Knitted fabrics containing stainless steel fibre yarns have the potential to act as sensors. Andrea Ehrmann, researcher at the Niederrhein University of Applied Sciences, explains how
Figure 1
Yarns or filaments with a conductive coating have a constant electrical resistance whereas spun yarns with thin stainless steel fibres exhibit different behaviour. When they are used to knit fabrics with different structures, their resistance reacts to pressure, elongation and internal relaxation processes, enabling use of such fabrics as a variety of sensors.
Polyester (PES) staple fibre yarns with an additional fraction of stainless steel (INOX) fibres can be used as conductive yarns. INOX fibres and PES filaments are spun together, so conductive wires are distributed randomly in the final yarn. The INOX fibres are very thin; in the yarn, S-Shield, produced by Schoeller, Austria, which was used in the following experiments, their diameter is 8?m.
In a woven fabric containing this stainless steel staple fibre yarn, the conductivity could be used, for example, to heat the fabric by applying a small voltage, to transmit data or to measure signals. However, in knitted fabrics, the strong resistance changes due to external forces and internal relaxation processes inhibit these applications.
Figure 2
Figure 3
On the other hand, this problematic behaviour implies the chance to use these knitted fabrics as textile sensors (Fig 1), for example, as elongation sensors. In contrast to integrated electronic sensors, knitted sensors keep the textile haptics and do not cause skin irritation. However, depending on the desired application, their stability in different applications with e.g. various washing cycles and the influence of external forces has to be proven.
When a knitted fabric made of stainless steel fibre yarn is stretched, changes in electrical resistance depend strongly on its structure. While the resistance of a single face fabric decreases significantly for small elongations and remains nearly constant for high elongations, a double face fabric shows an almost linear decrease of resistance with increasing elongation (Fig 2). This behaviour can be explained by the different mechanical reactions to the external stress applied to the fabric.
A single face fabric reacts, even on small elongations, by a dimensional change of the stitches. Conversely, a double face fabric first “unfolds” its rib structure and the dimensional change of the stitches is much less pronounced for small elongations. Thus, the measured response of resistance to elongation is significantly smaller and only approaches the values of the single face fabric for very high elongations.
If these elongation sensors were used as a breathing sensor, a constant correlation between elongation and resistance is of utmost importance. Unfortunately, for most structures, the difference between breathing in and out, measured by a cyclic force on the knitted fabric, vanishes after several breathing periods.
However, a single face fabric with conductive horizontal stripes - alternating courses of conductive and non-conductive yarns - always shows a difference of about 20% between breathing in and breathing out. Even better values are found for the single face fleece fabric, in which the conductive yarn is representing the fleecy yarn and does not form stitches at all.
Apparently, decoupling yarn elongation from influences due to stitch formation is an effective approach. This study proves that differences between breathing in and breathing out greater than 100% are possible. This makes it possible to create a breathing-sensor shirt by applying stainless steel yarns as single face fleece yarns in the front. Such a shirt allows for breathing detection over several hours.
The electrical resistance of a weft knitted fabric made from yarn with thin stainless steel fibres changes not only with elongation, but also with a pressure perpendicular to the fabric surface. But most structures are not well suited to creating pressure sensors, since low pressures are not detected exactly and higher pressures can deform the fabric for a certain time.
However, weft knitted spacer fabrics are suitable as pressure sensors. With the front and back plane produced from conductive staple fibre yarn, such a spacer fabric with normally separated planes, coming into electrical contact only under pressure (Fig 3), can be produced on a flatbed knitting machine with a fixed needlebed distance if an adequate structure is used.
The three-dimensional structures in the fabric can be compressed with nearly no lateral shift of one conductive plane with respect to the other. The necessary pressure for the electrical contact depends on the structure as well as the material of the monofil used as spacer yarn.
As depicted in Figure 4, the electrical resistance values along wales and courses differ within each plane. A mechanical contact of both planes at a certain position leads to finite electrical resistance values between the conductive planes (yellow) which as well as creating an electrical contact between front and back plane, also change the resistance network within both planes. Such a spacer fabric could be used to detect the position of contact or dimension of the contacted area. In the simplest case, it works as a switch which may serve as car seat upholstery and as an indicator as to whether there is a front passenger on the seat or sets off an alarm when a baby leaves its play rug. More intelligent electronics could, for example, be used to measure the pressure distribution put on a mattress by a sleeping person.
In hand flat knitting machines, the take-down normally consists of a comb which is weighted, pulling the knitwear down with a constant force. During knitting, the fabric shrinks laterally if no wires or the like are used as spreaders. This deformation can also be seen – in a reduced form – after removing the fabric from the machine and even after longer relaxation times.
Figure 4
In order to quantitatively describe this behaviour, knitted fabrics from stainless steel staple fibre yarn produced on a hand flat knitting machine with comb take-down have been examined in a defined measuring raster (distances of 10 courses or wales, respectively). The conductivity is measured in the wale direction. The results are presented in the colour-coded graph (Figure 5).
Figure 5
The figure shows that resistance differs between the vertically neighbouring measuring positions at the right and left sides of the fabric. This asymmetry cannot be recognised visually in the fabric. It is probably due to uneven takeup in the width.
Moreover, resistance in the middle of the fabric is distinctly higher than at the upper and lower edges. The explanation can be found in the stitch geometry by imagining that the stitches at the edges become longer during the shrinking process, and that after taking the fabric out of the machine, they are correspondingly looser than stitches in the middle of the fabric. The loose stitches touch each other less strongly at the binding points.
Measuring resistance in several sample positions is an easy tool to examine the regularity of a knitted fabric and to adjust machine parameters properly afterwards, if necessary.
As well as changing with elongation, the electrical resistance of a conductive fabric also alters over time in a constantly stretched knitted fabric. Figure 6 shows a time-resolved measurement of this process, performed on a double face and a cardigan fabric with typical stitch lengths.
Figure 6
Both knitted fabrics have been preelongated by 20% for a duration of 30 minutes in order to create a defined initial state, and afterwards stretched to 35%.
The time-dependent resistance can be fitted by a logarithmic function. This behaviour is typical for systems in which statistical processes play an important role. The fitting parameters depend on the structure, stitch length, different numbers of washing cycles etc.
In this way, a simple resistance measurement can, for example, easily provide information about how fast a certain fabric wears out under stress.
Immediately after the fabrication process, a knitted fabric starts to relax. Since the dimensional changes are not very distinct (about 10% for a double face fabric) and occur during long term periods (weeks to months), it is not easy to detect these changes.
It is not possible to measure washing relaxation using the resistance of conductive knitted fabrics. In these cases the wet fabric is influenced by large superposed effects caused by the mechanical impact on the surface and especially on the fine stainless steel wires.
In contrast to this, the dry relaxation Figure 5: Position-dependent resistance of a weft knitted fabric, produced with comb take-down. Irregularities in the fabric are more distinctive than by pure optical examination Figure 6: Time-dependent resistance of two fabrics pre-elongated by 20% for 30 minutes and afterwards stretched to 35%. Different time dependencies are attributed to different relaxation processes under tensile tension process can be examined using electrical resistance.
First experiments have shown that best results can be achieved for relaxation on a smooth surface. In other relaxation situations the measurement of a few peaks dominates the results and could lead to inaccurate measurements. This method allows for a more precise examination of the dry relaxation, leading to a better understanding and description of this process.
While knitted fabrics from stainless steel fibre yarns are inappropriate for applications which require a constant resistance, they can be used as pressure or elongation sensors, for examinations of machine parameters and to visualise internal processes, like relaxation. In this way, the examined conductive knitted fabrics offer new possibilities in the field of textile sensory as well as in basic research.
This project has been partly supported by the internal project funding of Niederrhein University of Applied Sciences. Co-researchers were Frank Heimlich, Andrea Brücken and Marcus O Weber FTB (Research Institute for Textile and Clothing), Niederrhein University of Applied Sciences.