Proc.

IX National Conference with International Participation "Electronica 2018", May 17 - 18, 2018, Sofia, Bulgaria

Analog Front End Multiplexing for Amperometric

Gas Sensors
Mаrin B. Marinov, Georgi T. Nikolov and Borislav Ganev
Technical University of Sofia, Faculty of Electronic Engineering and Technologies, 1756 Sofia, Bulgaria,
E-mail: {mbm, gnikolov, b_ganev}@tu-sofia.bg

The main goal of our study is to develop low-cost, mobile,

Abstract – Advances in modern gas sensors and in smart
systems development have made it possible to have new cost air quality sensor nodes and to explore whether such systems
effective and accessible air quality monitoring tools. This paper can provide reliable results and indications about air quality
presents a multiplexing of commercially available analog front- and can be used in practice. This approach should provide
end (AFE) for four-electrode amperometric gas sensors. This easier access to air quality monitoring data to a wider
approach allows for the easy expansion of the number of audience of citizens, scientists and control authorities.
channels for analogue-to-digital conversion, which is important
especially for multi-sensor systems for air quality monitoring.
Along with this additional air parameters can be measured and
II. MATERIALS AND METHODS
thus accuracy improvement can be achieved by applying
corrective algorithms. The proposed sensor interface has been A. Amperometric gas sensors
tested in the field with a wireless module and shows accuracy
comparable to that of the multi-channel DAQ in laboratory There are different new technologies for measuring gas
conditions. concentrations. One of the most promising near-term
Keywords – air quality monitoring; amperometric gas technologies for accurate detection of many atmospheric
sensors; real-time monitoring; wireless sensor networks. pollutant gases are the commercially available
amperometric sensors.
I. INTRODUCTION This sensor technology has a lot of advantages such as
linear response to concentration, low fabrication cost,
Wireless sensor networks (WSNs) have the advantage of (relative) fast response, low power consumption and light-
being inexpensive and easy to set up. They also enable real- weight, all of which are desirable for the so-called “next
time monitoring of air quality. It is important to make generation” of air monitoring tools.
accurate real-time monitoring in urban areas as air quality Sensor-based monitoring applications bring new
affects citizens’ health and necessitates making timely opportunities for air monitoring but there are important
public decisions to counteract any negative consequences. issues related to data quality in sensor applications, which
However, citizens do not often have the opportunity to need to be considered. Studies have demonstrated that
obtain relevant air quality data. Practically, the equipment sensor data are subjected to considerable influence from
used to monitor air quality is usually placed in fixed large environmental factors such as temperature, humidity, and
and expensive sensing stations in strategic locations, and is even interference of other air pollutants [4, 5].
managed by public authorities. Thus, monitoring is limited Four-electrode amperometric gas sensors from
to a few designated areas and sometimes citizens do not have Alphasense, in particular the A4 models, have been chosen
access to the measurements collected in this way. for this implementation. These sensors are intended for
Advances in embedded systems and new gas sensors mobile air quality monitoring and provide:
technologies make it possible for a new generation of low- • low noise and sensitivity down to the very low ppb
cost air quality monitoring systems to emerge. Low-cost air range;
quality sensors drive the development of new electronic • low cross-sensitivities to other gas species;
technologies, which are increasingly widely accepted and • near linear response across a wide detection range and
implemented. The portable and autonomous sensors have 1% gain accuracy from the factory;
the potential to take measurements with sufficient accuracy • built-in temperature compensation mechanism that
and in this way to capture effectively the spatial variability provides an offset error signal on a fourth auxiliary
of air pollutants. The number of these commercially electrode (AE) pins [6].
available devices has increased considerably over the last The sensing range, the typical sensitivity and the full scale
five years although the quality of the data which they output of the sensors used are given in Table 1.
provide is still questionable [1, 2]. The amperometric 4-electrode sensors are equipped with
To increase the spatio-temporal resolution of air pollution an auxiliary electrode (AE) for correction of background, or
information, researchers and local communities try to zero currents. In addition to the current created by the main
expand measurement capabilities by using, in addition to redox reaction involving the target gas, the background
traditional devices, low-cost, portable devices and sensors current is an additional current created by ”solid electrolyte
[3]. processes or from electrochemistry involving the
electrolyte”. AE is a catalyst similar to working electrode

978-1-5386-5801-7/18/$31.00 ©2018 IEEE

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on April 03,2024 at 12:19:20 UTC from IEEE Xplore. Restrictions apply.
(WE), but placed inside the sensor, i.e. not in contact with AFEs are analogue potentiostat circuits with on-board power
the target gas, thus, both catalysts should in theory be regulation and a set of reference voltages. The power
affected equally by changes in the background current. requirement is about 0.65 per channel. The negative
However, there may be a “contribution to the background supply is generated on the AFE so only power supply of
current from oxygen reduction at WE”, which is not 3.4 − 6.4 and analogue ground is required. Each AFE
captured by AE as the oxygen supply to AE will be limited includes a Pt1000 sensor for temperature compensation with
by solubility and diffusion rate in the electrolyte [6]. output of 1 / [7].
TABLE 1. DATASHEET INFORMATION FOR THE USED AMPEROMETRIC The AFE has a number of disadvantages:
SENSORS AND SENSITIVITY OF THE AFE. • AFEs are not user adjustable: offset voltage for each
Pollutant О sensor is defined in the calibration document (two offsets
Sensor NO2-A43F OX-A431 CO-A4 SO2-A4 for each sensor: working electrode offset and auxiliary
Sensitivity, / -0.253 -0.592 0.352 0.423 electrode offset) which you program into your software.
Warranty range, 20 20 500 50 AFE gain is preset. The calibration document also states
Circuit gain, mV/nA 20 0.423 SO2-A4 10 the / calibration for each WE, which you
Response time, , < 60 < 45 20 < 20 program into your software.
Operating life, months >24 >24 >36 >36 • All channels are all analog-only, and still require at least
Sensitivity, mV/ 0.186 0.441 0.226 0.356 good 12-bit - 16-bit (or better) analog-digital conversion
for use in a digital system.
The subtraction of AE from WE to correct for background • Requires a low noise precise (± 0.1 ) power supply.
currents and the subtraction of the zero current of the • The AFE boards are specifically customized and tuned at
individual electrodes is represented as [7]: the factory for each particular sensor model's bias and
gain requirements, and must be returned to the factory
( − )−( − )
= (1) for recalibration.
• The configuration of the measurement circuits is related
where is the measured gas concentration in , and to earth which increases the number of the channels for
are the measured signals in , and are the analog-digital conversion.
zero voltage offsets of and respectively in , S is
the total sensitivity, / (see Table 1). C. Multiplexor
A scaling factor has to be introduced as the areas of WE
and AE are not equal. Furthermore, the active area may There are no special requirements for the multiplexer
“vary with temperature because of gas and liquid expansion (MUX). For the needs of the present study it is necessary for
in the membrane pores” and “the electrochemistry could be the MUX to perform as follows:
different because of the different environments”. Thus, the • Low power consumption;
scaling factor, should be temperature dependent. Alphasense • Single-supply capability (3V to 5V);
Ltd. recommends the following interdependence: • Digital signals range from 3 V to 5 V;
( − )− ( − ) • Relatively low resistance in ON mode;
= , (2) • High resistance in OFF mode;
Since the change in the concentration of air pollutants is a
where n is a temperature dependent scaling factor, whose slowly changing value, there are no requirements for the
value has been determined in laboratory experiments. propagation delay time of the multiplexer. The present
Recommended factor values are given by the manufacturer project uses a multiplexer CD4051 from Texas Instruments
[6]. but ADG408 from Analog Devices can also be used and or
So the basic parameters of the amperometric sensors, such any other compatible device.
as sensitivity, zero current, response time have strong
temperature dependence. Pressure and humidity also affect D. Data acquisition modules
sensor output [8].
Combining measurements of amperometric gas sensors The NI 6289 high-accuracy multifunction data acquisition
with synchronous measurement of basic atmospheric module is selected as precision voltage measurement device.
parameters, such as pressure, temperature and humidity This device incorporates a low noise programmable gain
makes it possible to use corrective algorithms and improve instrumental amplifier and has 18 bit resolution with an
accuracy of the main pollutant measurements. This absolute accuracy of 28 at ±10 input range.
improvement leads to the necessity of increasing the number The selected wireless node is WSN 3202. This ZigBee
of measuring channels. based data acquisition system offers four 16 bit analog input
channels, with programmable ranges of ±0.5, ± 2, ± 5
B. Four-sensor Analogue Front End board and ±10 . The device incorporates four digital I/O
channels configured for sourcing output, sinking output and
The amperometric air quality sensors require high- input.
precision low noise electronics to optimize their Both devices are controlled in the LabVIEW graphical
performance. The family of Analogue Front End circuits programming environment.
from Alphasense are for up to four sensors and are designed The purpose of the study is to estimate the error caused by
for use with the A4 amperometric air quality sensors. The the multiplexing of the channels. An experimental set-up has

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on April 03,2024 at 12:19:20 UTC from IEEE Xplore. Restrictions apply.
been implemented in which the sensor signals are measured III. EXPERIMENTAL RESULTS
simultaneously with the 18-bit DAQ and the 16-bit wireless
module to which a multiplexer is attached (Fig. 1). A number of laboratory experiments were carried out to
Multiplexing increases the number of channels for determine the responses in different weather conditions.
analogue-to-digital conversion, and the number of measured For the purposes of the experiment, a graphical program
values from one wireless module. Thus, measurement code, or the so-called block diagram is created, which is
accuracy increase can be achieved through the use of shown in Fig. 3. This LabVIEW code consists of DAQmx
corrective algorithms. instrument drivers, “While loop” and “Shared variables”.
DAQmx drivers are used to control the precise
multifunctional data acquisition system NI USB 6289. These
drivers are located in the middle of the block diagram shown
in the figure. At the top of the while loop are positioned
shared variables to control the DIO0 to DIO2 digital outputs
of the wireless sensor node NI WSN 3202. Although
network-published shared variable communication is non-
deterministic, this approach is the easiest and is suitable for
the present experiment. The same approach with network-
published shared variable is also used to read the result from
analog input AIO of the sensor node. The simultaneous
measured voltages of the precise data acquisition system and
of the wireless sensor node are processed in a subVI and the
absolute and relative measurement errors are displayed. For
Fig. 1. Scheme of AFE for 4 sensors, each with 2 electrodes, illustrative purposes only, at the bottom of the block diagram
incorporation of multiplexor, WSN and DAQ is positioned a shared variable TCO for connection to a
wireless sensor module WSN 3212 for measuring ambient
Fig. 2 shows the circuit diagram of two channels of the air temperature.
potentiostat circuit, multiplexor and WSN 3202.

Fig. 3. LabVIEW block diagram for comparison of the results

between NI DAQ 6289 and WSN 3202

The front panel of the developed virtual system for

assessing the impact of additional circuitry on the accuracy
Fig. 2. Two channels of the AFE potentiostat circuit, of the amperometric sensors is shown in Fig. 4. The figure
multiplexor and WSN 3202 shows the channel selection buttons and indicators of
measured and manipulated data.

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on April 03,2024 at 12:19:20 UTC from IEEE Xplore. Restrictions apply.
 The results obtained show that the relative error rates of
the experiments do not exceed 0.5 %. Such a result is Although the quality of data obtained with low-cost
entirely acceptable considering the systematic error of the sensors is worse than that received from reference devices,
amperometric sensors. the fact that there is an increasing interest in the surrounding
environment can have a positive impact on human behavior
and put pressure on the authorities to improve air quality.
[11, 12]

REFERENCES

[1] R. Williams, R. Long, M. Beaver et al., "Sensor Evaluation

Report," Environmental Protection Agency, Office of
Research and Development, Washington, DC, May 2014.
[2] S. Brienza, A. Galli, G. Anastasi and P. Bruschi, "A Low-
Cost Sensing System for Cooperative Air Quality
Monitoring in Urban Areas," Sensors, vol. 15, pp. 12242-
12259, 2015.
[3] M. B. Marinov, I. Topalov, E. Gieva and G. Nikolov, "Air
Quality Monitoring in Urban Environments," in 39th
International Spring Seminar on Electronics Technology
(ISSE), 18-22 May, Pilsen Czech Republic, 2016.
[4] M. I. Mead, O. A. M. Popoola, G. B. Stewart, P. Landshoff,
M. Calleja, M. Hayes et al., "The use of electrochemical
sensors for monitoring urban air quality in low-cost, high-
density networks," Atmospheric Environment, vol. 70, pp.
186-203, 2013.
[5] L. Spinelle, M. Gerboles, M. G. Villani, M. Aleixandre and
F. Bonavitacola,, "Spinelle, L.; Gerboles, M.; Villani, M.G.;
Aleixandre, M.; Bonavitacola, F. Field calibration of a
cluster of low-cost available sensors for air quality
monitoring. Part A: Ozone and nitrogen dioxide," Sens.
Actuators B Chem., vol. 215, p. 249–257, 2015.
[6] Alphasense Ltd. , "Application Note AAN 803-01:
Correcting for backgorund currents in four electrode toxic
Fig. 4. Front panel of the virtual system gas sensors," Sensor Technology House, Essex CM77 7AA
United Kingdom, 2015.
[7] Alphasense Ltd., "Application Note 072-0281: Alphasense
IV. CONCLUSIONS A4 Analogoue Front End (AFE) Circuit Board Family; User
Manual.," Sensor Technology House, Essex CM77 7AA
United Kingdom, 2015.
The development of alternative air quality monitoring
[8] Alphasense Ltd., "Application Note AAN 110:
networks based on low-cost sensors is very important from
Environmental changes: temperatute, pressure numidity,"
the perspective of city residents. Current research focuses on Sensor Technology House, Essex CM77 7AA United
the concept of a next-generation air pollution monitoring Kingdom, 2015.
systems that use new measurement technologies and
[9] "Department for Health and Environment of the City of
techniques for communicating and delivering data. These Munich (Germany)," 1 Feb. 2014. [Online]. Available:
systems can be complementary to traditional air quality http://tinyurl.com/2gg7dl.
monitoring systems. [10] "Smart City-Strategie Berlin," Senatsverwaltung für
The use of multi-parameter air quality measurement Stadtentwicklung und Umwelt - Berlin, 2015.
systems makes it possible to have a detailed level analysis
[11] Hinov N., Ts. Hranov, "Smart training environment for
and localization of the main pollution sources. These power electronics," in AIP Conference Proceedings, Vol. №
measurement systems are important components of many 1910, 060017-1–060017-5 (2017), ISBN: 978-0-7354-1602-
actual Smart-City-projects [9, 10]. The usage of 4, doi.org/10.1063/1.5014011, 2017.
amperometric sensors together with a programmable analog [12] L. Klein, Sensor and Data Fusion, Washington: SPIE-
front end for low-power chemical-sensing applications International Society for Optical Engineering,, 2004.
offers more advantages, such as low power consumption,
low cost, fast response, ability to produce real-time
measurement, etc.
This work presents a set-up, which allows to development
experiments for the study of the influence of basic
atmospheric factors on the amperometric sensor responses.
Thus different approaches for environmental impact
correction of the sensor responses and the quality of sensor
data can be tested.

Authorized licensed use limited to: Indian Institute of Technology- Goa. Downloaded on April 03,2024 at 12:19:20 UTC from IEEE Xplore. Restrictions apply.