Circuit Topologies for MOS-Type Gas Sensor

electronics

Article Circuit Topologies for MOS-Type Gas Sensor

Javier Cervera Gómez 1 , Jose Pelegri-Sebastia 2,* and Rafael Lajara 3

1 Electronic Engineering Department, Universitat Politècnica de València, 46730 Gandia, Spain; [email protected] 2 Research Institute for Integrated Coastal Zone Management (IGIC Institute), Universitat Politècnica de València, Campus Gandia, 46730 Gandia, Spain 3 E.T.S.E, University of Valencia, Avda Universitat S/N, 46100 Burjassot, Valencia, Spain; [email protected] * Correspondence: [email protected]

Received: 27 February 2020; Accepted: 21 March 2020; Published: 23 March 2020

Abstract: Metal Oxide Semiconductor or MOS-type gas sensors are resistive sensors which can detect different reducible or volatile gases in atmospheres with oxygen. These gas sensors have been used in different areas such as food and drink industries or healthcare, among others. In this type of sensor, the resistance value changes when it detects certain types of gases. Due to the electrical characteristics, the sensors need a conditioning circuit to transform and acquire the data. Four different electronic topologies, two different MOS-type gas sensors, and different concentrations of a gas substance are presented and compared in this paper. The study and experimental analysis of the properties of each of the designed topology allows designers to make a choice of the best circuit for a specific application depending on the situation, considering the required power, noise, linearity, and number of sensors to be used. This study will give more freedom of choice, the more adequate electronic conditioning topology for different applications where MOS-type sensors are used, obtaining the best accuracy.

Keywords: electronic nose; sensor arrays; environmental analysis; Anderson loop; Wheatstone bridge

1. Introduction Metal Oxide Semiconductor or MOS-Type gas sensors are a type of sensors that can detect presence of some volatile, oxidizable or reducible substances in an oxygen environment. In recent years, these sensors have been used in different study areas like healthcare [1,2] or food industry [3], among others. These sensors are resistive sensors whose nominal resistance changes with the presence of different fuels, oxidizing gases or reducing gases [4,5]. This type of sensor has many advantages such as a high response, low cost and portability. Due to these characteristics, different fabrication processes of MOS-Type gas sensors have been studied to improve their sensitivity, power consumption, and response time [6–9]. Other studies have researched into the applications in fields like the Internet of Things (IoT) devices [10] and wearables [11] by using different techniques of power supply, such as triboelectric nano-generator (TENG) [12], and others. With these types of sensors, a voltage divider [13,14] is often used to measure the substance searched for. But other electronic topologies could be used to power resistive sensors. In this work, besides a voltage divider, the Wheatstone bridge, the Anderson loop, and a resistance-to-frequency converter were designed to be used with different MOS-Type gas sensors and compared among them. The difference of these MOS gas sensors from others resistive sensors (as platinum resistance temperature sensor or thermistors) are: they use a heater in the same sensor (need to work high temperatures), and the range of the variable resistance is higher than others. In this article, TGS2600 [15] and TGS2610 [16] sensors are used, whose sensing material is tin dioxide (SnO2), because SnO2 is a very popular sensing material in this area [17]. In addition, both

Electronics 2020, 9, 525; doi:10.3390/electronics9030525 www.mdpi.com/journal/electronics Electronics 2020, 9, x FOR PEER REVIEW 2 of 18

ElectronicsIn 2020this, article,9, 525 TGS2600 [15] and TGS2610 [16] sensors are used, whose sensing material2 is of tin 14 dioxide (SnO2), because SnO2 is a very popular sensing material in this area [17]. In addition, both sensors have similar heaters onto the reverse side of the substrate, which is made of RuO2 [15,16]. sensorsHowever, have their similar bases heaters are different: onto the TGS2600 reverse has side a Ni-plated of the substrate, steel sensor which base is made[15], whereas of RuO 2TGS2610[15,16]. However,sensor base their is made bases of are NiCu-plated different: TGS2600 steel [16]. has Mo a Ni-platedreover, another steel sensor difference base between [15], whereas both TGS2610sensors is sensorthe long-term base is made stability of NiCu-plated(Figure S1 and steel S2 [in16 the]. Moreover, supplementary another materials). difference TGS2610 between is both very sensors stable but is theTGS2600 long-term has stabilitya worst long-term (Figure S1 stability. and S2 in the Supplementary Materials). TGS2610 is very stable but TGS2600 has a worst long-term stability. 2. Materials and Methods 2. Materials and Methods This section deals with the design of different electronic conditioner topologies to maximize the This section deals with the design of different electronic conditioner topologies to maximize the accuracy of MOS-type gas sensor applications. The analysis and simulated results for the different accuracy of MOS-type gas sensor applications. The analysis and simulated results for the different topology are discussed. topology are discussed. The voltage divider and the Wheatstone bridge are the most commonly used topologies for The voltage divider and the Wheatstone bridge are the most commonly used topologies for measurement systems based on resistive sensors because they are easy to design and easy to get data measurement systems based on resistive sensors because they are easy to design and easy to get data from [4]. from [4]. The Anderson loop is more difficult to design because a current source and an active voltage The Anderson loop is more difficult to design because a current source and an active voltage subtractor are needed, as shown in Figure 1. However, unlike the voltage divider and the Wheatstone subtractor are needed, as shown in Figure1. However, unlike the voltage divider and the Wheatstone bridge, the output signal for this topology changes linearly with the sensor resistance. Furthermore, bridge, the output signal for this topology changes linearly with the sensor resistance. Furthermore, the Anderson loop can be easily designed for an array of resistive sensors [18–20]. the Anderson loop can be easily designed for an array of resistive sensors [18–20].

Figure 1. Anderson loop. Figure 1. Anderson loop. The resistance-to-frequency converter is the most complex topology we are going to study in this paper.The This resistance-to-frequency topology is based on the converter use of a is simulated the most capacitancecomplex topology using a we Generalized are going Impedanceto study in Converterthis paper. (GIC). This Thetopology capacitance is based changes on the linearly use of with a simulated the sensor resistance;capacitance an using oscillator a Generalized is created usingImpedance a 555 timerConverter with (GIC). a frequency The capacitance that depends changes on the linearly simulated with the capacitance sensor resistance; [21]. This an topology oscillator theoreticallyis created using has aa linear555 timer dependence with a frequency betweenthe that sensor depends resistance on theand simulated the oscillator capacitance frequency, [21]. likeThis thetopology Anderson theoretically loop. But ithas has a several linear advantagesdependence due between to the fact the of sensor using frequency:resistance itand is more the oscillator immune tofrequency, noise (but like it has the a Anderson lower Spurious-Free loop. But it Dynamichas severa Rangel advantages (SFDR), due the to transmission the fact of using distance frequency: can be higher,it is more and immune it could to be noise acquired (but it by has digital a lower system Spurious-Free without an Dynamic Analog toRange Digital (SFDR), Converter the transmission (ADC). distanceThe interfacecan be higher, to obtain and data it fromcould these be acquired designs is by the digital Red-Pitaya system STEMlab without 125-14, an Analog which hasto Digital 14-bit ADC,Converter with an(ADC). input voltage range from 20 V to 20 V. Furthermore, the sample rate is 125 MS/s [22]. − ThanksThe to interface these characteristics, to obtain data this from acquisition these designs board is can the be Red-Pitaya used with STEMlab all topologies 125-14, described which has above 14- (Figurebit ADC, S3 with and S4an in input the Supplementaryvoltage range from Materials). −20 V to 20 V. Furthermore, the sample rate is 125 MS/s [22]. Thanks to these characteristics, this acquisition board can be used with all topologies described 2.1. Voltage Divider and Wheatstone Bridge above (Figure S3 and S4 in the supplementary materials). The voltage divider was designed as indicated in the datasheet [15] and [16] for TGS2600 and TGS2610,2.1. Voltage respectively. Divider and Both Wheatstone datasheets Bridge indicate that the load resistance of the voltage divider should be greater than 0.45 kΩ. Another important thing to consider when choosing this resistance was the sensor resistance when detect clean air. In this case, this resistance can be from 10 kΩ to until 100 kΩ, Electronics 2020, 9, x FOR PEER REVIEW 3 of 18

The voltage divider was designed as indicated in the datasheet [15] and [16] for TGS2600 and TGS2610, respectively. Both datasheets indicate that the load resistance of the voltage divider should be greater than 0.45 kΩ. Another important thing to consider when choosing this resistance Electronicswas the2020 sensor, 9, 525 resistance when detect clean air. In this case, this resistance can be from3 of 10 14 kΩ to until 100 kΩ, it is a big range. For this reason, if the load resistance is close to 100 kΩ itand is athe big sensor range. has For a this sensor reason, resist if theance load with resistance clean air is closeclose to to 100 10 k ΩΩ,and the theoutput sensor signal has achange sensor resistanceis going to with be cleansmall air when close the to 10sensor kΩ, the resistance output signal changes change and is that going made to be smallthe resolution when the sensorof the resistancemeasure will changes be worst. and that Consequently, made the resolution the load of the resistance measure should will be worst.be close Consequently, to the small the values load resistanceof that range. should Finally, be close the to chosen the small load values resistance of that is range. 10 kΩ Finally, because the this chosen is suitable load resistance for bothis sensors 10 kΩ becausewhen measuring this is suitable clean forair. both sensors when measuring clean air. The WheatstoneWheatstone bridge bridge is is based based on theon the voltage voltage divider, divider, Figure Figure2. Unfortunately, 2. Unfortunately, with these with sensors these issensors diﬃcult is difficult to make to a balancedmake a balanced bridge, becausebridge, becaus althoughe although the same the model same ismodel used, is the used, resistance the resistance of two identicalof two identical sensors cansensors have acan high have variation. a high Forvariat example,ion. For a sensor example, TGS2600 a sensor can haveTGS2600 a resistance can have of 10 a kresistanceΩ when measuringof 10 kΩ when clean measuring air and another clean TGS2600 air and cananother have TGS2600 90 kΩ in can the samehave 90 conditions. kΩ in the Due same to thisconditions. reason, theDue Wheatstone to this reason, bridge th hase Wheatstone two 10 kΩ resistors bridge andhas atwo 100 k10Ω kpotentiometer,Ω resistors and which a 100 will k beΩ regulatedpotentiometer, at the which beginning will be of eachregulated measurement at the begi asnning a calibration of each step.measurement as a calibration step.

Figure 2. Voltage divider and Wheatstone bridgebridge designs.designs. 2.2. Anderson Loop 2.2. Anderson Loop The electronic system based on the Anderson loop needs a current source, which is designed The electronic system based on the Anderson loop needs a current source, which is designed taking advantage of the following characteristic of the GIC: the current flowing through R5 resistor in taking advantage of the following characteristic of the GIC: the current flowing through R5 resistor Figure3 is the same current that flows through R4. And, if all passive components are resistors and the in Figure 3 is the same current that flows through R4. And, if all passive components are resistors voltage Vcc [23,24]. and the voltage Vcc [23,24]. A current of 0.1 mA is selected for this design because the maximum power supported by the A current of 0.1 mA is selected for this design because the maximum power supported by the chosen sensors is 15 mW and the maximum resistance of the sensors is around 100 kΩ [13,14]. chosen sensors is 15 mW and the maximum resistance of the sensors is around 100 kΩ [13,14]. Once the current source is designed, the next step is to create an active subtractor by using instrumentation amplifiers. In this case, unlike the Anderson loop, the differential voltage of the reference resistance is not subtracted from the differential voltage of the sensor resistance; in this design, the differential voltage of the sensor resistance is subtracted from the differential voltage of the reference resistors, as Figure4 shows. For this reason, the sensors have maximum resistance when measuring clean air, and in this way, a negative voltage is not needed [25]. The R3 value must have a high value to assure good efficiency between the upper operational amplifier output current and the current source [24], with that, we selected R3 with a resistance value of 47 kΩ, this is because is close to the maximum value of the resistance sensor. In the other hand, R1 and R2 have been modified to achieve, in the lower values of the range, a properly behavior of the GIC, and these values are 3.3 kΩ to both. Finally, it is only necessary to create a non-inverter amplifier with a gain between 1 to 4.5, which is controlled by a potentiometer in the position of R9 in the circuit, and the resistance selected is R8 = 1 kΩ and the potentiometer R9 = 3.5 kΩ.

Electronics 2020, 9, 525 4 of 14 Electronics 2020, 9, x FOR PEER REVIEW 4 of 18

Electronics 2020, 9, x FOR PEER REVIEWFigure 3. Generalized Impedance Converter. 5 of 18

Once the current source is designed, the next step is to create an active subtractor by using instrumentation amplifiers. In this case, unlike the Anderson loop, the differential voltage of the reference resistance is not subtracted from the differential voltage of the sensor resistance; in this design, the differential voltage of the sensor resistance is subtracted from the differential voltage of the reference resistors, as Figure 4 shows. For this reason, the sensors have maximum resistance when measuring clean air, and in this way, a negative voltage is not needed [25]. The R3 value must have a high value to assure good efficiency between the upper operational amplifier output current and the current source [24], with that, we selected R3 with a resistance value of 47 kΩ, this is because is close to the maximum value of the resistance sensor. In the other hand, R1 and R2 have been modified to achieve, in the lower values of the range, a properly behavior of the GIC, and these values are 3.3 kΩ to both. Finally, it is only necessary to create a non-inverter amplifier with a gain between 1 to 4.5, which is controlled by a potentiometer in the position of R9 in the circuit, and the resistance selected is R8 = 1 kΩ and the potentiometer R9 = 3.5 kΩ.

FigureFigure 4.4. Electronic system based on the Anderson loop.loop.

2.3.2.3. Converter Resistance-to-Frequency TheThe lastlast topologytopology waswas proposedproposed inin [[21],21], andand itit isis basedbased onon thethe simulationsimulation ofof aa capacitancecapacitance usingusing thethe GeneralizeGeneralize ImpedanceImpedance Converter.Converter. In addition, a 555 timer is used in a stable mode toto obtainobtain aa signalsignal whosewhose frequencyfrequency changeschanges linearlylinearly withwith thethe sensorsensor resistanceresistance [[26].26]. InIn [[21],21], thethe sensorsensor resistanceresistance rangerange isis betweenbetween 500500 ΩΩ andand 3030 kkΩΩ,, butbut thethe rangerange isis widerwider inin inin thisthis proposal,proposal, fromfrom 500500 ΩΩ toto 100100 kkΩΩ.. DueDue toto this,this, somesome changeschanges inin thethe topologytopology werewere introduced.introduced. OnOn oneone hand,hand, thethe capacitorcapacitor C3C3 ofof thethe Figure5 5 changes changes its its place place for for thethe resistanceresistance R5, R5, and and to to growgrow their their values values untiluntil 3333 nFnF andand 1515 kkΩΩ;; thisthis changechange doesdoes notnot aaffectﬀect toto thethe theoreticaltheoretical behaviorbehavior ofof thethe GICGIC [[27],27], butbut thethe maximummaximum voltagevoltage drop drop is is only only 3 V 3 in V the in worstthe worst case. case. On the On other the hand,other ourhand, design our hasdesign higher has voltages higher voltages than the voltages used in the original paper; this is due to two reasons: the first reason is that the change in the resistance range of the sensor increases the voltage needed; and the second reason is that in [21] a microcontroller PSoC was used simulating a timer 555, with power voltage level of 3.3 V, whereas in our design a 555 timer integrated circuit [26] is used with a minimum power voltage of 5 V.

Electronics 2020, 9, 525 5 of 14 than the voltages used in the original paper; this is due to two reasons: the ﬁrst reason is that the change in the resistance range of the sensor increases the voltage needed; and the second reason is that in [21] a microcontroller PSoC was used simulating a timer 555, with power voltage level of 3.3 V, whereas in our design a 555 timer integrated circuit [26] is used with a minimum power voltage of 5 V. Electronics 2020, 9, x FOR PEER REVIEW 6 of 18

Figure 5.5. DesignDesign ofof thethe paper paper in in which which is basedis based the the design design of theof converterthe converter resistance-to-frequency resistance-to-frequency [21]. [21]. There was a calibration process when 30 resistors with different values where selected. Then, the differentThere circuits was a werecalibration activated process with when these resistors30 resistors and with 5 V different in order tovalues make where them stableselected. in termsThen, ofthe temperature. different circuits A 4-wires were activated measure waswith takenthese forresist eachors resistorand 5 V andin order each to circuit. make Withthem thesestable data,in terms the outputof temperature. characteristic A 4-wires (voltage measure or frequency) was taken vs. for Rsensor each resistor can be comparedand each circuit. with the With theoretical these data, value. the outputIf wecharacteristic consider the (voltage resistance-to-frequency or frequency) vs. circuit Rsensor in Figure can be6 ,compared the output with frequency the theoretical vs. capacitance value. is shownIf we in Figureconsider7, where the weresistance-to-frequency can see that the output frequencycircuit in doesFigure not change6, the linearlyoutput withfrequency the sensor vs. resistancecapacitance along is shown a wide in range,Figure but 7, where it still we can can be usedsee that taking the intooutput account frequency this particularity. does not change In the linearly range ofwith the the sensor sensor resistance, resistance from along 500 Ωa (wideC range,52.89 nFbut) toit 40still k Ωcan(C be used0.657 takingnF), frequency into account changes this GIC ≈ GIC ≈ linearlyparticularity. with theIn the sensor range resistance, of the sensor but if resistance, the resistance from is bigger500 Ω ( than 40 52.89 kΩ this) behavior to 40 kΩ changes. ( Moreover,0.657), frequency it shows that, changes when linearly the capacitance with the simulated sensor resist by theance, GIC but is lowerif the resistance than 0.65 nFis bigger and the than sensor 40 resistancekΩ this behavior is smaller changes. than 40 Moreover, kΩ, the change it shows of thethat frequency, when the regards capacitance the change simulated of the by capacitance the GIC is simulatedlower than by 0.65 the nF GIC and has the an sensor hyperbolic resistance function, is smaller and given than that 40 CkΩ, the= 0.394417change /of( Rthesensor frequency15000), GIC ∗ Figureregards7 confirmsthe change the of behavior the capacitance of the output simulated frequency by the explained GIC has above.an hyperbolic function, and given

that = 0.394417⁄ ∗ 15000, Figure 7 confirms the behavior of the output frequency explained above.

ElectronicsElectronics2020 2020, 9, ,9 525, x FOR PEER REVIEW 67 of of 14 18 Electronics 2020, 9, x FOR PEER REVIEW 7 of 18

Figure 6. Final circuit of the converter resistance-to-frequency design. Figure 6. Final circuit of the converter resistance-to-frequency design.design.

FigureFigure 7. Capacitance7. Capacitance simulated simulated by by Generalized Generalized Impedance Impedance Converter Converter (GIC) (GIC) versus versus frequency frequency of theof outputFigure signal 7. Capacitance of the design. simulated bythe Generalized output signal Impeda of thence design. Converter (GIC) versus frequency of the output signal of the design. 2.4.2.4. Methods Methods 2.4. Methods TheThe output output voltage voltage was was measured measured using using a Red-Pitaya a Red-Pitaya STEMlab STEMlab 125-14 [ 22125-14] board, [22] whose board, operative whose systemoperativeThe and output system Field voltage and Programmable Field was Programmable measured Gate Array using Gate (FPGA) aArra Red-Pitayay program(FPGA) STEMlab program were modiﬁed were125-14 modified to[22] use board, itto asuse awhose it data as a acquisitionoperativedata acquisition system card using cardand Field theusing project Programmable the project made by made NilsGate by Roos Arra Nils [y28 Roos(FPGA)]. The [28]. board program The sends board were all sends datamodified throughall data to usethrough a TCP-IP it as a a data acquisition card using the project made by Nils Roos [28]. The board sends all data through a

Electronics 2020, 9, 525 7 of 14

Electronics 2020, 9, x FOR PEER REVIEW 8 of 18 connection, to a computer that saves all data in a ﬁle using a LabVIEW program. The experimental TCP-IPElectronics connection, 2020, 9, x FOR to PEER a computerREVIEW that saves all data in a file using a LabVIEW program.8 of 18The setup of electronics is shown in Figure8 and the frontal panel is shown in Figure9. experimental setup of electronics is shown in Figure 8 and the frontal panel is shown in Figure 9. TCP-IP connection, to a computer that saves all data in a file using a LabVIEW program. The experimental setup of electronics is shown in Figure 8 and the frontal panel is shown in Figure 9.

Figure 8. Experimental setup including the Red-Pitaya, the designed circuit with the vial including FigureFigure 8. 8.Experimental Experimental setup setup including including the the Red-Pitaya, Red-Pitaya, the the designed designed circuit circuit with with the the vial vial including including the the gas sampling. gasthe sampling. gas sampling.

FigureFigure 9. 9. Graphical Graphical user user interfaceinterface of the LabVIEW LabVIEW program program used used to toto store storestore measures. measures.measures.

3 TheTheThe substancesubstance substance toto to bebe be detecteddetected detected waswaswas absoluteabsoluteabsolute ethanolethanolhanol of ofof Scharlau, Scharlau,Scharlau, whose whosewhose density densitydensity is isis0.79 0.790.79 g/cm gg/cm/cm in33 inin water,water,water, andand and dissolutionsdissolutions dissolutions withwith with didifferent differentﬀerent concentrations concentrationsconcentrations werewere applied; applied; from fromfrom 500 500500 ppm ppmppm to to to6000 60006000 ppm. ppm.ppm. Each EachEach sample contained 2 cm3 of one dissolution in a vial, where the sensor was placed. All the measures samplesample containedcontained 22 cmcm33 of one dissolution in a vial,vial, wherewhere thethe sensorsensor waswas placed.placed. AllAll thethe measuresmeasures were taken at temperatures from 296 to 297 °K. This temperature was measured using a thermocouple werewere takentaken atat temperaturestemperatures fromfrom 296296 toto 297297 °K.K. This temperature was measured using a thermocouple type K and a digital thermometer TES-1302.◦ typetype KK andand aa digitaldigital thermometerthermometer TES-1302.TES-1302. Typical curves obtained in the acquisition system are shown in Figure 10 for illustrative TypicalTypical curvescurves obtained obtained in thein the acquisition acquisition system syst areem shown are shown in Figure in 10Figure for illustrative 10 for illustrative purposes; purposes; in this case, the condition circuit was the voltage divider, and the sensor was the TGS2600, purposes; in this case, the condition circuit was the voltage divider, and the sensor was the TGS2600, inwith this case,different the concentr conditionations circuit of ethanol. was the voltage divider, and the sensor was the TGS2600, with diwithﬀerent different concentrations concentrations of ethanol. of ethanol.

Electronics 2020, 9, x FOR PEER REVIEW 9 of 18 Electronics 2020, 9, 525 8 of 14 Electronics 2020, 9, x FOR PEER REVIEW 9 of 18

Figure 10. Voltage output for the voltage divider design with the sensor TGS2600.

MeasuresFigureFigure were 10. 10. takenVoltage Voltage following output output for for this the the voltage procedurevoltage divider divide (Figurer design design S5 with with in the thethe sensor sensor supplementary TGS2600. TGS2600. materials): firstly, a measure of clean air is obtained to take a voltage reference. Secondly, the dissolution sample MeasuresMeasures were were taken taken following following this this procedure procedure (Figure (Figure S5 in theS5 in Supplementary the supplementary Materials): materials): ﬁrstly, is measured, in this step two characteristics of the measure were considered: the rising edge of the afirstly, measure a measure of clean of air clean is obtained air is obtained to take to a take voltage a voltage reference. reference. Secondly, Secondly, the dissolutionthe dissolution sample sample is voltage, and the voltage value when it is stable. The last step is to measure clean air again. All sensors measured,is measured, in thisin this step step two two characteristics characteristics of of the th measuree measure were were considered: considered: the the rising rising edge edge of of the the had been working previously at least 24 hours because they need to be warmed up before being used. voltage,voltage, and and the the voltage voltage value value when when it it is is stable. stable. The The last last step step is is to to measure measure clean clean air air again. again. All All sensors sensors One of the characteristics measured in the voltage divider, Wheatstone bridge and Anderson hadhad been been working working previously previously at at leastleast 2424 hhours because beca theyuse they need need to be to warmed be warmed up before up before being being used. used. loop designs was the slew rate when the sensor starts to measure the sample. The voltage rise OneOne of of the the characteristics characteristics measured measured in thein the voltage voltage divider, divider, Wheatstone Wheatstone bridge bridge and Andersonand Anderson loop transition is approximated to two lines, whose slopes are calculated as follow: the first one from t = 0 designsloop designs was the was slew the rate slew when rate the when sensor the starts sensor to measure starts to the measure sample. the The sample. voltage riseThe transitionvoltage rise is and instant when voltage is 60% of the stable value; and the second one from 60% to 90%, as shown approximatedtransition is approximated to two lines, whoseto two slopeslines, whose are calculated slopes are as follow:calculated the as ﬁrst follow: onefrom the first t = 0one and from instant t = 0 in Figure 11. whenand instant voltage when is 60% voltage of the is stable 60% value;of the andstable the value; second and one the from second 60% one to 90%, from as 60% shown to 90%, in Figure as shown 11. in Figure 11.

Figure 11. How slopes are measured. Figure 11. How slopes are measured. The characteristic measured in the resistance-to-frequency converter design is the main frequency of theThe output characteristic signal. For thismeasured reason,Figure ain Fast the 11. Fourier Howresistance slopes Transform-to-frequency are measured. (FFT) of theconverter data was design calculated is the to obtain main thefrequency main frequency of the output component. signal. TheFor stepsthis reason, followed a Fast for theseFourier measures Transform are the(FFT) same of asthe explained data was The characteristic measured in the resistance-to-frequency converter design is the main previouslycalculated (Figureto obtain S5 the in themain Supplementary frequency componen Materials).t. The steps followed for these measures are the frequency of the output signal. For this reason, a Fast Fourier Transform (FFT) of the data was same as explained previously (S5 in the supplementary materials). calculated to obtain the main frequency component. The steps followed for these measures are the same as explained previously (S5 in the supplementary materials).

Electronics 2020, 9, 525 9 of 14 ElectronicsElectronics 20202020,, 99,, xx FORFOR PEERPEER REVIEWREVIEW 1010 ofof 1818

3.3. Results Results FigureFigure 121212 shows showsshows thethethe valuesvaluesvalues whenwhenwhen thethethe sensorsensorsensor isisis stableststableable withwithwith thethethe TGSTGSTGS 2600.2600.2600. AlthoughAlthough allall measuresmeasures areare ininin the thethe same samesame graph, graph,graph, the thethe resistance-to-frequency resistance-to-frequencresistance-to-frequenc converteryy converterconverter design designdesign uses the usesuses right thethe vertical rightright verticalvertical scale (output scalescale signal(output(output frequency) signalsignal frequency)frequency) whereas whereaswhereas the other thethe three otherother plots threethree use plpl theotsots vertical useuse thethe left verticalvertical scale leftleft (output scalescale voltage). (output(output voltage).voltage).

FigureFigure 12.12. ValuesValues measuredmeasured forfor allall designsdesigns usingusing thethe TGS2600TGS2600 withwith didifferentdifferentﬀerent dissolutions. dissolutions.dissolutions.

FigureFigure 12 12 confirms confirmsconfirms thatthatthat allall designsdesigns cancan bebe usedused toto measuremeasure didifferentdifferentfferent concentrationsconcentrationsconcentrations ofofof ethanolethanolethanol inin thethethe air,air,air, butbutbut theretherethere are areare other otherother characteristics characteristicscharacteristics tototo consider,considerconsider,, such suchsuch as asas the thethe signal signalsignal noise. noise.noise. ThisThisThis characteristiccharacteristiccharacteristic isisis measuredmeasured when whenwhen the thethe voltage voltagevoltage output outputoutput is isis stable. stable.stable. In InIn case cacasese of ofof the thethe voltage voltagevoltage divider dividerdivider and andand the thethe Anderson AndersonAnderson loop, loop,loop, the noisethethe noisenoise is considered isis consideredconsidered as the asas thethe floor floorfloor noise noisenoise of of theof thethe ADC, ADC,ADC, but butbut in inin the thethe Wheatstone WheatstoneWheatstone bridge bridgebridge the thethenoise noisenoise is isis greatergreater thanthanthan ininin thethethe others.others.others. AnotherAnother characteristiccharacteristic ofof thethe outputoutput signalsignal fromfrfromom thethethe voltagevoltagevoltage divider,divider,divider, WheatstoneWheatstoneWheatstone bridge,bridge,bridge, andandand AndersonAndersonAnderson looplooploop designsdesignsdesigns isisis thethethe rising risingrising edge edgeedge modeled modeledmodeled as asas two twotwo lines, lines,lines, as asas explained explainedexplained previously. previously.previously. Figures FigureFigure 131313 andand 14 FigureFigure proves 1414 thatprovesproves the thatthat value thethe of valuevalue the first ofof slopethethe firstfirst can slsl beopeope used cancan tobebe know usedused thetoto knowknow ethanol thethe concentration ethanolethanol concentrationconcentration in the air ininin thethethe threeairair inin designs. thethe threethree The designs.designs. second TheThe slope secondsecond can only slopeslope be cancan used onlyonly for be thebe usedused voltage forfor dividerthethe voltagevoltage and fordividerdivider Wheatstone andand forfor bridgeWheatstoneWheatstone designs, bridgebridge but notdesigns,designs, for the butbut others notnot forfor because thethe otothershers of the becausebecause lack of ofof monotonicity. thethe lacklack ofof monotonicity.monotonicity.

FigureFigure 13.13. ValueValue ofof ﬁrstfirstfirst slopeslope measuredmeasured withwithwith thethethe TGSTGSTGS 2600 26002600 in inin the thethe rising risingrising edge. edge.edge.

Electronics 2020, 9, 525 10 of 14 Electronics 2020, 9, x FOR PEER REVIEW 11 of 18

Figure 14. Value of second slope measured with the TGS 2600 in the rising edge. Figure 14. Value of second slope measured with the TGS 2600 in the rising edge. The last characteristic considered is the power consumption of each design. The power consumed by theThe Wheatstone last characteristic bridge and considered voltage divider is the is power very small consumption and similar of ineach our design. circuits becauseThe power of ourconsumed experimental by the setup: Wheatstone both designs bridge shareand voltage components divider in is the very same small Printed and similar Circuit in Board our circuits (PCB). Moreover,because of it our had experimental the lowest power setup: consumptionboth designs share of all components the designs, in 2.28 the mW same maximum Printed Circuit by voltage Board divider(PCB). andMoreover, 4.56 mW it had maximum the lowest by Wheatstone power cons bridge.umption Others of all designsthe designs, have 2.28 active mW electronics, maximum and by thisvoltage increases divider the powerand 4.56 consumption. mW maximum In the Andersonby Wheatstone loop, thebridge. current Others source, designs this design have has active the higherelectronics, power and consumption this increases (114.91 the mWpower at most),consumptio but itn. allows In the to Anderson use the same loop, current the current for some source, sensors this indesign array has conﬁguration, the higher power Table1 consumption. The consumption (114.91 of mW the resistive-to-frequencyat most), but it allows to design use the is insame a middle current pointfor some (41.85 sensors mW). in array configuration, Table 1. The consumption of the resistive-to-frequency design is in a middle point (41.85 mW). Table 1. Comparison of obtained data. Table 1. Comparison of obtained data. Type Power Noise Linearity Array Type Power Noise Linearity Array Voltage Divider 2.28 mW 80 dB No No WheatstoneVoltage Bridge Divider 4.56 mW 2.28 mW 55 dB 80 dB No No No No AndersonWheatstone loop Bridge 114.91 mW 4.56 mW 80 dB 55 dB Yes No Yes No Converter resistance-to-frequencyAnderson loop 41.85 mW 114.91 mW 42.5 dB 80 dB Rs < 40 k YesΩ No Yes Converter resistance-to-frequency 41.85 mW 42.5 dB Rs < 40kΩ No Results were repeated for the sensor TGS2610 and results similar to sensor TGS2600 were obtained, Results were repeated for the sensor TGS2610 and results similar to sensor TGS2600 were as Figures 15–17 show that. obtained, as Figures 15, 16 and 17 show that.

Electronics 2020, 9, x FOR PEER REVIEW 11 of 18

Figure 14. Value of second slope measured with the TGS 2600 in the rising edge.

The last characteristic considered is the power consumption of each design. The power consumed by the Wheatstone bridge and voltage divider is very small and similar in our circuits because of our experimental setup: both designs share components in the same Printed Circuit Board (PCB). Moreover, it had the lowest power consumption of all the designs, 2.28 mW maximum by voltage divider and 4.56 mW maximum by Wheatstone bridge. Others designs have active electronics, and this increases the power consumption. In the Anderson loop, the current source, this design has the higher power consumption (114.91 mW at most), but it allows to use the same current for some sensors in array configuration, Table 1. The consumption of the resistive-to-frequency design is in a middle point (41.85 mW).

Table 1. Comparison of obtained data.

Type Power Noise Linearity Array Voltage Divider 2.28 mW 80 dB No No Wheatstone Bridge 4.56 mW 55 dB No No Anderson loop 114.91 mW 80 dB Yes Yes Converter resistance-to-frequency 41.85 mW 42.5 dB Rs < 40kΩ No

ElectronicsResults2020 were, 9, 525 repeated for the sensor TGS2610 and results similar to sensor TGS2600 were11 of 14 obtained, as Figures 15, 16 and 17 show that.

Electronics 2020, 9, x FOR PEER REVIEW 12 of 18 Electronics 2020, 9, x FOR PEER REVIEW 12 of 18 Figure 15. Value of second slope measured with the TGS 2610 in the rising edge. FigureFigure 15. 15.Value Value of of second second slope slope measured measured withwith thethe TGSTGS 2610 in the rising rising edge. edge.

Figure 16. Value of first slope measured with the TGS 2610 in the rising edge. FigureFigure 16.16. ValueValue ofof ﬁrstfirst slopeslope measuredmeasured with withthe the TGS TGS 2610 2610 in in the the rising rising edge. edge.

FigureFigure 17. 17.Value Value of of second second slope slope measured measured with with the the TGSTGS 26102610 when a measure of of a a sample sample starts. starts. Figure 17. Value of second slope measured with the TGS 2610 when a measure of a sample starts. The acquisition system behavior does not depend on the gas used in the measure, but it only dependsThe acquisition on the electronic system behaviorused (sensors, does operationot dependnal amplifiers,on the gas among used in others). the measure, For this but reason, it only dependsmeasuring on theother electronic gases is not used relevant (sensors, from operatio the pointnal of viewamplifiers, of the electronicamong others). conditioning, For this only reason, the sensor, and it does not affect the results. measuring other gases is not relevant from the point of view of the electronic conditioning, only the Finally, an equation, which models the behavior of each measurement system, is found (Figure sensor, and it does not affect the results. S6 and S7 in the supplementary materials). The method to obtain this is a fitting process by using Finally, an equation, which models the behavior of each measurement system, is found (Figure functions fit() and fittype() in MATLAB; it was tested a polynomic, logarithmic and exponential S6 and S7 in the supplementary materials). The method to obtain this is a fitting process by using fitting with different orders. After different combinations, the best results were obtained by using a functionspolynomic fit() division, and fittype() with thein MATLAB;coefficient ofit determinationwas tested a polynomic,(R-Squared) logarithmicshown in Table and 2. exponential This fact fittingconfirms with thatdifferent all systems orders. analyzed After different in this paper combin canations, be used the to best measure results different were obtained concentrations by using of a polynomicethanol in division, the air. These with modelsthe coeffi followcient two of typesdetermination of equations: (R-Squared) shown in Table 2. This fact confirms• that Whether all systems the outputanalyzed voltage in this changes paper ca linearln be usedy with to measurerespect todifferent the sensor concentrations resistance of ethanol(Anderson in the loop air. Theseand Converter models follow resistance-to-frequency two types of equations: when the sensor resistance is under 40 kΩ [when• TGS2610 Whether is the used]), output their modelvoltage has changes an equation linearl suchy aswith equation respect (1). to the sensor resistance (Anderson loop and Converter resistance-to-frequency when the sensor resistance is under 40 kΩ [when TGS2610 is used]), their model has an equation such as equation (1).

Electronics 2020, 9, 525 12 of 14

The acquisition system behavior does not depend on the gas used in the measure, but it only depends on the electronic used (sensors, operational amplifiers, among others). For this reason, measuring other gases is not relevant from the point of view of the electronic conditioning, only the sensor, and it does not affect the results. Finally, an equation, which models the behavior of each measurement system, is found (Figure S6 and S7 in the Supplementary Materials). The method to obtain this is a fitting process by using functions fit() and fittype() in MATLAB; it was tested a polynomic, logarithmic and exponential fitting with different orders. After different combinations, the best results were obtained by using a polynomic division, with the coefficient of determination (R-Squared) shown in Table2. This fact confirms that all systems analyzed in this paper can be used to measure different concentrations of ethanol in the air. These models follow two types of equations:

Whether the output voltage changes linearly with respect to the sensor resistance (Anderson loop • and Converter resistance-to-frequency when the sensor resistance is under 40 kΩ [when TGS2610 is used]), their model has an equation such as Equation (1). However, if the system has not these characteristics (voltage divider and Wheatstone bridge), • their equations correspond to (2).

2 p2R + p1Rsen + p0 out = sen (1) 2 Rsen + q1Rsen + q0

3 2 p3R + p2R + p1Rsen + p0 out = sen sen (2) 3 2 Rsen + q2Rsen + q1Rsen + q0

Table 2. Coeﬃcient of determination of model of each system.

Converter R2 Voltage Divider Wheatstone Bridge Anderson Loop Resistance-to-Frequency TGS2600 0.999981 0.996644 0.999854 0.999379 TGS2610 0.999989 0.999841 0.999542 0.999999

The coeﬃcients px and qx for each equation are shown in Table3.

Table 3. Coeﬃcient of determination of model of each system.

Wheatstone Converter Voltage Divider Anderson Loop Bridge Resistance-to-Frequency Coeﬃcients [p0,p1,p2,p3] [p0,p1,p2] [p0,p1,p2,p3] [p0,p1,p2] [q0, q1, q2] [q0, q1] [q0, q1, q2] [q0, q1] [2.483, 204.5, 3.067, [2.342, 125.7, [4.203, 4.34 104], p TGS2610 − − × [ 227.9, 5.7 107,6.6 109] 0.549] 298.5, 0.433] 4.75 105] − × × − − × q TGS2610 [617.4, 1.844, 0.865] [1369, 61.32, 12.57] [ 8975, 5.3*106] [4930, 3.9 105] − − − × [4.389, [3.068, 122.8, p TGS2600 − [3.038, 22.6, 0.296] [0.559, 2669, 1.486, 1.292] 957.5,8.028,0.908] 245.5, 1.113] − − − q TGS2600 [611, 18.15, 0.789] [440, 83.91, 12.4] [86.83, 0.728] [324.2, 17.69, 0.105] −

4. Conclusions Each circuit has benefits and drawbacks that the designer should consider. The voltage divider is the topology with less power consumption, and the noise is small enough to obtain good results. This makes this topology the best option to implement in a portable device. Furthermore, as it does not need the rising edge but only the voltage level, the frequency sampling can be much lower. The Wheatstone bridge is very difficult to implement using this type of sensor due to the big sensors’ nominal resistance variability. For this reason, this topology is not recommended. The Anderson loop has some strengths: the output voltage linearly depends on the resistance sensor; and in case of the Electronics 2020, 9, 525 13 of 14 design presented in this paper, the resistance range is quite high, from 10 kΩ up to 90 kΩ, which makes it useful for these sensors and others with lower ranges; and finally, it can be easily redesigned to have more than one sensor. However, this design has the higher power consumption, and it needs at least two different voltage sources. For this reason this topology is only recommended when the system is permanently connected to the power source, or for sensor arrays. The last topology presented is GIC-based. This is recommended when the voltage output cannot be measured with a good ADC because it provides a digital output which can be measured easily using a timer/counter.

Supplementary Materials: The following are available online at http://www.mdpi.com/2079-9292/9/3/525/s1. Author Contributions: Conceptualization, J.C.G. and J.P.-S.; methodology, J.P.-S.; software, J.C.G.; validation, J.C.G., J.P.-S. and R.L.; formal analysis, J.C.G. and J.P.-S.; investigation, J.P.-S. and J.C.G.; resources, J.P.-S.; data curation, J.C.G.; writing—original draft preparation, J.C.G. and J.P.-S.; writing—review and editing, J.P.-S. and R.L.; visualization, J.C.G.; supervision, J.P.-S. and R.L.; project administration, J.P.-S. funding acquisition, J.P.-S. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the I+D+i Program of the Generalitat Valenciana [AICO/2016/046], Spain. Acknowledgments: The Authors thanks Monica Catala from the Chemical department of Universitat Politècnica de Valencia, for his helpful technical assistance in the creation of the samples to measure in this paper. Conflicts of Interest: The authors have no conflict of interest either financial or personal that affects the objectivity of the results. Rafael Lajara works for Analog Devices S.A.

References

1. Chilo, J.; Sebastia, J.P.; Cupane, M.; Sogorb, T. E-nose application to food industry production. IEEE Instrum. Meas. Mag. 2016, 19, 27–33. [CrossRef] 2. Cupane, M.; Sebastia, J.P.; Climent, E.; Guarrasi, V.; Sogorb, T.; Germana, M.A. Application of MOOSY32 eNose to Assess the Effects of Some Post Harvest Treatments on the Quality of ‘Salustiana’ Orange Juice. J. Biosens. Bioelectron. 2015, 6, 4–7. 3. Olarte, O.; Chilo, J.; Sebastia, J.P.; Barbé, K.; Moer, W. Glucose detection in human sweat using an electronic nose. In Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, Japan, 3–7 July 2013; pp. 1462–1465. 4. Pearce, T.C.; Schiffman, S.S.; Nagle, H.T.; Gardner, J.W. Handbook of Machine Olfaction; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2004. 5. Figaro. Operating Principle-MOS-Type Gas Sensor; Figaro USA Inc.: Arlington Heights, IL, USA, 2018; Available online: http://www.figaro.co.jp/en/technicalinfo/principle/mos-type.html (accessed on 19 July 2018). 6. Yang, Z.; Huang, Y.; Chen, G.; Guo, Z.; Cheng, S.; Huange, S. Ethanol gas sensor based on Al-doped ZnO nanomaterial with many gas diffusing channels. Sens. Actuators B Chem. 2009, 140, 549–556. [CrossRef]

7. Mirzaei, A.; Park, S.; Sun, G.; Kheel, H.; Lee, S.; Lee, C. Fe2O3/CO3O4 composite nanoparticle ethanol sensor. J. Korean Phys. 2016, 69, 373–380. [CrossRef] 8. Du, H.; Xie, G.; Su, Y.; Tai, H.; Du, X.; Yu, H.; Zhang, Q. A New Model and Its Application for the Dynamic Response of RGO Resistive Gas Sensor. Sensors 2019, 19, 889. [CrossRef][PubMed] 9. Xie, D.; Chen, D.; Peng, S.; Yang, Y.; Xu, L.; Wu, F. A Low Power Cantilever-Based Metal Oxide Semiconductor Gas Sensor. IEEE Electron Device Lett. 2019, 40, 1178–1181. [CrossRef] 10. Su, Y.; Xie, G.; Tai, H.; Li, S.; Yang, B.; Wang, S.; Zhang, Q.; Du, H.; Zhang, H.; Du, X.; et al. Self-Powered

Room Temperature NO2 Detection Driven by Triboelectric Nanogenerator under UV illumination. Nano Energy 2018, 47, 316–324. [CrossRef] 11. Wang, S.; Jiang, Y.; Tai, H.; Liu, B.; Duan, Z.; Yuan, Z.; Pan, H.; Xie, G.; Du, X.; Su, Y. An integrated ﬂexible self-powered wearable respiration sensor. Nano Energy 2019, 63, 103829. [CrossRef] 12. Chen, J.; Wang, Z.L. Reviving Vibration Energy Harvesting and Self-Powered Sensing by a Triboelectric Nanogenerator. Joule 2017, 1, 480–521. [CrossRef] 13. Figaro. TGS 2610—for the Detection of LP Gas; Figaro USA Inc.: Arlington Heights, IL, USA, 2014. 14. Figaro. TGS 2600—for the Detection of Air Contaminants; Figaro USA Inc.: Arlington Heights, IL, USA, 2013. 15. Figaro. Technical Information for TGS2600; Figaro Engineering Inc.: Mino, Japan, 2004. 16. Figaro. Technical Information for TGS2610; Figaro Engineering Inc.: Mino, Japan, 2006. Electronics 2020, 9, 525 14 of 14

17. Kakoty, P.; Bhuyan, M. SnO2 based gas sensors: Why it is so popular? In Proceedings of the IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), Coimbatore, India, 5–7 March 2015. 18. Anderson, K.F. Your Successor to the Wheatstone Bridge? NASA’s Anderson Loop. IEEE Instrum. Meas. Mag. 1998, 1, 5–15. [CrossRef] 19. Anderson, K.F. The loop technique for strain gage rosette signal conditioning. Exp. Tech. 2000, 24, 21–23. [CrossRef] 20. Shelake, V.G.; Kudale, A.B.; Mujawar, M.U.; Raicar, S.P.; Kamat, R.K.; Sawant, S.R. Evaluation of Anderson Loop for Resistive Sensors using PSPICE. In Proceedings of the Control Instrumentation System Conference (CISCON2005), MIT, Manipal, India, 11–12 November 2005. 21. De Albornoz, A.D.C.; Muñoz, D.R.; Moreno, J.S.; Berga, S.C.; Antón, E.N. A new gas sensor electronic interface with generalized impedance converter. Sensors Actuators B Chem. 2008, 134, 591–596. [CrossRef] 22. 3.1.1. STEMlab 125-10 vs. STEMlab 125-14 (Originally Red Pitaya v1.1)—Red Pitaya STEMlab 0.97 Documentation. Available online: https://redpitaya.readthedocs.io/en/latest/developerGuide/125-10/vs.html (accessed on 20 July 2018). 23. Muñoz, D.R.; Moreno, J.S.; Escrivà, C.R.; Berga, S.C.; Antón, A.E.N. Constant current drive for resistive sensors based on Generalized Impedance Converter. IEEE Trans. Instrum. Meas. 2008, 57, 2290–2296. [CrossRef] 24. Ramírez-Muñoz, D.; Sánchez, J.; Casans, S.; Reig, C.; Navarro, A.E. Series sensor current loop from a Generalized Impedance Converter circuit with reference current input. In Proceedings of the 2006 IEEE Instrumentation and Measurement Technology Conference Proceedings, Sorrento, Italy, 24–27 April 2006; pp. 2265–2270. 25. Cuenca, A.D.; Beltran, L.O.; Vilela, J.P.T.; Miranda, J. Comparison of microstrain indicators measurements based on Anderson’s Loop and Wheatstone Bridge. In Proceedings of the 16th International Conference on Electronics, Communications and Computers (CONIELECOMP’06), Puebla, Mexico, 1 March–27 February 2006. 26. Texas Instruments. TLC555 LinCMOS TM Timer; Texas Instruments: Dallas, TX, USA, 2016. 27. Franco, S. Active Filters: Part II” in Design with Operational Ampliﬁers and Analog Integrated Circuits, 3rd ed.; McGraw-Hill: New York, NY, USA, 2001; pp. 160–210. 28. New Feature: High Speed Continuous Recording—Redpitaya Forum. Available online: https://forum. redpitaya.com/viewtopic.php?f=7&t=317 (accessed on 25 July 2019).

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).