Intermediate Mechatronics interview questions

150 questions with answers (MTX-151 to MTX-300). Try each one aloud before reading the answer.

Mechatronic System Design

MTX-151 · How do you select a sensor for an application?

Match measured quantity, range, resolution, accuracy, bandwidth and environment (temperature, dust, moisture, EMI). Then check output type and interface, power, size, mounting, life and cost. Choose the simplest sensor that meets the requirement with margin.

MTX-152 · How do you size a motor for an axis?

Define the motion profile, then compute load torque from friction, gravity and acceleration (including reflected inertia). Check peak torque against the motor's peak rating, RMS torque against continuous rating, maximum speed, and the load-to-motor inertia ratio.

MTX-153 · What is inertia ratio and why does it matter?

The load inertia reflected to the motor shaft divided by the motor rotor inertia. A large ratio makes the servo loop sluggish and prone to resonance; many designers aim to keep it below about 10:1 for responsive servo axes.

MTX-154 · How does a gearbox change reflected inertia?

Load inertia seen at the motor shaft is divided by the square of the gear ratio: J_reflected = J_load/N². A 10:1 gearbox reduces reflected inertia by a factor of 100, which is why gearing helps match high-inertia loads.

MTX-155 · What gear ratio gives maximum acceleration of an inertial load?

The ratio that matches reflected load inertia to motor inertia: N_opt = √(J_load/J_motor). At this ratio, half the motor torque accelerates the motor and half the load, giving maximum load acceleration for a given torque.

MTX-156 · How do you choose between electric, hydraulic and pneumatic actuation?

Electric suits precise, programmable, clean motion at small to medium force. Hydraulic suits very high force and power density with stiff holding. Pneumatic suits fast, cheap, simple end-to-end motion with low force and limited precision.

MTX-157 · Why does mechanical stiffness matter in a servo system?

Compliance in couplings, belts and frames creates resonances between motor and load. These limit the achievable control bandwidth and gain, so a stiff mechanical design allows faster, more accurate control.

MTX-158 · How do you size a power supply for a mechatronic system?

Add the continuous power of all loads (motors, electronics, heaters) with efficiency losses, then check peak current during simultaneous acceleration. Add margin (often 20–30%), check inrush, and plan for regenerated energy returning from motors.

MTX-159 · What is EMC and why does it matter in mechatronic design?

Electromagnetic compatibility means a device neither emits excessive interference nor is disturbed by it. PWM drives and motors produce strong noise, so layout, filtering, shielding and grounding must be designed in to keep sensors and communications reliable.

MTX-160 · What is a ground loop and how do you avoid it?

A ground loop forms when two points are grounded at different potentials, causing current through signal grounds and noise in measurements. Avoid it with star grounding, a single ground reference point, differential inputs or galvanic isolation.

MTX-161 · Why is thermal management important in drives and electronics?

Losses in power semiconductors, windings and resistors cause heating, which reduces efficiency, lifetime and rated output. Heat sinks, fans, derating and temperature monitoring keep components within limits.

MTX-162 · What is a morphological chart?

A concept-generation table listing the required sub-functions in rows and alternative solutions for each in columns. Combining one solution per row generates candidate system concepts for evaluation.

MTX-163 · What is hardware-software partitioning?

Deciding which functions are implemented in mechanics or electronics and which in software. For example, a mechanical cam can be replaced by an electronic cam profile in a servo drive, adding flexibility at the cost of sensing and computation.

Sensors

MTX-164 · What is the difference between incremental and absolute encoders?

An incremental encoder outputs pulses as it moves, so position is counted from a reference and lost on power loss unless re-homed. An absolute encoder outputs a unique digital code for every position, so position is known immediately at power-up.

MTX-165 · What is the difference between sensor resolution and accuracy?

Resolution is the smallest change in the measured quantity the sensor can detect. Accuracy is how close the reading is to the true value. A sensor can have fine resolution yet poor accuracy because of offset, gain or linearity errors.

MTX-166 · How does quadrature encoding work?

Two channels, A and B, are offset by 90° electrical. Direction comes from which channel leads. Counting every edge of both channels (4× decoding) gives four counts per line, so a 1000-PPR encoder gives 4000 counts per revolution.

MTX-167 · How does an LVDT indicate direction of motion?

The output amplitude gives the displacement magnitude, and the output phase relative to the excitation (in phase or 180° out) shows which side of null the core is on. A phase-sensitive demodulator converts this to a signed DC voltage.

MTX-168 · What is thermocouple cold-junction compensation?

A thermocouple measures the difference between its hot and reference junctions. The reference junction temperature is measured with another sensor such as a thermistor, and its equivalent voltage is added to obtain the absolute hot-junction temperature.

MTX-169 · Why use 3-wire or 4-wire RTD connections?

Lead resistance adds to the RTD resistance and causes error. A 3-wire connection compensates for lead resistance if the leads match. A 4-wire connection forces current on two wires and senses voltage on the other two, eliminating lead error.

MTX-170 · How does a capacitive sensor measure displacement?

Capacitance C = ε·A/d depends on plate area, gap and dielectric. Changing the gap or overlap area with the target changes capacitance, which is converted to a voltage, giving very high resolution at short range.

MTX-171 · How does a resolver work?

A resolver is a rotary transformer with an excited rotor winding and two stator windings at 90°. The stator outputs are amplitude-modulated by sin θ and cos θ of shaft angle, which a resolver-to-digital converter decodes. It is rugged and tolerates heat and vibration.

MTX-172 · What is sensor hysteresis?

The difference in output for the same input value depending on whether the input is increasing or decreasing. It is caused by friction, magnetic effects or material behaviour and is expressed as a percentage of full scale.

MTX-173 · What is a sensor's dynamic response?

How quickly the sensor output follows changes in the input. It is described by time constant, rise time, bandwidth or natural frequency. A slow sensor adds lag in a control loop and distorts fast signals.

MTX-174 · How does an ultrasonic sensor calculate distance?

It emits a sound pulse and measures the echo's round-trip time t. Distance d = v·t/2, where the speed of sound v ≈ 343 m/s in air at 20 °C. A 5.83 ms echo corresponds to about 1 m.

MTX-175 · How does a MEMS gyroscope work?

A vibrating proof mass experiences a Coriolis force perpendicular to its vibration when the device rotates. The resulting secondary deflection, sensed capacitively, is proportional to angular rate.

MTX-176 · What is linearity error in a sensor?

The maximum deviation of the actual input-output curve from a reference straight line (best-fit or end-point), expressed as a percentage of full scale. It can be reduced by calibration tables or polynomial correction.

Actuators

MTX-177 · How do you calculate the thrust from a ball screw?

F = 2π·η·T/L, where T is motor torque (N·m), L is lead (m) and η is efficiency. With T = 1 N·m, L = 5 mm and η = 0.9: F = 2π × 0.9 × 1/0.005 ≈ 1131 N.

MTX-178 · Why choose a ball screw over a lead screw?

Ball screws have rolling contact, giving efficiency above 90%, low wear, high precision and preloading to remove backlash. Lead screws are cheaper, quieter and can be self-locking, but have high friction and wear, with efficiency often below 50%.

MTX-179 · What is back-driving and self-locking?

Back-driving is when a load force on the output drives the input to rotate. A screw or worm gear is self-locking when friction is high enough to prevent this, typically when the lead angle is smaller than the friction angle.

MTX-180 · How does solenoid force vary with plunger position?

Force rises sharply as the air gap closes because the magnetic reluctance drops; ideally it varies roughly as 1/gap². This is why solenoids pull weakly at the start of stroke and strongly when nearly closed.

MTX-181 · Why is a flyback diode placed across a relay or solenoid coil?

When the switch turns off, the coil's inductance tries to keep current flowing, producing a high voltage spike (V = L·di/dt) that can destroy the transistor. The diode gives the current a safe path to decay.

MTX-182 · What are the limitations of piezoelectric actuators?

They give very short stroke (tens of micrometres for a stack), need high drive voltages, and show open-loop hysteresis that can be around 10–15% of stroke, plus creep. Closed-loop control with a position sensor is used for precision.

MTX-183 · What is a harmonic drive?

A strain wave gear with a flexible spline, a rigid circular spline and an elliptical wave generator. It gives high reduction ratios (typically 30:1 to 160:1) in a compact package with near-zero backlash.

MTX-184 · How does a linear motor compare with a ball screw drive?

A linear motor drives the load directly, with no backlash, very high speed and acceleration, and no wear. It costs more, has lower force density, needs cooling, and does not hold a vertical load without a brake.

MTX-185 · How does a shape memory alloy actuator work?

At low temperature, the alloy is in a soft martensite phase and can be deformed. Heating it, usually by passing current, transforms it to austenite and it recovers its original shape with useful force. Cooling is slow, limiting cycle speed.

MTX-186 · What is actuator duty cycle rating?

The percentage of time an actuator can be energised in a given period without overheating. A solenoid rated 25% duty can be on for 15 s in each minute; continuous duty (100%) needs a larger or lower-power design.

MTX-187 · What are the advantages of planetary gearboxes?

Load is shared between several planet gears, giving high torque density, coaxial input and output, high efficiency and low backlash in precision versions. They are common with servo motors.

MTX-188 · How does a solid-state relay differ from an electromechanical relay?

An SSR switches with semiconductors (triac, thyristor or MOSFET) and optical isolation: no moving parts, silent, fast and long-lived. It has a voltage drop and heat dissipation, some leakage current when off, and usually tends to fail shorted.

MTX-189 · How do you calculate actuator force to lift a load vertically with acceleration?

F = m(g + a) + friction. For m = 10 kg and a = 2 m/s², F = 10 × (9.81 + 2) ≈ 118 N plus friction and a safety margin.

Electric Motors & Drives

MTX-190 · How do stepper, servo and BLDC motors compare?

A stepper is usually run open-loop with precise step increments and high low-speed torque. A servo uses feedback for accurate closed-loop control of position, speed or torque. A BLDC is an efficient, electronically commutated motor often used for speed control.

MTX-191 · What is an H-bridge?

Four switches (transistors) arranged in an H around the motor. Turning on diagonal pairs applies voltage in either polarity, so the DC motor can run forward or reverse, and PWM on the switches controls speed. Both switches on one side must never be on together.

MTX-192 · What is back-EMF?

The voltage generated by a spinning motor acting as a generator, opposing the applied voltage. It is proportional to speed, E = Ke·ω, and it limits current as the motor speeds up.

MTX-193 · What is the relationship between torque constant and back-EMF constant?

In SI units, a permanent-magnet DC motor's torque constant Kt (N·m/A) equals its back-EMF constant Ke (V·s/rad). Torque T = Kt·I and back-EMF E = Ke·ω both come from the same magnetic flux linkage.

MTX-194 · What is the steady-state speed equation of a DC motor?

From V = I·R + Ke·ω, the speed is ω = (V − I·R)/Ke. Speed rises with voltage and falls as load current increases.

MTX-195 · How is a BLDC motor commutated with Hall sensors?

Three Hall sensors spaced around the stator give six distinct states per electrical revolution. Each state selects which two of the three phases are energised (six-step trapezoidal commutation), keeping the stator field ahead of the rotor.

MTX-196 · What is microstepping?

Driving a stepper's two phases with stepped sinusoidal currents instead of full on/off, so the rotor takes positions between full steps. It gives smoother motion and less resonance, though the extra resolution is not matched by equal positional accuracy.

MTX-197 · What is the difference between full-step and half-step drive?

In full-step, one or two phases are energised at a time, giving the rated step angle. Half-step alternates between one and two phases on, halving the step angle (0.9° for a 1.8° motor) with some torque ripple.

MTX-198 · What is V/f control of an induction motor?

The drive keeps the ratio of voltage to frequency constant below base speed so the air-gap flux and torque capability remain roughly constant while speed changes. It is simple scalar control without accurate torque control.

MTX-199 · How does regenerative braking work in a drive?

When the load drives the motor faster than its command, the motor acts as a generator and returns energy to the DC bus. The energy is stored, fed back to the supply by an active front end, or dissipated in a braking resistor.

MTX-200 · How do you calculate mechanical power of a motor?

P = T·ω with ω in rad/s. A motor delivering 2 N·m at 3000 rpm has ω = 3000 × 2π/60 ≈ 314 rad/s, so P ≈ 628 W.

MTX-201 · What is the difference between unipolar and bipolar stepper motors?

A unipolar stepper has centre-tapped windings, so current flows one way through each half and the driver is simple. A bipolar stepper reverses current through whole windings using H-bridges, giving more torque for the same size.

MTX-202 · What is cogging torque?

Torque ripple in permanent-magnet motors caused by the attraction between rotor magnets and stator teeth, felt as detent positions when turned by hand. It affects smooth motion at low speed and is reduced by skewing or slot design.

Microcontrollers & Interfacing

MTX-203 · How is PWM frequency set on a microcontroller timer?

f_PWM = f_clk/((PSC + 1)(ARR + 1)), where PSC is the prescaler and ARR the auto-reload value. With 72 MHz, PSC = 71 and ARR = 999, f_PWM = 72 × 10⁶/(72 × 1000) = 1 kHz. Duty = compare value/(ARR + 1).

MTX-204 · What is the trade-off between PWM frequency and resolution?

For a fixed timer clock, resolution in steps equals f_clk/f_PWM. Raising PWM frequency reduces the number of duty-cycle steps. A 16 MHz timer gives 16,000 steps at 1 kHz but only 800 at 20 kHz.

MTX-205 · What is the difference between polling and interrupts?

Polling repeatedly checks a flag or input in software, which is simple but wastes CPU time and can miss short events. Interrupts let hardware signal the CPU only when an event occurs, giving faster response and better efficiency.

MTX-206 · What are good practices for writing an ISR?

Keep it short and fast, avoid blocking calls and long loops, set flags or buffer data for the main loop, clear the interrupt flag, and declare shared variables volatile while protecting multi-byte data from race conditions.

MTX-207 · What does the volatile keyword do in embedded C?

It tells the compiler that a variable can change outside the normal program flow, such as in an ISR or hardware register, so every access must read or write memory instead of using a cached copy.

MTX-208 · When would you choose I²C over SPI or the other way round?

I²C uses only two wires and supports addressing many devices, which suits slow sensors and configuration chips. SPI is faster, full-duplex and simpler at the protocol level, suiting ADCs, displays and flash memory, but needs a chip select per device.

MTX-209 · Why does I²C need pull-up resistors?

I²C devices use open-drain outputs that can only pull the line low. Pull-up resistors return SDA and SCL to high. Their value balances rise time (bus capacitance) against sink current; typical values are 2.2–10 kΩ.

MTX-210 · What is DMA?

Direct Memory Access lets a peripheral transfer data to or from memory without CPU involvement. It is used for continuous ADC sampling, UART buffers and SPI transfers, freeing the CPU for control tasks.

MTX-211 · How do you interface a 5 V sensor output to a 3.3 V microcontroller?

Use a resistor voltage divider for slow analog or digital signals, a level-shifter IC for bidirectional or fast signals, or check whether the pin is 5 V tolerant. Never exceed the input's absolute maximum voltage.

MTX-212 · How do you drive an inductive load from a GPIO pin?

Use a transistor or logic-level MOSFET as a switch, with a gate or base resistor, and a flyback diode across the load. The GPIO only supplies the small control current; the load current comes from a separate supply.

MTX-213 · What is CAN bus?

Controller Area Network: a robust differential two-wire bus with priority-based arbitration and error checking. Classic CAN runs up to 1 Mbit/s and needs 120 Ω termination at each end. It is used in vehicles and machines.

MTX-214 · When do you need an RTOS?

When a system has several concurrent tasks with different timing requirements, such as control loops, communications and user interface. An RTOS provides prioritised scheduling, timing services and inter-task communication with predictable timing.

MTX-215 · How do you convert an ADC reading to voltage?

V = (code/(2ⁿ − 1)) × Vref, or approximately code × Vref/2ⁿ. For a 12-bit ADC with a 3.3 V reference, a reading of 2048 corresponds to about 1.65 V.

Signal Conditioning & Data Acquisition

MTX-216 · What is the output of a quarter-bridge strain gauge circuit?

Vout ≈ Vex·GF·ε/4 for small strain. With 5 V excitation, GF = 2 and 1000 µε: Vout ≈ 5 × 2 × 0.001/4 = 2.5 mV, which must then be amplified.

MTX-217 · Why use half-bridge or full-bridge strain gauge configurations?

Placing active gauges in more arms increases output (2× for half, 4× for full bridge with opposite strains) and cancels temperature effects, because temperature changes affect all gauges equally while strains add.

MTX-218 · How is the gain of a three-op-amp instrumentation amplifier set?

Gain G = (1 + 2R/Rg) × (R3/R2), and with a unity-gain difference stage G = 1 + 2R/Rg. A single external resistor Rg sets the gain without degrading common-mode rejection.

MTX-219 · What is CMRR?

Common-Mode Rejection Ratio: the ratio of differential gain to common-mode gain, usually in dB. A high CMRR (for example 100 dB) means noise appearing equally on both inputs is strongly rejected.

MTX-220 · What are the gain equations for inverting and non-inverting op-amp amplifiers?

Inverting: G = −Rf/Rin, with input impedance equal to Rin. Non-inverting: G = 1 + Rf/Rg, with very high input impedance. A non-inverting amplifier with Rf = 0 is a unity-gain buffer.

MTX-221 · What is the difference between passive and active filters?

Passive filters use only R, L and C, need no power and can handle high power, but cannot provide gain and are loaded by the next stage. Active filters use op-amps to give gain, buffering and sharper responses without inductors.

MTX-222 · How do Butterworth, Chebyshev and Bessel filters differ?

Butterworth has the flattest passband. Chebyshev gives a steeper roll-off with passband ripple. Bessel has the most linear phase (constant group delay), preserving pulse shape, but with the slowest roll-off.

MTX-223 · Why is sampling usually much faster than twice the signal frequency?

Real anti-aliasing filters are not ideal and need a transition band, and capturing waveform shape or feeding a control loop needs several samples per cycle. Sampling at 5–10× the highest frequency of interest is common in practice.

MTX-224 · What is the difference between single-ended and differential inputs?

A single-ended input measures relative to a common ground and picks up ground noise. A differential input measures the voltage between two lines, rejecting noise common to both, which suits low-level signals and long cables.

MTX-225 · Why is the 4–20 mA current loop widely used?

Current is not affected by cable resistance or voltage drops, is robust to noise over long distances, and the 4 mA live zero detects broken wires (0 mA). Two-wire transmitters can also be powered from the loop.

MTX-226 · How do you convert a 4–20 mA signal to a voltage?

Pass it through a precision resistor. A 250 Ω resistor gives 1–5 V, since V = I·R. Engineering value = (I − 4)/16 × span + minimum.

MTX-227 · What is a sample-and-hold circuit?

A circuit that captures the input voltage at the sampling instant and holds it constant while the ADC performs conversion. It prevents errors from the input changing during conversion.

MTX-228 · What is a moving-average filter?

A digital filter that outputs the average of the last N samples. It reduces random noise by about √N but adds a delay of (N − 1)/2 samples and has poor frequency selectivity.

Control Systems

MTX-229 · What does each term of a PID controller do?

P gives a correction proportional to error, speeding response but leaving steady-state error in many systems. I accumulates error to remove steady-state error but adds lag and overshoot. D reacts to the rate of change, adding damping but amplifying noise.

MTX-230 · What is integral windup and how is it prevented?

When the actuator saturates, the integrator keeps accumulating error, causing large overshoot when the error reverses. Anti-windup methods clamp the integrator, stop integration during saturation, or use back-calculation.

MTX-231 · What is the Ziegler–Nichols tuning method?

In the ultimate gain method, set I and D to zero and raise Kp until sustained oscillation occurs at gain Ku with period Tu. A classic PID setting is Kp = 0.6Ku, Ti = Tu/2, Td = Tu/8. It gives aggressive tuning with notable overshoot.

MTX-232 · How are overshoot and damping ratio related for a second-order system?

Percent overshoot Mp = exp(−πζ/√(1 − ζ²)) × 100%. For ζ = 0.5, Mp ≈ 16.3%; for ζ = 0.7, Mp ≈ 4.6%. Undamped natural frequency ωn sets the speed of response.

MTX-233 · How is steady-state error found using the final value theorem?

e_ss = lim(s→0) s·E(s). For unity feedback with a step input, e_ss = 1/(1 + Kp_static), where Kp_static = lim(s→0) G(s). A type-1 system (one integrator) has zero step error.

MTX-234 · What is the Routh–Hurwitz criterion?

An algebraic test that determines the number of roots of the characteristic polynomial in the right-half s-plane from a table of its coefficients, without solving for roots. The system is stable if there are no sign changes in the first column.

MTX-235 · What are gain margin and phase margin?

Gain margin is how much loop gain can increase before instability, measured at the phase crossover (−180°). Phase margin is the extra phase lag allowed at the gain crossover (0 dB) before instability. Typical targets are GM > 6 dB and PM of 45–60°.

MTX-236 · What is a Bode plot?

Plots of magnitude (dB) and phase (degrees) of a system's frequency response against logarithmic frequency. It shows bandwidth, resonances and stability margins and is used to design compensators.

MTX-237 · What is cascade control in servo drives?

Nested loops: an inner current (torque) loop, a middle velocity loop and an outer position loop. Each inner loop is faster than the next outer one (often 5–10 times), which simplifies tuning and improves disturbance rejection.

MTX-238 · Why is the derivative term filtered in practical PID controllers?

Pure differentiation amplifies high-frequency noise. A low-pass filter on the D term (for example with time constant Td/N, N ≈ 5–20) limits high-frequency gain. Taking the derivative of the measurement instead of error also avoids set-point "kick".

MTX-239 · How is a PID controller implemented digitally?

The controller runs at a fixed sample time Ts: the integral is approximated as a sum (I += Ki·e·Ts) and the derivative as a difference ((e − e_prev)/Ts). A consistent, jitter-free sample time is essential.

MTX-240 · What is a root locus?

A plot of how closed-loop poles move in the s-plane as a gain varies from zero to infinity. It shows the gain range for stability and helps choose gain for desired damping and speed.

PLC & Industrial Control

MTX-241 · How does a start-stop motor control rung work in ladder logic?

A NO start contact in parallel with a NO contact of the motor output, all in series with the stop contact, drives the motor coil. Pressing start energises the coil, which seals itself in; pressing stop breaks the circuit.

MTX-242 · Why are stop buttons wired as normally closed?

If the wire breaks or a terminal loosens, the input goes off just as if stop were pressed, so the machine stops. With an NO stop button, a broken wire would silently prevent stopping. This is fail-safe design.

MTX-243 · What is the difference between TON, TOF and retentive timers?

TON delays turning on and resets when its input drops. TOF delays turning off after its input drops. A retentive timer (such as RTO or TONR) keeps its accumulated time when the input goes off and needs a separate reset.

MTX-244 · What is a one-shot (edge detection) instruction?

An instruction (such as R_TRIG, OSR or a P contact) that is true for only one scan when its input changes from false to true. It ensures an action, such as incrementing a counter, happens once per button press.

MTX-245 · How does scan time affect fast input signals?

An input pulse shorter than the scan time may go unseen, because inputs are only read once per scan. Fast signals like encoders need high-speed counter modules or hardware interrupt inputs that work independently of the scan.

MTX-246 · What is Sequential Function Chart (SFC)?

A graphical IEC 61131-3 language derived from GRAFCET that describes sequential processes as steps with actions, linked by transitions with conditions. It suits machine sequences and makes complex sequences easy to follow.

MTX-247 · When would you use Structured Text instead of ladder logic?

ST is a Pascal-like text language suited to mathematics, loops, arrays, data handling and complex algorithms. Ladder is better for discrete interlocks and is easier for maintenance staff to troubleshoot online.

MTX-248 · How are analog inputs scaled in a PLC?

The raw ADC count is linearly scaled to engineering units: value = (raw − raw_min)/(raw_max − raw_min) × (EU_max − EU_min) + EU_min. For example, a Siemens S7 analog input uses 0–27648 counts for the nominal range.

MTX-249 · How does a safety PLC differ from a standard PLC?

A safety PLC has redundant processors, self-diagnostics, safe I/O with test pulses and certified firmware, and is rated for functional safety (SIL or PL). It is used for emergency stops, light curtains and guard interlocks.

MTX-250 · What is the difference between Modbus RTU and Modbus TCP?

Modbus RTU is a binary master-slave protocol over serial links like RS-485, with CRC error checking. Modbus TCP carries the same function codes over Ethernet TCP/IP on port 502, with no CRC since TCP handles errors.

MTX-251 · How does a star-delta starter work and why is it used?

The motor starts with windings in star, so each winding sees 1/√3 of line voltage and starting current and torque fall to about one-third of direct-on-line values. After a timed delay, the PLC or contactors switch to delta for normal running.

MTX-252 · How do you interlock forward and reverse motor contactors?

Place an NC contact of the reverse output in the forward rung and an NC contact of forward in the reverse rung (software interlock). Also wire NC auxiliary contacts in each contactor coil circuit, or use a mechanical interlock, to prevent a phase-to-phase short.

Hydraulics & Pneumatics

MTX-253 · How do you calculate hydraulic cylinder speed?

v = Q/A. For Q = 20 L/min = 3.33 × 10⁻⁴ m³/s and a 50 mm bore (A = 1.963 × 10⁻³ m²), v ≈ 0.17 m/s on extension.

MTX-254 · How do you calculate hydraulic power?

P = p × Q. At 100 bar (10⁷ Pa) and 30 L/min (5 × 10⁻⁴ m³/s), P = 5 kW of fluid power. Electric motor power = fluid power/overall pump efficiency.

MTX-255 · Why does a double-acting cylinder retract faster than it extends?

On retraction, fluid acts on the annulus area (piston area minus rod area), which is smaller. For the same flow, speed v = Q/A is higher, but available force F = p·A is lower.

MTX-256 · What is the difference between meter-in and meter-out flow control?

Meter-in throttles the flow entering the actuator; it suits loads that resist motion. Meter-out throttles the flow leaving the actuator, holding back the load; it gives better control of overrunning or pulling loads and is common in pneumatics.

MTX-257 · How do proportional valves differ from servo valves?

Proportional valves use a proportional solenoid to move the spool in proportion to an electrical signal; they are cheaper and more dirt-tolerant but have deadband and lower bandwidth. Servo valves use a torque motor and pilot stage, with high bandwidth and precision but need very clean oil.

MTX-258 · What is pump cavitation?

When inlet pressure drops below the fluid's vapour pressure, vapour bubbles form and then collapse violently in high-pressure zones. This causes noise, vibration, erosion and loss of flow. Causes include clogged suction filters, high suction lift or high viscosity.

MTX-259 · What is a regenerative hydraulic circuit?

The oil leaving the rod end during extension is routed back into the cap end instead of to the tank. This increases extension speed (effective area equals rod area) but reduces available force.

MTX-260 · What is a counterbalance valve?

A pressure-control valve in the return line that prevents a vertical or overrunning load from falling or running away. It holds back pressure until pilot pressure from the driving side opens it.

MTX-261 · Why is hydraulic fluid viscosity important?

Too low a viscosity increases leakage and wear; too high increases pressure losses, heat and cavitation risk. Viscosity changes with temperature, so the fluid grade and operating temperature range must be matched.

MTX-262 · What is a pressure-compensated variable-displacement pump?

A pump (usually axial piston) whose displacement automatically reduces as system pressure approaches a set value. It delivers only the flow needed, reducing heat and energy waste compared with a fixed pump dumping flow over a relief valve.

MTX-263 · How does a pneumatic vacuum generator work?

Compressed air passes through a venturi nozzle, where high velocity creates low pressure that draws air from the suction port. It provides vacuum for suction cups without a vacuum pump, with no moving parts.

MTX-264 · Why are pneumatic cylinders difficult to stop accurately mid-stroke?

Air is compressible, so the cylinder behaves like a spring and the load can overshoot or drift. Friction, stick-slip and changing load add to the problem. Mechanical stops, brakes or servo-pneumatic control are used when mid-stroke positions are needed.

System Modeling & Simulation

MTX-265 · What are the governing equations of an armature-controlled DC motor?

Electrical: V = L·di/dt + R·i + Ke·ω. Mechanical: J·dω/dt + b·ω = Kt·i − T_load. These are coupled by back-EMF and torque, with Kt = Ke in SI units.

MTX-266 · What is the step response of a first-order system?

For G(s) = K/(τs + 1) with a step of size A, y(t) = K·A·(1 − e^(−t/τ)). It reaches 63.2% at t = τ and about 98% at 4τ, with no overshoot.

MTX-267 · What are the natural frequency and damping ratio of a mass-spring-damper?

ωn = √(k/m) and ζ = c/(2√(km)). For m = 1 kg, k = 100 N/m and c = 4 N·s/m: ωn = 10 rad/s and ζ = 0.2, so the system is underdamped.

MTX-268 · What is linearisation?

Approximating a nonlinear model by a linear one around an operating point using a first-order Taylor expansion. The linear model is valid for small deviations and allows transfer function analysis and linear controller design.

MTX-269 · How do you convert a transfer function into state-space form?

Choose state variables, often the output and its derivatives (controllable canonical form). For example, for Y/U = 1/(s² + a₁s + a₀), set x₁ = y, x₂ = ẏ, giving ẋ₁ = x₂ and ẋ₂ = −a₀x₁ − a₁x₂ + u.

MTX-270 · How do Euler and Runge–Kutta integration methods differ?

Forward Euler uses only the slope at the start of the step; it is simple but first-order accurate and can be unstable for large steps. Fourth-order Runge–Kutta (RK4) uses four slope evaluations per step and is much more accurate for the same step size.

MTX-271 · How is a simulation model validated?

Compare model outputs with measured data from the real system under the same inputs, ideally data not used for parameter fitting. Check steady-state values, transient response and frequency response, and quantify the error.

MTX-272 · What is system identification?

Building or fitting a dynamic model from measured input-output data. Test signals such as steps, chirps or pseudo-random binary sequences excite the system, and methods like least squares estimate the model parameters.

MTX-273 · How do you model a gear train between motor and load?

Reflect load quantities to the motor side: inertia J/N², damping b/N², torque T/N (ideal gearbox), plus efficiency losses. Then total motor-side inertia is J_motor + J_load/N².

MTX-274 · How is a simple thermal system modelled?

As a thermal resistance R (K/W) and capacitance C (J/K): C·dT/dt = P − (T − T_amb)/R. It is first-order, with time constant τ = R·C and steady-state temperature rise ΔT = P·R.

MTX-275 · What are MIL and SIL testing?

Model-in-the-loop tests the controller model against a plant model entirely in simulation. Software-in-the-loop compiles the generated or hand-written controller code and runs it against the plant model on a PC to check the implementation.

MTX-276 · What is a stiff system in simulation?

A system with dynamics on very different time scales, such as fast electrical and slow thermal effects. Explicit solvers need tiny steps for stability, so implicit stiff solvers are used to simulate efficiently.

Machine Vision

MTX-277 · How do you calculate the camera resolution needed for an inspection?

Pixels = FOV/(smallest feature) × pixels per feature. To detect 0.1 mm defects across a 100 mm FOV with 3 pixels per feature, you need 100/0.1 × 3 = 3000 pixels along that axis.

MTX-278 · How do focal length and working distance determine field of view?

Using the thin-lens approximation, FOV ≈ sensor size × working distance/focal length. A 7.2 mm sensor width, 300 mm working distance and 16 mm lens give FOV ≈ 135 mm.

MTX-279 · What is blob analysis?

Grouping connected pixels of a binary image into regions (blobs) and measuring properties such as area, centroid, perimeter and bounding box. It is used for counting parts, presence checks and locating objects.

MTX-280 · What is template matching?

Searching an image for regions that resemble a stored reference pattern, often using normalised cross-correlation, which is robust to uniform brightness changes. Geometric pattern matching extends this to rotation and scale.

MTX-281 · What are morphological operations?

Shape-based operations on binary or grayscale images using a structuring element. Erosion shrinks objects, dilation grows them, opening (erosion then dilation) removes small noise, and closing (dilation then erosion) fills small holes.

MTX-282 · What is the difference between global and adaptive thresholding?

Global thresholding uses a single threshold for the whole image, possibly chosen automatically by Otsu's method. Adaptive thresholding computes a local threshold for each region, handling uneven lighting.

MTX-283 · How do Sobel and Canny edge detectors differ?

Sobel computes image gradients with 3 × 3 kernels and gives thick, noisy edges. Canny smooths with a Gaussian, computes gradients, thins edges by non-maximum suppression and links them with hysteresis thresholds, giving thin, continuous edges.

MTX-284 · What is a telecentric lens?

A lens whose chief rays are parallel to the optical axis in object space, so magnification does not change with object distance and there is no perspective error. It is used for precise dimensional measurement.

MTX-285 · What is the Hough transform?

A technique that detects parametric shapes like lines and circles by having each edge pixel vote for all shapes it could belong to in parameter space. Peaks in the accumulator correspond to detected shapes, even with gaps and noise.

MTX-286 · How does exposure time cause motion blur?

Blur length = object speed × exposure time. A part moving at 1 m/s with 1 ms exposure blurs by 1 mm. Short exposures or strobed lighting freeze motion, but need more light.

MTX-287 · How does structured light 3D vision work?

A projector casts known patterns (lines, stripes or grids) onto the object, and a camera at an angle observes their deformation. Triangulation converts the pattern shift into depth, building a 3D profile.

MTX-288 · Why use the HSV colour space instead of RGB for colour inspection?

HSV separates hue (colour), saturation and value (brightness). Colour segmentation on hue is less affected by lighting intensity changes than thresholds on R, G and B channels, which all change with brightness.

Industry 4.0, IoT & Digital Twins

MTX-289 · What are the MQTT QoS levels?

QoS 0: at most once, no acknowledgement. QoS 1: at least once, acknowledged, duplicates possible. QoS 2: exactly once, using a four-step handshake. Higher QoS adds reliability but more overhead and latency.

MTX-290 · How does OPC UA differ from Modbus?

Modbus exchanges raw registers and coils with no meaning attached and no built-in security. OPC UA provides a rich information model with named, typed variables and methods, discovery, and built-in authentication and encryption.

MTX-291 · How does a digital twin differ from a normal simulation model?

A simulation model usually runs offline with assumed inputs. A digital twin is linked to a specific physical asset and is continuously updated with its operating data, so it reflects the asset's current state and history.

MTX-292 · What are the benefits of edge computing in IIoT?

It reduces latency for local decisions, cuts network bandwidth by pre-processing and filtering data, keeps working when the connection fails, and keeps sensitive data on site.

MTX-293 · How is Overall Equipment Effectiveness (OEE) calculated?

OEE = Availability × Performance × Quality. With 90% availability, 95% performance and 99% quality, OEE = 0.9 × 0.95 × 0.99 ≈ 84.6%.

MTX-294 · How is vibration data used for condition monitoring?

Accelerometers on bearings and gearboxes record vibration, and FFT analysis reveals characteristic frequencies of faults such as imbalance (1× running speed), misalignment (often 2×) and bearing defects. Trend changes trigger maintenance.

MTX-295 · What is the role of an IIoT gateway?

It connects field devices using industrial protocols (Modbus, PROFINET, OPC UA) to IT or cloud systems using protocols like MQTT or HTTPS. It translates protocols, buffers data, adds security and may do edge processing.

MTX-296 · What is Time-Sensitive Networking (TSN)?

A set of IEEE 802.1 Ethernet standards that add time synchronisation, scheduled traffic and bounded latency. It lets real-time control traffic and normal IT traffic share the same Ethernet network deterministically.

MTX-297 · What is RAMI 4.0?

The Reference Architectural Model Industrie 4.0: a three-dimensional model combining architecture layers (asset to business), product life cycle and value stream, and hierarchy levels (product to connected world). It gives a common framework for Industry 4.0 components.

MTX-298 · What is the Asset Administration Shell?

The standardised digital representation of an asset in Industry 4.0. It holds the asset's information and functions in submodels, such as nameplate, technical data and operating data, so different systems can exchange information about it.

MTX-299 · Why is cybersecurity critical in IIoT?

Connecting machines to networks exposes control systems to attacks that can stop production, damage equipment or endanger people. Standards such as IEC 62443 define zones, conduits, security levels and secure development practices.

MTX-300 · What is a digital thread?

A connected data flow linking a product's information across its life cycle, from requirements and design through manufacturing to service. It gives traceability and lets feedback from operation improve design.

Other levels: Basic · Advanced · Tricky · All Mechatronics questions