Papadakis, Nikolaos (2002) Strain rate dependency of the properties of a unidirectional thermoplastic composite material. PhD thesis, University of Warwick.

Preview

PDF WRAP_Theses_Papadakis_2002.pdf - Unspecified Version - Requires a PDF viewer. Download (20Mb) | Preview

Request Changes to record.

Abstract

This research work established the strain rate dependency of the the mechanical properties of a highly orientated glass fibre/thermoplastic composite lamina and validated a model for Computer Aided Engi­neering (CAE). The mechanical properties examined for strain rate dependency were elasticity, strength and damage evolution at a ply level.
A rigourous statistical methodology were established and implemented through mechanical testing to­gether with processing of the results for the development of semi-empirical strain rate models.
Two different methods of data acquisition were considered, specifically strain measurement using videoex- tensometry and contacting extensometry. The resulting strain measurements were then computed. Video extensometry appeared to have clear advantages, however, scatter in the response was appreciably higher compared to the contacting extensometry. This was due to the much smaller scale of gauge length for strain measurement.
A rigourous validation methodology was further complemented through a statistical procedure and tool kit (utilising statistical tools and procedures like density distributions plot, hypothesis testing, analysis of variance). The statistical tool kit was developed to enable objective assessment of strain rate dependency and to establish the quality of a relationship (model) should one exist for the range of mechanical properties tested. Using this validation methodology, a semi empirical strain rate dependent model was validated for elasticity strength and damage evolution.
The effect of strain rate on the above mechanical properties was investigated for Plytron1 Ai. The Plytron™material was supplied by Borealis as a lOOmm-wide, 0.22[mm]-deep tape, comprising aligned continuous glass fibres in a polypropylene matrix. To manufacture a laminate, the tape was laid-up ply-by-ply into an unconsolidated stack. This stack was then consolidated using under pressure and heat according to a Warwick Manufacturing Group’s proprietary membrane-forming process [!]. For the purposes of this study, specimens were machined in accordance with ISO-527-4 from 4 different layup sequences: [0°]4, [±45°]2,, [+45°]s and [±67.5°]2„. The specimens were tested at 5, 50 and 500[min/min] crosshead displacement rates using monotonic and cyclic loading.
FYom this investigation, over the examined strain rate range, the longitudinal tensile modulus increased with strain rate, while the shear modulus and Poisson’s ratio decreased. The transverse tensile modulus did not exhibit any statistically significant difference. The shear failure stress and the longitudinal tensile failure strain and stress appeared to increase for increasing strain rate, while the shear failure strain were not strain rate dependent. The transverse tensile failure stress and strain did not exhibit any statistically significant strain rate dependency.
The characterisation parameters of the damage evolution were based on the global composite ply model for composites in the framework of continuum damage mechanics (CDM). This model was developed by Ladeveze et al. (.’), [It], for thermosetting composites. It was established that shear damage evolution of the thermoplastic materials exhibits different behaviour compared to thermosets. It also was established that the rate of shear damage evolution decreases with increasing strain rate and that the point that shear damage initiates increases with increasing strain rate.
All testing was conducted with INSTRON 4505 universal testing machine instrumented with a 100[kN] load cell. Contacting extensometry and videoextensometry has been examined as data acquisition meth­ods. It was established in this work that contacting extensometry provided data with less scatter, however the videoextensometry exhibited significant advantages.
Having established and validated semi-empirical rate dependent models, for the characterisation parame­ters to service the CDM models, CAE models were established and validated using a well known explicit FE numerical simulation. To maintain rigour, the validation methodology employed new metrics to en­able objective comparison between FE and experimental results. These metrics are Pearson correlation coefficient and correlation range ratio. The comparison of experimental to FEM results revealed that the available models predict adequately well the stiffness of laminates as expected.
The onset of failure is predicted at significantly lower strains compared to the experimental results (depended on layup - usually 30% of the total failure strain). The premature failure is attributed to the failure criterion implementation at ply level and/or the definition of the boundary conditions.

Item Type:	Thesis (PhD)
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Library of Congress Subject Headings (LCSH):	Composite materials, Strains and stresses, Thermoplastic composites, Composite materials -- Computer simulation
Official Date:	October 2002
Dates:	Date Event October 2002 Submitted
Institution:	University of Warwick
Theses Department:	Department of Engineering
Thesis Type:	PhD
Publication Status:	Unpublished
Supervisor(s)/Advisor:	Wood, Paul , Pharaoh, Mark
Extent:	2 volumes (li, 281, 204 leaves) : illustrations
Language:	eng

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads