Experimental simulation of friction and wear of carbon yarns during the weaving process

https://doi.org/10.1016/j.compositesa.2015.07.024Get rights and content

Abstract

Nowadays many mechanical composite parts are reinforced by carbon weaving material. During the weaving process some damage to the carbon yarns appears that decreases the final mechanical properties of the part. This study focused on the friction interactions that occurred between warp yarns. A specific kinematic experiment has been done in order to simulate the weaving movement of these yarns, which induces friction. By varying experimental parameters, initial normal load, oscillation frequency and oscillation angle, it is possible to understand their influence on friction phenomena and fibre damage. The energies induced during friction have also been studied. The results obtained show that the coefficient of friction decreases with the increase of normal load, but is not influenced by the oscillation frequency. Yarn wear, i.e. filament breaks, appears beyond a certain normal load. The friction energy increases with normal load and oscillating angle but does not depend on oscillation frequency.

Introduction

Nowadays, carbon is often used in mechanical industries for reinforcement of composite materials. Most of the time, this carbon is in the form of yarns. These yarns, made of multifilaments, are woven. According to the application, the interlacement of the yarns, i.e. the kind of weave, can change. The main goal of these materials is to assure the required mechanical properties in specific directions whilst minimising the global mass of the mechanical part. The most important domains for use of these materials are transportation (automotive, airplane, rail) and space.

During the weaving of these carbon yarns, friction phenomena occur, breaking fibres and affecting the production process [1]. There is friction between yarns and metallic parts of the weaving machine but also friction between only yarns. This damage decreases the mechanical properties of these carbon woven fabrics, and are more particularly common in 3D-weaving [2], for interlocking woven fabrics. The final composite part then shows some defects [3]. Mechanical properties of yarns after each step of the weaving process can be evaluated [4]. Studies have been conducted for different types of yarns (carbon, polyester, glass, etc.) and for different weaving parameters. For instance, the effect of the tensioning step [5] or the shedding parameters [6], [7] have been studied in order to reduce the yarn damage during the weaving process.

This study concerns only the friction phenomena between yarns and more particularly between warp yarns (Fig. 1). Friction between warp and weft, which occurs during the weft insertion, is not considered in this paper, because it occurs once for each yarn portion during the process and is therefore not considered to be crucial.

Several studies have been done to quantify friction between fibres or yarns. Most of them are summarised by Hong and Jayaraman [8]. For these methods, the contact can be punctual or linear. In Howell’s method [9], [10] the contact is punctual between two fibres. The first fibre part is hung and the tension is given by a dead mass. The other fibre is taut on a horizontal bow with a lateral movement, which drags a displacement of the hung fibre. This displacement is linked to the static coefficient of friction (COF) between these fibres. The hanging fibre method [11], [12], [13] is close to the previous one. The horizontal fibre is fixed with a tension and the hung fibre has a vertical movement. The apparent load of the vertical fibre changes according to the coefficient of friction between these fibres obtained from the capstan equation. This method is also applied to estimate the coefficient of friction of yarns [14]. Mercer and Makinson [15] and Gralen and Olofsson [16] have measured friction between two fibres taut by two perpendicular bows. They studied the friction evolution relative to the sliding distance and the stick–slip.

Others methods have a linear contact. The fibre twist method [17] consists of twisting two yarns or fibres together. The force to extract a fibre is used to calculate the COF from the capstan equation. Another method used is based on the capstan method by covering a roller with the tested yarn or fibre [18]. The measurement can be done by measuring the force for moving the yarn against this roller. This method is used for the measurement of yarn to yarn friction for carbon by Cornelissen et al. [19], [20] and Chakladar et al. [21] by wrapping friction pulleys with carbon tows. The influence of the experimental conditions, and more particularly fibre orientation and tow size, is evaluated. These two methods (i.e. fibre twist and the capstan method) are used in ASTM D3412-01 [22].

However, no method proposes to study the friction between yarns with respect to the particular kinematic of warp yarns with an angular movement occurring during weaving process. The experiment developed in this study consists of reproducing this movement and analysing the influence of the experimental parameters: the oscillation angle, the oscillation frequency and the normal force between yarns.

Section snippets

Carbon yarn investigated

The study is carried out on 3 K carbon yarn (HTA40) manufactured by Toho Tenax Europe GmbH. That means that there are 3000 filaments of carbon per yarn section, whose diameter is 7 μm. The tensile modulus is 240 GPa and the tensile strength is 4100 MPa (datum from the producer).

Friction experiment

The objective is to simulate the weaving process in terms of friction stress. Friction measurements are performed by means of a NTR2 nanotribometer (CSM Instrument Company, Peseux, Switzerland). This device is originally a

Influence of the initial normal load

The influence of the normal load on the friction behaviour and wear phenomena are studied through the results obtained for the three values of initial normal loads: 200, 500 and 800 mN. The oscillation frequency is 3 Hz and the oscillation angle is 34°, i.e. ±17°.

It is interesting to note the evolution of the friction phenomena from the first cycle to the last cycle towards a quasi-constant cycle (Fig. 6). This trend is the same for all the friction tests. In the study only the average cycles are

Influence of the normal load

The obtained decrease of the coefficient of friction with the increase of the normal load is well known for some materials including textile materials. In fact for textile materials, the coefficient of friction (COF) follows the commonly used empirical law established by Bowden and Young [28]: the COF μ  Wn−1, where W is the normal force and n is a coefficient between 2/3 and 1 [29], [30], [31], [32], [33].

Moreover, the evolution of the normal force mean value during a friction cycle relative to

Conclusion

During the weaving of carbon yarn that is often used in composite materials, some damage appears that decreases the mechanical characteristic of the final part. The study allows a better understanding of the friction phenomena that occurs during the weaving process and more particularly, between warp yarns during shedding. A specific experiment has been developed in order to reproduce this yarn crossing kinematic. The friction phenomena occurring between carbon yarns have been studied.

The

Acknowledgment

The authors want to thank Toho Tenax Europe GmbH for providing carbon filament yarns.

References (33)

  • C. McHugh

    The use of recent developments in conventional weaving & shedding technology to create 3D one piece woven carbon preforms

    (2009)
  • J. Hong et al.

    Friction in textile

    Text Prog

    (2003)
  • H.G. Howell

    27 – Inter-fibre friction

    J Text Inst (Trans)

    (1951)
  • V. Subramaniam et al.

    Frictional properties of siro spun yarns

    Text Res J

    (1990)
  • Y.E. El Mogahzy et al.

    Friction in fibrous materials: Part II: experimental study of the effects of structural and morphological factors

    Text Res J

    (1993)
  • B.J. Briscoe et al.

    A study of the friction and adhesion of polyethylene-terephthalate monofilaments

    J Phys D Appl Phys

    (1979)
  • Cited by (37)

    • A mesoscopic model for inter-yarn friction

      2024, Composites Part A: Applied Science and Manufacturing
    • Modeling of yarn interactions for non-axisymmetric biaxial overbraiding simulations

      2023, Composites Part A: Applied Science and Manufacturing
    • Mechanics of textiles used as composite preforms: A review

      2023, Composite Structures
      Citation Excerpt :

      Degradation of mechanical properties for different fibers during the 3D warp interlock weaving has been reported in [50]. Up to now, many friction test methods have been proposed for characterizing the friction behavior of yarns or textile preforms, which can be divided into five categories: the capstan test [51], fiber twist test [52], reciprocating test [53–55], pull-out test [56,57] and hanging fiber test [58], as shown in Fig. 5. All the test methods have their practical advantages and drawbacks, which are summarized in Table 1.

    • Study on wear characteristics of carbon fiber at needle end in prefabricated composite weaving

      2021, Journal of Materials Research and Technology
      Citation Excerpt :

      Broken ends and surface hairiness of carbon fiber preforms during weaving will directly affect the mechanical properties of the composites [6,7]. In order to explore the damage mechanism of carbon fiber in the weaving process, domestic and foreign scholars have done a lot of research, the methods involved mainly include winch method [8,9], plate friction method [10–12], rotary reciprocating friction method [13,14], etc. Meng [15] obtained the conclusion that water-soluble dimethylsilicone oil can reduce the wear of carbon fiber bundle through quantitative analysis and modification study on the damage of carbon fiber bundle in three-dimensional orthogonal weaving preform yarn replacement process.

    View all citing articles on Scopus
    View full text