Textile touch sensors for wearable and ubiquitous interfaces

Abstract

The design of textile touch sensing interaction was explored with the new metal composite embroidery yarns (MCEYs) and a simple and easy fabrication technique aimed towards robust and reliable pressure sensitive position sensors for wearable tangible interfaces. In this paper, the resistive sensing method of a potentiometer as an accurate positional indicator was chosen to make simple prototypes of MCEY embroidered touch sensors. A simple structure of embroidered potentiometer to create textile switches as an input device in a smart textile system was tested. Both one- and two-point sensing method were successfully demonstrated. A complete success rate on switching was observed. These simple but ingenious embroidered touch sensors showed the possibility of a soft, lightweight, flexible, freely foldable touchpad as a ubiquitous solution. It was also shown that these minimal fabrication technologies may be highly valued in the smart textile field thanks to their simplified interconnections, customizability and tailorability on double curvature surfaces.

Keywords

Textile touch sensor textile switches wearable interfaces embroidered potentiometer smart textiles e-textiles

Textile touch sensor technologies have been developed for smart environments, wearable interfaces and ubiquitous tangible computing. These will be embedded in clothes and other everyday life goods, such as control devices for smart clothing, remote control cushions, touchtone musical keyboards, interactive walls, haptic communication between human and computers, full-body tactile sensors for robots, capacitively sensing textile pressure sensors for sitting posture classification and smart carpets for human tracking, and so on.^1–12 Inaba et al. presented a full-body tactile sensor suit with 192 sensing regions, each of which was wired as an individual binary switch, using copper and nickel plated conductive fabrics and yarns.¹ Post et al. developed a musical jacket with a capacitive touch keypad embroidered with a stainless steel and polyester composite thread, and controlled through stitch placement.⁸ Embroidering with conductive threads offers advantages for interactive e-textile development including the abilities to stitch multiple layers of fabric in one step and to precisely specify circuit layout with CAD. Swallow and Thompson introduced sensory fabric having a layer construction of two conductive fabric sheets and a nonconductive mesh sandwiched between them, which can detect the distance of an interaction from a connection point; thus the position of an interaction on the fabric can be ascertained by triangulation using the distance data from many connection points made to the conductive sheet.⁹ Freed and Roh have experimented with multi-touch sensing flexible fabric substrates for musical controllers.^5,10

Actually textile touchpads which detect weighted contact or pressure have attracted considerable commercial interests because their flexibility and pliability make easy incorporation into garments and textile products with lightweight, warm and soft wearing comfort. ILC Dover, Inc., worked with Softswitch, Ltd., to create pressure sensitive textile switches and cabling that was embedded into an astronaut’s arm glove gauntlet using quantum tunneling composites technology, which was designed for rugged use in harsh changeable, thermal and dusty environments.¹¹ This technology came into the market with the QIO ElekTex® six-button textile touchpad.¹² For the Elektex® and similar products, the number of connections from the sensing surfaces to the controller is equal to the number of buttons plus one ground line; as the number of required sensing positions increases, so does the number of connection lines. The Elektex® programmable textile touchscreen is an upgraded version of a textile touch sensor designed with greater customizability to meet the needs of DIY users.¹³ This resistive touchpad senses x- and y-coordinates and gives a relative pressure reading (z). It can be programmed to create a variety of machine interfaces including keyboards, number pads for mobile phones, MP3 player interfaces, virtual rotating scroll wheels, and so on, though they still have limited customizability for individual creative design.

There are three types of conductive yarns for embroidery, plied yarn of stainless steel filaments, plied yarn of metal-coated polymer filaments, and composite yarn consisting of metal filaments and polymer filaments.²⁰ A three-plied yarn of stainless steel filaments (Bekinox® VN 12/3 × 275/175 S/316 L, BEKAERT) has a very high linear density of 6840 denier for a resistance of 9.31 Ω/m. A variety of silver coated polymer yarns are available as a conductive embroidery yarn, but they have considerably less than ideal conductivity. For instance, two plies of Shieldex®, about 210 denier, showed 350 Ω/m. Moreover, the metal coating has poor durability due to its low abrasion and corrosion resistance. Composite yarn composed of silver-plated copper filaments and polyester filaments has some advantages over stainless steel and silver coated yarn due to its lower resistance and solderable interconnection. Ohmatex has developed a bi-component yarn composed of two strands of silver-plated copper filaments (diameter of 40 µm or 63 µm) and polyester filaments. But the long loops of the two metal filaments hang loose from the polyester filaments, so it is easy to get frictional damage during the machine embroidery process.

The novel contributions reported here primarily involve an easy and simple integration technique using metal composite embroidery yarns (MCEYs) with the aim of developing robust and reliable fabric touch sensing surfaces for wearable smart interactive textile systems. This technique allows an improvement in fabrication technology for e-textiles from hand craft to automated production thanks to a computer numerical control (CNC) embroidery process using MCEYs, and thus it can lead to mass production of e-textiles as well as specially customized production. Moreover, the embroidered circuits of this conductive yarn can be part of a sensing surface with double curvature.

This paper focuses in particular on a minimal design as well as an energy efficient design to offer opportunities for creative customizable designs to DIY users. The textile circuits presented here started with a voltage-centric view of resistive sensing with a potentiometer because measuring voltage is more energy efficient than measuring current. This voltage-centric view can be reinforced by energy efficient design.⁵

In this paper, two different types of MCEYs were developed for a flexible sensing surface, and their physical and electrical properties were investigated. The electrical characteristics of the MCEY embroideries were also investigated. Embroidered touch sensors were fabricated using the two types of MCEYs and one-point and two-point sensing methods were designed. Finally, the performance and functionality of textile touch sensors as textile switches for wearable tangible interfaces in smart textile systems were evaluated.

Materials and methods

Preparation of MCEYs

The textile sensors proposed in this paper are based on potentiometric resistive embroidery which is fabricated by using MCEYs. Two different types of MCEYs were developed by wrapping and twisting superfine metal filaments (diameter of 0.040 mm, silver plated copper filament, TW-O, Textile Wire of Elektrisola Feindraht AG, Switzerland) with polyester filaments (75 denier with 36 filaments). The first step was to wrap a silver-plated copper filament around the surface of a strand of polyester yarn in an S-direction with 150 T/m (twists per meter). Then the silver-plated copper filament wrapped polyester yarn was additionally twisted in an S-direction with 700 T/m. Lastly, three strands of the S-twisted yarns were plied together in the Z-direction with 550 T/m (MCEY1, Figure 1(a)). During high-speed machine embroidery, the embroidery yarns, especially the upper yarns, are subjected to high stresses, so better mechanical properties are demanded for yarns to be used as the upper yarn.^15,16 Thus, in the above mentioned last part of the plying process, the three strands of the S-twisted yarns were wrapped with an additional polyester filament yarn of 20 denier in the Z-direction with 90 T/m and then plied together in the Z-direction with 550 T/m to improve the yield strength of the MCEY and to protect the metal filaments within MCEY from the frictional stress induced by needle-eye movement (MCEY2, Figure 1(b)).¹⁷

Figure 1.

Structures of the MCEYs: cross-sectional SEM images (upper) and side views (lower) of the MCEY1 (a) and the MCEY2 (b).

Design and fabrication of embroidered touch sensors

To make a simple textile touch sensor, the resistive sensing method of a potentiometer as an accurate positional indicator was taken into account. A potentiometer is a manually adjustable resistor or variable voltage divider with a shaft or slide control for setting the division ratio.¹⁸ One terminal of a potentiometer is connected to a power source and another is hooked up to ground, while the other terminal runs across a strip of resistive material. This resistive strip generally has a low resistance at one end and its resistance gradually increases to a maximum resistance at the other end. Voltage drop can be calculated using Ohm's law; E = RI, where E is voltage drop (Volts), R is electrical resistance (Ohms) and I is current (Amps). The third terminal serves as the connection between the power source and ground. The user can adjust the position of the third terminal along the resistive strip in order to manually increase or decrease resistance.

The main ideas of the embroidered touch sensors are based on the principle of a membrane potentiometer which consists of a resistive path that is printed onto a membrane base and a collector with a printed short-circuit path that is applied on top of this base, and both paths are separated by a circumferential spacer.¹⁹ When pressure is applied to the collector foil, electrical contact is simultaneously applied to the resistive path and a voltage can be tapped into movement via the collector foil. Once the pressure being applied to the collector foil ceases, the voltage can no longer be tapped. Positional information can be detected from the relationship between input voltage and position of a touch point.

Figure 2 illustrates the design principle of an embroidered potentiometer and its position sensing circuit diagram. In order to make a fabric resistive trace, a variety of creative, beautiful space-filling structures using MCEY can be embroidered on the CNC embroidery process. Zig-zag or meander-shaped stitches of MCEY can effectively fill a resistive trace of a textile sensor without any intersection of the embroidered resistive path. To measure the incoming voltage at the pressed point, highly conducive woven fabric, Shieldex® Kassel,²⁰ anticorrosion coated copper-silver plated nylon fabric with a surface resistance of less than 0.03 ohms/square and a thickness of 0.11 mm, is used as a conductive fabric sensor. A hexagonal mesh fabric with a thickness of 0.24 mm and mesh openness of 9 mm² is inserted between the resistive embroidery and the fabric collector in order to make space between them. An ultra-thin nylon fabric with a thickness of 0.04 mm and an area density of 1960 × 1860 is inserted between the resistive embroidery and the conductive fabric in order to prevent unwanted electrical contact outside of the buttons. The total thickness of this textile sensor is about 1.25 mm.

Figure 2.

Illustration of an embroidered potentiometer and its position sensing circuit diagram.

Equation (1) explains the incoming voltage, V_in, while pressing position P in Figure 2. If no touch is occurring there is no incoming voltage. When touched the layers are compressed together to form an electronic circuit and a voltage can be tapped into movement via the conductive fabric collector. From this the position or distance from the measuring point can be detected through the change in resistance. By pressing down on various parts of the resistive MCEY embroidery, the resistance linearly changes along the resistive trace allowing the user to very accurately calculate the relative position on it. In this case, electric current flows continuously on this resistive embroidery causing energy loss. To improve the energy efficiency, an alteration to the Microcontroller is needed to let the current flow only when there is contact between the resistive embroidery and conductive fabric collector.

V_{in} = \frac{R_{1}}{R_{0} + R_{1}} V

(1)

Interestingly, Figure 3 represents duo-touch sensing using the indented bar symbol.⁵ When touched there can be no potential difference between the outermost pair of points of contact as they are both shorted together and connected to ground. The positions of these outer touch points can be estimated by applying the standard potential divider formula to the resistances to the left and right of the contact points and respective pull-up resistors, R₀. In Figure 3, if no touch is occurring, there is no path for current to ground so both measurement nodes must be at a high potential. This sensing method is more energy efficient than the fore-mentioned method because the current flows only when a contact occurs between the conductive thread and the conductive fabric.

Figure 3.

Two-point sensing circuit diagram with an embroidered potentiometer.

Characterization

SEM and a video microscope system were used for the surface morphologies and cross-sectional images of the MCEYs. Tensile tests of the MCEYs have been conducted on a universal testing machine (Instron 5543), according to ASTM D 2256 with a gauge length of 250 mm.²¹

The electrical resistance of the MCEYs and the MCEY embroidered circuits were measured with a low resistance meter (Miliohmmeter, Agilent 4293B). To investigate the electrical resistance of the MCEY embroidered circuits and the effect of CNC embroidering directions on the electrical resistance of the MCEY embroidered circuits, 20 samples of MCEYs running stiches in each of the four following directions were prepared: direction 1 (front-to-back), direction 2 (left-to-right), direction 3 (back-to-front) and direction 4 (right-to-left) (see Figure 6, below). These samples were prepared in the size of 10 cm in length with a 2.5 mm running stitch size on a plain weave cotton fabric (thickness of 0.24 mm). The sensing performance of the embroidered switches was measured by checking the incoming voltage while pressing by using a multimeter (HP 34401A).

Figure 4.

Resistance linearity of the MCEYs as a function of the yarn length: (a) MCEY1 and (b) MCEY2.

Figure 5.

A soldering interconnection method: left is a soldered pad for interconnection and right is an interconnection pad before soldering.

Figure 6.

Surface images of the MCEY embroidered circuit lines for both (a) MCEY1 and (b) MCEY2 captured with a video microscopic system: four directions of the running stitches of front-to-back (D1), left-to-right (D2), back-to-front (D3) and right-to-left (D4).

The voltage input of the MCEY embroidered touch sensors was measured with a multimeter (Agilent 34401A). The term “input” means the direction of “entering into-the-microcontroller” and the output means “coming out of-the-microcontroller.” Data acquisition and analysis of the relationship between input voltage and position of a touch point can be measured. The embroidered touch sensors are subject to a predetermined output voltage of 5 V from a microcontroller and when a certain position on the resistive embroidery is pressed by a finger, an analogue voltage input can be transferred and converted to digital values ranging from 0 to 255 (for 8 bits) through a microcontroller. The value is dependent on the voltage where a value of 0 is 0 V incoming and 255 is 5 V incoming for one-point sensing. In the case of two-point sensing, the maximum voltage of the AD converter in the microcontroller was set to 2.8 V because the voltage input values from two-point sensing method were below 2.8 V. Thus, a value of 0 is 0 V incoming and 255 is over 2.8 V incoming. The sampling rate was less than 10 ms. Sensing performance was displayed on the screen of the controller.

Results and discussion

Flexible, robust, and solderable MCEYs

Morphologies of MCEYs

The physical and electrical characteristics of the MCEYs used in this paper are listed in Table 1.

Table 1.

Characteristics of the MCEYs

I.D.	Composition (wt%)	Resistance (Ω/cm)	Linear density (denier)	Thickness (µm)	Load at yield (Offset 0.2%) (N)	Max. load (N)	Strain (%)
MCEY1	Ag–Cu: Polyester = 56.4 : 43.6	0.04873	547	199	3.92	13.3	13.1
MCEY2	Ag–Cu: Polyester = 54.5 : 45.5	0.04915	570	231	4.45	13.2	12.4

The smooth surface morphologies of the MCEYs will provide advantageous conditions for less friction during the embroidering process and provide uniform conductivity. The side views in Figure 1 show well the structures of MCEYs where the three strands of metal filaments are electrically coupled in a spiral. The helical structure of the metal filament wrapped over the polyester yarn provides strain relief during the machine embroidery process and the best conditions for electric contact between resistive embroidery and conductive fabric when pressed.

Tensile properties

As shown in Table 1, both MCEYs showed a very high tensile strength of over 13 N. And as expected, the MCEY2 showed a 13.5% higher yield strength than the MCEY1 by using an additional polyester filament yarn of 20 denier, even though the elongation of the MCEY2 was less than MCEY1.

Electrical properties

As shown in Table 1, the MCEYs in this paper are not only very thin and light, but also offer the lowest resistance in comparison with the currently developed conductive embroidery yarns. The electrical resistance of the MCEYs was measured to examine the resistance linearity of the MCEYs as a function of the yarn length. Mean values of 20 measurements are plotted and the standard error bars are also displayed in Figure 4. The result shows the resistance of the MCEYs increased linearly with the length of MCEYs. Both of the MCEYs showed very good regularity of the electrical resistance. In the case of the samples 40 cm in length, the standard deviations of both MCEYs were less than 0.3%. Moreover, unlike silver plated polymer yarn, metal composite embroidery yarn is less influenced by corrosion and friction thanks to its much thicker and higher metal content.

Soldering interconnection

At a beginning or an end of embroidered circuits an MCEY can be connected to the controller via direct soldering. As shown in Figure 5, an embroidered soldering pad made of small dense satin stitches of the MCEY can also be used for a quick reliable interconnection. The polyester filaments melted and deposited on the edge of the soldering pad and the Ag-coated copper filaments remained during the soldering interconnection process. This may be the most distinctive advantage over silver-plated polymer yarn, stainless steel yarn and conductive ink. MCEYs may be soldered which enormously simplifies the challenge of reliably connecting the textile sensor with conventional electronic circuits.

Characterization of MCEY embroidered circuits

Morphologies of MCEY embroideries

The embroidery morphologies and the resistance of the prepared MCEY embroidery samples were compared. A CNC embroidery machine was set up with a conductive MCEY top thread and insulating cotton bottom thread on a commercial CNC embroidery machine (Tajima TFM) with a motor resolution of 0.001 mm. The MCEY circuits were embroidered on a nonconductive cotton woven fabric at the lowest rate of sewing speed (250 rpm) to reduce the force induced from a needle penetrating and withdrawing movement.

Figure 6 shows the surface morphologies of the MCEY embroideries with four different running directions. The first direction (D1) was a stitch running backward (from front to back), the second direction (D2) was a stitch running from left to right, the third direction (D3) was a stitch running forward (from back to front) and the fourth direction (D4) was a stitch running from right to left. The results showed that in D1 and D4, the MCEY thread got untwisted once for each stitch and in D2 and D3, one extra twist occurred. Therefore, in D1 and D4, metal filaments in the MCEY thread were slightly loosened. When an MCEY thread is untwisted, the polyester yarns in the MCEY thread contract due to their visco-elastic properties. However the metal filament would stay slightly stretched due to its plasticity. The metal filaments of MCEY2 become looser than MCEY1 because MCEY2 had one more strand of 20 denier polyester yarn resulting in a stronger contraction.

Electrical properties of MCEY embroideries

Table 2 shows the electrical resistance and the effect of embroidering directions on the electrical resistance of the MCEY embroideries for both MCEY1 and MCEY2. The values are mean values of the 20 samples. The results showed the characteristic differences among them. There were differences in resistance for each direction. D1 and D4 presented higher resistance than D2 and D3 for both MCEY1 and MCEY2. MCEY2 showed a bigger difference in resistance for each direction because MCEY2 has a higher yield strength and lower stretchability than MCEY1 (Table 1). For this reason, MECY2 showed better regularity of electrical resistance. Coefficient of variation (CV%) of the MCEY2 was approximately three times lower than MCEY1 (Table 2).

Table 2.

Characteristics of the MCEY embroidered circuit line (2.5 mm running stitches)

ITEMs	Running direction	D1. ↑ (front-to-back)	D2. → (left-to-right)	D3. ↓ (back-to-front)	D4. ← (right-to-left)	Average
MCEY1	Resistance (Ω/cm)	0.08437	0.07986	0.07870	0.08454	0.0819
MCEY1	CV%	6.02	5.53	5.69	5.38	n.a.
MCEY2	Resistance (Ω/cm)	0.08772	0.06804	0.06700	0.08400	0.0769
MCEY2	CV%	1.82	2.26	1.36	2.11	n.a.

Application for textile switches

Design of embroidered switches

The technology on the MCEY embroidered potentiometer as an accurate positional indicator can be used for straight user input, i.e. textile switches where multiple button areas translate to given resistance levels. The same method for designing a resistive strip can be sectioned off to create an array of buttons using three fabric layers: resistive embroidery, insulating mesh spacer and conductive fabric. Six switch buttons are located as shown in Figure 7. Unlike other commercial textile switches,¹² there is no need to make an additional connection line as the number of buttons increases which enormously simplifies the fabrication process. To make a resistive embroidery for textile switches, a meandering resistive trace in the shape of a filled rounded rectangle (Figures 7 and 8) was chosen. The MCEYs were used for the upper embroidery yarn and a cotton thread was set up for the bottom yarn. The rounded rectangle area of this resistive embroidery was effectively filled with about 2.5 mm stitches which run from side to side with up to 8 mm width movement and 0.4 mm forward movement without a short between the adjacent embroidered lines (Figure 8). The resistance of the MCEY resistive embroideries was 65.57 Ω for MCEY1 and 65.27 Ω for MCEY2. The top and bottom circuits were separated by an insulating mesh spacer of 0.24 mm thickness and an insulating nylon fabric of 0.04 mm thickness with openings for the button areas. Electric contact between the MCEY thread of the resistive embroidery and conductive fabric forming a circuit occurred by normal tapping pressure with a finger to create a potentiometric output. By pressing down on various parts of the embroidered resistive trace, they produced a corresponding electrical input which changed depending on resistance allowing the user to very accurately detect the relative position on the resistive trace.

Figure 7.

Design of MCEY embroidered switches and the location of six buttons for switching: (a) MCEY resistive embroidery, (b) one-point sensing and (c) two-point sensing.

Figure 8.

Photograph of an MCEY embroidered potentiometer which consists of a conductive fabric, mesh spacer and insulating fabric, and MCEY2 resistive embroidery.

The principle of this switching system is illustrated in Figures 7 and 9. By imposing voltage (+5 V) to the textile switch unit, the microcontroller measures the corresponding switch button from the resulting voltage transmitted by the AD converter. Equation (2) explains the incoming voltage, V_in, while tapping each of the buttons. In the case of one-point sensing circuit, R₀ and R_n are variables of the resistance produced by nth button. For the two-point sensing circuit, a 65 Ω resister was used for R₀, and R_n was the resistance produced by nth button.

V_{in} = \frac{R_{n}}{R_{o} + R_{n}} V_{cc}

(2)

Figure 9.

Circuit diagram of the six button switches: (a) one-point sensing and (b) two-point sensing.

The software in this paper analyzed the relationship between the input voltage and position being pressed and provided precise detection and effective visualization of the sensing performance. To localize six switch buttons, the incoming voltage values were mapped. The values ranged from 0 to 255 (8 bits) and were mapped to six corresponding values depending on the button area using the simple function below;

If (cutoff 0 < input voltage < cutoff 1) then output switch 1,

if (cutoff 1 < input voltage < cutoff 2) then output switch 2,

if (cutoff 2 < input voltage < cutoff 3) then output switch 3,

if (cutoff 3 < input voltage < cutoff 4) then output switch 4,

if (cutoff 4 < input voltage < cutoff 5) then output switch 5,

if (cutoff 5 < input voltage < cutoff 6) then output switch 6.

This was easily done by finding the incoming voltage at each button point and the input voltage data around those points was mapped to one of the six switches.

Evaluation of the sensing performance

Figure 10 shows the simple mapping functions between the numbers of the MCEY embroidered switches and the values of the input voltages which were transferred pressing the areas designated for switching button numbers by using one-point sensing method illustrated in Figure 2. The mean values and standard errors of 20 input voltage measurements at each number are plotted. The standard deviations of the input voltage values were narrow enough to use a simple mapping function between the switching number and the input voltages for both of the MCEY1 embroidered switches and the MCEY2 embroidered switches. Therefore, they showed a complete success rate on switching performance. The result also shows that the MCEY2 yarn with the additional 20 denier polyester had enhanced yield strength and did not interrupt electric contact between the MCEY thread and conductive fabric.

Figure 10.

Input voltage mapping function and sensing scheme of MCEY embroidered switches based on the one-point sensing method for the resistive embroidery made of (a) MCEY1 and (b) MCEY2.

Figure 11 shows the simple mapping functions between the number of the MCEY embroidered switches and the values of the input voltages which were transferred while pressing one point or two points designated for switching button numbers by using the two-point sensing method illustrated in Figure 3. As illustrated in Figure 3, if no touch is occurring there is no path for current to ground so both measurement nodes must be at a high potential. The mean values and standard errors of 20 input voltage measurements at each number are plotted. Figures 11(a) and (c) are the results of the voltage input transferred to the left, V_in(L) (Figures 7 and 9), while pressing a button at a time from one to six on the two-point sensing circuits. The voltage input values did not change linearly along the resistive trace as expected from equation (2) and circuit diagram in Figure 9(b). But the standard deviations of the voltage input values corresponding to the switching number from one to six were narrow enough to use this simple mapping function between them for both of the MCEY1 embroidered switches and the MCEY2 embroidered switches.

Figure 11.

Input voltage mapping function and sensing scheme of MCEY embroidered switches based on two-point sensing method: (a) one-point sensing switches and (b) two-point sensing switches using the MCEY1 resistive embroidery; and (c) one-point sensing switches and (d) two-point sensing switches using the MCEY2 resistive embroidery.

Figured 11(b) and (d) represent the results of the two-point sensing performance of the MCEY1 embroidered switches and MCEY2 embroidered switches while pressing the number one and another number from two to six at the same time. When they touched, the number one was mapped by using the voltage input to the left and the other number touched was mapped by using the voltage input to the right. As illustrated in Figure 3, when two different switching buttons are touched there can be no potential difference between them as they are both shorted together and connected to ground. The two outer buttons touched were easily mapped by using the voltage input data to the left and right of the two contact points.

Figure 12 shows a controller display demonstrating the sensing and switching performance of the MCEY2 embroidered switches with six buttons using the two-point sensing method. The maximum voltage of the AD converter in the microcontroller was set to 2.8 V because the voltage input values from two-point sensing were below 2.8 V. Thus, a value of 0 is 0 V incoming and 255 is over 2.8 V incoming. The digital input value of the input voltage transferred to the left, V_in(L), was displayed on the left scoreboard of the controller and the digital input value of the input voltage transferred to the right, V_in(R), was displayed on the right scoreboard of the controller. And the six LEDs display the switching performance of the six button switches. There was a complete success rate during testing.

Figure 12.

Displays of switching performance: (a) no contact, (b) one-point sensing while pressing button number one and (c) while pressing button number four, and (d) two-point sensing while pressing both one and four.

Conclusions

In this paper, the methods for fabricating robust, reliable and low-cost textile touch sensors was successfully developed by using novel MCEYs with a computer numerical control embroidery process. Their feasibility as an accurate positional indicator was demonstrated by testing textile switches with multiple buttons. An energy efficient method of circuit design was also successfully presented. The structures of the MCEYs were highly effective for reliable touch sensing performance. MCEY2 has much more electrical regularity due to its higher yield strength than MCEY1; thus, MCEY2 may be a better choice for smart textile systems due to its greater reliability.

Textile touch sensors using the simplified design and fabrication method proposed in this paper can be easily integrated into a variety of textile products from jackets and jeans to textile accessory items to make these products interactive in new, creative ways for individual users. With the machine CNC embroidery method, a resistive path can be precisely constructed on a textile substrate and repairs to broken conductive yarn may be easily carried out. The geometry of a sensor can be easily customized by an individual designer.

Further studies will be carried out to investigate large surface area sensing fabrics, stretchable sensing surfaces for softer and more comfortable tangible interfaces, and interconnection and integration within a whole smart textile system, among other applications.

Footnotes

Funding

This work was supported by the Ministry of Knowledge Economy (MKE), Rep. of Korea, under the IT R&D program supervised by the KOREA Evaluation Institute of Industrial Technology (KEIT) (10041059).

References

Inaba M, Hoshino Y, Nagasaka K, Ninomiya T, Kagami S and Inoue H. A full-body tactile sensor suit using electrically conductive fabric and strings. In: IEEE/RSJ international conference on intelligent robots and systems (IROS 96), Osaka, Japan, 8 November 1996, vol. 2, pp. 450–457.

Pan Z, Cui H and Zhu Z. A flexible full-body tactile sensor of low cost and minimal connections. In: IEEE international conference on systems, man, and cybernetics (SMC 03), Washington DC, 5–8 October 2003, vol. 3, pp. 2368–2373.

Glaser R, Lauterbach C, Savio D, Schnell M, Karadal S, Weber W, Kornely S and Stohr A. Smart Carpet: A textile-based large-area sensor network, http://www.futureshape.de/publications_lauterbach/SmartFloor2005. pdf (2005, accessed 30 May 2012).

Graumann D, Raffa G and Quirk M. Large surface area electronic textiles for ubiquitous computing: A system approach. In: Fourth annual international conference on mobile and ubiquitous systems: Networking and services (MobiQuitous 2007), Philadelphia, PA, August 2007, vol. 4, pp. 1–8.

Freed A. Novel and forgotten current-steering techniques for resistive multitouch, duotouch and polytouch position sensing with pressure. In: New interfaces for musical expression (NIME 09), Pittsburgh, PA, 3–6 June 2009.

Sergio

Manaresi

Tartagni

Canegallo

. A textile based capacitive pressure sensor. Sens Lett 2004; 2(2): 153–160.

Meyer

Arnich

Schumm

. Design and modeling of a textile pressure sensor for sitting posture classification. IEEE Sens J 2010; 10(8): 1391–1398.

Post

Orth

Russo

Gershenfeld

. E-broidery: Design and fabrication of textile-based computing. IBM Syst J 2000; 39(3–4): 840–860.

Swallow

Thompson

. Sensory fabric for Ubiquitous Interfaces. Int J Hum–Comput Int 2001; 13(2): 147–159.

10.

Roh J, Mann Y, Freed A and Wessel D. Robust and reliable fabric, piezoresistive multitouch sensing surfaces for musical controllers. In: The 2011 international conference on new interfaces for musical expression (NIME2011), Oslo, Norway, 30 May–1 June 2011, pp. 393–398.

11.

Cadogan DP and Shook LS. Manufacturing and performance assessments of several applications of electrotextiles and large-area flexible circuits. In: Materials research society symposia 2002, San Francisco, CA, 2–5 April 2002, pp. 49–60.

12.

Eleksen. ElekTex textile buttons & switches, http://www.eleksen.com/?page=products/elektexproducts/buttonswitches.php (accessed 30 May 2012).

13.

Eleksen. ElekTex programmable textile touchscreen, http://www.eleksen.com/?page=products/elektexproducts/touchscreen.php (accessed 30 May 2012).

14.

Roh

Chi

Kang

. Wearable textile antennas. Int J Fashion Design Technol Educ 2010; 3(3): 135–153.

15.

Sundaresan

Hari

Salhotra

. Strength reduction of sewing threads during high speed sewing in an industrial lockstitch machine: Part I—Mechanism of thread strength reduction. Int J Cloth Sci Tech 1997; 9: 334–345.

16.

Roh

Chi

Lee

Nam

Kang

. Characterization of embroidered inductors. Smart Mater Struct 2010; 19(11): 115020–115020.

17.

Roh J. Manufacturing method of metal composite yarn with enhanced yield strength, electrically conductive metal composite yarn with enhanced yield strength and embroidered circuit using thereof. Patent application 10-2013-0044843, Korea, 2013.

18.

Kuphaldt TR. DC circuit experiments—Basic electronic tutorials; Chapter 3: DC circuits. Lessons in Electric Circuits 2000–2002, http://www.opamp-electronics.com/tutorials/experiments_ch_003.htm (2002, accessed 30 May 2012).

19.

Koch R. Membrane potentiometer: A competitive alternative to traditional linear and rotary position sensors, Hoffmann + Krippner Inc, http://www.laboratoryequipment.com/uploadedFiles/Resources/White_Papers/021210_Membrane_Potentiometer_USA_JAN2010.pdf (2010, accessed 30 May 2012).

20.

Shieldex® Kassel technical data sheet, http://www.statex.biz/datasheets/Kassel.pdf (2008, accessed 30 May 2012).

21.

ASTM D 2256. Test method for breaking load and elongation of yarn by the single strand method.