Influence of weave structure on delamination when milling CFRP
Introduction
The use of components made of carbon-fiber-reinforced plastic (CFRP) has risen sharply in recent years. Studies assume that this trend will continue in the future (Roberts, 2011). Lässig (2012) reported that in the automotive sector, in addition to being used for weight-reduction measures, CFRP with a woven fabric is also being used for visual effects in decorative elements. In this case, the visible surfaces of the components must meet very high standards (Schubel et al., 2006).
In many instances, the components are manufactured using a resin transfer molding (RTM) process or an autoclave process. In both cases, it is commonly necessary for the components to undergo further processing to provide the final shape (Sheikh-Ahmad, 2009). This is frequently done using machining processes such as milling or drilling. As reported by Yashiro et al. (2013), the non-homogeneous material structure and the low temperature resistance of the matrix present considerable challenges for the machining processes in terms of achievable edge quality.
When milling CFRP with a weave structure, the undulation of the fiber rovings results in a heterogeneous cutting behavior at the cutting edge. Depending on the position of the milled edge relative to the undulation, this will either promote or prevent edge damage in the form of delamination. This influence is the subject of the following study.
Most published studies involving milling of CFRP are focused on non-crimp fabrics. According to Hintze and Hartmann (2013), delamination in the top layers occurs during the cutting action of the milling tool, which bends the fibers outward and deflects them away from the plane of the laminate. This induces stress between the layers of the laminate, causing the layers to separate. Colligan and Ramulu (1991) have classified the resulting delamination in various types. Type I delamination describes a surface break-out of fiber bundles at the trimmed edge that extends in the plane of the top layer. Type II, by contrast, is characterized by fibers that protrude beyond the trimmed edge without causing perceptible damage on the surface. While the type II fiber protrusions can easily be rectified, correcting type I break-outs is critical (König and Graß, 1989). A mixed form of types I and II involving both mechanisms is frequently encountered in which the fiber bundles protrude beyond the trimmed edge accompanied by significant surface damage in the material. Which of the respective types occurs depends to a considerable degree on the fiber orientation in the top layer (Puw and Hocheng, 1993).
According to Hintze et al. (2011), in full-groove cuts, the delamination is induced at the point of the initial contact between the fiber and the cutter, where the first cutting of the fiber occurs. It therefore follows that the maximum length of the fiber protrusion is the distance between the initial contact and the milled edge in the longitudinal direction of the fiber.
Another significant factor for the quality of the workpiece edge is the wear condition of the tool. Ferreira et al. (1999) has demonstrated that the feed force increases significantly with increasing wear. Karpat et al. (2012) found that the increased feed force is accompanied by a greater degree of delamination. Due to the high abrasiveness of the fibers, Teti (2002) and Karpat et al. (2012) have suggested using polycrystalline diamond (PCD) as a cutting material.
The edge quality is also significantly influenced by the parameters of feed per tooth and cutting speed as well as by tool geometry. Colligan and Ramulu (1991) found that the surface roughness and delamination are negatively affected by increasing feed per tooth. According to Davim and Reis (2005) as well as Puw and Hocheng (1993), an increase of the cutting speed can result in greater degrees of delamination and surface roughness. When it comes to the tool, a helix angle unequal 0∘ produces an axial force perpendicular to the laminate surface (Kalla et al., 2010). According to Karpat and Polat (2013), this force produces a significantly better machining result at the edge where the axial force is oriented inward. But on the opposite side, an increase in delamination occurs, as reported by Hosokawa et al. (2014).
Only a few studies on machining of CFRP have used a woven fabric. König and Graß (1989) studied the edge quality in drilling of plain and satin weaves. They observed a dispersion of the delamination depending on the random position of the hole relative to the weave. The delamination extended at most to the closest intersecting point of the yarns since the crossing fibers prevent the damage from propagating further. Faraz (2011) came to the same conclusion when studying delamination on the exit-ply when drilling. He also observed an elliptical spread of damage in the direction of the fiber. No previous study has focused on the influence of the woven structure on the occurrence of delamination in milling. In a variance analysis, Davim and Reis (2005) studied the influence of various process parameters on different quality variables in the milling of CFRP with twill weave. Azmi et al. (2013) recognized feed rate, depth of cut, number of teeth and cutting speed as main influencing parameters on delamination when milling GFRP with plain weave using unworn tools. Both studies do not consider any influence by the position of the milled edge relative to the yarn undulation.
Textile technology differentiates weaves based on their float. This refers to the number of crossing yarns that lie under or over an unbound length of yarn. A float of one results in a plain weave. When examining the unit cell of a plain weave, Naik and Shembekar (1992) as well as Tang and Whitcomb (2003) derived further parameters for characterizing the weave (Fig. 1).
A basic distinction is drawn between warp yarn and fill yarn; in a plain weave with uniform yarns, there are no differences in the individual properties. Starting from the surface of the workpiece, the first element is a thin resin layer with a thickness of tm. The thickness varies over the path of the yarn. The yarn follows with a thickness of . The wavelength occurs in the laminate plane, after which the fiber path repeats itself. A half wavelength is composed of the flattened region at the uppermost point and the undulation region . If the laminate is observed from above, the flattened region, the upper undulation region, and part of the lower undulation region are visible. The two undulation regions are separated by the turning point of the yarn, which is situated in the middle between the likewise visible weave interstices , which represent local agglomerations of resin. According to Huang (2000), the undulation path of warp and fill yarns in a plain weave can be approximately described with a sine function.
Section snippets
Experimental setup and materials
In order to study the influence of the woven structure on the edge quality full-groove cuts were milled into CFRP with a plain weave fabric. The milling path was oriented at an angle of 1∘ relative to the fill yarn. This was used to study the influence of the undulation at discrete distances xd at fiber orientation angles ψ of approximately 0∘ and 90∘, respectively (Fig. 2). The slight deviation of the fiber orientation angle of 0∘ and 90∘, respectively, exerts no significant influence on the
Influence of the trimmed edge position relative to yarn undulation on delamination
With a fiber orientation angle of ψ = 90∘, significant edge damage in the form of delamination is produced in some areas of the trimmed edges. At ψ = 0∘, however, no damage or only slight damage is observed. This coincides with the conclusions of Hintze et al. (2011), who identified a critical cutting orientation in the use of non-crimp fabrics at ψ = 90∘, whereas at ψ = 0∘, no damage occurs. For this reason, only the edge quality at ψ = 90∘ in the top layer is studied in the following. Furthermore, the
Conclusions
This experimental study investigates the influence of the trimmed edge position relative to the weave on delamination when milling CFRP with a woven fabric. On the basis of the study, the following conclusions have been drawn for a constant fiber orientation angle of ψ = 90∘:
- •
Unlike non-crimp fabrics, the undulation of the woven yarn, which results in a varying thickness of the top resin layer, leads to delamination of different types and magnitudes. The types can be subdivided into fiber
Acknowledgments
The authors would like to thank ‘Daimler AG’ and in particular Dr. Wilfrid Polley, senior manager process development machining technologies, and Dr. Waldemar Zielasko, manager process development machining with defined edge, for their valuable input and for providing tools, test material and measuring equipment.
References (22)
- et al.
Damage and dimensional precision on milling carbon fiber-reinforced plastics using design experiments
J. Mater. Process. Technol.
(2005) - et al.
Machining optimisation in carbon fibre reinforced composite materials
J. Mater. Process. Technol.
(1999) - et al.
Modeling of delamination during milling of unidirectional CFRP
Proc. CIRP
(2013) - et al.
Occurrence and propagation of delamination during the machining of carbon fibre reinforced plastics (CFRPs) – an experimental study
Compos. Sci. Technol.
(2011) - et al.
High-quality machining of CFRP with high helix end mill
CIRP Ann. Manuf. Technol.
(2014) The mechanical properties of composites reinforced with woven and braided fabrics
Compos. Sci. Technol.
(2000)- et al.
Prediction of cutting forces in helical end milling fiber reinforced polymers
Int. J. Mach. Tools Manuf.
(2010) - et al.
Mechanistic force modeling for milling of unidirectional carbon fiber reinforced polymer laminates
Int. J. Mach. Tools Manuf.
(2012) - et al.
Mechanistic force modeling for milling of carbon fiber reinforced polymers with double helix tools
CIRP Ann. Manuf. Technol.
(2013) - et al.
Quality definition and assessment in drilling of fibre reinforced thermosets
CIRP Ann. Manuf. Technol.
(1989)