1.1.1. Bias Extension Test

The bias extension test (BET) is widely used because of its simplicity. A 45◦ oriented textile sample is fixed into a tensile testing device and loaded with tension [20–22]. Additionally, the BET can be performed as a biaxial tensile test with a modified sample geometry and clamping in the second direction [23]. In a standard clamping arrangement, there are different shear zones, Figure 1a. In zone A, pure and constant shear is induced, due to the free edges of the relevant yarns. In zone B, only half the shear of zone A is induced due to the partly clamped yarns. Finally, in zone C, no shear occurs due to the completely clamped yarns. During testing, the force/displacement diagram is recorded. However, the measured force reflects the whole effort of sample shearing. As this effort depends on the sample size, a normalized shear force was introduced [12]. Calculating the normalized shear force requires a comprehensive knowledge of engineering mechanics.

**Figure 1.** Principle of shear deformation in the bias extension test (**a**) and in the picture frame test (**b**) [1].

As mentioned above, the BET is widely used and well-investigated. Most of the research only considers woven fabrics as to their "pin-joint" arrangement. In several studies, a material-specific locking angle was determined, describing the shear angle when "locking" occurs. The locking phenomenon is observed when warp and weft yarns cannot rotate further without inducing non-negligible in-plane compressions on their neighboring yarns. As it is not possible to measure the moment of locking directly, it is derived from the shear force/shear angle diagram or by additional optical measurements. Theoretically, the shear force/shear angle diagram shows a steep rise at the moment of locking, so it can be derived quite easily. However, it was shown in [24] that folding was observed before locking. This additionally supports the distinction in the terminology used. In [25], it is shown that folding depends on textile properties, whole strain and stress distribution and on boundary conditions. Folding is mentioned as a multiscale problem. Thus, the determination of the locking angle does not seem to be a reliable criterion for deriving the critical shear angle needed for kinematic draping simulations.

In addition, there are a few studies also investigating multi- and uni-directional non-crimp textiles (NCF and UD-NCF, respectively) [26–29]. For NCFs, slippage was observed starting at a shear angle of 40◦ [27]. This is a major disadvantage of BET and leads to a minor deviation between the mathematically calculated shear angle and the optically measured shear angle. Consequently, the applicability of BET is questionable for NCFs, because of the missing pin joint arrangemen<sup>t</sup> and the spuriously made assumption regarding slippage. Nevertheless, such effects are also encountered in the practice of manufacturing FRP products and need to be considered.

Optical measurements of the out-of-plane deformation behavior during BET are rare. In [23], a multistep biaxial bias extension test was performed to measure the force required to unfold an already sheared woven fabric sample. At each step, the resulting deformation was 3D scanned. Therefore, the sample and its surroundings needed to be additionally prepared, as compared to the standard bias extension test. The scans were edited to achieve smooth evaluable surfaces and the dimensions of the fold were derived. This procedure was not aimed at determining a critical shear angle as needed for kinematic draping simulations, but it can be used for the comparison of materials. It is not a continuous measurement of folding, i.e., the beginning of folding cannot be determined. In addition, sample preparation, the test procedure, data preparation, and evaluation are time consuming and cost intensive.
