当前位置：文档库 › Resistive fibre-meshed transducers

Resistive fibre-meshed transducers

Resistive Fibre-Meshed Transducers

R.Wijesiriwardana, T.Dias, S.Mukhopadhyay

William Lee Innovation Centre, UMIST

P.O Box 88 Sackville Street, Manchester, UK M60 1QD

R.Wijesiriwardana@https://www.wendangku.net/doc/fc1590699.html,,T.Dias@https://www.wendangku.net/doc/fc1590699.html,, S.Mukhopadhyay@https://www.wendangku.net/doc/fc1590699.html,

Abstract

The paper demonstrates the preliminary research carried out on fibre meshed transducers for wearable computing applications, their constructions using electro- conductive fibres via the route of modern electronic flatbed knitting technology. The paper reports the construction of resistive strain and displacement transducers using electroconductive polymeric fibres (resistivity of 104?cm-1) and metallic fibres (resistivity of 10?cm-1) and the modelling of electrical equivalent circuits of the fibre meshed transducers. The models are discussed in order to demonstrate the variation of electrical parameters under static and dynamic planar loads. Two methods of transducer construction are also discussed; in the first method the electroconductive fibres are stretched out in the base fibre meshed structure, and

in the second method the electroconductive fibres are intermeshed as an integral section of the base fibre meshed structure.

1.Introduction

F ibre-meshed transducers are emerging as a new generation of flexible transducer family due to the requirements in the field of wearable electronics. These transducers are mainly to be used in real time personalized information gathering systems [9]. The traditional transducers used in these systems have limitations including discomfort, bulkiness, unbreathability, low reliability and data integrity due to poor contact with the body, high power requirement, unwashable and high cost

of production. Currently there are several approaches used for their construction; sewing electroconductive fibres on

to a textile fabric in a secondary stage [5],[7], application

of piezoresistive coatings [9] and stretchable electroconductive stripes [8]. However, above transducers have inherent disadvantages such as low dynamic range, poor data integrity, high production cost and manufacturing time, inferior performance after washing, poor repeatability and aging effects. Moreover they are mostly used as switches. Due to above reasons we decided on a different approach. Our concept is based on constructing fabric transducers by integrating electroconductive materials in to the structure of a fibrous structure during its manufacturing process, which is based on the electronic flat-bed knitting technology.

At present major part of the research in this area is focussed on electroconductive woven structures. These structures are used as touch switches [5], proximity sensors [7] and flexible biopotential electrodes [22], surface electrodes for antennae [6] and circuit boards [5],[7],[9]. The literature also demonstrates that the fabrication of transducers, especially strain and displacement transducers has not been fully exploited. A recent publication reports the advantage of developing strain gauges from knitted structures [8]; however these strain gauges are not fully optimized for their performance due to the incomplete knowledge on the electrical behaviour of the electroconductive fibre meshed (knitted) structures under relaxed, static and dynamic loading conditions. On the other hand the positioning of these transducers on a garment is also not optimized, and this has resulted on poor performance and limited applications. The above reasons motivated us to carryout a programme of research on a comprehensive study on fibre meshed transducers. The main aim of the research is to create the knowledge base on the electrical characteristics of fibre meshed structures made from electroconductive fibres and to evaluate the feasibility of the application of the technology for constructing unobtrusive transducers (Knitted Fibre Transducers, KFTs). Our current objectives are to characterise the electrical properties of KFTs using electrical models (DC & AC) based on analytical & empirical studies, which would be utilised to study the behaviour of KFTs under mechanical loading conditions. Our present research methodology is based on the use of mesoscophic scale [14] mechanical and geometrical models of fibre meshed structures to develop the necessary electrical models.

The KFTs demonstrated in this paper can be used to monitor important bodily information such as respiratory patterns, motion and gestures. Arrays of KFTs can be employed to identify the context of the body using gait analysis techniques. We intend to discuss the applications and the performance of KFTs in garments in future publications. We also wish to highlight the fact that our current work is limited to polymeric electroconductive fibres and metal wires. The procedure of constructing the electrical equivalent circuits discussed in the following chapters can be used to model the other types (woven, etc) of electroconductive fibre assemblies as well.

2. Significance of knitting technology

Our ultimate objective is to develop unobtrusive transducers that can primarily be used in wearable real time personalized monitoring systems and in noninvasive patient monitoring systems. As such these need to be integrated into the undergarments so that the required information can be obtained without causing any discomfort to the wearer/patient. I n addition, such garments must also be washable. It is a well accepted fact among textile engineers that the knitted structure is the most suitable fibrous structure for the manufacture of undergarments due to:

x good tensile recovery properties;

x superior drapability, which would provide excellent skin contact;

x breathability: the structures are air permeable hence better comfort.

I n view of the above it was decided to develop strain gauges and displacement transducers using knitting technology, which consists of several different production routes; circular weft knitting, warp knitting and flat knitting. One technology in particular, i.e. the modern electronic flat-bed technology, has made quantum leaps during the last few decades, thanks to the extensive use of mechatronics for all mechanical movements and CAD/CAM systems. The most important features of the technology are:

x the precision positioning of fibers in 3D space: very useful in transducer construction and

placing them in appropriate positions;

x the freedom of constructing structures with different pattern elements: useful for the

optimization of transducers for maximum

sensitivity and performance;

x the ability to create 3D structures: adds an additional dimension to the transducer

construction;

x multilayer structures: very useful for constructing transducers with integrated power and data lines;

x seamless garment knitting;

x“Scan2Knit” technology [21].

The above technology provides an excellent platform for the development of multilayer seamless undergarments with multitude of KFTs and power and data lines. The KFTs and the power and data lines can be created as an integral part of the knitted structure of the undergarment; different layers can be used for creating the arrays of KFTs and conductive pathways for power and data lines. Such a garment would be washable, and the technology would enable them to be manufactured with a high degree of precision at low cost. 3. Configuration of resistive fibre meshed

transducers

KFTs are constructed using electro conductive fibres. The generic method of construction of the transducers is to knit a predetermined area of a knitted structure (base structure) with the electro conductive fibres, herein after referred as Electro Conductive Area (ECA). In addition to the material properties the size, shape and the fibre binding methods (stitches, tuck loops and floats; known as binding elements) and their organization, orientation in the base structure would determine the overall electrical characteristics of the ECA, and its response to structural deformation(s) of the structure. This variation of the electrical characteristics of the ECA would determine the type of KFT and its function.

One of the measurable electrical properties of the KFTs is the electrical resistance of the ECA and its variation. The variation of resistance can be captured using two different approaches. Generally, when a knitted structure is deformed the structural deformations are due to the fibre deformations and/or slippages between the fibre contact areas of stitches (stitches are the building blocks of a knitted structure). Fibre deformations may be due to stretching, bending, twisting and compression. The first method of capturing the electrical resistance of a KFT is by considering only the length variation of the electroconductive fibre in the KFT, which will result in a variation of resistance. Therefore named fibre meshed resistive strain transducer in the following text. The second method is to consider the structural deformations of the stitches in the ECA which will results in a variation in electrical resistance. Under small loads this functions as a potentiometer, and, therefore, we have defined these as displacement transducers.

3.1. Resistive fibre meshed strain transducer

A fibre meshed resistive strain gauge is made using an electroconductive elastomeric fibre (e. g. carbon filled polymeric fibre, typically with a specific resistance of 5-10 K?cm-1) and non conductive base fibre (which could be an elastomeric fibre). The base structure is formed using the nonconductive fibre and the conductive fibre is laid in the course direction (rows of stitches) of the base structure according to a predetermined configuration, which could be in the geometrical form of a rectangle, a square, a triangle, a circle, an ellipse or any other arbitrary shape (Figure 1).

Due to the above configuration the variation of resistance will only be due to the stretching of the conductive fibre. In Figure 1 the white base structure has been knitted with the non conductive fibre with the laid in

conductive elastomeric fibre (black fibre). Depending on the application the size and the shape of the transducer can be selected.

Figure 1: Resistive fibre meshed strain transducers The electrical equivalent circuit of the strain transducer was modeled by considering the geometrical path of the conductive elastomeric fibre. Since the conductive elastomeric fibre is secured within the courses of the base structure, which has been knitted from the non conductive fibre, the conductive elastomeric fibre lies electrically insulated, i.e. there are no cross connections in the conductive elastomeric fibre. Therefore, the electrical path will be the same as the geometrical path of the laid in conductive elastomeric fibre. The equivalent resistance was calculated, assuming it was electrically powered from the two ends (Figure 2) of the conductive elastomer. Therefore it is arranged such that when the structure is stretched in the direction of Y the electrical resistance of the electroconductive elastomeric fibre would increase with respect to the extension.

3.1.1. Operation of the fibre meshed strain transducer When the fibre meshed strain transducer was stretched in Y direction by using planar loads (Figure 2) an increase in the electrical resistance was observed (Figure 5b). The stretching of the transducer resulted in elongating the electro conductive fibre. Such an extension of the fibre length would cause a change in the electrical resistance of the fibre. This is due to the piezoresistive behavior of the composite elastomer [18]. However when the transducer was stretched in the X direction an insignificant variation of the electrical resistance was observed. Moreover the resistance is reduced (Figure 4b). Further it is reduced because of the loss of curvature at bending regions (Figure 2). This was expected due to unaccountable deformation of the electroconductive fibre.

The resistance R and the variation in resistance 'R with the deformation are calculated by using the following Equations ((1)-(2)).

(

U u

??????????????????(1)

Figure 2: Electroconductive fibre path

???????(2) Where m is the number of columns (parallel paths), r is the radius, y is the column length and U is the resistivity of the fibre.

Due to the non linear mechanical properties and nonlinear electrical properties of the electroconductive elastomeric fibres the poison’s ratio V is a function of extension ()

V). I n addition U is also a variable function of extension ()

U). The sensitivities of the transducer, when it is stretched in Y (courses) and X (wales) directions are given by O1 and O2 respectively. They are obtained by using Equation (3).

}

{

O????????????????????(3) From Equations (1) and (2);

}

)

{

R???????(4) From Equations (3) and (4);

}

]

)

1[(

)

{

O?????

(5) Figure 3: Testrig of the Fibre-Meshed Strain Gauge

1.4cm 2cm

The above models were validated with practical results. A fibre meshed strain transducer was clamped between the cross bars of a tensile tester (I NSTRON 5500). The upper crossbar was programmed to carry out a dynamic extension of the transducer (the positive values of ?Y = Asin(?t)), and the variation of the electrical resistance of the conductive elastomeric fibre was measured. The testing was carried out for both coursewise (X ) and walewise (Y ) directions (Figure 2). The results are summarised in Figures 3-4. A low cross sensitivity was observed (the maximum of ?2/?1 was 12.5%) and this was

expected due to the negligible small extension of the

conductive elastomeric fibre in the X direction (refer

Equation (3)). Reproducible results were observed after a

few cycles, which were caused by the mechanical hysteresis of the overall fibre meshed structure. The number of cycles necessary to obtain consistence results is expected to be frequency dependent. Also a biaxial off axis strain gauge can be constructed from two such

transducers arranged orthogonally.

Figure 4 (a): Load vs walewise extension (X)

Figure 4 (b): Resistance vs Walewise Extension (X)

Figure 5 (a): Load vs coursewise extension (Y)

Figure 5 (b): Resistance vs coursewise extension (Y) 3.2. Resistive fibre meshed displacement transducer Fibre meshed displacement transducers were constructed as explained earlier by using electroconductive fibres of different conductivity levels and non conductive fibres. Stainless steel wire (10:cm -1)

was used to power the transducer. An example of a displacement transducer is shown in Figure 6.

Figure 6: Fibre meshed displacement transducer

L 1 L 2

Figure 7: Electroconductive fibre paths

Figure 7 highlights the electroconductive fibre paths of the displacement transducer (Figure 6). The structure is

comprised of three different regions of fibres and hence it

High conductive fibres

Low/Semi conductive fibres

2.5cm

4.5cm

Wale direction

Course direction

can be divided into three different segments. The first is the sensing component (ECA), the second powering

component and the third is the base structure. Where L1 and L2 are the free lengths of the high conductive

(stainless steel) fibres used for powering, m is the number

of courses (rows) and n is the number of wales (columns) of the low conductive fibres of the sensing segment

(ECA).

3.2.1. Functional behaviour

The functional behaviour depends on the initial physical orientation of the sensing segment (ECA). The ECA

should follow the three dimensional surface contour of the

human body. But during our initial research we have only considered about the planar biaxial deformation of the

ECA. Therefore when the fibre meshed structure (Figure 6) is stressed in the direction of the wales the lengths of the legs (Figure 8a)of the stitches L Leg s are increased and

lengths of the heads (Figure 8a)of the stitches L Head s are

decreased. This is due to the slipping of the fibres at contact regions. The explanation of this deformation depends on characteristics of the applied load as well. The slipping is continued until the fibres get jammed. After that the slipping is stopped and then onwards fibres are compressed at the contact regions and stretched from the legs. During this period the L Head s are zero and L Leg s are increased producing a near linear variation of resistance with the extension. When the load is released due to the elasticity the structure returns back. The mechanism can be explained similarly when it is stressed in the course direction.

3.2.2. Modelling of the electrical equivalent circuit

The mathematical modeling of the electrical equivalent circuit was carried out in four steps. I n the first step the unit cell of the resistance mesh was derived. In the second the sensing area (low/semi conductive area Figure 6) was modeled by using points and springs in order to study the deformations under the planar loads. Then in the third step; localized information (wale separation w’ and course separation c’(Figure 14) of the deformations of the lengths of the springs was used to construct the geometry of the loop and then in the final step the electrical parameters were calculated by using the geometrical parameters (L Legs and L Heads). Then the variation of the electrical parameters can be analyzed.

3.2.2.1. DC equivalent circuit

Weft knitted fibre mesh structures are created by forming fibres into rows of loops and then interconnecting these with loops of a similar row of loops [16]. Hence a loop of a fibre meshed structure [16] is connected to the rest of the structure by intermeshing with the adjacent rows which results in intertwining of fibres at four regions (Figure 8a) of a single stitch. These regions are called binding or contact regions. The contact mechanics at these regions are very complex and their behaviour is not yet fully understood. I f the fibres are incompressible and inextensible then there are two points of contact in each region [13],[19]. However this becomes a line or an area contact depending on the applied load and the fibre properties. For the fibres that we were used in the construction of the sensing segment of the structure (ECA) it was assumed to be a line contact at the relaxed state, and changing into an area contact when a planar force was applied to the structure. This is due to the compressibility of the elastomeric fibres used in the sensing segment (ECA). Therefore it was assumed that the fibre contact region is to be a short circuiting point in modelling the electrical equivalent circuit. Thus it would act as a node of a resistor mesh. For each loop there are four contact regions or points. Therefore the DC equivalent circuit of a stitch (unit cell) was formulated considering the resistances of the free electroconductive fibre paths between these four nodes (Figure 8b). The lengths (L Leg s, L Head s) were calculated by considering the yarn path geometry of the loop. Therefore the DC equivalent circuit of a unit cell consists of four resistors; two resistors representing the heads (R H) of the interwining loops of the adjacent courses and two resistors representing the two legs (R L). They are connected to each other as shown in Figure 8b. The resistor mesh of the structure was constructed by repeating this four resistance model (Figure 15). The total equivalent resistance (R eq) can be calculated for a given powering configuration.

Head

Legs

Regions

(a)(b)

Figure 8: (a) Fibre intersections of weft knitted loop

(b) Four resistance model of the unit cell

3.2.2.2. Point spring model of the fabric structure

Since these transducers are mainly to be utilized in dynamic measurements (including movements of the kinematical joints) a dynamic model is needed to estimate the w’ and c’ of each stitch and a geometrical model to estimate the fibre paths. By using this information the resistance (R eq), the variation of resistance (¨R eq) and the sensitivity('R eq 'L) can be calculated.

R H

R L R L

The geometry of the fibre paths depends on the applied load characteristics [13], [15], [19], [20]. The approach was to take the fabric structure including the ECA as masses and springs and estimate the deformation of the whole fabric under applied load. Moreover since the bodily movements are lower in speed (very much lower than the deformation speeds of the fabric structures where the inertia effects due to the masses come into the picture) it is sufficient enough to analyse the deformations with quasi static approach rather than structural dynamics approach. These structures were modelled with discrete spring elements thus neglecting the masses. The modified model we had used was used by Eberhart et al [17] for modelling the knitted patterns. They had modelled each unit cell with four springs and four points arranged in rectangular orientation. We have represented each stitch by six springs and four points (Figure 9). This was taken as the unit cell and was repeated throughout the structure (Figure 10). I n addition to the four springs used in the Eberhart’s model we have used additional two springs across the points of the diagonals of the unit cell to represent the compression at the contact regions.

For m number of courses and n number of wales there are )1()1( u n m number of contact points (CPs ). The resultant force F i,j acts on the i th ,j th CP ]}1,1[{]}1,1[{ n N j m N i is given by Equation (6). (x (i,j),y (i,j))is the position of the i th ,j th CP.

Legs

Contact

Regions

Figure 9: Point spring model of a stitch

Figure 10: Point spring model of the structure

i n k k

j i z y x F

z y x F ,0

,),,(),,(??????????????????????(6)

Where n i,j is the number of the neighbours in the neighbourhood of the i th ,j th CP .F k is either the stretching, the repelling or the compressing force on the k th neighbour of the i th ,j th CP .

D H )(),,(k k C z y x F ????????????????????????????????(7)

Where C k and D were determined by using empirical data of the fibres and H is the difference in lengths of the springs between the initial and the next equilibrium state under a given load. Moreover, the bending of a fibre loops were simulated via another spring force across two CPs .The slipping of the fibres at a contact point was simulated via a rough heuristic. Therefore the dynamics of the structure was resolved into a particle system with spring forces between them. Therefore the equilibrium state of the system under a given load was determined by using the principle minimum potential energy. The total potential energy under a given loading condition U was calculated by Equations (8) and (9); | springs

all energy spring U __????????????????????????(8)

|||| n

compressio stretch repel bend U U U U U ???(9) Where U bend is the bending energy, U repel is the repelling energy,U stretch is the stretching energy and U compression is the compression energy at CPs . The trajectories of the CPs were obtained through a numerical solution of the well known Euler-Lagrange differential Equation (10).

j i j i j i j i y

L y L dt d x L x L dt

d ,,,,w w ??1

¨¨?§w w w w ??1·¨¨?§w w ???(10) E is the total kinetic energy of the system, U friction Total frictional energy loss.

friction U U E L ?????????????????????????????? (11)

The w’ and c’ were calculated (Equation (12)) with respect to the globalize coordinate system and this information was passed to the next step to estimate the geometry of each stitch.

}''{),(),1(),()1,(),(j i j i j i j i j i y y c x x w ????(12)

3.2.2.3. Geometrical model of the stitch

The fibre path geometry of a weft knitted stitch is very complex and it depends on the mechanical properties of the fibre, knitting machine parameters, the environmental conditions, position of the loop on the fabric and the applied loads. The models put forward over the years [13], [14], [15] only had discussed about the geometry under relaxed situations. For the analysis of the fibre meshed transducer a generalized geometry of a stitch under any loading condition was needed. Here we have proposed a simple generalized geometrical model for the stitch based on principle of minimisation of energy. We have used it for estimating the resistance (R eq ).

The proposed model takes

w’ and c’ and the diameter and the mechanical properties of the fibre as the input

parameters. Let the stitch height be c, the stitch width be w and the thickness be A(Figures 11-12). The localize axis system of the loop was defined as shown in the Figures 11-13. For the simulation w=2.5mm, c=2.9mm and A=0.7mm were used (Figures 11-13).

a c-a

Figure 12: Elevation of the loop

The geometry of the axis of loop was assumed to be

symmetrical under relaxed conditions about the X axis on

the XY plane therefore the parametric equation of the axis

of the loop was obtained by dividing one half of the loop

into three segments. The first segment Y(x), a

x,0

represented by using quarter ellipse who’s major axis are

a,b and the centre (a,w). The second segment

Y(x),)

,(a

is represented by a 7th order polynomial.

Figure 13: 3D View of the loop

The third segment Y(x), c a

is again represented

by a quarter ellipse having axis a,b and the centre (0,c-a).

On the XZ plain it (Z(x)) was taken as a polynomial of 4th

order. Let R(x) is the position vector of the loop. The

geometrical curvature (K) and the geometrical torsion (W)

were calculated by using Equations (13).

}

)''

'.(

{

W?????????(13)

The bending rigidity modulus (EI) and the torsional

rigidity modulus (GJ) of the fibres were taken from the

experimental data (Standard notation is used). The

bending (U B) and the torsional (U T) strain energies were

calculated for a one half of the loop (Equations (14)).

3dx

U c

)

(

{}

)

(

W??(14)

The tensile energy U E and the compression energy the

U C were assumed to be zero since the fabric is in its

relaxed state. The total potential energy (U stitch) of the

loop at relaxed state was given by;

)}

(

)

(

{

stitch

u??(15)

The Equation (15) is taken as the objective function.

a and

b were found for a given w,

c and

. Moreover a and b were determined by using the

x=a1(i,j) x=a2(i,j) defining the contact points of the i th,

th loop(>@

i,1

j,1

) with

(i+1,j) and (i-1,j). The leg length

L Leg(i,j) and the head length L Head(i,j) were calculated by

using Equations (16).

3),(10

),(

)('

{j i a

j i

head

L3),(),(21

),(

}

)('

j i

Leg

L?(16)

),(

)

j i

c )}

(*2

),(

),(j i

j i

w ??(17)

Where )

(

),(

),(j i

j i

Y and B(i,j) is a value depend on

the diameter of the fibre and the applied load (Figure 14).

Figure 14: Plan of (i-1),(i),(i+1) stitches of j th column

Figure 15: Resistor mesh of the sensing segment 3.2.2.4. Determination of resistances (R L s and R H s)

The R L(i,j) and R H(i,j) were calculated by using the proposed geometrical model of the stitch. The leg lengths and the head lengths between the contact points were

calculated by using the 3Dspace curve )}),0({)((c x x R

of a single stitch. The lengths (leg length L Leg(i,j)and head length L Head(i,j)) were calculated for each of the loop separately and for each length the resistances was calculated (Equations (18)). For the simulation m =n =20 was used.

}{)

,(),()

,(),(A

L R A

L R j i Head L j i H j i Leg L j i L U U

??(18)

The head resistance matrix R H and the leg resistance matrix R L were obtained. The equivalent resistance of the structure R eq was calculated by using electrical network theory of mesh analysis (Equations (19)-(20)). The Voltage Vector V M ={V,0,0,0,…,0}m*n+1,1

The Current Vector I M ={ I 0, I 1,1,I 1,2,…, I 2,1 ,…I m,n }m*n+1,1

M M M M M M V Z I I Z V u u 1??????????????(19)

c eq R z R

)

1,1(11

????????????????????????????????(20)

Where z -1(1,1)is the (1,1) element of Z -1M .The total equivalent resistance R T is given by;

eq C T H H C R R R L L R U U 2122?????????(21) Where p H is the linear resistivity of the high conductive fibre. The current distribution of heads of the loops I_H (m+1,n) and the current distribution of the legs of the loops I_L (m+1,n+1) were calculated by using Euation (22) The normalized results are shown in Figures 16-17. Also

the voltages at the nodes (V N ) were calculated by using Equation (23).

)},,(_),,(_{M M I n m G L I I n m F H I ????(22)

),,,,_,_(M C L H N V R R R L I H I E V ??????????(23)

The power distribution of the heads (P_H ) and legs (P_L ) were calculated by using Equations (24).

}

____{),(2

),(),(),(2),(),(j i j i L j i j i j i H j i L I R L P H I R H P u u ??(24) The temperature distributions of the heads (T_H ) and legs (T_L ) were calculated assuming that the heat was lost only by radiation (Equations (25)-(26)). Form the Boltzmann law of radiation;

4/14

),(),()/)_(_a c j i j i T K H P H T ???????????(25)

4/14

),(),()/)_(_a c j i j i T K L P L T ???????????(26) Where K c is the radiation constant of the conductive fibre and the T a is ambient temperature (Figures 18-19).

Figure 16: I_H distribution

Figure 17: I_L distribution

The current distributions were justified by observing the temperature distributions of the structure which were

measured by using a thermocouple.

Figure 18: Normalized temperature distribution of the

heads (T_H)

Figure 19: Normalized temperature distribution of the

legs (T_L)

Figure 20: Observed temperature distribution 3.2.2.5.AC equivalent circuit of the resistive displacement fibre meshed transducer

The I mpedance was measured by using an impedance analyser (Agilent 4192A). I t was observed that the structure is having a band pass characteristics under relaxed conditions (Figure 21a ). Therefore we have assumed that in addition to R eq it has a series inductance and a shunt capacitance (Figure 21b ). The observed resonance frequency also depends on the dimensional parameters of the ECA.

Figure 21: (a) Impedance spectrum of the ECA

Figure 21: (b) AC equivalent circuit of the ECA

Therefore the expected transfer function (T(S)) of the structure;

???

o???a c eq eq R CS R LCS SL R S T 1)

()(2

?????????????(27) 4. Final remarks

4.1. Resistive fibre meshed strain transducer

The Observed results showed that the structure should be pretensioned to be used in strain gauge applications. The experiments on the bidirectional strain gauge are still to be carried out. The impedance spectrum has to be feather experimented. Also the strain gauge has to be tested and characterized for the unloading. In addition the humidity, temperature, washing effects and over tensioned effects have to be experimented.

4.2.Resistive fibre meshed displacement transducer

The model discussed here describes the variation of the impedance has to be revisited and replaced by more accurate model. The mechanical model is still under research and we are using finite element approach for the modelling. Since these transducers are to be used in measuring dynamic variations the electrical hysteresis of the structures has to be taken in to account as well. Moreover calibration of the transducer is to be achieved through dynamic calibration during a particular operation. The impedance variation of the fabric transducers can be derived by using the scheme outlined in Figure 22. Also the model discussed is to be experimentally tested. The dynamic tests of the transducer are to be carried out. I n addition the metal semiconductor junctions formed at the connection points (ECA and powering fibres) of the transducer has to be investigated. From the observed results it is reasonable to assume that the junctions are behaving as Ohmic contacts. Here we only discussed the impedance of the plain weft knitted structures. The other knitted structures can also be modelled using a similar procedure. Then the best structure for a particular application can be selected. Moreover this mesoscophic scale study also can be extended to analyse woven conductive fibre-mesh structures as well.

5. Applications

The strain gauges are to be used in wearable real time personalised information gathering systems such as physiological condition monitoring systems for monitoring and measuring breathing, movements of

kinematical joints of the body and body gestures.

6. Acknowledgements

The authors would like to appreciate the help given by Dr Paul Beatty Department of I maging Science and Biomedical Engineering at University of Manchester, Mr. K. Mitcham Department of Textiles at UMI ST for constructing the samples. Mr. A.G. Handley Department of Paper Science at UMIST.

Figure 22: Procedure for Analysis of Impedance of Electroconductive Fibre Meshed Structures

7. Bibliography & References

[1] Xiaoming Tao “Smart Fibres, Fabrics and Clothing”CRC Press ISBN 0-8493-1172-1,2001

[2] S. Jayaraman, “Designing a Textile Curriculum for the 1990s

A Rewarding Challenge”, Journal of the Textile I nstitute, Volume 81, Number 2,185-194, 1990.

[3] Cliff Randell Henk Muller Chris Djallis Constance Fleuriot Mike McCarthy Andrew Moss “The Cyber Jacket MkII Further Development of a Wearable Computing Research Platform” IEE Eurowear 2002,University of Bristol, Bristol 2002

[4] Anthony Schwirtz Chris Baber James Knight “Sens Vest” IEE Eurowear 2002,University of Birmingham, Bristol 2002

[5] E. R. Post, M.Orth, P. R. Russo, and N. Gershenfeld “E-broidery: Design and fabrication of textile-based computing”

I BM systems journal,Vol.39,Nos3&4,Armonk,NY,I BMCo,2000

[6] P.J.Massey “Fabric Antennas for Mobile Telephony

I ntegrated Within Clothing” LCS2000,Philips Research Laboratories, Redhill, UK,University College London 2000

[7] M. Orth, Rehmie. Post Emaly Cooper “Fabric Computing Interfaces”,human factors in computing,LA,ACM press 1998

[8] Jonny Farringdon, Andrew J. Moore, Nancy Tilbury, James Church & Pieter D. Biemond “Wearable Sensor Badge & Sensor Jacket for Context Awareness” Philips Research Laboratories, UK,IEEE ISWC’99,1999

[9] Dr. Eric J. Lind, Dr. Robert Eisler, Dr. S. Jayaraman, Mr. Tony McKee “A Sensate Liner for Personnel Monitoring Applications” Georgia I nstitute of Technology USA,I EEE ISWC’97,1997

[10] Yoseph Bar-Cohen “Electro-active Polymers: Current Capabilities and Challenges” Jet Propulsion Laboratory/Caltech 4800 Oak Grove Drive, M/S 82-105, Pasadena, CA 91109, USA.

[11] “Softswitch”https://www.wendangku.net/doc/fc1590699.html,/

[12] Timothy G.Gutowski “Advanced Composites Manufacturing” John Wiley & Sons Inc ISBN 0-471 15301-X. [13] Barbara Hepworth & G.A.V Leaf “The Mechanics of an idealized Weft Knitted Structure” journal of textiles institute 67,241-248, 1976.

[14] Olivier Nocent, Jean-Michel Nourrit, Yannick Remion “Towards Mechanical Level of Detail for Knitwear Simulation” proceedings of WSCG 2001, Laboratoire d’Etudes et de Recherches I nformatiques, Université de Reims Champagne-Ardenne, Rue des crayères, BP 1035 51687 Reims Cedex 2 France.

[15] Takahiro Wada, Shinichi Hirai, Tatsuya Hirano, and Sadao Kawamura “Modeling of Plain Knitted Fabrics for Their Deformation Control” Proc. I EEE I nt. Conf. on Robotics and Automation, Vol.3, pp.1960-1965, Albuquerque, April, 1997 [16] David J. Spencer “Knitting Technology” A Wheaton & Co. Ltd. ISBN 0-08-024763-6.

[17] M. Mei?ner and B. Eberhardt “The Art of Knitted Fabrics, Realistic & Physically Based Modelling Of Knitted Patterns” EUROGRAPHI CS ’98 / N. Ferreira and M. G?bel Volume 17, (1998),Number 3.

[18] M. Taya ,W.J. Kim , K. Ono “Piezoresistivity of a Short Fibre Elastomer Matrix Composite” Mechanics of Materials 28 1998 53–59.

[19] W.J Shanahan and R.Postle “ A Theoretical Analysis of the Tensile Properties of Plain Knitted Fabrics” Textile Research Journal (1972),23,200.

[20] M. Konopasek and J.W.S. Hearle “Computational Theory

of Bending Curves” Fibre Sci.Technol. 5,1,(1972).

[21] Tilak Dias “Pressure Garment” Patent No: WO03040449 15-05-2003.

[22] Jayaraman Sundaresan and Park Sungmee “A Novel Fabric Based Sensor For Monitoring Vital Signs” Patent No:WO 02/40091 A2, 23-05-2003.