Mechanical and Motion Sensors

Introduction and Context

Mechanical and motion sensors convert position, displacement, velocity, acceleration, force, pressure, strain, orientation, or contact into an electrical signal that a controller can interpret. They are fundamental to mechatronic systems because feedback is required to compare commanded behavior with actual behavior. A servo motor, for example, cannot regulate position accurately without a position sensor; a robotic gripper cannot detect contact reliably without tactile or force sensing; and a vehicle airbag controller cannot estimate a crash event without accelerometers.

A sensor is not merely a physical component. It is part of a measurement chain consisting of the measurand, sensing element, signal conditioning, analogue-to-digital conversion, computation, and decision or control action. Sensor selection therefore requires consideration of range, sensitivity, resolution, accuracy, bandwidth, linearity, hysteresis, environmental compatibility, reliability, cost, and ease of calibration.

This chapter examines common mechanical and motion sensors used in mechatronics: potentiometers, encoders, resolvers, LVDTs, Hall-effect and inductive proximity sensors, strain gauges, load cells, piezoelectric sensors, accelerometers, gyroscopes, tactile sensors, and machine-vision position systems.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Explain the operating principles of major mechanical and motion sensors.
2. Distinguish analogue, incremental, absolute, contact, and non-contact measurement methods.
3. Select sensors according to range, resolution, accuracy, bandwidth, environment, and interface requirements.
4. Apply basic equations for displacement, strain, force, acceleration, and angular motion measurement.
5. Identify error sources, conditioning requirements, and calibration procedures.
6. Integrate sensors into practical mechatronic systems and evaluate their limitations.

1. Measurement Concepts and Performance Terms

The measurand is the physical quantity being measured, such as shaft angle or linear displacement. The sensor output may be voltage, current, resistance, frequency, pulse count, phase, or a digital data word.

Important performance terms include:

• Range: the interval between minimum and maximum measurable values.
• Sensitivity: change in output per unit change in input.
• Resolution: smallest distinguishable input change.
• Accuracy: closeness of a measurement to the accepted value.
• Repeatability: closeness of repeated measurements under identical conditions.
• Linearity: closeness of the input-output relation to a specified straight line.
• Hysteresis: different output values for the same input depending on whether the input is increasing or decreasing.
• Bandwidth: frequency range over which the sensor responds adequately.
• Response time: time required to reach a specified fraction of the final output.
• Drift: gradual change in output caused by time, temperature, ageing, or mechanical change.

Accuracy and resolution are not equivalent. A sensor may have fine resolution but poor accuracy because of calibration error. Conversely, a coarse sensor may be accurate within its limited resolution.

For a linear sensor, sensitivity is commonly expressed as

where S is sensitivity, ∆y is output change, and ∆x is measurand change. Units depend on the sensor, such as V/mm or mV/N. This expression is a local approximation when the transfer characteristic is nonlinear.

2. Potentiometers: Resistive Position Measurement

A potentiometer measures position through a sliding or rotating contact, called a wiper, moving along a resistive track. If a constant voltage V_s is applied across a uniform track and the wiper position is x over total travel L, the ideal output is

This assumes a uniform track, negligible wiper resistance, negligible loading, and a wiper that moves linearly with the measured displacement. For a rotary potentiometer, x÷L is replaced by the fraction of angular travel.

Potentiometers are inexpensive and provide an absolute output: the voltage indicates position immediately after power-up. Their disadvantages include mechanical wear, electrical noise from the wiper, limited speed, friction, and sensitivity to contamination. Conductive-plastic and wire-wound versions are selected according to life, resolution, power, and linearity requirements.

The connected electronics must have sufficiently high input resistance. If the load resistance is comparable with the track resistance, loading causes a nonlinear output. A buffer amplifier or high-impedance ADC input reduces this error.

Application: A low-cost motorized valve may use a rotary potentiometer to report shaft position. It is appropriate where moderate accuracy and limited rotational speed are acceptable, but not where continuous high-speed rotation or long service life is required.

3. Encoders

An encoder converts rotary or linear motion into coded electrical information.

3.1 Incremental encoders

An incremental rotary encoder commonly provides two square-wave channels, A and B, separated by approximately 90° electrical phase. The phase relationship identifies direction. A third index channel, Z, supplies one reference pulse per revolution.

If an encoder has N counts per revolution and the controller counts C state changes per revolution, the ideal angular resolution is

For quadrature decoding, C may be one, two, or four times the number of cycles per revolution, depending on the decoding method. The count must not be confused with the manufacturer’s stated pulses per revolution.

For a screw-driven linear axis with lead p metres per revolution, the ideal displacement per count is

These equations assume no mechanical backlash, slip, missed counts, or interpolation error.

Incremental encoders lose their absolute position when power is removed unless a homing or index procedure is used. Their strengths are high speed, good resolution, and simple digital interfaces. Potential errors include missed pulses, electrical interference, incorrect edge interpretation, and mechanical coupling errors.

3.2 Absolute encoders

An absolute encoder outputs a unique code for each shaft position. Optical discs may use Gray code, in which adjacent positions differ by only one bit. This reduces the risk of a transient multi-bit error during transitions. Magnetic and capacitive absolute encoders are also common.

Absolute encoders retain position information through power interruption, although some systems require a backup power source or multi-turn mechanism. They are preferred in robotics, machine tools, and safety-related axes where homing is undesirable or unsafe.

4. Resolvers

A resolver is an electromechanical transformer that measures angular position through sinusoidal and cosinusoidal signals. An AC excitation is applied to the rotor or stator winding, and two output windings produce signals approximately proportional to

where V_e is excitation amplitude, k is a scale factor, and θ is shaft angle. An electronic resolver-to-digital converter estimates angle using the ratio of the two signals, commonly represented conceptually as

The two-argument arctangent is required because it identifies the correct quadrant. Resolver operation assumes sinusoidal excitation, adequate winding balance, and proper demodulation.

Resolvers are rugged, tolerant of high temperature, vibration, and contamination, and suitable for aircraft, electric vehicles, and industrial servo drives. They require excitation and specialized signal conditioning, making them more complex than many digital encoders.

5. LVDTs

A linear variable differential transformer (LVDT) measures linear displacement without mechanical electrical contact. It consists of a movable ferromagnetic core and one primary winding surrounded by two secondary windings connected in opposition. AC excitation of the primary induces voltages in the secondaries.

At the electrical null position, the secondary voltages are equal and their differential output is approximately zero. Movement toward one secondary increases its voltage and changes the phase; movement toward the other produces the opposite phase. After demodulation, output magnitude indicates displacement and phase indicates direction.

An idealized differential output can be written as

where K is sensitivity and x is displacement within the specified linear region. This is an approximation, not a universal law over unlimited travel.

LVDTs offer excellent repeatability, low friction, high resolution, and long life. They require AC excitation, demodulation, shielding, and attention to magnetic materials near the sensor. They are widely used in hydraulic actuators, test machines, valve position systems, and precision displacement measurement.

6. Proximity Sensors

Proximity sensors detect an object without physical contact. They usually provide a switching output rather than a continuous position value.

6.1 Hall-effect sensors

A Hall sensor detects magnetic flux density. When current flows through a semiconductor and a magnetic field is applied perpendicular to that current, a transverse Hall voltage appears. In simplified form,

where V_H is Hall voltage, R_H is the Hall coefficient, I is bias current, B is magnetic flux density, and t is active-layer thickness. The expression assumes a simplified uniform-field model.

Digital Hall switches detect a threshold and are used for gear-tooth sensing, brushless DC motor commutation, limit detection, and speed measurement. A rotating target with n magnetic events per revolution and measured event frequency f has ideal rotational speed

where ω is angular speed in rad/s. Hall sensors are non-contact and durable, but require a suitable magnet or ferromagnetic target and may be affected by magnetic interference and air gap.

6.2 Inductive proximity sensors

An inductive proximity sensor creates an alternating electromagnetic field using a coil. A conductive target causes eddy currents, changing coil impedance and reducing oscillation amplitude. Detection electronics convert this change into a switching signal.

Inductive sensors detect metals reliably and are resistant to oil, dust, and vibration. They do not generally detect plastic, wood, or glass unless a metal target is added. Sensing distance depends on target material, size, orientation, coil geometry, and mounting arrangement. Flush and non-flush installation have different specified distances and clearance requirements.

7. Strain Gauges and Load Cells

A strain gauge measures deformation through a change in electrical resistance. Normal strain is

where ε is dimensionless strain, ∆L is change in length, and L is original length. Gauge resistance changes approximately according to

where R is nominal resistance and GF is gauge factor. For metallic gauges, gauge factor is commonly near 2, but the actual value is specified by the manufacturer.

Because resistance changes are small, strain gauges are usually connected in a Wheatstone bridge. For a quarter-bridge with one active gauge and small strain, the bridge output is approximately

where V_ex is bridge excitation. This approximation assumes a balanced bridge, small resistance changes, and ideal resistors. Half- and full-bridge arrangements increase sensitivity and can compensate for temperature or unwanted bending.

A load cell is a calibrated mechanical structure containing strain gauges. Its elastic element converts force into strain. For a uniaxial loading arrangement within the elastic range,

where σ is stress, F is force, A is loaded cross-sectional area, and E is Young’s modulus. Real load cells require calibration because geometry, shear, bending, temperature, creep, and installation influence output.

Worked example: A full-bridge load cell has rated output 2 mV/V and excitation 10 V. At rated load, output is

At half rated load, assuming linearity and zero offset, output is approximately 10 mV. The result is small, so low-noise differential amplification and shielding are important.

8. Piezoelectric Sensors

Piezoelectric materials generate electric charge when mechanically stressed. The constitutive behavior is coupled: mechanical stress can produce charge, and electric field can produce strain. For a simplified direct-effect model,

where Q is charge, d is the piezoelectric charge coefficient, and F is applied force. The coefficient and exact tensor form depend on material orientation and loading direction.

With sensor capacitance C, the open-circuit voltage is approximately

assuming negligible leakage and a quasi-static equivalent circuit. In practice, piezoelectric sensors are excellent for dynamic force, vibration, impact, and acoustic measurements, but poor for truly static force because charge leaks through finite insulation and amplifier input resistance. Charge amplifiers improve stability by making the measured voltage primarily dependent on charge and feedback capacitance.

9. Accelerometers and Gyroscopes

9.1 Accelerometers

A basic inertial accelerometer contains a proof mass, suspension, and damping. When the sensor housing accelerates, the proof mass deflects relative to the housing. In a simplified mass-spring model,

where m is proof mass, c is damping coefficient, k is spring stiffness, x is relative displacement, and a is case acceleration. The negative sign depends on the chosen coordinate convention.

At frequencies well below the natural frequency and with suitable damping, displacement is approximately proportional to acceleration:

MEMS accelerometers use microscopic beams and capacitive, piezoresistive, or piezoelectric readout. They measure vibration, tilt, shock, machine condition, and vehicle motion. A static accelerometer also responds to gravity, so orientation can be estimated from the gravity vector when dynamic acceleration is small.

9.2 Gyroscopes

A gyroscope measures angular velocity. MEMS vibratory gyroscopes use the Coriolis effect. A vibrating mass moving with velocity v experiences Coriolis force

where m is mass and Ω is angular velocity; the multiplication is a vector cross product. The force is perpendicular to both vibration velocity and rotation axis. Practical sensors use capacitive detection and closed-loop control.

Integrating angular velocity estimates angle:

where ω is measured angular velocity. Bias produces accumulating drift, so gyroscopes are often combined with accelerometers, magnetometers, encoders, or external references. Sensor fusion methods such as complementary filters and Kalman filters can exploit the complementary strengths of these sensors.

10. Tactile Sensors

Tactile sensors detect contact, pressure distribution, slip, or texture. Technologies include force-sensitive resistors, piezoresistive arrays, capacitive taxels, optical skins, piezoelectric films, and magnetic microstructures.

A tactile array contains many sensing elements, or taxels. A controller scans rows and columns, converts local signals into pressure estimates, and identifies contact location, shape, and sometimes shear direction. Force-sensitive resistors are inexpensive but often nonlinear, hysteretic, and strongly dependent on loading history. Capacitive sensors can provide better repeatability but require careful shielding and electronics.

Tactile sensing is valuable in robotic grasping, prosthetics, human-machine interfaces, and collision detection. Design must address spatial resolution, force range, sampling rate, mechanical protection, calibration, and cross-talk between adjacent taxels.

11. Machine-Vision-Based Position Measurement

A camera measures image coordinates rather than physical position directly. A pinhole camera model relates a three-dimensional point to image coordinates:

where (X,Y,Z) is the point in camera coordinates, f is focal length in pixel units, (u,v) is image position, and (c_x,c_y) is the principal point. This model assumes ideal perspective projection; lens distortion requires calibration correction.

Position measurement may use thresholding, edges, feature matching, fiducial markers, stereo vision, structured light, or learned object detectors. A calibrated planar system can convert image coordinates to physical coordinates through a homography. Three-dimensional measurement requires stereo, depth cameras, structured light, or known geometric constraints.

Vision offers non-contact, multi-object measurement and large spatial coverage. Its limitations include sensitivity to lighting, occlusion, reflections, lens distortion, computational latency, and calibration drift. Industrial systems should control illumination and use fixtures or fiducial references whenever possible.

12. Sensor Selection and Signal Conditioning

Selection should begin with the required measurand and control objective rather than with a preferred technology.

Signal conditioning may include amplification, filtering, bridge excitation, demodulation, linearization, isolation, debouncing, and analogue-to-digital conversion. Sampling frequency should be sufficiently higher than the highest signal frequency of interest. For a sampled system, frequencies above half the sampling frequency can alias into lower frequencies; this is the Nyquist limitation. An anti-alias low-pass filter is therefore required before conversion when significant high-frequency content exists.

Calibration establishes the relationship between sensor output and the accepted measurand. A linear calibration can be represented as

where a is scale factor and b is offset. Nonlinear sensors may require polynomial, piecewise, lookup-table, or model-based compensation. Calibration should include zero, span, repeatability, temperature, mounting, and uncertainty where the application demands traceability.

13. Illustrative Case: Position and Force Control of a Robotic Gripper

Background

Consider an illustrative industrial gripper that must approach a component, stop at a commanded position, close until contact, and apply a controlled gripping force.

Problem

A position-only system may crush fragile parts or fail to detect a misplaced component. The gripper must distinguish free motion from contact and must operate repeatedly in an environment containing dust and vibration.

Analysis

An incremental motor encoder provides high-resolution joint motion but does not directly measure contact force. A potentiometer would be simple but may wear under frequent cycling. A strain-gauge load cell in the gripper structure can measure static gripping force, while tactile elements can identify contact distribution. An inductive proximity sensor can confirm that a metal component is present before closure.

Approach and solution

The motor encoder is used for closed-loop position and velocity control. The load cell output is amplified with a differential instrumentation amplifier and sampled by a high-resolution ADC. A digital filter suppresses electrical noise without introducing unacceptable delay. The controller uses a state sequence: approach by position, detect contact from force rise, switch to force control, and stop if force exceeds a safety threshold. Tactile sensing is added when part alignment and slip detection are important.

Outcome

The combined architecture separates the functions of motion feedback, object detection, and force feedback. This improves repeatability and protects parts, although it increases wiring, calibration effort, software complexity, and fault-diagnosis requirements.

Learning points

1. No single sensor necessarily satisfies every control requirement.
2. A sensor should measure the variable that must actually be controlled.
3. Mechanical mounting and signal conditioning can be as important as the sensing principle.
4. Redundant or complementary sensing improves safety and diagnosis.

14. Limitations, Safety, and Contemporary Developments

Mechanical sensors are affected by temperature, vibration, contamination, electromagnetic interference, ageing, mounting stress, and mechanical backlash. A sensor specification should therefore be interpreted in the context of the complete installation. For example, encoder resolution does not guarantee axis accuracy if the transmission has backlash, compliance, or thermal expansion.

Safety-related applications require documented fault assumptions, diagnostic coverage, suitable architecture, and compliance with relevant machinery or functional-safety standards. Sensor failure modes include open circuits, short circuits, stuck outputs, drift, intermittent connections, and physically damaged mounting. Plausibility checks, redundant channels, range monitoring, and safe-state behavior are common safeguards.

Modern developments include miniature MEMS inertial sensors, integrated sensor fusion, wireless condition monitoring, event-based vision, flexible tactile skins, fibre-optic strain sensing, and intelligent sensors with embedded diagnostics. Artificial intelligence can assist vision-based detection and anomaly classification, but training data, lighting variation, explainability, cybersecurity, and validation remain engineering concerns. AI should not replace calibration, physical safety limits, or independent protective functions.

Chapter Summary

Mechanical and motion sensors provide the feedback required for measurement, automation, and control. Potentiometers offer simple absolute position measurement but involve contact and wear. Incremental encoders provide high-speed digital motion information, whereas absolute encoders retain position through power interruption. Resolvers are rugged angular sensors suited to harsh environments, and LVDTs provide precise contactless linear displacement measurement.

Hall and inductive proximity sensors detect magnetic or metallic targets without contact. Strain gauges and load cells convert deformation into small resistance changes and require bridge circuits and careful calibration. Piezoelectric sensors are particularly effective for dynamic force and vibration but are unsuitable for long-term static measurement without specialized conditioning. Accelerometers measure inertial acceleration, while gyroscopes measure angular rate and require bias management when integrated. Tactile arrays measure contact distribution, and machine vision provides non-contact, multi-object position measurement when calibration and illumination are controlled.

Effective sensor application depends on matching the sensing principle to the measurand, dynamics, environment, interface, safety requirements, and total system accuracy. Signal conditioning, sampling, calibration, mechanical design, and fault diagnosis are integral parts of sensor engineering.

Key Terms

Measurand: Physical quantity intended to be measured.

Sensitivity: Change in output per unit change in input.

Resolution: Smallest distinguishable input change.

Hysteresis: Dependence of output on the direction of input change.

Potentiometer: Resistive position sensor with a movable wiper.

Incremental encoder: Encoder that reports relative motion through pulses.

Absolute encoder: Encoder that provides a unique code for position.

Quadrature: Two phase-shifted encoder signals used to determine direction.

Resolver: Electromagnetic rotary position sensor producing sine and cosine signals.

LVDT: Transformer-based linear displacement sensor with differential output.

Hall effect: Generation of transverse voltage in a current-carrying material exposed to magnetic flux.

Inductive proximity sensor: Non-contact detector using electromagnetic eddy-current effects.

Strain gauge: Resistive element whose resistance changes with mechanical strain.

Load cell: Calibrated force transducer commonly based on strain gauges.

Piezoelectric sensor: Sensor producing electric charge in response to mechanical stress.

Accelerometer: Sensor that measures specific force or inertial acceleration.

Gyroscope: Sensor that measures angular velocity.

Taxel: Individual sensing element in a tactile array.

Calibration: Process of relating sensor output to an accepted reference.

Aliasing: Misinterpretation of high-frequency content caused by insufficient sampling.

Review Questions

Short Answer Questions

1. Distinguish accuracy from resolution.
2. Why does an incremental encoder usually require a homing procedure?
3. State one advantage and one limitation of an LVDT.
4. Why are strain gauges commonly used in Wheatstone bridges?
5. Why are piezoelectric sensors unsuitable for many static-force measurements?

Long Answer Questions

1. Explain the construction, operation, advantages, and limitations of incremental and absolute encoders.
2. Describe how a strain-gauge load cell converts force into an electrical signal, including the roles of strain, gauge factor, bridge excitation, and calibration.
3. Compare Hall-effect and inductive proximity sensors in terms of operating principle, target material, installation, and applications.
4. Explain the mass-spring model of an accelerometer and discuss the effects of bandwidth, damping, bias, and gravity.
5. Describe a complete machine-vision position-measurement system from image acquisition through calibration and coordinate conversion.

Analytical / Application Questions

1. A motor encoder has 2,000 cycles per revolution and is decoded at four edges per cycle. Determine the angular resolution in degrees per count and discuss two practical effects that could reduce actual axis accuracy.
2. Design a sensor set for a robotic gripper that must measure motor position, detect a metallic workpiece, and regulate gripping force. Justify each choice and identify likely signal-conditioning requirements.
3. A vibration-monitoring system must measure both low-frequency machine motion and short-duration impacts. Evaluate whether an accelerometer, piezoelectric sensor, or combination of both is most appropriate, explaining bandwidth, static response, and integration issues.

Multiple Choice Questions with Answers

Question 1

Which sensor provides an absolute position output immediately after power is applied?

A. Incremental encoder

B. Potentiometer

C. Piezoelectric accelerometer

D. Inductive proximity switch

Correct Answer: B. Potentiometer

Question 2

In a quadrature encoder, the phase relationship between channels A and B is primarily used to determine what quantity?

A. Temperature

B. Force

C. Direction of rotation

D. Magnetic permeability

Correct Answer: C. Direction of rotation

Question 3

An LVDT measures displacement by detecting changes in which quantity?

A. Differential transformer secondary voltage

B. Wiper resistance only

C. Photodiode current

D. Fluid pressure

Correct Answer: A. Differential transformer secondary voltage

Question 4

A metallic target is especially suitable for detection by which sensor?

A. Inductive proximity sensor

B. Piezoelectric film only

C. Potentiometer

D. Gyroscope

Correct Answer: A. Inductive proximity sensor

Question 5

A strain gauge has gauge factor 2 and strain 500 µε. What is the approximate fractional resistance change?

A. 0.00025

B. 0.001

C. 0.01

D. 0.25

Correct Answer: B. 0.001

Question 6

A 2 mV/V load cell is excited with 10 V. What is its ideal full-scale output?

A. 2 mV

B. 5 mV

C. 10 mV

D. 20 mV

Correct Answer: D. 20 mV

Question 7

Which sensor is generally most appropriate for measuring a short-duration impact?

A. Piezoelectric sensor

B. Potentiometer

C. Static load cell only

D. Absolute encoder

Correct Answer: A. Piezoelectric sensor

Question 8

What is a major source of long-term angle error when gyroscope output is integrated?

A. Encoder index pulse

B. Bias drift

C. Bridge completion resistor

D. Lens distortion

Correct Answer: B. Bias drift

Question 9

In a tactile array, a taxel is best described as what?

A. A camera lens

B. An individual tactile sensing element

C. A resolver winding

D. A calibration weight

Correct Answer: B. An individual tactile sensing element

Question 10

In the ideal pinhole camera equation u = f × X ÷ Z + c_x, what does Z represent?

A. Object depth along the camera optical axis

B. Sensor supply voltage

C. Encoder count

D. Strain-gauge resistance

Correct Answer: A. Object depth along the camera optical axis