First Year Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Sensors and transducers

Definitions, principles, characteristics, types, formulas, and examples for sensors and transducers used to convert physical quantities into electrical signals, including static/dynamic behavior and signal conditioning.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Engineers use sensors and transducers to measure temperature, force, displacement, speed, light, pressure, flow, and many other physical quantities and convert them into electrical signals for monitoring, control, data logging, and automation. In labs, factories, automobiles, healthcare, and IoT systems, correct selection, calibration, and interfacing of sensors decide the accuracy and reliability of the whole system.

Key ideas

  • The measurand is the physical quantity being measured (e.g., temperature in kelvin).
  • The output signal is the electrical quantity produced (voltage, current, resistance, capacitance, frequency, or digital code).
  • Primary sensing element directly interacts with the measurand. Secondary transduction element converts the primary effect to an electrical form (e.g., mechanical displacement to voltage via an LVDT).
  • Signal conditioning includes amplification, filtering, linearization, excitation (supply for passive sensors), isolation, and conversion (ADC/DAC), to make the output suitable for display, control, or digital processing.

Classification (common in exams):

  • By energy source: Active sensors generate their own electrical signal from the measurand (e.g., thermocouple, piezoelectric, photovoltaic cell). Passive sensors require external excitation (e.g., RTD, LVDT secondary requires excitation of primary, strain gauge bridge).
  • By output: Analog sensors provide continuous-valued outputs (e.g., RTD resistance). Digital sensors provide discrete pulses or codes (e.g., optical encoder, I2C temperature sensor module).
  • By contact: Contact sensors physically touch the object (thermistor probe). Non-contact sensors sense remotely (infrared thermometer, Hall sensor for magnetic field).

Core devices and working principles (what to draw and label in diagrams):

  1. Wheatstone bridge (for resistive sensors like strain gauges, RTDs): Draw a diamond with four resistors R1–R4, top node to DC supply +Vs, bottom to ground, left node VA, right node VB to differential output Vout = VA − VB. Show sensor as one or more arms. Label Vs, Vout, and nominal resistor values. Principle: small resistance change unbalances the bridge, producing a differential voltage proportional to the change.
  2. LVDT (Linear Variable Differential Transformer): Draw a cylindrical core centered in a transformer with one primary coil (excited by AC) and two equal secondary coils in series opposition. Label primary, secondaries S1 and S2, movable ferromagnetic core, and output Vout = VS1 − VS2. Principle: displacement of core changes mutual inductance, producing differential AC output proportional to displacement and indicating direction by phase.
  3. Capacitive sensor: Draw two parallel plates area A separated by distance d with dielectric ε between; show that displacement or level changes d or A. Label C = εA/d. Principle: measurand changes geometry/dielectric, changing capacitance.
  4. Strain gauge: Draw zig-zag metallic foil bonded to a test specimen with arrows for principal strain direction. Label nominal resistance R, gauge factor GF. Principle: resistance changes with strain: ΔR/R = GF·ε.
  5. Piezoelectric sensor: Draw a crystal with electrodes, applied force F, and generated charge Q at terminals. Label d (piezoelectric coefficient). Principle: mechanical stress generates charge and hence voltage across a capacitor.
  6. Hall-effect sensor: Draw a thin rectangular semiconductor with current I along x, magnetic field B into the page, and Hall voltage VH across y. Label thickness t, Hall coefficient RH. Principle: Lorentz force deflects carriers to create transverse voltage VH ∝ I·B.

Static characteristics (state clearly):

  • Accuracy: Closeness of the measured value to the true value; often specified as ± percentage of full-scale or reading.
  • Precision: The degree of agreement among repeated measurements under identical conditions; relates to repeatability, not truth.
  • Sensitivity (S): Ratio of change in output to change in measurand, S = Δoutput/Δinput (units depend on sensor, e.g., mV/mm).
  • Resolution: Smallest change in input that produces a detectable change in output (e.g., 0.1 °C).
  • Linearity: Degree to which the input–output relation is a straight line; expressed as maximum deviation from a best-fit line as a percentage of full-scale.
  • Hysteresis: Difference in output for the same input when approached from increasing versus decreasing direction.
  • Repeatability: Closeness of agreement among consecutive measurements under the same conditions.
  • Range (span): Difference between maximum and minimum measurable input values.
  • Zero drift and sensitivity drift: Change in zero output or sensitivity with time/temperature.
  • Loading effect: Alteration of the measurand or sensor output due to the measuring system drawing energy (e.g., finite input resistance of a voltmeter or amplifier).

Dynamic characteristics (basic model and statements):

  • Many sensors can be modeled as a first-order system with time constant τ and static gain K. For a step input, output y(t) = K(1 − e^(−t/τ))·u(t). Assumptions: small-signal linear region, single dominant energy-storage element (thermal, mechanical, or electrical).
  • Time constant (τ): Time for the step response to reach 63.2% of final value. Rise time (10–90%) ≈ 2.2τ for first-order. Settling time (within ±2%) ≈ 4τ.
  • Some transducers (e.g., piezoelectric) behave as high-pass or differentiating sensors with negligible DC response, useful for dynamic force/vibration.

Common sensors and exact relations typically examined:

  • RTD (Resistance Temperature Detector, usually platinum): R(T) ≈ R0[1 + α(T − T0)] for limited range near T0. Standard α for Pt is approximately 0.00385 °C⁻¹ (verify depending on standard). For wider ranges, use Callendar–Van Dusen (often beyond first-year scope).
  • Thermistor (NTC): R(T) = R∞·e^(B/T), or in practical form R(T) = R1·exp[B(1/T − 1/T1)], where T in kelvin, B is material constant. Highly nonlinear; needs linearization or lookup.
  • Thermocouple: V ≈ S·ΔT (open-circuit Seebeck voltage), where S is the Seebeck coefficient (e.g., ≈ 41 µV/°C for Type K near room temperature). Requires a reference (cold) junction and compensation.
  • Strain gauge: ΔR/R = GF·ε, where ε is strain (dimensionless), GF ≈ 2 for metal foil gauges (verify actual). Requires bridge excitation and temperature compensation.
  • LVDT: Vout (rms) ∝ displacement x near null; direction indicated by phase relative to primary excitation. Sensitivity given by datasheet (e.g., mV/V/mm).
  • Capacitive sensor: C = εA/d (parallel plate); ΔC relates to change in d, A, or ε. Used for small displacement, level, or humidity (changing dielectric constant).
  • Piezoelectric transducer: Q = d·F (scalar form for a given mode), V = Q/Cp, where Cp is sensor capacitance. No DC response; best for dynamic force/acceleration (with charge or IEPE amplifier).
  • Hall-effect sensor: VH = (RH·I·B)/t for a rectangular slab of thickness t; RH depends on carrier type and concentration.
  • Optical sensors: Photodiode (photovoltaic or photoconductive), LDR (photoresistor), optical encoder (digital pulses per revolution for angle/speed). Use biasing and transimpedance amplifier for photodiodes.
  • Tachogenerator: Voltage proportional to rotational speed (active sensor); often used for speed feedback.

Signal conditioning essentials (what to show in block diagrams):

  • For resistive sensors: Wheatstone bridge → instrumentation amplifier (high input impedance, high common-mode rejection, gain set by resistor) → low-pass filter → ADC.
  • For LVDT: AC excitation oscillator → primary coil → differential secondary → demodulator/rectifier → low-pass filter → DC proportional to displacement.
  • For capacitive sensors: AC excitation and bridge or oscillator-based measurement (capacitance-to-frequency) → counter or demodulator.
  • For piezoelectric: charge amplifier (op-amp with feedback capacitor) converts charge to voltage; ensures stable low-frequency response.
  • For photodiode: transimpedance amplifier converts photocurrent to voltage; add anti-alias filter before ADC.
  • Ensure proper impedance bridging (high input impedance stage after the sensor) to minimize loading. Use isolation where needed for safety.

Calibration and error analysis:

  • Calibration: Compare sensor output with a standard over the range; record input–output pairs; obtain best-fit line or curve; determine zero and span adjustments.
  • Types of errors: systematic (bias), random (noise), environmental (temperature, humidity), drift, quantization (in ADC), hysteresis, and loading. Combine independent uncertainties by root-sum-square (RSS) if uncorrelated.

Safety and standards:

  • Observe maximum ratings: excitation voltage/current, temperature limits, and insulation. Use standardized color codes/terminals where applicable and follow grounding and shielding practices to reduce EMI.

Formulas

Sensitivity S = Δy/Δx — y: output signal (e.g., V), x: measurand (e.g., mm). Applies near operating point. Bridge output (near balance) Vout ≈ (Vs/4)·(ΔR/R) — Vs: supply voltage (V), R: arm resistance (Ω), ΔR: small change in one arm (Ω). For a single active arm in a Wheatstone bridge, small-signal approximation. Strain gauge ΔR/R = GF·ε — GF: gauge factor (dimensionless), ε: strain (m/m). Linear range before yield. RTD (linearized) R(T) = R0·[1 + α·(T − T0)] — R0 at T0 (Ω), α (°C⁻¹), T and T0 in °C. Approximate linear region. Thermistor R(T) = R1·exp[B·(1/T − 1/T1)] — R1 at T1 (Ω), B (K), T, T1 (K). NTC behavior. Capacitance C = ε·A/d — ε (F/m), A (m²), d (m), C (F). Parallel-plate model, fringe fields neglected. Piezoelectric Q = d·F — Q: charge (C), d: piezoelectric coefficient (C/N or m/V in alternative forms; use C/N here), F: force (N). For specified mode (e.g., d33). Piezoelectric V = Q/Cp — V: voltage (V), Cp: sensor capacitance (F). Open-circuit or with high input impedance. Hall voltage VH = (RH·I·B)/t — RH: Hall coefficient (m³/C), I: current (A), B: magnetic flux density (T), t: thickness (m). Uniform field and current. LVDT Vout (rms) = S·Vs·x — S: sensitivity (per unit excitation per displacement, e.g., mV/V/mm), Vs: primary excitation (V rms), x: displacement (mm). Near-null linear region. First-order step response y(t) = K·(1 − e^(−t/τ)) — K: static gain (output per input), τ: time constant (s). Small-signal, single-pole model. Time constant τ = R·C (for thermal/electrical analogues) — R: resistance (Ω), C: capacitance (F); for thermal analogues use R_th (K/W) and C_th (J/K). Applies to simple RC-like sensors. Uncertainty (independent) u_total = √(u1² + u2² + …) — ui: standard uncertainties (same units). For uncorrelated errors.

Worked examples

Example 1: LVDT displacement from output Given: LVDT sensitivity S = 2.0 mV/V/mm (rms). Primary excitation Vs = 5.0 V rms. Measured differential output Vout = 30 mV rms, in-phase with primary. Steps:

  1. Formula: x = Vout/(S·Vs)
  2. Substitute: x = 30 mV / (2.0 mV/V/mm × 5.0 V) = 30 mV / (10 mV/mm)
  3. Compute: x = 3.0 mm. Phase in-phase indicates positive direction. Answer: Displacement = 3.0 mm in positive direction. Final: 3.0 mm.

Example 2: Strain gauge in quarter-bridge, small-signal output Given: Metal foil strain gauge with R = 120 Ω, GF = 2.0, applied strain ε = 500 µε = 500 × 10⁻⁶. Wheatstone bridge with one active arm (quarter-bridge) excited by Vs = 5.0 V. Find differential output Vout (approximate). Steps:

  1. Resistance change: ΔR/R = GF·ε = 2.0 × 500 × 10⁻⁶ = 0.001.
  2. Small-signal bridge output (single active arm): Vout ≈ (Vs/4)·(ΔR/R).
  3. Substitute: Vout ≈ (5.0 V/4) × 0.001 = 1.25 V × 0.001 = 1.25 mV. Answer: Bridge output ≈ 1.25 mV. Final: 1.25 mV.

Example 3: RTD temperature using linear approximation Given: Platinum RTD (Pt100) with R0 = 100.00 Ω at T0 = 0 °C. Assume α = 0.00385 °C⁻¹. Measured resistance R = 119.25 Ω. Find temperature T (°C). Steps:

  1. Formula: R(T) = R0[1 + α(T − T0)]. Rearranged: T = T0 + (R/R0 − 1)/α.
  2. Substitute: T = 0 °C + (119.25/100.00 − 1)/0.00385 = (1.1925 − 1)/0.00385.
  3. Compute numerator: 0.1925. Divide: 0.1925/0.00385 ≈ 50.0 °C (since 0.00385 × 50 = 0.1925). Answer: Temperature ≈ 50.0 °C. Final: 50.0 °C.

Common mistakes

  • Ignoring excitation requirements (treating passive sensors as self-powered) or forgetting polarity/phase information (LVDT, Hall sensor).
  • Using RTD linear approximation far outside the limited range; not stating the assumption of linearity and reference temperature.
  • Forgetting units (e.g., using °C in the thermistor equation that requires kelvin).
  • Neglecting bridge completion resistors’ tolerances and temperature effects; not mentioning temperature compensation for strain gauges (quarter, half, full bridge).
  • Loading the sensor with low input impedance amplifiers and then blaming “sensor error.” Always consider loading and use instrumentation amplifiers.
  • Treating piezoelectric sensors as DC force sensors; they respond to dynamic changes and need charge or IEPE conditioning.
  • Skipping calibration and uncertainty discussion in long answers; universities expect a note on error sources and calibration method.

For your exam

  • Short notes/diagrams: Wheatstone bridge for strain gauge (label all nodes, write Vout expression), LVDT construction and working (show primary, secondaries, core, phase), capacitive displacement sensor (parallel plate formula), Hall effect sensor (derive VH expression with symbols defined), piezoelectric transducer (Q = dF, V = Q/Cp and note dynamic nature).
  • Derivations: Small-signal Wheatstone bridge output Vout ≈ (Vs/4)(ΔR/R). First-order step response y(t) and definition of τ. Hall voltage formula from Lorentz force balance (outline only if asked; usually the final formula suffices with definitions).
  • Numericals: Compute displacement from LVDT output given sensitivity and excitation; strain from measured bridge output; temperature from RTD or thermistor resistance (convert to kelvin for thermistor). Show formula first, substitute with units, and box the final answer.
  • Interfacing: Block diagram of signal conditioning chain for a chosen sensor (bridge → INA → filter → ADC; or LVDT → demodulator → LPF). Explain why each block is used.
  • Calibration and errors: Describe method with standard references, plotting, best-fit line; list systematic, random, hysteresis, drift, loading, and quantization errors; show RSS combination for independent uncertainties.

Quick check

  1. State the difference between an active and a passive sensor, with one example each.
  2. Write the small-signal output of a quarter-bridge when one arm changes by ΔR from R, with supply Vs.
  3. A Pt100 (α = 0.00385 °C⁻¹) reads 107.70 Ω. What is the temperature (assume T0 = 0 °C)? Answers: 1) Active generates its own signal (e.g., thermocouple, piezoelectric); passive needs excitation (e.g., RTD, strain gauge). 2) Vout ≈ (Vs/4)(ΔR/R). 3) T = (107.70/100 − 1)/0.00385 ≈ 20.0 °C.

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?