Sensor Calibration
Sensor Calibration
Sensor Calibration
TEMPERATURE SENSOR
CALIBRATION
Resistors whose resistance changes dramatically with changes in temperature are called as thermistors.
The thermistor begins self-heating its components as the ambient temperature changes. With relation to
this change in temperature, its resistance value changes. As a result, the voltage across the thermistor
changes, allowing for the detection of temperature differences.
To model the resistance-temperature behavior of a thermistor we can use The Steinhart-Hart equation
which is given by,
1/T = A + B ln(R/R₀) + C ln(R/R₀)^3
Where: T = Kelvin units (°C + 273.15), and ln(R) = natural logarithm of a resistance in ohms. A, B, C,
D, and E are the Steinhart-Hart coefficients that vary depending on the type of thermistor used and the
range of temperature being detected.
To ensure accuracy and reproducibility, measurement equipment such as sensors and measuring systems
are calibrated. To get precise, dependable, and repeatable measurement results, sensors must be
calibrated. One of the essential prerequisites for successful quality assurance is calibration.
Sensor calibration, in the context of measurement technology, describes the procedures used to establish,
under specific conditions, the correspondence between the values of a measuring output produced by a
measurement instrument such as a sensor or measuring system and the corresponding values of a
measuring instrument established with standards.
3. The new temperature sensor is to be calibrated using the Wheatstone bridge principle. Design
a circuit to calibrate the temperature sensor and sketch it using the values calculated.
PROCEDURE
P
Q
.
Obtained voltage readings using the Wheatstone bridge by considering appropriate temperature
steps
4. Based on your calibration equation obtained, determine the corresponding temperature value
for the sensor output of 1 V
X = 177.5 °C or 25.5 °C
1. What are the issues in using the higher-order equations for calibration?
• Higher-order equations are often more complex than linear or polynomial equations, which
makes them potentially more challenging to use and computationally expensive.
• Higher-order equations can better match the data since they are more flexible and complex than
linear or polynomial equations. However,this might result in overfitting, when the model
underperforms on fresh, untested data because it becomes strongly customized to the particular
data used for calibration.
• Higher-order equations can be challenging to understand, particularly when there are numerous
terms included. Understanding the underlying relationship between the input and output
variables can be difficult as a result.
• The calibration data's range may limit how accurately higher-order equations extrapolate. This
indicates that predictions produced using the model outside of the calibration data's range may
not be accurate.
2. To make this calibration curve into a linear model, what type of approach do you suggest?
Briefly explain and determine the equation(s). Determine the maximum non-linearity error
based on the percentage of the full-scale deflection.
To make this equation in to a linear model we can use piecewise linear functions defined on a
collection of intervals. This method involves dividing the range of the independent variable (x) into
several intervals and fitting a straight line to each interval.
we can divide the range of x(temperature) into five intervals: [0, 20], [20, 40], [40, 60],[60,80] and
[80,100]. Then, we can fit a straight line to each interval using the method of least squares
For the first interval, we have the following data points: (0, -0.3575), (5,-0.0605), (10,0.2215), and
(15,0.4885).
For the fifth interval, we have the following data points(80,2.5945), (85,2.6515), (90,2.6935), and
(95,2.7205).
Therefore, the piecewise linear approximation for the calibration curve is:
In the above part we have obtained a linear equation for the range 40<=x<60.
y = 0.0309x+0.3925
In case of a linear model sensitivity is equal to the slope of the linear equation.
Hence the sensitivity of the sensor in range of 40°C to 60°C = 0.0309V/°C
4. What are the other types of circuits that you can construct to measure temperature using
thermistors?
Self-Heating Circuit:
This circuit measures temperature using the thermistor's ability to self-heat. The thermistor is wired
in series with a constant current source, a fixed resistor, and other components. The temperature is
determined by keeping track of the thermistor's change in resistance as the circuit's current heats it
up.
5. If you want to use a thermocouple instead of a thermistor, what type of signal conditioning
circuits you require?
A signal conditioning circuit is needed to amplify and transform the small thermocouple voltage
into a useable signal that can be read by a data acquisition system or a temperature controller in
order to measure temperature using a thermocouple.
Thermocouple Amplifiers:
The small voltage produced by the thermocouple is amplified by these amplifiers to a greater
voltage that can be monitored with ease. Different varieties of thermocouple amplifiers, including
differential, non-differential, and instrumentation amplifiers, are available.
The reference junction, also known as the cold junction, of the thermocouple must be kept at a
constant temperature to accurately measure the temperature. The temperature of the cold junction
is measured, and the temperature differential between the cold junction and the measurement point
is made up for, using a cold junction compensation circuit.
Linearization Circuits:
A temperature-dependent nonlinear voltage signal is generated by thermocouples. The nonlinear
signal is linearized into a temperature-proportional output using a linearization circuit.
Bishop, C. M. (2006). Pattern Recognition and Machine Learning (1st ed.). Springer.
https://doi.org/10.1007/978-0-387-84858-7
Draper, N. R., & Smith, H. (1998). Applied Regression Analysis (3rd ed.). Wiley.
https://doi.org/10.1002/9781118625590
https://www.ti.com/lit/ml/slyp161/slyp161.pdf