Process induced distortions of composite structures

Traiforos, Neoklis (2023) Process induced distortions of composite structures. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (48MB) | Preview


A significant problem encountered during the manufacturing process of thermoset composite structures is the distortion of their shape from their Computer-Aided Design (CAD) nominal geometry.

In this research the effect of stacking sequence, specimen thickness, tooling material and curing cycle on Process Induced Distortion (PID) of L-shape composite structures is investigated experimentally and numerically. The specimens were manufactured from high temperature Vacuum Assisted Resin Transfer Moulding (VARTM) process with the EPIKOTETM System 600 and IMS65 fibres. The spring-in angle of the parts was measured with a 3D scanner and a Coordinate Measuring Machine (CMM).

A thermo-chemo-mechanical simulation approach was developed to simulate the experiment, which is composed of a thermo-chemical and a chemo-mechanical part. The purpose of the thermo-chemical part, is to calculate at every time increment the temperature distribution across the structure. The calculated temperature field can then be used as input to the chemo-mechanical module, which purpose is to calculate the stress state of the material at every time increment of the curing history. Two chemo-mechanical modules were developed. The first one employs a modified Cure Hardening Instantaneously Linear Elastic (CHILE) material model, and the second one a linear viscoelastic material model. The material modelling developed is applicable from the gelation point of the resin, and is implemented using the ABAQUS Finite Element Analysis (FEA) software. The effect of various Boundary Conditions (BCs) on simulation results were investigated. Simulation results were compared against analytical results and the measured spring-in angles.

The spring-in angle of the parts predicted by the CHILE material model was found to be higher than that predicted by the viscoelastic model. The tool material significantly affects the distortion of the parts and contact boundary conditions should be employed for an accurate prediction of the part shape. Balanced, symmetric and thick laminates contribute to a reduction in part distortion, while elevated curing temperatures slightly affect the distortion level of the parts.

The chemo-mechanical modules developed were also applied to predict PID of a composite test frame from the door surround structure of A350 aircraft. The material systems and processes used to manufacture this part are the same as those of the L-shape structures studied. The 3D scanned geometry of the frame was compared against simulation results from the two chemo-mechanical modules (CHILE and

viscoelastic material model).

It was found that manufacturing the product without any compensation will produce a part not meeting its manufacturing specifications. Furthermore, a tool geometry compensation approach based on the shape of the part after demolding will produce inaccurate results as the trimming operations release stresses and affect the distortion field of the structure. Finally, scripts were developed to automate the mirroring process of calculated distortions.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Turner, Thomas
Matveev, Mikhail
Keywords: Process induced distortion; Thermoset composite materials; Finite element analysis (FEA); Process simulation; Dimensional control Aerospace structures
Subjects: T Technology > TP Chemical technology > TP1080 Polymers and polymer manufacture
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 73913
Depositing User: Traiforos, Neoklis
Date Deposited: 27 Jul 2023 07:34
Last Modified: 27 Jul 2023 07:34

Actions (Archive Staff Only)

Edit View Edit View