sensors Communication Electromagnetic Encoders Screen-Printed on Rubber Belts for Absolute Measurement of Position and Velocity Ferran Paredes 1, * , Cristian Herrojo 1 , Ana Moya 2 , Miguel Berenguel Alonso 2 , David Gonzalez 3 , Pep Bruguera 3 , Claudia Delgado Simao 2 and Ferran Martín 1 1 CIMITEC, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; cristian.herrojo@uab.es (C.H.); ferran.martin@uab.es (F.M.) 2 Eurecat, Centre Tecnològic de Catalunya, Parc Científic i de la Innovació TecnoCampus, 08302 Mataró, Spain; ana.moya@eurecat.org (A.M.); miguel.berenguel@eurecat.org (M.B.A.); claudia.delgado@eurecat.org (C.D.S.) 3 Hohner Automation, P.I. Cal Batlle, 17400 Breda, Spain; dgonzalez@encoderhohner.com (D.G.); bruguera@encoderhohner.com (P.B.) * Correspondence: ferran.paredes@uab.cat Abstract: This paper presents, for the first time, an absolute linear electromagnetic encoder consisting of a rubber belt with two chains of screen-printed metallic inclusions (rectangular patches). The position, velocity, and direction of the belt (the moving part) is determined by detecting the inclu- sions when they cross the stator (the static part). The stator is a microstrip line loaded with three complementary split ring resonators (CSRRs), resonant elements exhibiting a resonance frequency perturbed by the presence of inclusions on top of them (contactless). The line is fed by three har- monic signals tuned to the resonance frequencies of the CSRRs. Such signals are generated by a   voltage-controlled oscillator (VCO) managed by a microcontroller. The sensed data are retrieved from Citation: Paredes, F.; Herrojo, C.; the pulses contained in the envelope functions of the respective amplitude modulated (AM) signals Moya, A.; Berenguel Alonso, M.; (caused by the belt motion) generated at the output port of the line. One of the signals provides the Gonzalez, D.; Bruguera, P.; Delgado absolute belt position, determined by one of the chains, the encoded one. The information relative to Simao, C.; Martín, F. Electromagnetic the velocity and motion direction is contained in the other AM signals generated by the motion of the Encoders Screen-Printed on Rubber other chain, periodic, and thereby, uncoded. The spatial resolution of the system, a figure of merit, is Belts for Absolute Measurement of 4 mm. Special emphasis is devoted to the printing process of the belt inclusions. Position and Velocity. Sensors 2022, 22, 2044. https://doi.org/10.3390/ Keywords: electromagnetic encoders; screen-printing; motion control; microstrip technology s22052044 Academic Editors: Yong Zhu and Ali Bazaei 1. Introduction Received: 4 February 2022 Accepted: 4 March 2022 Motion control is today present in a wide variety of industrial scenarios, such as Published: 5 March 2022 robotics, the automotive and aeronautics industry, the textile industry, logistics, the food industry, mechatronics, space, and medical instrumentation, to cite some examples. In Publisher’s Note: MDPI stays neutral any motion control system, a sensor is used to provide the necessary feedback signal to with regard to jurisdictional claims in the controller and actuator (e.g., a step motor), in order to eventually compensate for published maps and institutional affil- any potential motion error. Proximity, displacement (either linear or angular), velocity, iations. and acceleration are typical motion variables that need to be retrieved in many applica- tions. For that purpose, there are a wide variety of motion sensors on the market. In particular, optical encoders (either linear or angular) are used in many industrial systems Copyright: © 2022 by the authors. (e.g., servomechanisms, conveyor belts, pointing mechanisms, reaction wheels, elevators, Licensee MDPI, Basel, Switzerland. etc.) [1–3]. This article is an open access article Optical encoders, although very accurate, exhibit an intrinsic drawback, i.e., a limited distributed under the terms and robustness against the effects of dirtiness, grease, and pollution (present in most industrial conditions of the Creative Commons systems). By contrast, electromagnetic (or microwave) encoders, reported recently [4–10], Attribution (CC BY) license (https:// are tolerant against such pernicious effects, since these devices utilize microwaves for creativecommons.org/licenses/by/ sensing. Magnetic encoders, proposed in [11,12], are also robust against hostile environ- 4.0/). ments, but such sensors are relatively complex, since they need inductive coils. In the Sensors 2022, 22, 2044. https://doi.org/10.3390/s22052044 https://www.mdpi.com/journal/sensors Sensors 2022, 22, 2044 2 of 8 electromagnetic encoder systems reported so far, the movable element (a dielectric disk in rotary encoders and a dielectric tape in linear encoders) should be attached to the moving target, e.g., a shaft (rotary encoders) or a conveyor belt (linear encoders). In this paper, we investigate the possibility of using the moving target as the encoder, provided it is made of a dielectric material, in particular, rubber. This is of interest, since many industrial and logistic systems utilize belts made of rubber (e.g., conveyor belts, elevator belts, etc.). Thus, the encoder presented in this paper is implemented by screen-printing (using commercial conductive inks) chains of rectangular patch inclusions on a rubber belt, specifically a commercial elevator belt. Nevertheless, besides the implementation of the encoder in rubber material by means of conductive ink using screen-printing (for the first time), there are other aspects that represent an innovation with regard to the previously proposed encoders. Probably, the most important one concerns the fact that in the synchronous system proposed in this paper, where the encoder velocity, quasi-absolute position, and motion direction can be determined, the three harmonic signals needed for that purpose are generated by means of a VCO managed by a microcontroller. Moreover, the microcontroller processes and separates the three signals providing that information. In the encoders presented in [8–10], a synchronous reading is also proposed, but the proof-of-concept demonstrators are based on obtaining the required (envelope function) signals by injecting the necessary harmonic signals sequentially, so that each signal is inferred at different times, rather than from the processing action of the microcontroller. On the other hand, the encoder presented in [7] exhibits very good resolution, thanks to the use of transversely oriented metallic strips, but it does not operate synchronously. However, in the paper [10], the system works similarly to [7], but it is able to provide the direction of motion and operates synchronously (nevertheless, the resolution is not as good as in [7]). For that purpose, the reader used three double stub pairs, thereby representing a significant size for the reader. Concerning the encoder system presented in [8], it was the precursor of the one proposed in this paper, but we have demonstrated for the first time system its functionality by including the necessary elements for retrieving the signals containing the relevant information, i.e., the VCO, the microcontroller, and the associated circuitry. Moreover, in [8], the encoder was implemented in a rigid microwave substrate by means of a drilling machine. Let us finally mention, in regard to other encoders available in the literature, that those reported in [4,5] are rotary, rather than linear, and are not absolute encoders but incremental-type encoders. Nevertheless, in the prototype presented in [5], the direction of motion (clockwise or counterclockwise) can be detected. There are many other systems for the measurement of displacements, velocities, and proximity. For example, there are sensors based on coupling modulation [13–17] and frequency variation [18], where a resonant (movable) element is displaced over a transmission line based structure (typically, the resonator in motion is displaced in a plane parallel to the one of the lines, but there are also realizations, where the resonator displaces vertically [17,19]). However, these systems offer a very limited input dynamic range, as compared to linear electromagnetic encoders, where such dynamic range is potentially unlimited. On the other hand, there many displacement and proximity sensors based on the Hall Effect [20–25], and there are commercially available devices [26]. In general, these devices offer excellent resolution, even by considering significant input dynamic ranges, but they are based on the use of magnets. The system proposed in this paper cannot compete against optical encoders and Hall effect sensors in terms of resolution, but, in many applications, the resolutions demonstrated in this paper suffice, in particular in industrial scenarios (elevators, conveyor systems, etc.). 2. Sensor System The proposed linear displacement and velocity sensor consists of two parts: the movable belt with two screen-printed chains of rectangular metallic inclusions (the encoder) and a planar microwave structure (stator) able to detect the inclusions when the belt is in Sensors 2022, 22, x FOR PEER REVIEW 3 of 9 2. Sensor System Sensors 2022, 22, 2044 The proposed linear displacement and velocity sensor consists of two parts: the mov- 3 of 8 able belt with two screen-printed chains of rectangular metallic inclusions (the encoder) and a planar microwave structure (stator) able to detect the inclusions when the belt is in relative relative motion motionwith withregard regardtotothethestator. stator.TheThesketch sketch ofof the complete the complete system system is depicted is depictedin Figure 1. One of the encoder chains is periodic and is used to determine in Figure 1. One of the encoder chains is periodic and is used to determine the velocity the velocity (ve- locity chain). (velocity The The chain). other chain other is encoded, chain that is, is encoded, thatcertain inclusions is, certain are present inclusions at the at are present same the axial samepositions of the of axial positions inclusions of the of the inclusions velocity chain (‘1′ the velocity logic chain (‘1’state), whereas logic state), other inclu- whereas other sions are not inclusions present are (‘0′ state). not present (‘0’ With state).thisWith chain (position this chain), it is chain (position possible chain), it istopossible determine to the absolute determine position the absolute ofposition the encoder of thebyencoder “reading” the state (bit) by “reading” corresponding the state to that po- (bit) corresponding to that position. That bit, plus the bits corresponding to the previous N − 1 positions, provide sition. That bit, plus the bits corresponding to the previous N − 1 positions, provide a unique a unique code codethat thatidentifies identifiesunivocally univocallythe theposition positionofofthe theencoder. encoder.In Inorder order to to guarantee guarantee that that any any subset subset of of N N bits bits does does not not repeat, repeat, itit is is necessary necessary to to encode encode the the whole whole position position chain chain following the following the De De Bruijn Bruijn sequence sequence [27]. [27]. The The variable variable N N is is the the number number of of bits bits necessary necessary to to univocally determine univocally determine the the different different positions positions of of the the encoder encoder (in (in turn turn determined determined by by the length of length the encoder, L, and by the spatial resolution, p, or chain period), i.e., L/p << 2NN.. of the Note also Note also that that the the instants instants ofof time time necessary necessary to to read read the the position position chain chain areare determined determined by by the velocity chain (also called clock chain). the Amplifier DAC Microcontroller 1 1 0 0 x3 12-bits ID 1 1 1 1 CSRRc CSRRd CSRRp Direct. VCO 1 1 1 1 Clock Vin V’in Env. Det. V3 V2 V1 ts 2ts 3ts t p ts 2ts 3ts t Clock and ID Direction Figure 1. Figure Sketch of 1. Sketch of the the encoder encoder sensor sensor system. system. The static The static part partofofthe system the systemis aismicrostrip a microstrip transmission transmissionline loaded with three line loaded withCSRRs three tuned to different frequencies (see Figure 2). In order to determine CSRRs tuned to different frequencies (see Figure 2). In order to determine the belt velocity, the belt velocity, it is necessary to inject a harmonic signal into the input port of the line it is necessary to inject a harmonic signal into the input port of the line tuned to the reso-tuned to the resonance frequency nance of the resonator frequency identified of the resonator as CSRRas identified c . This CSRRfrequency, f 0,c , varies c. This frequency, f0,cwhen a patch , varies when of a the velocity chain is contactless on top of the CSRR c . The presence patch of the velocity chain is contactless on top of the CSRRc. The presence of the patch of the patch modifies the transmission coefficient at f 0,c (see Figure 2c). Consequently, by encoder motion, an modifies the transmission coefficient at f0,c (see Figure 2c). Consequently, by encoder mo- amplitude modulated (AM) signal is generated at the output port of the line, and the tion, an amplitude modulated (AM) signal is generated at the output port of the line, and instantaneous velocity of the belt is determined from the time lapse between adjacent peaks the instantaneous velocity of the belt is determined from the time lapse between adjacent in the envelope function of that AM signal. The resonator designated as CSRRp is devoted peaks in the envelope function of that AM signal. The resonator designated as CSRRp is to determining the ID code of the position chain, and for that purpose, a harmonic signal devoted to determining the ID code of the position chain, and for that purpose, a harmonic tuned to f 0,p , the resonance frequency of the unloaded CSRRp , is required. Finally, the signal tuned to f0,p, the resonance frequency of the unloaded CSRRp, is required. Finally, resonator called CSRRd , with resonance frequency f0 ,d , is also sensitive to the velocity chain, the resonator called CSRR d, with resonance frequency f0,d, is also sensitive to the velocity but it is located in such a way that the generated envelope function is either lag or lead chain, but it is located in such a way that the generated envelope function is either lag or with regard to the envelope function of the AM signal with carrier frequency f 0,c . Thus, it lead with regard to the envelope function of the AM signal with carrier frequency f0,c. is possible to determine the motion direction. Thus, it is possible to determine the motion direction. For the generation of the harmonic signal, the HMC391LP4 VCO (Analog Devices, For theMA, Norwood, generation USA) was of used. the harmonic The control signal, the of voltage HMC391LP4 this component VCO is(Analog managed Devices, by the from Massachussets (United States)) was used. The control voltage ATmega328 microcontroller (Atmel Corporation, San Jose, CA, USA) (Arduino development of this component is managed by the ATmega328 microcontroller (Atmel Corporation, platform). As shown in Figure 1, the microcontroller output was connected to the 12-bit from California (United States)) MCP4725 (Arduino development DAC (Sparkfun platform). Electronics, As shown Boulder, in Figure CO, USA), 1, theby followed microcontroller out- an amplifier stage, put was connected to the 12-bit MCP4725 DAC (Sparkfun Electronics, in order to cover the full operation range of the VCO. A linear displacement system (model from Colorado Thorlabs LTS300/M (Thorlabs Inc., Newton, NJ, USA)) was employed to displace the encoder over the stator. The air gap, the separation between the stator and encoder, was set to 1 mm. Finally, the ADL5511 envelope detector was connected between the output port of Sensors 2022, 22, x FOR PEER REVIEW 4 of 9 (United States)), followed by an amplifier stage, in order to cover the full operation range of the VCO. A linear displacement system (model Thorlabs LTS300/M (Thorlabs Inc, from Sensors 2022, 22, 2044 4 of 8 Newton (United States))) was employed to displace the encoder over the stator. The air gap, the separation between the stator and encoder, was set to 1 mm. Finally, the ADL5511 envelope detector was connected between the output port of the stator and the input port the stator of the and the inputThe microcontroller. portphotograph of the microcontroller. The photograph of the experimental of the experimental setup is depicted in Figure 3. setup is depicted in Figure 3. Sensors 2022, 22, x FOR PEER REVIEW 5 of 9 as mentioned later). However, it should be mentioned that depending on the microcon- troller, the time ts can be reduced to few μs. Thus, in this case, the velocities of the encoder can Figurebe,2.2.atPhotograph Figure least, two orders Photograph of ofthe of thestator.magnitude stator. (a)(a) Top Top higher. and (b)(b) and bottom. bottom.(c) The frequency (c) The response frequency of theofstator response the stator when the patches rely on the top of each resonator. The stator was fabricated on RO4003C when Itthe patches should be rely on the mentioned top of that each the resonator. frequencies The stator of the was fabricated harmonic on signals a Rogers necessary for a Rogers substrate system with a thickness functionality were of chosen h = 0.81 mm,in dielectric the vicinity constant of 4 GHzεr =as 3.38, a and loss factor tradeoff. That tanδ is, = 0.0027. reducing RO4003C substrate with a thickness of h = 0.81 mm, dielectric constant εr = 3.38, and loss factor Thefrequency the dimensionsincreases are: transmission the size linethewidths w1 = 6.4 of mm theand w2 =part 1.9 mm; CSRR widths wr = 2.9 tan δ = 0.0027. The dimensions are: of resonators transmission line widths static w of the sensor, 1 = 6.4 mm and w2 = 1.9 mm; CSRR or reader mm (for the three resonators); resonators lengths lc = ld = 10.5 mm and lp = 14.5 mm; ring splits sd = (i.e., the widths wrssize = 2.9ofmm the CSRRs) (for the threeand, consequently, resonators); resonatorsthelengths size oflcthe ldpatches, =mm. =The with and alpc,penalty 10.5sub-index mm =d,14.5 in 0.4 mm, c = 1.6 mm and sp = 6.2 mm; and CSRR slot width c = 0.5 andmm; p in terms ring of the variables encoder resolution. splits sd =corresponding 0.4 mm, sc = 1.6 Thus, to mm for sensor and sp = 6.2length the resonator’s resolution, mm; andand CSRR operation slot width ring splits at are used high c = to frequencies 0.5differentiate mm. The sub- is the preferred, index CSRRs. but, in this case, all the associated sensor electronics were more c, d, and p in the variables corresponding to the resonator’s length and ring splits are used to complex and expensive. the CSRRs. differentiate The procedure to generate the signals and process the data was as follows. First, the microcontroller set a specific control voltage V1 (so that the VCO generated a tone at f0,p) and read the output signal of the envelope (AM) detector. After the sample period, ts = 1.25 ms, the control voltage was changed to V2 (corresponding to the unmodulated tone f0,d), and the procedure was repeated (the microcontroller retrieved the signal given by the AM detector) until the end of the sample period, 2ts. Finally, at 2ts, the voltage was set to V3 (with the VCO giving the tone at f0,c), and this voltage was kept until t = 3ts. This se- quence was constantly repeated while the encoder was in motion, and the three AM mod- ulated signals (envelope functions) were obtained. Note that the total sweep time was 3ts = 3.75 ms. Let us assume that it is reasonable that the time period between the crossing of two adjacent patches above the resonators was at least 10 times higher than 3ts = 3.75 ms (this is a conservative value taking into account the sampling Nyquist criterion). This gives 37.5 ms, and, taking into account that the distance between adjacent patches (period) in the considered encoder was 4 mm, the maximum encoder velocity was vmax = 4 mm/37.5 ms = 106 mm/s (note that in our case, the considered encoder velocity was set to 40 mm/s, Figure 3. Photograph of the experimental setup. Figure 3. Photograph of the experimental setup. 3. Encoder Design and Fabrication The encoders were implemented by screen-printing the two chains of patch inclu- sions on a rubber belt. The considered belt had a steel core in order to provide mechanical strength against possible tensile stress. Nevertheless, the presence of the steel core did not seem to have an appreciable effect on sensor functionality, particularly in the envelope Sensors 2022, 22, 2044 5 of 8 The procedure to generate the signals and process the data was as follows. First, the microcontroller set a specific control voltage V 1 (so that the VCO generated a tone at f 0,p ) and read the output signal of the envelope (AM) detector. After the sample period, ts = 1.25 ms, the control voltage was changed to V 2 (corresponding to the unmodulated tone f 0,d ), and the procedure was repeated (the microcontroller retrieved the signal given by the AM detector) until the end of the sample period, 2ts . Finally, at 2ts , the voltage was set to V 3 (with the VCO giving the tone at f 0,c ), and this voltage was kept until t = 3ts . This sequence was constantly repeated while the encoder was in motion, and the three AM modulated signals (envelope functions) were obtained. Note that the total sweep time was 3ts = 3.75 ms. Let us assume that it is reasonable that the time period between the crossing of two adjacent patches above the resonators was at least 10 times higher than 3ts = 3.75 ms (this is a conservative value taking into account the sampling Nyquist criterion). This gives 37.5 ms, and, taking into account that the distance between adjacent patches (period) in the considered encoder was 4 mm, the maximum encoder velocity was vmax = 4 mm/37.5 ms = 106 mm/s (note that in our case, the considered encoder velocity was set to 40 mm/s, as mentioned later). However, it should be mentioned that depending on the microcontroller, the time ts can be reduced to few µs. Thus, in this case, the velocities of the encoder can be, at least, two orders of magnitude higher. It should be mentioned that the frequencies of the harmonic signals necessary for system functionality were chosen in the vicinity of 4 GHz as a tradeoff. That is, reducing the frequency increases the size of the resonators of the static part of the sensor, or reader (i.e., the size of the CSRRs) and, consequently, the size of the patches, with a penalty in terms of encoder resolution. Thus, for sensor resolution, operation at high frequencies is preferred, but, in this case, all the associated sensor electronics were more complex and expensive. 3. Encoder Design and Fabrication The encoders were implemented by screen-printing the two chains of patch inclusions on a rubber belt. The considered belt had a steel core in order to provide mechanical strength against possible tensile stress. Nevertheless, the presence of the steel core did not seem to have an appreciable effect on sensor functionality, particularly in the envelope detector (as will be shown in Section 4). It is believed that the rubber material present between the steel core and the patches sufficed to minimize the effects of the steel core. The length, width, and thickness of the belt were 20 cm, 3 cm, and 3 mm, respectively. The metallic patches were designed, along with the stator, considering the maximum printable width of the belt (3 cm), and taking into account that two chains must be assembled. The screen-printing process consisted of the deposition of pastes or inks printed on a substrate through a screen. The screen, based on polyester threads, was placed at a certain distance from the substrate, and it contained the layout to be printed. When a scraper was pressed down, the screen reached the substrate, depositing the ink. A screen- printing technique must consider three factors: (i) the layout to be printed in terms of the resolution of strips and slots, (ii) the desired conductivity of the ink when printed, and (iii) the substrate. In the present design, the resolution of the rectangular inclusions was not critical. However, it was important to determine the adhesion of the ink to the substrate, in this case the rubber belt, as well as the conductivity of the silver ink. Thus, previous to implementing the inclusions, some patterns consisting of strips of different widths and length were considered, in order to measure the conductivity and to verify the adhesion of the ink. The patterns were screen-printed using the Norcote ELG conductive silver ink. Then the ink was UV cured by a halogen light (5 s at 500 W), and the printed belt was finally introduced in an air flow oven (10 min at 130 ◦ C) to cause thermal curing. The ink was correctly adhered to the rubber belt; therefore, the conductivity was measured (σ ≈ 1 × 106 S/m), and this value ensured the system functionality. Finally, the metallic inclusions were screen-printed using the same process, and the fabricated encoder was depicted in Figure 4. the ink was UV cured by a halogen light (5 s at 500 W), and the printed belt was finally introduced in an air flow oven (10 min at 130 °C) to cause thermal curing. The ink was correctly adhered to the rubber belt; therefore, the conductivity was measured (σ ≈ 1.106 S/m), and this value ensured the system functionality. Finally, the metallic inclusions were Sensors 2022, 22, 2044 screen-printed using the same process, and the fabricated encoder was depicted in Figure6 of 8 4. Figure4.4. Photograph Photograph of of the the 48-bits 48-bits encoder encoderwith withclock clockchain. chain.The Thedimensions are: dimensions d1 d are: = 11.5 mm; d2 = Figure 1 = 11.5 mm; 15.9 mm, w = 3 mm, s = 1 mm, and g = 1.9 mm. The chain period is p = 4 mm. Note that d2 = 15.9 mm, w = 3 mm, s = 1 mm, and g = 1.9 mm. The chain period is p = 4 mm. Note that L/pL/p = 48; =thus, 48; N = 6. thus, N = 6. 4.4.Results Results Thefunctionality The functionality of the proposed proposed encoder encodersystemsystemwas wasvalidated validated byby means means of of thethe ex- perimental setup experimental setupofofFigure Figure3.3.The Theencoder encoderwas wasdisplaced displacedatata avelocity velocityofofvv==4040mm/smm/s in the in theupwards upwards direction. direction. Figure Figure5 shows 5 shows thethe measured measured signal signalat the at theoutput outputport of the port en- of the velope (AM) envelope (AM)detector, detector,which which contained contained the the three AM AM demodulated demodulated signals signals(envelope (envelope functions). The functions). The microcontroller microcontroller was was inin charge charge of ofprocessing processingand andseparating separatingthe thesignals signals (Figure (Figure6). Agood 6). A goodsynchronism synchronismbetween between thethe clock clock andand thethe ID code ID code waswas visible. visible. The The direc- direction tion signal signal was delayed was delayed with regard with regard to the to the signal, clock clock signal, thus indicating thus indicating that thethat the encoder encoder moved movedupwards. upwards. The separation The separation between between peaks peaks was 100 wasms,100which ms, which corresponded corresponded to an to an encoder encoder velocity velocity of 40of 40 mm/s, mm/s, since since the period the period of theof the metallic metallic inclusions inclusions (resolution) (resolution) was p was p = 4 mm. Finally, the peaks in the ID code signal perfectly = 4 mm. Finally, the peaks in the ID code signal perfectly correlated with the presence correlated with the of presence rectangular of rectangular patches in the patches in the position chainposition chain4).(see (see Figure WithFigure these4). Withthe results, these results, functional- the ity functionality of the system of thevalidated. was system was validated. Nevertheless, Nevertheless, it can be it can be appreciated appreciated that that the the discrimination discrimination capability of the capability systems of thelimited was systems wasthe since limited since voltage the voltage difference difference between between the logic states the logic states ‘0 0 and ‘10 was not very significant. One possibility to improve this aspect ‘0′ and ‘1′ was not very significant. One possibility to improve this aspect is to include an Sensors 2022, 22, x FOR PEER REVIEWis to include an amplifier stage after the VCO. By doing this, the difference will increase, 7 of 9 amplifier stage after the VCO. By doing this, the difference will increase, thereby favoring thereby favoringHowever, discrimination. discrimination. However, this aspect thisfor was left aspect was work. a future left for a future work. Another important aspect that determined the functionality of the system was the vertical distance between the patches of the encoder and the static part (reader), particu- larly the CSRRs. The patches should be located at a maximum vertical distance (or air gap) that guarantees that such patches perturb the resonance frequency of the resonant ele- ments (CSRRs) of the static part of the sensor when they are on top. In our case, the nom- inal distance was 1 mm, but it was concluded from electromagnetic simulations that tol- erances in the air gap between 0.5 mm and 1.5 mm were compatible with system func- tionality. Indeed, the proposed system, despite this nominal air gap, was subjected to cer- tain variations in the air gap distance, unavoidable in practice. However, it should be mentioned that as the air gap increased, the modulation index decreased, thereby having a negative impact on the capability of the system to discriminate between the ‘0′ and the Figure Figure5.5.Measured signal at the output of the envelope (AM) detector. ‘1′ logic Measured states. signal at the output of the envelope (AM) detector. Another important aspect that determined the functionality of the system was the vertical distance between the patches of the encoder and the static part (reader), particularly the CSRRs. The patches should be located at a maximum vertical distance (or air gap) that guarantees that such patches perturb the resonance frequency of the resonant elements (CSRRs) of the static part of the sensor when they are on top. In our case, the nominal dis- tance was 1 mm, but it was concluded from electromagnetic simulations that tolerances in the air gap between 0.5 mm and 1.5 mm were compatible with system functionality. Indeed, the proposed system, despite this nominal air gap, was subjected to certain variations in the air gap distance, unavoidable in practice. However, it should be mentioned that as the air gap increased, the modulation index decreased, thereby having a negative impact on the capability of the system to discriminate between the ‘0’ and the ‘1’ logic states. Sensors 2022, 22, 2044 7 of 8 Figure 5. Measured signal at the output of the envelope (AM) detector. Figure 6. Figure Measured envelope 6. Measured envelopefunction functionfor forthe theclock, clock,direction, direction,and andposition positionsignals signalsafter afterthe themicro- micro- controller controller separated separated the the signal signal of Figure 5. 5. Conclusions 5. Conclusions In conclusion, In conclusion, aa linear linear displacement displacement andand velocity velocity sensor sensor based based onon an an electromagnetic electromagnetic encoder was proposed. The system was able to provide the absolute position andand encoder was proposed. The system was able to provide the absolute position the the in- instantaneous velocity of the movable part (the encoder) with regard to the stantaneous velocity of the movable part (the encoder) with regard to the static part (sta- static part (stator). tor). SuchSuch functionality functionality was achieved was achieved by designing by designing the encoder the encoder with twowith twoofchains chains of metallic metallic patches. One chain was periodic, with all the patches present at their patches. One chain was periodic, with all the patches present at their predefined positions. predefined positions. The other chain was provided with a certain identification code (achieved by The other chain was provided with a certain identification code (achieved by the presence the presence or absence of patches at their predefined positions in the chain). This code or absence of patches at their predefined positions in the chain). This code followed the followed the so-called De Bruijn sequence, where any subset of N bits did not repeat along so-called De Bruijn sequence, where any subset of N bits did not repeat along the wholeN the whole encoder, N being the number of bits necessary to univocally determine the 2 encoder, N being the number of bits necessary to univocally determine the 2N different different positions of the encoder belt (N = 6, in our case). Nevertheless, the main relevant positions of the encoder belt (N = 6, in our case). Nevertheless, the main relevant aspect of aspect of the present contribution concerned the implementation of the encoder on a belt the present contribution concerned the implementation of the encoder on a belt made of made of rubber, by means of screen-printing, using commercial conductive inks. Despite rubber, by means of screen-printing, using commercial conductive inks. Despite the lim- the limited electromagnetic characteristics of the considered materials, as compared to ited electromagnetic characteristics of the considered materials, as compared to those of those of previous electromagnetic encoders, typically implemented on rigid microwave previous electromagnetic encoders, typically implemented on rigid microwave sub- substrates, or on flexible polymeric substrates, the functionality of the system was validated strates, or on flexible polymeric substrates, the functionality of the system was validated by considering a displacement of the encoder at constant velocity. It can be envisaged by considering that a displacement the proposed encoders may of the encoder find at constant application velocity. in many It cansystems industrial be envisaged that equipped the proposed encoders may find application in many industrial systems with dynamic (movable) elements made of rubber, such as conveyor or elevator belts, equipped with dynamic (movable) elements made of rubber, such as conveyor or elevator belts, among among others. others. Author Contributions: Conceptualization, F.P., C.H. and F.M.; methodology, A.M., F.P., C.H.; Design, F.P., C.H.; fabrication A.M., M.B.A. and C.D.S.; software, D.G.; experimental validation, D.G., P.B., F.P. and C.H.; paper writing, F.P., F.M.; general supervision, P.B., C.D.S. and F.M. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by MCIN/AEI 10.13039/501100011033, Spain, through the projects PID2019-103904RB-I00, RTC-2017-6303-7 (ERDF European Union) and PDC2021-121085- I00 (European Union Next Generation EU/PRTR), by the AGAUR Research Agency, Catalonia Government, through the project 2017SGR-1159, and by Institució Catalana de Recerca i Estudis Avançats (who awarded Ferran Martín). Conflicts of Interest: The authors declare no conflict of interest. Sensors 2022, 22, 2044 8 of 8 References 1. Petriu, E.M. Reconsidering natural binary encoding for absolute position measurement application. IEEE Trans. Instrum. Meas. 1989, 38, 1014–1016. [CrossRef] 2. Ueda, T.; Kohsaka, F.; Iino, T.; Kazami, K.; Nakayama, H. Optical absolute encoder using spatial filter. In Proceedings of the SPIE Photomechanics and Speckle Metrology, San Diego, CA, USA, 17 August 1987; pp. 217–221. 3. Eitel, E. Basics of Rotary Encoders: Overview and New Technologies. Machine Design Magazine, 7 May 2014. 4. Mata-Contreras, J.; Herrojo, C.; Martín, F. Application of split ring resonator (SRR) loaded transmission lines to the design of angular displacement and velocity sensors for space applications. IEEE Trans. Microw. Theory Tech. 2017, 65, 4450–4460. [CrossRef] 5. Mata-Contreras, J.; Herrojo, C.; Martín, F. Detecting the rotation direction in contactless angular velocity sensors implemented with rotors loaded with multiple chains of split ring resonators (SRRs). IEEE Sens. J. 2018, 18, 7055–7065. [CrossRef] 6. Herrojo, C.; Muela, F.; Mata-Contreras, J.; Paredes, F.; Martín, F. High-density microwave encoders for motion control and near-field chipless-RFID. IEEE Sens. J. 2019, 19, 3673–3682. [CrossRef] 7. Herrojo, C.; Paredes, F.; Martín, F. Double-stub loaded microstrip line reader for very high data density microwave encoders. IEEE Trans. Microw. Theory Tech. 2019, 67, 3527–3536. [CrossRef] 8. Paredes, F.; Herrojo, C.; Martín, F. Microwave Encoders with Synchronous Reading and Direction Detection for Motion Control Applications. In Proceedings of the IEEE MTT-S International Microwave Symposium Digest (IMS’20), Los Angeles, CA, USA, 21–26 June 2020; pp. 472–475. 9. Herrojo, C.; Paredes, F.; Martín, F. 3D-printed all-dielectric electromagnetic encoders with synchronous reading for measuring displacements and velocities. Sensors 2020, 20, 4837. [CrossRef] 10. Herrojo, C.; Paredes, F.; Martín, F. Synchronism and Direction Detection in High-Resolution/High-Density Electromagnetic Encoders. IEEE Sens. J. 2021, 21, 2873–2882. [CrossRef] 11. Zhang, Z.; Dong, Y.; Ni, F.; Jin, M.; Liu, H. A Method for Measurement of Absolute Angular Position and Application in a Novel Electromagnetic Encoder System. J. Sens. 2015, 2015, 503852. [CrossRef] 12. Zhang, Z.; Ni, F.; Dong, Y.; Jin, M.; Liu, H. A novel absolute angular position sensor based on electromagnetism. Sens. Actuators A Phys. 2013, 194, 196–203. [CrossRef] 13. Naqui, J.; Durán-Sindreu, M.; Martín, F. Alignment and position sensors based on split ring resonators. Sensors 2012, 12, 11790–11797. [CrossRef] 14. Karami-Horestani, A.; Fumeaux, C.; Al-Sarawi, S.F.; Abbott, D. Displacement sensor based on diamond-shaped tapered split ring resonator. IEEE Sens. J. 2013, 13, 1153–1160. 15. Horestani, A.K.; Naqui, J.; Abbott, D.; Fumeaux, C.; Martín, F. Two-dimensional displacement and alignment sensor based on reflection coefficients of open microstrip lines loaded with split ring resonators. Electron Lett. 2014, 50, 620–622. [CrossRef] 16. Horestani, A.K.; Abbott, D.; Fumeaux, C. Rotation sensor based on horn-shaped split ring resonator. IEEE Sens. J. 2013, 13, 3014–3015. [CrossRef] 17. Elgeziry, M.; Costa, F.; Genovesi, S. Wireless Monitoring of Displacement Using Spiral Resonators. IEEE Sens. J. 2021, 21, 17838–17845. [CrossRef] 18. Mandel, C.; Kubina, B.; Schüßler, M.; Jakoby, R. Passive chipless wireless sensor for two-dimensional displacement measurement. In Proceedings of the European Microwave Conference, Manchester, UK, 10–13 October 2011. 19. Paredes, F.; Herrojo, C.; Mata-Contreras, J.; Martín, F. Near-field chipless-RFID sensing and identification system with switching reading. Sensors 2018, 18, 1148. [CrossRef] [PubMed] 20. Jezný, J.; Čurilla, M. Position Measurement with Hall Effect Sensors. Am. J. Mech. Eng. 2013, 1, 231–235. 21. Granell, P.N.; Wang, G.; Cañon Bermudez, G.S.; Kosub, T.; Golmar, F.; Steren, L.; Fassbender, J.; Makarov, D. Highly compliant planar Hall effect sensor with sub 200 nT sensitivity. Npj Flex. Electron. 2019, 3, 3. [CrossRef] 22. Lidozzi, A.; Solero, L.; Crescimbini, F.; Di Napoli, A. SVM PMSM Drive with Low Resolution Hall-Effect Sensors. IEEE Trans. Power Electr. 2007, 22, 282–290. [CrossRef] 23. Scelba, G.; De Donato, G.; Scarcella, G.; Giulii Capponi, F.; Bonaccorso, F. Fault-Tolerant Rotor Position and Velocity Estimation Using Binary Hall-Effect Sensors for Low-Cost Vector Control Drives. IEEE Trans. Ind. Appl. 2014, 50, 3403–3413. [CrossRef] 24. Zhang, X.; Mehrtash, M.; Khamesee, M.B. Dual-Axial Motion Control of a Magnetic Levitation System Using Hall-Effect Sensors. IEEE/ASME Trans. Mechatron. 2016, 21, 1129–1139. [CrossRef] 25. Liu, G.; Chen, B.; Song, X. High-Precision Speed and Position Estimation Based on Hall Vector Frequency Tracking for PMSM With Bipolar Hall-Effect Sensors. IEEE Sens. J. 2019, 19, 2347–2355. [CrossRef] 26. Rotary and Linear Motion Sensors. Available online: https://www.rls.si/eng/hilin (accessed on 17 January 2022). 27. de Bruijn, N.G. Acknowledgement of Priority to C. Flye Sainte-Marie on the Counting of Circular Arrangements of 2n Zeros and Ones That Show Each n-Letter Word Exactly Once; EUT Report 75-WSK-06; Technological University Eindhoven: Eindhoven, The Netherlands, 1975; pp. 1–16.