.
- Process control for PET powder coating sintering and rotational molding, where shrinkage determines coating continuity and wall uniformity.
- Material development and grade comparison for virgin, recycled, and filled PET powders.
- Quality assurance and incoming inspection that verifies powder lots against established shrinkage rules before production release.
- Calibration input for warpage and residual stress simulation models used in tooling and part design.
- Recycling programs that track shrinkage drift over repeated reuse cycles to set safe powder refresh ratios.
FAQ
What is the typical linear shrinkage of sintered PET?
Indicative values for unfilled PET in laser sintering fall roughly between 1.5 and 4 percent in the in-plane directions and between 3 and 6 percent in the build direction. Loose-pack furnace sintering can produce linear shrinkage of 10 percent or more because the powder bed starts at much lower packing density. Exact figures must always be established for the specific powder, geometry, and process parameters.
Why is PET shrinkage anisotropic in laser sintering?
Anisotropy arises from three combined effects: the powder bed packs less densely than a solidified layer, each new layer applies a thermal cycle that further consolidates underlying layers, and heat dissipates preferentially downward through previously sintered material. The result is systematically higher shrinkage along the build direction compared with the scan plane.
Which standards are referenced in PET sintering shrinkage testing?
There is no single standard dedicated exclusively to polymer sintering shrinkage, so testing is built from recognized frameworks. These include ISO 291 for specimen conditioning, ISO 1183 for density, ISO 11357 for DSC crystallinity analysis, ISO/ASTM 52900 terminology and ISO/ASTM 52902 test artifacts for additive manufacturing dimensional evaluation, with ASTM D955 and ISO 294-4 serving as methodological analogies adapted from mold shrinkage testing.
How is measured shrinkage compensated in production?
The established shrinkage rule for each axis is inverted into a scale factor applied to the CAD geometry. Because shrinkage can differ among axes and feature types, compensation is often iterative: a first calibration build is measured, scale factors are updated, and a verification build confirms that dimensions fall within tolerance before series production begins.
How does moisture affect the test results?
PET is hygroscopic, and residual moisture triggers hydrolysis at sintering temperatures. This degrades molecular weight, reduces melt viscosity, and generates erratic shrinkage, bubbles, and weak parts. Powder must therefore be dried to below about 0.02 percent moisture and handled in a controlled atmosphere to obtain reproducible shrinkage data.
How does sintering shrinkage differ from injection molding shrinkage?
Molding shrinkage occurs in a fully molten material inside a rigid cavity and is dominated by pressure and crystallization effects. Sintering shrinkage originates from powder densification and particle coalescence, depends on energy input and packing density, and is inherently more anisotropic in layer-wise processes. The two phenomena therefore require different measurement procedures and compensation strategies.
