Conference PaperPDF Available

Novel and low-cost temperature compensation technique for piezoresistive pressure sensors

Authors:

Abstract and Figures

This paper proposes a low-cost technique to compensate for the temperature dependence of piezoresistive pressure sensors configured in a Wheatstone bridge. The sensor is treated as a resistive circuit and three equivalent resistances are measured by setting appropriately the bridge terminals. One of these equivalent resistances depends on temperature but not on pressure and, hence, it can be used to compensate for the temperature dependence of the output parameter. In such a way, neither data of a previous temperature calibration nor additional components are required to compensate for the temperature dependence. The proposed technique is applied to a commercial pressure sensor, which is measured by means of a direct sensor-to-microcontroller interface circuit. Experimental results show that temperature effects on the pressure measurement decrease more than ten times when the proposed technique is applied. Semiconductor sensors are widely used in current measurement systems because of their low cost, small size and easy integration with electronic circuits [1]. In the automotive industry, for example, mechanical semiconductor sensors have become very widespread, especially those intended for the measurement of pressure and acceleration [2]. Such mechanical sensors generally rely on either piezoresistances or capacitances. Semiconductor pressure sensors based on piezoresistances are usually configured in a Wheatstone bridge with four active arms, as shown in Fig. 1. In the event of a pressure change, two piezoresistors (R 1 and R 4) undergo a resistance change that is equal but opposite to that of the other two piezoresistors (R 2 and R 3). In such a way, the sensitivity of the bridge is higher than in bridges with only one or two active arms [3]. Figure 1. Piezoresistive pressure sensor in a Wheatstone-bridge configuration.
Content may be subject to copyright.
XIX IMEKO World Congress
Fundamental and Applied Metrology
September 611, 2009, Lisbon, Portugal
NOVEL AND LOW-COST TEMPERATURE COMPENSATION TECHNIQUE
FOR PIEZORESISTIVE PRESSURE SENSORS
Ferran Reverter 1, Goran Horak 2, Vedran Bilas 2, Manel Gasulla 1
1 Instrumentation, Sensors and Interfaces Group, Castelldefels School of Technology,
Universitat Politècnica de Catalunya, Castelldefels (Barcelona), Spain, reverter@eel.upc.edu
2 Faculty of Electrical Engineering and Computing, University of Zagreb, Zagreb, Croatia, Vedran.Bilas@fer.hr
Abstract This paper proposes a low-cost technique to
compensate for the temperature dependence of
piezoresistive pressure sensors configured in a Wheatstone
bridge. The sensor is treated as a resistive circuit and three
equivalent resistances are measured by setting appropriately
the bridge terminals. One of these equivalent resistances
depends on temperature but not on pressure and, hence, it
can be used to compensate for the temperature dependence
of the output parameter. In such a way, neither data of a
previous temperature calibration nor additional components
are required to compensate for the temperature dependence.
The proposed technique is applied to a commercial pressure
sensor, which is measured by means of a direct sensor-to-
microcontroller interface circuit. Experimental results show
that temperature effects on the pressure measurement
decrease more than ten times when the proposed technique
is applied.
Keywords: pressure sensor, piezoresistive sensor,
temperature compensation, sensor electronic interface,
microcontroller.
1. NOMENCLATURE
P Pressure applied to the sensor
Patm Atmospheric pressure
T Actual temperature
T0 Reference temperature defined by the sensor
manufacturer
T0 Reference temperature defined by the sensor user
R1-R4 The four piezoresistors of the bridge sensor
Rn Nominal value of the piezoresistors at P = 0
R0 Resistance value of Rn at T = T0
R0 Resistance value of Rn at T = T0
α
R Temperature coefficient of Rn using T0 as a reference
α
R Temperature coefficient of Rn using T0 as a reference
S Piezoresistive sensitivity
S0 Piezoresistive sensitivity at T = T0
α
S Temperature coefficient of S using T0 as a reference
M Output parameter that combines the measurement of
the equivalent resistances without compensation
M* Output parameter that combines the measurement of
the equivalent resistances with compensation
2. INTRODUCTION
Semiconductor sensors are widely used in current
measurement systems because of their low cost, small size
and easy integration with electronic circuits [1]. In the
automotive industry, for example, mechanical
semiconductor sensors have become very widespread,
especially those intended for the measurement of pressure
and acceleration [2]. Such mechanical sensors generally rely
on either piezoresistances or capacitances.
Semiconductor pressure sensors based on
piezoresistances are usually configured in a Wheatstone
bridge with four active arms, as shown in Fig. 1. In the event
of a pressure change, two piezoresistors (R1 and R4) undergo
a resistance change that is equal but opposite to that of the
other two piezoresistors (R2 and R3). In such a way, the
sensitivity of the bridge is higher than in bridges with only
one or two active arms [3].
Figure 1. Piezoresistive pressure sensor in a Wheatstone-bridge
configuration.
A remarkable drawback of piezoresistive pressure
sensors is their temperature dependence [1] and, for this
reason, several temperature compensation techniques have
been proposed. Generally, these techniques involve
additional components [4] (e.g. a thermistor) that increase
the effective voltage applied to the bridge when the
temperature increases so that the loss of sensitivity is
compensated [5]. The temperature dependence can also be
compensated by measuring the sensor temperature by means
of the sensor itself. For example, references [6, 7] propose a
2084
ISBN 978-963-88410-0-1 © 2009 IMEKO
relaxation oscillator in which the frequency of the output
signal carries information about the bridge unbalance
whereas the duty cycle carries information about the bridge
resistance and, hence, the temperature. Reference [8]
proposes (for metallic strain gages, not for semiconductor
ones) a technique based on two current sources and two
differential voltage measurements, one with information
about the mechanical stress and the other with information
about the temperature.
This paper proposes a novel technique to compensate for
the temperature dependence of piezoresistive pressure
sensors by using a simple and low-cost microcontroller-
based interface circuit. The proposed technique uses the
same pressure sensor to measure the temperature and, hence,
it does not require either data of a previous temperature
calibration or additional components.
3. TEMPERATURE DEPENDENCE
The four piezoresistors of the pressure sensor (Fig. 1)
can be modelled by [9]
[
]
14n
1
RRR SP
== +
(1a)
[
]
23n
1
RRR SP
, (1b)
in which both Rn and S depend on temperature.
Consequently, if we assume a first-order temperature
dependence and the same temperature coefficients for the
four piezoresistors, these can be modelled by
(
)
(
)
(
)
140 R 0 0 S 0
111
RRR TT S TTP
αα
⎡⎤
== + + +
⎡⎤
⎣⎦
⎣⎦
(2a)
(
)
(
)
(
)
230 R 0 0 S 0
111
RRR TT S TTP
αα
⎡⎤
== + − +
⎡⎤
⎣⎦
⎣⎦
.(2b)
According to [1],
α
R is a positive temperature coefficient
(i.e.
α
R > 0) whereas
α
S is a negative temperature coefficient
(i.e.
α
S < 0). Data sheets of commercial pressure sensors
usually specify
α
R and
α
S by means of the “temperature
coefficient of input resistance” and the “temperature
coefficient of span”, respectively. These coefficients depend
on the temperature used as a reference by the manufacturer
(T0 according to our nomenclature).
If a constant supply voltage (VDD) is applied to the bridge
sensor and then the differential output voltage (vo) is read, as
shown in Fig. 1, the measurement result is
(
)
(
)
oDD0 S 0
1
vVS TTP
α
=+
, (3)
which depends on pressure but also on temperature. Since
α
S < 0, for a given value of P, the value of vo decreases with
temperature. Generally, this temperature dependence is
compensated by using additional components that increase
the effective voltage applied to the bridge when the
temperature increases [4, 5].
Bridge sensors can also be measured by determining
three equivalent resistances of the bridge [10]. To do so, the
bridge terminals are appropriately set to ground or in high-
impedance state (HZ), as shown in Fig. 2, and the resulting
equivalent resistances are:
(
)
41 2 3
eq,A
1234
RR R R
R
RRRR
++
=+++ (4a)
(
)
1234
eq,B
1234
)(
RRRR
R
RRRR
++
=+++ (4b)
(
)
21 3 4
eq,C
1234
RR R R
R
RRRR
++
=+++ . (4c)
Replacing (2) in (4) yields
() ()
()
0
eq,A R 0 0 S 0
1321
4
R
RTTSTTP
αα
⎡⎤
≈+ − + +−
⎡⎤
⎣⎦
⎣⎦
(5a)
(
)
eq,B 0 R 0
1
RR TT
α
=+ −
⎡⎤
⎣⎦
(5b)
() ()
()
0
eq,C R 0 0 S 0
1321
4
R
RTTSTTP
αα
⎡⎤
≈+ − − +
⎡⎤
⎣⎦
⎣⎦
. (5c)
Then, the measurand can be estimated by calculating the
parameter M that combines the measurement of the
equivalent resistances as follows [10]:
eq,A eq,C
eq,B
RR
MR
=. (6)
Replacing now (5) in (6) yields
(
)
(
)
0S 0
1
MS TTP
α
=+ − , (7)
which, the same as in (3), depends on pressure but also on
temperature. Therefore, since
α
S < 0, for a given value of P,
the value of M also decreases with temperature.
4. COMPENSATION TECHNIQUE
To compensate for the temperature dependence shown in
(7), we propose to estimate the sensor temperature by means
of the sensor itself. This can be done by using appropriately
the measurement of Req,B, which just depends on
temperature, as shown in (5b). Then, in order to achieve an
output independent of temperature (i.e. M* = S0P), we could
calculate the following parameter:
*
eq,B
S
R0
11
M
MR
R
α
α
=
⎛⎞
+−
⎜⎟
⎝⎠
, (8)
where M is calculated by (6),
α
S and
α
R are specified by the
sensor manufacturer, Req,B is the measured equivalent
resistance at a given temperature T, and R0 is the value of
Req,B at T = T0. Regrettably, the tolerance of R0 is either very
large (say, 20 % or higher) or not specified at all in sensor
data sheets. Of course, R0 could be measured at T = T0, but
this involves time and specific temperature-calibration
instrumentation.
2085
Figure 2. Bridge sensor treated as a resistive circuit and measurement of the equivalent resistances (a) Req,A, (b) Req,B and (c) Req,C.
To avoid the previous limitations, we propose to use T0
as a reference instead of T0; T0 can be, for example, room
temperature whenever it does not differ significantly from
T0. In such conditions, we can assume
α
R
α
R and, hence,
Req,B can be expressed and approximated to
(
)
(
)
'' '' '
eq,B 0 R 0 0 R 0
11RR TTR TT
αα
⎡⎤⎡⎤
=+ −≈+ −
⎣⎦⎣⎦
. (9)
Then, instead of (8), we propose to calculate M* as follows:
*
eq,B
S
'
R0
11
M
MR
R
α
α
=
⎛⎞
+−
⎜⎟
⎝⎠
, (10)
where R0 is the value of Req,B registered at T = T0.
Replacing now (7) in (10) yields
(
)
(
)
*'
0S00
1
MS TTP
α
≈+ − . (11)
which just depends on pressure because the term (T0 T0) is
constant.
This compensation technique can also be applied to
conventional conditioning circuits based on the
measurement of the differential output voltage (Fig. 1). If
the circuit is able to measure both vo and Req,B, then the
information provided by the latter can be used to
compensate for the temperature dependence of the former
(see (3)).
5. MATERIALS AND METHOD
The proposed temperature compensation technique has
been applied to a commercial pressure sensor (SX15AD2,
SensorTechnics), which is intended for absolute pressure
measurements. This sensor is internally configured in a
Wheatstone bridge with four active arms (Fig. 1) and does
not include either compensation electronics or conditioning
circuits inside. Table I summarises the main features of this
sensor.
Table 1. Features of the SX15AD2 sensor.
Feature Value
Operating pressure range [0, 103] kPa
Sensitivity 214 µV/V/kPa (typ)
Operating temperature range [-40, +85] ºC
Temperature coefficient of input
resistance (
α
R) using T0 = 25 ºC
(not 100 % tested)
+690 ppm/ºC (min)
+750 ppm/ºC (typ)
+810 ppm/ºC (max)
Temperature coefficient of span
(
α
S) using T0 = 25 ºC
(not 100 % tested)
-2550 ppm/ºC (min)
-2150 ppm/ºC (typ)
-1900 ppm/ºC (max)
The sensor was subjected to atmospheric pressure, which
was equal to 101 kPa according to the Croatian national
meteorological institute. The temperature applied to the
sensor was controlled by a climatic chamber (Weiss Technik
125 SB) from -40 ºC to 70 ºC. For each test condition, the
three equivalent resistances (Fig. 2) were measured using
the direct sensor-to-microcontroller interface circuit
proposed in [10], which was controlled by the
dsPIC30F3012 microcontroller (Microchip). For the
temperature compensation, we used T0 26 ºC as a
reference, which is near to T0 = 25 ºC.
6. EXPERIMENTAL RESULTS AND DISCUSSION
Fig. 3 (solid line and crosses) shows the normalized
value of M versus temperature at P = Patm when the
compensation technique was not applied (Eq. (6)). The
parameter M clearly depended on temperature, to be precise,
the temperature coefficient was -2700 ppm/ºC. As predicted
by (7), the temperature coefficient of M is negative, however
its (absolute) value is slightly higher than the maximum one
expected (-2550 ppm/ºC according to Table 1). In the worst
case (i.e. for T -40 ºC), the relative error in M (and, hence,
in the pressure measurement) due to temperature was about
17 %.
2086
Fig. 3 (dashed line and squares) also shows the
normalized value of M* versus temperature at P = Patm when
the compensation technique was applied (Eq. (10)). The
parameter M* was more insensitive to temperature, as
predicted by (11). The use of the typical values of
α
R and
α
S
in (10) and/or the temperature dependence of the interface
circuit can explain why the response in Fig. 3 is not
completely flat. Even so, in the worst case (i.e. for
T 70 ºC), the relative error due to temperature was about
1.6 %, which is more than ten times smaller than that
obtained when the compensation technique was not applied.
-40 -20 020 40 60 80
0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
Temperature (ºC)
Normalized output parameter
Without compensation
With compensation
Figure 3. Temperature dependence of the output parameter at
P = Patm = 101 kPa.
7. CONCLUSIONS
A novel and low-cost technique to compensate for the
temperature dependence of piezoresistive pressure sensors
has been proposed. This compensation technique relies on
estimating the temperature of the sensor by means of the
sensor itself, without using any additional component. Once
the temperature is known, the output parameter can easily be
corrected. This idea has been tested in a commercial
pressure sensor and the experimental results agree with the
theoretical predictions. Other semiconductor sensors (such
as acceleration sensors and magnetoresistive sensors) can
also benefit from the results of this work.
ACKNOWLEDGMENTS
This work has been funded by the Spanish Ministry of
Education and Science and the European Regional
Development Fund through Project DPI2006-04017, and by
the Croatian Ministry of Science, Education and Sport
through Project 036-0361621-1625. Authors appreciate the
initial proposal provided by Prof. Ramon Pallàs-Areny.
REFERENCES
[1] S.M. Sze, Semiconductor Sensors, Wiley, New York, 1994.
[2] G.C. Meijer, Smart Sensor Systems, Wiley, Chichester (UK),
2008 (Chapter 3).
[3] R. Pallàs-Areny and J.G. Webster, Sensors and Signal
Conditioning, John Wiley & Sons, New York, 2001, 2nd Ed.
[4] J. Fraden, Handbook of Modern Sensors. Physics, Designs,
and Applications, Springer, New York, 2004, 3rd Ed.
[5] D. Ramírez Muñoz, J. Sánchez Moreno, S. Casans Berga, E.
Castro Montero, C. Reig Escrivà and A. Edith Navarro
Antón, “Temperature compensation of Wheatstone bridge
magnetoresistive sensors based on generalized impedance
converter with input reference current”, Rev. Sci. Instrum.,
vol. 77, 105102, 2006.
[6] V. Ferrari, D. Marioli and A. Taroni, “Oscillator-based
interface for measurand-plus-temperature readout from
resistive bridge sensors”, IEEE Trans. Instrum. Meas., vol.
49 (3), pp. 585-590, 2000.
[7] V. Ferrari, A. Ghisla, Zs. Kovács Vajna, D. Marioli and A.
Taroni, “ASIC front-end interface with frequency and duty
cycle output for resistive-bridge sensors”, Sensors and
Actuators A, vol. 138, pp. 112-119, 2007.
[8] A. Idzkowski, J. Makal and Z.L. Warsza, “Application of
double current bridge-circuit for simultaneous measurements
of strain and temperature”, Proceedings of the 24th IEEE
IMTC 2007, pp. 1-4, Warsaw, Poland, May 2007.
[9] S. Vlassis, S. Siskos and T. Laopoulos, “A piezoresistive
pressure sensor interfacing circuit”, Proceedings of the 16th
IEEE IMTC 1999, Volume 1, pp. 303 – 308, Venice, Italy,
May 1999.
[10] E. Sifuentes, Ò. Casas, F. Reverter and R. Pallàs-Areny,
“Direct interface circuit to linearise resistive sensor bridges”,
Sensors and Actuators A, vol. 147, pp. 210-215, 2008.
2087
... The measurement of temperature is also employed to indirectly estimate other non-thermal magnitudes [5], for example: 1) the direction and speed of wind can be determined by means of a set of thermal sensors circularly distributed with a heater at the center [6], and 2) the mass flow can be measured through a couple of thermal sensors and an intermediate heater correctly aligned inside a pipeline. Another essential role of thermal sensors is to compensate for the thermal drifts of other non-thermal sensors, such as mechanical sensors [7]. ...
Article
Two noteworthy events associated to the physics of thermal sensors were demonstrated and announced in 1821, exactly two hundred years ago. The first event was the Seebeck effect, which led to the development of thermocouples. The second was the study of the thermal dependence of the resistivity of pure metals, which led to the design of resistance temperature detectors (RTD). Sensors’ science and technology has experienced remarkable advances in the last decades, but these two types of thermal sensors, whose physics were announced 200 years ago, are still nowadays the main sensor technology in many industrial applications that require the measurement of extreme temperatures. In such a 200th200^{th} anniversary, this tutorial aims to explain the operating principle, subtypes, input-output characteristic, limitations, and new trends under research of the four main types of thermal sensor: RTDs, thermistors, silicon sensors, and thermocouples. These are also compared with each other, highlighting the main advantages and disadvantages of each sensor technology.
... Such sensors, exhibit high sensitivity but have significant temperature dependency of pressure reading [2]. Different techniques for temperature compensation of piezoresistive sensor can be found in the literature [3], [4]. This influence of temperature should also be taken into account when evaluating measurement uncertainty of calibration. ...
Article
Full-text available
This paper investigates accuracy, influence of temperature variations and time drift of water level data loggers, operating on a principle of hydrostatic pressure measurement. The laboratory setup for calibration of such devices, equipped with self-designed pressure chamber and commercial climatic chamber is presented. Calibration procedure for logger’s built-in pressure transducer, with ensured traceability to the primary standards of pressure and temperature, is proposed. Influence of temperature on the pressure reading was examined by conducting calibration at 10 °C, 15 °C and 25 °C.
... Direct interface circuits (DICs) [5] are a series of circuits based on methods that use a small number of additional components to make the connection between a sensor and a programmable digital device (PDD). There are DICs for resistive sensors [6][7][8][9][10][11][12][13][14], capacitive [15][16][17] or inductive [18][19][20][21][22] sensors, and even DICs built to measure any of them [23]. ...
Chapter
Full-text available
Direct interface circuits are a simple, inexpensive alternative for the digital conversion of a sensor reading, and in some of these circuits only passive calibration elements are required in order to carry out this conversion. In the case of resistive sensors, the most accurate methods of calibration, namely two-point calibration method (TPCM) and fast calibration methods I and II (FCMs I and II), require two calibration resistors to estimate the value of a sensor. However, although FCMs I and II considerably reduce the time necessary to estimate the value of the sensor, this may still be excessiveincertainapplications,suchaswhenmakingrepetitivereadingsofasensororreadingsofa large series of sensors. For these situations, this paper proposes a series of calibration methods that decrease the mean estimation time. Some of the proposed methods (quasi single-point calibration methods)arebasedontheTPCM,whileothers(fastquasisingle-pointcalibrationmethods)makethe most of the advantages of FCM. In general, the proposed methods significantly reduce estimation times in exchange for a small increase in errors. To validate the proposal, a circuit with a Xilinx XC3S50AN-4TQG144C FPGA has been designed and resistors in the range (267.56 Ω, 7464.5 Ω) have been measured. For 20 repetitive measurements, the proposed methods achieve time reductions of up to 61% with a relative error increase of only 0.1%.
... As an important intelligent material, piezoelectric quartz has attracted extensive attention of researchers for its high stiffness, high natural frequency and excellent static characteristics, especially for high-frequency dynamic force measurement (Reverter et al. 2009;Totis and Sortino 2011). Therefore, taking vertical large VFCD (vector force calibration device) as an example, a multi-station calibration method is proposed. ...
Article
Full-text available
High-accuracy measurement for vector force is essential in the process of industrial manufacture and control. Piezoelectric sensor has the characteristics of high frequency and stiffness, great dynamic response and resistance to harsh conditions, which are widely used to measure vector force. To solve the problem that vector force calibration device (VFCD) is difficult to comprehensively calibrate vector force test system (VFTS), a multi-station calibration method is proposed. Taking the six degree-of-freedom of vertical large VFCD as an example, mechanical model of vector force is established after matching output voltage and force value of piezoelectric dynamometer. Multi-station deflection calibration experiment is performed, and a fitting circle method is proposed to fit vector force parameters of multi-station under different step-force loadings. Finally, the results show that error of multi-station vector force calibration is less than 2%, which proves rationality of the calibration method. The error of fitting radius obtained by fitting circle method is less than 3.4%, which further verifies reliability of the calibration method, and provides a guarantee for VFTS to accurately measure vector force.
... As a result, the sensitivity of four active arms is higher than any one or two active arms. For this reason, Wheatstone bridge circuit configuration is considered [35]. When the pressure is applied to the diaphragm, it gets deformed. ...
Article
Full-text available
This paper deals with the design and simulation of a stripped-shaped piezoresistive pressure sensor. The sensor is based on the Wheatstone bridge configuration. The piezoresistive effect is a change in the electrical resistivity of a semiconductor or metal when mechanical strain is applied. For designing the sensor, piezoresistors are placed in series on each side of the square. The number of the piezoresistor arms is varied in two proposed designs where the sensitivity and the operable pressure range are analyzed with their characteristics. The COMSOL Multiphysics software is used for this work. The simulated sensitivity is about 7.736 mV/V/MPa over the range of pressure from 0 to 5 MPa. The simulated result reveals that the sensitivity significantly depends on the placement of the resistors, combination of the materials, shape of diaphragm, and size of diaphragm.
... Direct interface circuits (DICs) [5] are a series of circuits based on methods that use a small number of additional components to make the connection between a sensor and a programmable digital device (PDD). There are DICs for resistive sensors [6][7][8][9][10][11][12][13][14], capacitive [15][16][17] or inductive [18][19][20][21][22] sensors, and even DICs built to measure any of them [23]. ...
Article
Full-text available
Direct interface circuits are a simple, inexpensive alternative for the digital conversion of a sensor reading, and in some of these circuits only passive calibration elements are required in order to carry out this conversion. In the case of resistive sensors, the most accurate methods of calibration, namely two-point calibration method (TPCM) and fast calibration methods I and II (FCMs I and II), require two calibration resistors to estimate the value of a sensor. However, although FCMs I and II considerably reduce the time necessary to estimate the value of the sensor, this may still be excessive in certain applications, such as when making repetitive readings of a sensor or readings of a large series of sensors. For these situations, this paper proposes a series of calibration methods that decrease the mean estimation time. Some of the proposed methods are based on the TPCM and quasi single-point calibration methods, while others make the most of the advantages of FCM II and fast quasi single-point calibration methods. In general, the proposed methods significantly reduce estimation times in exchange for a small increase in errors. To validate the proposal, a circuit with a Xilinx XC3S50AN-4TQG144C FPGA has been designed and resistors in the range (267.56 Ω, 7464.5 Ω) have been measured. For 20 repetitive measurements, the proposed methods achieve time reductions of up to 61% with a relative error increase of only 0.1%.
... Compared to strain sixaxis force/torque sensors, piezoelectric six-axis force/ torque sensors with high stiffness, high sensitivity, high natural frequency, and excellent dynamic characteristics are suitable for static and dynamic vector force measurement. [14][15][16] Piezoelectric six-axis force/torque sensors generally adopt multi-point support combined measuring structure, and dimension coupling is an important factor limiting the test accuracy. To improve the test accuracy of piezoelectric six-axis force/torque sensors, some articles have researched dimension coupling. ...
Article
Full-text available
In the process of manufacturing and operation of mechanical equipment, accurate measurement of vector force provides an important basis for real-time control of working conditions and process adjustment to ensure the performance. Piezoelectric three-component force unit can be used to measure three-axis force with the characteristics of high frequency and stiffness, great dynamic response, and resistance to harsh conditions. Piezoelectric dynamometer consisting of four three-component force units can measure six-axis force/torque, which is widely used in vector force measurement. However, because of the limitation of manufacturing capability and assembly process, dimension coupling caused by assembly error of three-component force unit restricts improvement of test accuracy for itself and piezoelectric dynamometer. To reduce dimension coupling caused by displacement and angle assembly error, this article establishes a theoretical model of assembly error of three-component force unit. To achieve that the dimension coupling of three-component force unit is less than 3%, stiffness of relevant components is obtained by simulation analysis and experiment, and assembly tolerance range of three-component force unit is proposed. Four three-component force units are assembled to meet assembly tolerance range, and the calibration experiment is performed. The results show that the maximum of dimension coupling is less than 2.5%, which verifies the rationality of assembly tolerance range and further proves the reliability of the theoretical model. The theoretical model of assembly error clarifies the assembly accuracy of three-component force unit to ensure its test accuracy, which can further improve test accuracy of dynamometer. This article will also provide theoretical reference basis for this kind of sensors.
... From the table 5.4 it clearly knows that the silicon based piezoresistive pressure sensor achieves low sensitivity as compared to Silicon On insulator based Piezoresistive pressure sensor. Because, of the high sensitivity the Silicon On insulator based Piezoresistive pressure sensor is widely used in high temperature applications [14]. ...
Article
Full-text available
In this paper we designed Piezoresistive pressure sensor using silicon on insulator based diaphragm. The silicon on insulator material is used to improve the sensitivity and also used to archive high reliability. The main aim of this SOI technology is silicon does not withstand high temperature in many applications so we use SOI instead of silicon and it will also act as an oxide layer between, piezoresistor and the substrate providing insulation. Piezoresistive pressure sensor employs a change in resistance values of piezoresistive elements placed on the surface of the diaphragm. The simulation and modeling of proposed pressure sensors is done using COMSOL Multiphysics. Based on simulation results, SOI based Piezoresistive pressure sensor provides high linearity and better sensitivity.
Article
Accurate measurement of thrust in rocket engine, as a core parameter of attitude control of rocket engine, is an important guarantee for reliable operation of rocket engines. Therefore, it is very important for accurate calibration. In this paper, a novel rotation calibration method for rocket thrust engine is proposed based on piezoelectric force measurement system. Based on the calibration matrix, sensitivity analysis model of calibration accuracy to coupling relationship between axes is established, and error mechanism and offset mechanism of force are studied. Starting from the deviation principle between actual axis of thrust system and theoretical one, an elliptical geometrical distribution law of lateral force caused by main force is obtained. According to the periodicity of ellipse distribution of calibration force, a new calibration method, rotation calibration method, is obtained based on an idea of homogenization error. Finally, vector force verification is designed, and experimental data are calibrated with the rotation calibration method compared with the traditional linear calibration method, verifying feasibility and rationality of the calibration method. The novel calibration can be also used in force measurement in any rotation devices like manipulator in robot,thrust vector engine, manufacturing force, rocket engine,etc.
Article
This paper presents a smart pressure transmitter using bellow as primary sensor. The deflection of bellow is converted into electrical output using hall probe sensor as secondary sensor. The output Hall voltage is affected by change in input parameters like temperature. So firstly the effect of temperature on Hall voltage is derived mathematically and then experimentally analyzed. This effect of temperature on output Hall voltage is then compensated using artificial neural network. The compensated output Hall voltage is then converted into (4–20) mA current signal using signal conditioning circuit. The proposed design, experimental and testing results are reported in this paper.
Article
A signal conditioning circuit based on a relaxation oscillator is proposed for use with resistive bridge sensors. The circuit provides a rectangular-wave output whose frequency is related to the bridge imbalance and the duty-cycle is a function of the overall sensor bridge resistance, and hence of its operating temperature. In this way, two segments of information are simultaneously and independently carried on the same output signal. The circuit makes use of a constant current bridge excitation which enables the connection of remote sensors without accuracy degradation, and, for silicon piezoresistive sensors, provides a first order temperature compensation. A frequency-doubling output stage significantly reduces the nonlinearity due to switching delay times, at parity of output center frequency and span. Experimental results are reported on the characterization of both the circuit alone and interfaced to a silicon pressure sensor
Article
An application specific integrated circuit (ASIC) front-end interface in 0.7-μm CMOS for resistive-bridge sensors is proposed. The circuit is based on a relaxation oscillator where the frequency of the rectangular-wave output is related to the fractional bridge unbalance, and the duty cycle depends on the overall bridge resistance, which typically is related to temperature. In this way, two independent pieces of information are simultaneously and cost-effectively carried on the same output signal. The bridge is driven at constant current, this avoids accuracy degradation with remotely placed sensors and enables a first-order thermal compensation for piezoresistive semiconductor sensors. The circuit has been characterized by means of a 1-kΩ reference bridge showing frequency and duty cycle sensitivities of 60.4Hz/(1000ppm) and 0.276%/(mΩ/Ω), respectively, at a central frequency of about 6.4kHz. The circuit has also been tested with a piezoresistive SiC sensor operated at temperatures up to 150°C, showing results in agreement with theoretical predictions.
Article
A compensation method of the sensitivity drift of the Wheatstone bridge sensor is proposed. The technique was carried out by placing a temperature sensor and the bridge to be compensated inside a generalized impedance converter with input reference current. No internal modification of the bridge arms is required so that the circuit is capable to compensate practical industrial sensors. The method is based on the temperature modulation of the current supplied to the bridge, which improves previous solutions based on constant current compensation. Experimental results are shown using a commercial magnetoresistive bridge sensor.
Conference Paper
A four-terminal (4T) bridge-circuit unconventionally supplied by current sources connected in parallel to opposite arms, named double current bridge (2J) is presented. It has two different outputs from both diagonals. Their output voltages as the functions of arm resistance increments in absolute and in relative units are given. Example of its application is proposed. Signals conditioning formulas of the two-parameter (2D) measurement - of strain and temperature are discussed in details. Some results of experimentally design instrumentation of this bridge-circuit are shortly presented.
Article
This paper proposes the direct connection of different configurations of resistive sensor bridges to a microcontroller without any intermediate active component. Such a direct interface circuit relies on measuring the discharging time of a RC network that includes the resistances of the sensor bridge. For quarter-, half-, and full-bridge circuits, we combine the discharging times to estimate the fractional resistance change x of the bridge arms. Experimental results for half- and full-bridge circuits emulated by resistors yield a nonlinearity error below 0.3%FSR (full-scale range) for x between 0 and 0.1 and an effective resolution of 11 bit. Measurements on two commercial magnetoresistive sensors yield higher nonlinearity errors: 1.8%FSR for an AMR (Anisotropic Magnetoresistive) sensor and 5.8%FSR for a GMR (Giant Magnetoresistive) sensor, which are mainly due to the nonlinearity of the sensors themselves. Therefore, the nonlinearity of the measurement is limited by the sensors, not by the proposed interface circuit and linearisation algorithm.
Conference Paper
A signal conditioning circuit with digital output for piezoresistive pressure sensors based on CMOS current mode building blocks is presented. The proposed circuit uses an instrumentation amplifier (IA) based on operational floating amplifiers (OFA), adapted for resistive sensors. This IA offers a output current which is independent of tracking and offset errors. Two piezoresistors with current supply are used against four piezoresistors of the basic Wheatstone bridge. A new temperature compensation technique for the pressure sensitivity is described. Also an accurate SPICE model for piezoresistive sensors is introduced. Simulation as well as experimental results are included to demonstrate the performance of the circuit