**4.1.5 Production of shell fabrics**

Shell fabrics are a special class of 3D woven fabrics since the structure of these fabrics may have only one layer or multiple layers; however the end product is always three dimensional. The importance of these types of fabrics is increasing since these types of woven fabrics are widely used in helmets, bra cups in fashion and clothing, female body armour and car door lining material (Chen&Tayyar, 2003).

Shell (a.k.a. doomed) fabrics can be produced with weaving, or cut and sew. Cut and sew technique has been the most commonly method used to produce shell fabrics but seams are a big disadvantage in technical applications, where the continuity of fibres is important. Seams definitely reduce the level of reinforcement and protection. Furthermore, cut and sew creates extra waste of materials and labour (Chen&Tayyar, 2003).

Fig. 15. An example of a shell fabric woven with conventional weaving (Chen, 2011) Fig. 15. An example of a shell fabric woven with conventional weaving (Busgen, 1999)

In conventional weaving, shell fabrics can be produced with using discrete take up and combination of different weaves. In order to produce a shell woven fabric, one can use a mixture of weaves with long and short floats. For instance, the plain weave, the tightest, is arranged in the middle, where a 2/2 twill is used in middle ring, and a five-end satin, with the longest average float length which is the loosest weave of the three types, is used for the outer ring. In a fabric with constant sett (the same warp and weft densities), the areas woven with plain weave tends to occupy a larger area and therefore will grow out of the fabric plane; the part of the fabric with the five-end satin tends to be squeezed, thus enhancing the domed effect. Consequently, the height difference between the lower and higher planes forms a dome. This method is a quick, easy, and economical way to produce fabrics that require relatively small domed effects. However, it appears that for fabrics requiring larger domed effects, the weave combination method is not sufficient (Chen&Tayyar, 2003).

Fig. 16. Weave combination to produce shell fabric (redrawn from Chen&Tayyar, 2003)

again beats up the second row and once again the heald frame is lowered to insert the

Fig. 17. An example of a uniaxial 3-D fabric and the modified conventional weaving loom

Uniaxial 3D variable shaped fabric can also be produced on conventional weaving machines (Mohamed&Zhang, 1992). In this system, two groups of weft yarns, Y1 and Y2 are used for weft insertion with one weft group (Y1) being inserted from one side for the flange and other weft yarn group (Y2) being inserted from the other side for the web portion of the inverted T cross-shape (Figure 18.b). Two selvage yarns, Sa and Sb, are required to hold the

 a. b. Fig. 18. Production of a uniaxial 3D variable shaped fabric on modified weaving machine

warp binder vertically (Greenwood, 1974).

(Greenwood, 1974)

(Mohamed&Zhang, 1992)

There is a balance between warp let-off and fabric take up of the weaving machine in normal weaving. Otherwise, there will be variations in the weft density which is an undesired situation. However, in weaving of shell fabrics, the variations of the weft density are required. But these variations have to be under control. In order to achieve a controlled variation in the weft density, the width of the loom is needed to be divided into several sections and each section is required to be maintained in its own balanced fabric take up and warp let off. This can be achieved by a method in which the fabric is taken forward at different rates across the sections (Chen&Tayyar, 2003). To produce the shell fabric given in Figure 16 as an example, the loom has to be divided into three parts and three warp beams are required controlled individually. By this way, the ratio of fabric take up and warp let off will be constant in each section.

A special fabric take up system which consists of many discs electronically controlled to perform individual take-up movement, besides individually controlled warp let off system on conventional weaving machines allow producing 3D shell woven fabrics (Busgen, 1999).

#### **4.1.6 Production of non-interlaced uniaxial orthogonal 3D fabrics**

Khokar (2002) named this kind of fabrics as NOOBed fabrics which is an acronym for Noninterlacing, Orthogonally Orientating and Binding. These fabrics are different in structure when compared to the classical weaving structures (Figure 17.a). There are three different yarns that are positioned in three coordinates (x, y, and z). However, the yarns are not interlaced with each other as in the conventional weaving. Orthogonal fabrics are divided into two groups as uniaxial and multiaxial.

A conventional weaving machine is modified to produce a fabric which has three dimensions of yarns (Figure 17). In this fabric, Ground warps GW are arranged in rows and columns and are positioned in the x direction. In y direction, weft yarns are positioned and these yarns are used to bind ground warps in the row direction. In z direction, extra binder warp threads are used which are supposed to bind ground warps in the column direction (Figure 17.a). Ground warp yarns pass over a roller RL and through two sets or columns of horizontal spacing or separating bars SBR (1-5) and SBL (1-5) arranged in pairs. They pass then between heald wires. The horizontal bars SBR, SBL are designed to open a warp gap between adjacent horizontal rows or layers of ground warps to facilitate the weft insertion. for each such horizontal row, two bars are required; one of which is under appropriate row of threads and serves to raise it SBR, and the other SBL above which serves to lower it. Binder warp threads are controlled by heald frame by raising or lowering the frame. The weaving is performed as follows; when the heald frame HF is in the lowered position, all the bars SBR and SBL are also in the lowered position except SBR1 and SBL1 which are in the raised position in order to form a warp gap for the weft insertion. Thus, the first weft is inserted in the fabric. After the first weft insertion, second bars of SBR and SBL join SBR1 and SBL1 in the raised position are raised and the second weft is inserted into the warp gap. This process is continued until the last weft thread of the first vertical row of weft threads has been inserted. Then reed beats up the inserted wefts by moving forward. The heald frame HF is moved up to insert the vertical binder warp. The lowest SBR5 and SBL5 bars are lowered to form a gap for the weft insertion. Pair by pair all the bars are lowered and after each lowering, the corresponding weft is inserted until the second vertical row of weft threads is completed. Then the reed

There is a balance between warp let-off and fabric take up of the weaving machine in normal weaving. Otherwise, there will be variations in the weft density which is an undesired situation. However, in weaving of shell fabrics, the variations of the weft density are required. But these variations have to be under control. In order to achieve a controlled variation in the weft density, the width of the loom is needed to be divided into several sections and each section is required to be maintained in its own balanced fabric take up and warp let off. This can be achieved by a method in which the fabric is taken forward at different rates across the sections (Chen&Tayyar, 2003). To produce the shell fabric given in Figure 16 as an example, the loom has to be divided into three parts and three warp beams are required controlled individually. By this way, the ratio of fabric take up and warp let off

A special fabric take up system which consists of many discs electronically controlled to perform individual take-up movement, besides individually controlled warp let off system on conventional weaving machines allow producing 3D shell woven fabrics (Busgen, 1999).

Khokar (2002) named this kind of fabrics as NOOBed fabrics which is an acronym for Noninterlacing, Orthogonally Orientating and Binding. These fabrics are different in structure when compared to the classical weaving structures (Figure 17.a). There are three different yarns that are positioned in three coordinates (x, y, and z). However, the yarns are not interlaced with each other as in the conventional weaving. Orthogonal fabrics are divided

A conventional weaving machine is modified to produce a fabric which has three dimensions of yarns (Figure 17). In this fabric, Ground warps GW are arranged in rows and columns and are positioned in the x direction. In y direction, weft yarns are positioned and these yarns are used to bind ground warps in the row direction. In z direction, extra binder warp threads are used which are supposed to bind ground warps in the column direction (Figure 17.a). Ground warp yarns pass over a roller RL and through two sets or columns of horizontal spacing or separating bars SBR (1-5) and SBL (1-5) arranged in pairs. They pass then between heald wires. The horizontal bars SBR, SBL are designed to open a warp gap between adjacent horizontal rows or layers of ground warps to facilitate the weft insertion. for each such horizontal row, two bars are required; one of which is under appropriate row of threads and serves to raise it SBR, and the other SBL above which serves to lower it. Binder warp threads are controlled by heald frame by raising or lowering the frame. The weaving is performed as follows; when the heald frame HF is in the lowered position, all the bars SBR and SBL are also in the lowered position except SBR1 and SBL1 which are in the raised position in order to form a warp gap for the weft insertion. Thus, the first weft is inserted in the fabric. After the first weft insertion, second bars of SBR and SBL join SBR1 and SBL1 in the raised position are raised and the second weft is inserted into the warp gap. This process is continued until the last weft thread of the first vertical row of weft threads has been inserted. Then reed beats up the inserted wefts by moving forward. The heald frame HF is moved up to insert the vertical binder warp. The lowest SBR5 and SBL5 bars are lowered to form a gap for the weft insertion. Pair by pair all the bars are lowered and after each lowering, the corresponding weft is inserted until the second vertical row of weft threads is completed. Then the reed

**4.1.6 Production of non-interlaced uniaxial orthogonal 3D fabrics** 

will be constant in each section.

into two groups as uniaxial and multiaxial.

again beats up the second row and once again the heald frame is lowered to insert the warp binder vertically (Greenwood, 1974).

Fig. 17. An example of a uniaxial 3-D fabric and the modified conventional weaving loom (Greenwood, 1974)

Uniaxial 3D variable shaped fabric can also be produced on conventional weaving machines (Mohamed&Zhang, 1992). In this system, two groups of weft yarns, Y1 and Y2 are used for weft insertion with one weft group (Y1) being inserted from one side for the flange and other weft yarn group (Y2) being inserted from the other side for the web portion of the inverted T cross-shape (Figure 18.b). Two selvage yarns, Sa and Sb, are required to hold the

Fig. 18. Production of a uniaxial 3D variable shaped fabric on modified weaving machine (Mohamed&Zhang, 1992)

Fig. 20. Special designed 3D weaving machine that produces a uniaxial 3-D fabric (Fukuta et

In order to pick in weft yarns X and vertical yarns Z into the horizontally and vertically aligned warp yarns, the weft inserting device 6 is first picked transversely or perpendicularly to the warp yarns while maintaining the upper and lower vertical yarn inserting devices 4 and 5 in the upper and lower retracted positions as shown in Figure 20. Each of the weft yarns X being inserted between the warp layers in double fold forming a loop at the fore end thereof. The weft inserting device 6 is temporarily stopped when the looped fore ends of the weft yarns are projected out of the warp yarns on the opposite side

As shown in Figure 21, base 10 supports movable upper and lower frames 12 and 13 with holes for supporting a plurality of filaments 15 that extends in the vertical (Z-axis) orientation. Identically working filament feed units 20 and 20' alternately insert yarns in the X- and Y-axes directions, respectively. First, filaments 21 from supply bobbins are woven through the spaced rows between filaments 15 along the X-axis by advancing the needles 22 by pushing rods 25. A pin 30 is inserted in the Y-axis direction to lie across the top of filaments 21 outside the last row of filaments 15 to tamp filaments 21 down. Needles 22 are then retracted from filaments 15, forming a tightly looped first course of Xaxis filaments that is restrained by pin 30. Similarly, the course of Y-axis filaments is woven next by advancing threaded needles 22', inserting pin 30' on top of filament 21' in the X-axis direction and retracting needles 22'. As the filament layers build up, pins 30 and 30' are removed. To increase the fabric's density, all the filament layers are compressed. The fabric integrity results primarily from inter-yarn friction (King, 1976 as

In Weinberg's special designed 3D weaving machine, it is possible to form sheds between layers of planar warp yarns, so that the orthogonal weft yarns can easily be inserted in any predetermined directions. Planar warp yarns are threaded through two parallel and perforated plates. The distance between these two plates is enough to accommodate the shedding and weft insertion. The top plate can slide on the warp yarns. The base plate is

al., 1974).

cited Khokar, 2002).

for threading a binder yarn P (Fukuta et al., 1974).

used to anchor the ends of the warp yarns (Weinberg, 1995).

fore end loops formed by the two groups of filling yarns, Y1 and Y2, respectively. Preferably, four harnesses, 11a, 11b, 12a, 12b, are used to control two sets of vertical Z yarns, Za-Zd. One of set of Z yarns, Za, Zb, is inserted for the flange portion of the inverted T shape fabric, and the other set of Z yarns, Zc, Zd, is inserted for the web portion of the inverted T cross-sectional shape fabric (Mohamed&Zhang, 1992).
