Signal Conditioning PDF

Signal Conditioning Prepares signals for accurate data acquisition in ABE. Ensures reliable measurements of biological and environmental data. Includes amplification, filtering, and noise reduction. Electrical output sensing devices advantages over mechanical devices 1\. Ease of transmi...

Signal Conditioning Prepares signals for accurate data acquisition in ABE. Ensures reliable measurements of biological and environmental data. Includes amplification, filtering, and noise reduction. Electrical output sensing devices advantages over mechanical devices 1\. Ease of transmitting the signal from measurement point to the data collection point 2\. Ease of amplifying, filtering, or otherwise modifying the signal 3\. Ease of recording the signal Why perform Signal Conditioning? 1\. Enhances measurement precision. 2\. Filters out unwanted interference. 3\. Ensures data reliability and consistency. Signal Amplification Signals: millivolt range are common, few in microvolts. Difficulty: Losses over wires of great length and many systems require 1 to 10 V input. Solution: Amplifier Purpose: Boosts weak signals to usable levels. Types: Includes voltage, current, and power amplification. Recall: Signal Amplification Gain is the factor by which a device, like an amplifier, increases the strength of an input signal. Common instrumentation amplifiers gain: 1 to 1000 It is often expressed as a ratio or in decibels (dB) and defined as: Example: G =10 would have a decibel gain (GdB) of 20 dB while G =1000 would have a Gdb of 60 dB. If a signal is attenuated the decibel gain will have a negative value. Recall: Signal Amplification (cont.) Amplifiers are designed to operate effectively within specific frequency bands (e.g., audio, RF). The range of frequencies an amplifier can handle without significant loss of gain or distortion. Application-Specific: Different amplifiers target different ranges, such as audio amplifiers (20 Hz - 20 kHz) or RF amplifiers (MHz to GHz). Always use the op-amp at its operating bandwidth Phase angle is the time shift between input and output signals in an amplifier, often caused by circuit design and signal frequency. If the voltage input signal to the amplifier is in the form of a sine wave and expressed as Vi (t)= Vmi sin (2πft) where, f is the frequency and Vmi is the amplitude of the input sine wave. The output signal will be Vo \(t) = GVmi( 2πft +φ ) where, φ is called the phase angle. Phase angle is the time shift between input and output signals in an amplifier, often caused by circuit design and signal frequency. If the voltage input signal to the amplifier is in the form of a sine wave and expressed as Vi (t)= Vmi sin (2πft) where, f is the frequency and Vmi is the amplitude of the input sine wave. The output signal will be Vo \(t) = GVmi( 2πft +φ ) where, φ is called the phase angle. Wheatstone Bridge (cont.) R1 R3 R2 R4 DC Vdc \+ V0 \- a b V V V 0 = −a b 2 1 2 a dc R V V R R   =     + 4 3 4 b dc R V V R R   =     + 2 4 0 1 2 3 4 dc R R V V R R R R   = −     + + //Chapter 2.3 Signal Conditioning 12 A Wheatstone bridge itself does not amplify a signal; rather, it converts small resistance changes into a measurable voltage difference (Vo ). Wheatstone Bridge (cont.) R1 R3 R2 R4 DC Vdc \+ V0 \- a b 2 4 0 1 2 3 4 dc R R V V R R R R   = −     + + 0 2 4 dc 1 2 3 4 V R R V R R R R = − \+ + ( ) 0 2 3 4 4 dc 1 2 V R R R R V R R   \+ − = −     + Solving for R3 3 4 2 0 1 2 1 1 dc R R R V R R V       = −         −     +   //Chapter 2.3 Signal Conditioning 13 Strain Gauge L L L L   = Definition of strain,  Force Force The most common application of Wheatstone bridge is the conversion of the change in resistance of a strain gauge into voltage. In the figure below, recall that: //Chapter 2.3 Signal Conditioning 14 Strain Gauge (cont.) A l = length Resistance R l A =  Recall: Wire total resistance (R) is a function of length (l), cross-sectional area (A) and material resistivity () Under axial strain, l increases and A decreases so the resistance R increases //Chapter 2.3 Signal Conditioning 15 Strain Gauge (cont.) 0 0 F R R R R G L L    = =  Gauge factor, GF R1 R3 R2 R4 DC Vdc \+ V0 \- a b R R R R R R 3 0 0 = +  = +  (1 0 ) R R G 3 0 = + (1 F  ) //Chapter 2.3 Signal Conditioning 16 Strain Gauge (cont.) R1 R3 R2 R4 DC Vdc \+ V0 \- a b 2 4 0 1 2 3 4 dc R R V V R R R R   = −     + + ( ) 2 4 0 1 2 0 4 1 dc F R R V V R R R G R    = −     + + +   R R G 3 0 = + (1 F  ) Solving for  4 0 2 0 1 2 1 1 1 F F dc R G R G R V R R V        = − −         −     +   //Chapter 2.3 Signal Conditioning 17 Arduino-Based Load Cell Using an HX711 amplifier module, Arduino Uno, 16x02 LCD and a Load Cell, one can easily make a weighing machine. //Chapter 2.3 Signal Conditioning 18 I2V I2V (Current-to-Voltage Conversion) converts a current signal into a proportional voltage signal, typically using an operational amplifier (op- amp) Applications: Ideal for sensors like photodiodes and biosensors in ABE. Benefit: Increases sensitivity and simplifies data processing. //Chapter 2.3 Signal Conditioning 19 https://www.hackatronic.com/wp- content/uploads/2020/10/I-to-V- converter-.jpg VCIS (Transresistance Amplifier) VCIS means Voltage Controlled Current Source and is an amplifier used for I2V Most signal processors, such as amplifiers, have headers that only utilize VOLTAGE SIGNAL //Chapter 2.3 Signal Conditioning 20 VCIS (cont.) Transresistance Amplifiers are used for low-power applications to produce an output voltage proportional to the input current. VCIS are commonly used for photodiodes and phototransistors (used in the production of solar power)which are commonly modeled as current sources //Chapter 2.3 Signal Conditioning 21 VCIS Calculations //Chapter 2.3 Signal Conditioning 22 Thermocouple Thermocouples are a popular type of temperature measurement device. A thermocouple is a cable of two wires made from two dissimilar conductors (usually alloys) that are soldered or welded together at one end All thermocouple operate thru the thermoelectric or Seebeck effect //Chapter 2.3 Signal Conditioning 23 Seebeck Effect Seebeeck Effect is when a conductor generates a voltage when it is subjected to a temperature gradient The voltage difference of the two dissimilar metals can be measured and related to the corresponding temperature gradient Nickel-Chromium Alloy Copper-Nickel Alloy The relationship between temperature difference and voltage varies with materials \+ \- VS = SΔT //Chapter 2.3 Signal Conditioning 24 Measuring Temperature To measure temperature using a thermocouple, one can't just connect the thermocouple to a voltmeter since the voltmeter will only measure voltage due to temperature difference between the primary junction (hot) and the junction where the voltage is being measured (cold junction) SOURCE: http://www.pcbheaven.com/wikipages/images/thermocouples\_1271330366.png To determine the absolute temperature at the hot junction\... You need to know the temperature at the cold junction! How can we determine the temperature at the reference junction? //Chapter 2.3 Signal Conditioning 25 Cold-Junction The cold junctions are formed since they connect to some form of terminal block that transitions from the thermocouple alloys into the traces used on the printed circuit board (PCB), which is usually copper. Thermocouples do not measure an absolute temperature and only measure the temperature difference between hot and cold junctions. Vmeasured = VHot -- Vcold //Chapter 2.3 Signal Conditioning 26 Cold-Junction Compensation Thermocouples measure the voltage difference between two points To know the absolute temperature at the hot junction, one must know the temperature at the Cold junction VHot = Vmeasured + Vcold //Chapter 2.3 Signal Conditioning 27 Cold-Junction Compensation (cont.) In many applications, the temperature at the cold junctions are measured using a diode, thermistor, or RTD. As with any form of cold-junction compensation, two conditions must be met to achieve accurate thermocouple measurements: 1\. Junctions B and C must remain isothermal or be held at the same temperature. 2\. The isothermal temperature of junctions B and C must be accurately measured SOURCE: http://www.industrial- electronics.com/DAQ/images/10\_13.jpg //Chapter 2.3 Signal Conditioning 28 Cold-Junction Compensation (cont.) The thermocouple temperature measurements involve two important factors: 1\. measurement of the thermal gradient from the cold junction available from the thermocouple based on the Seebeck effect 2\. The actual cold junction measurement. The thermocouple output coefficient (S) is a few μV/oC and is highly non-linear VS = SΔT //Chapter 2.3 Signal Conditioning 29 Cold-Junction Compensation (cont.) Thermocouple measurements depend on the temperature difference detected by the thermocouple (Seebeck effect) and (2) accurately measuring the cold junction temperature. Thermocouples produce a small output of only a few microvolts per degree Celsius (μV/°C). For accurate readings, the system must detect very small voltage changes in microvolts. A low-noise amplifier can boost the signal, improving measurement accuracy. //Chapter 2.3 Signal Conditioning 30 Thermocouple Temperature Conversion Equation The thermocouple temperature conversion equation relates the voltage output of a thermocouple to temperature using a polynomial or piecewise function, depending on the thermocouple type: T = a0 + a1V + a2V2 + \.... + anVn //Chapter 2.3 Signal Conditioning 31 Voltage difference of the hot and cold junctions: VD = 3.409 mV What is the temperature of the hot junction if the cold junction is at 22 oC? At 22 oC, the reference junction voltage is 0.870 mV The hot junction voltage is therefore 3.409 mV + 0.870 mV = 4.279 mV The temperature at the hot junction is therefore 100 oC Using Look-up Table for Type T Thermocouple //Chapter 2.3 Signal Conditioning 32 Acquiring Data Using Thermocouple //Chapter 2.3 Signal Conditioning 33 Arduino-Based Temperature Measurement Using Thermocouple Arduino and ThermoCouple K MAX6675 //Chapter 2.3 Signal Conditioning 34 \> Comparison Among Temperature Sensors //Chapter 2.3 Signal Conditioning 35 Moving Average Smooths out signal fluctuations by averaging recent data points. Calculates the average over a sliding window of data points. Noise Reduction: Reduces random noise, making the signal cleaner. Window Size Effect: Larger window = more smoothing smaller window = more responsiveness. //Chapter 2.3 Signal Conditioning 36 Moving Average The formula for the moving average of a signal x over a window of N data points is: MAt is the moving average at time t, N is the window size (number of data points to average), xi is the value of the signal at time i. //Chapter 2.3 Signal Conditioning 37 MAt = 1 N ෍ i=t−N+1 t xi Arduino MA function float calculateMovingAverage(float newReading) { // Update total by subtracting the oldest reading and adding the new one total -= readings\[index\]; readings\[index\] = newReading; total += readings\[index\]; // Update the index and wrap around if necessary index = (index + 1) % windowSize; // Calculate and return the moving average return total / windowSize;} //Chapter 2.3 Signal Conditioning 38 Single-Channel Kalman Filter Provides a powerful way to filter out noise and estimate the true state of a single variable (e.g., temperature, position) over time. Combines predictions from a mathematical model with actual noisy measurements, adjusting the estimate based on the uncertainty of each. Delivers smooth and accurate signal estimation, even with noisy input, by continuously updating based on new data. Ideal for real-time tracking and measurement in ABE, like smoothing soil moisture readings or tracking the position of agricultural equipment. //Chapter 2.3 Signal Conditioning 39 Single-Channel Kalman Filter https://user-images.githubusercontent.com/7277887/49056311-4699b880-f1c9-11e8- 9150-a6d52251f731.png //Chapter 2.3 Signal Conditioning 40 Arduino SKF function float kalmanFilter(float measurement) { // Prediction step (in this simplified case, no control input is considered) // Estimate remains the same, so we skip state transition here // Update error covariance for prediction step errorCovariance = errorCovariance + processNoise; // Calculate Kalman Gain kalmanGain = errorCovariance / (errorCovariance + measurementNoise); // Update estimate with the measurement estimate = estimate + kalmanGain \* (measurement - estimate); // Update error covariance errorCovariance = (1 - kalmanGain) \* errorCovariance; return estimate; // Return the filtered (estimated) value} //Chapter 2.3 Signal Conditioning 41 MA vs SKF //Chapter 2.3 Signal Conditioning 42 Window size: 3, fast response but does not effectively filter noise MA vs SKF //Chapter 2.3 Signal Conditioning 43 Window size: 10, smoother but lags