Electrical and Electronics Foundations for Mechanical Engineers

Electrical Quantities, Safety, and Measurement

Electrical/electronics knowledge shows up constantly in mechanical engineering—motors and drives, sensors and instrumentation, embedded control, robotics, HVAC controls, automotive systems, and even “purely mechanical” products that need power, wiring, and compliance. Before you can analyze or design anything electrical, you need the core physical quantities and the practical habits that keep you safe and keep your measurements trustworthy.

The core quantities: charge, voltage, current, resistance

Electric charge is a property of matter that causes electrical forces. In most engineering circuits you don’t track individual charges—you work with how charge flow behaves.

Current is the rate at which charge flows.

$I = \frac{dQ}{dt}$

• $I$ is current in amperes $\text{A}$.
• $Q$ is charge in coulombs $\text{C}$.
• $t$ is time in seconds $\text{s}$.

A common misconception is thinking “current gets used up.” In a steady circuit, current entering a component equals current leaving it—what’s “used” is energy, not current.

Voltage (electric potential difference) is the “push” that drives current. It is energy per unit charge:

$V = \frac{W}{Q}$

• $V$ in volts $\text{V}$.
• $W$ in joules $\text{J}$.

In mechanical terms, voltage is like pressure difference in a pipe system, and current is like volumetric flow rate. That analogy is useful—but imperfect: electrical systems respond much faster, and components like capacitors/inductors store energy in fields rather than as fluid volume.

Resistance describes how strongly a component opposes current flow (for resistive behavior). For an ohmic resistor, resistance is constant (within limits):

$R = \frac{V}{I}$

• $R$ in ohms $\Omega$.

Electrical power and energy

Electrical power is the rate of energy transfer:

$P = VI$

For resistors, you’ll also see:

$P = I^2R$

$P = \frac{V^2}{R}$

These aren’t “new laws”—they come from substituting $V = IR$ into $P = VI$. In practice, power is where mechanical engineers often focus: heat dissipation, battery life, motor sizing, and wire/connector limits.

DC vs AC (why you care)

DC (direct current) means voltage and current are (ideally) constant with time. Batteries and most electronics internally operate on DC.

AC (alternating current) means voltage and current vary with time (often sinusoidally). Mains power is AC, and many motors and power systems are designed around AC.

You’ll frequently convert between them: mains AC gets rectified to DC for electronics; motor drives convert DC to controlled AC for variable-speed motors.

Safety fundamentals (practical and testable)

Electrical safety is not just “don’t touch wires.” The hazard depends on the path through the body, the current level, and the duration. Typical engineering controls include:

• Isolation (transformers, optocouplers, isolated supplies) to prevent hazardous potentials from being referenced to touchable metal.
• Fusing and circuit breakers to interrupt overcurrent conditions.
• Grounding/earthing to provide a low-impedance fault path so protective devices trip.
• Enclosures and insulation to prevent access to live parts.

A key idea: protective devices generally respond to current, but a shock risk is often about voltage creating that current through the body. Wet skin lowers resistance dramatically—so “low voltage” systems can still be dangerous in some environments.

Measurement tools: multimeter basics

A digital multimeter (DMM) typically measures voltage, current, resistance, continuity, and sometimes capacitance, frequency, or diode drop.

• Voltage measurement: put the meter in parallel with the component/nodes of interest. The meter is designed to have high input resistance so it doesn’t load the circuit much.
• Current measurement: put the meter in series so all the branch current flows through the meter. This is where mistakes happen—putting a meter in current mode across a voltage source is effectively a short circuit.
• Resistance measurement: measure only on an unpowered circuit, because the DMM injects a small test current.

Worked example: power in a resistive load

A heater coil is modeled as $R = 12\,\Omega$ and connected to $V = 24\,\text{V}$ DC.

1) Find current:

$I = \frac{V}{R} = \frac{24\,\text{V}}{12\,\Omega} = 2\,\text{A}$

2) Find power:

$P = VI = 24\,\text{V} \times 2\,\text{A} = 48\,\text{W}$

This is the number you’d use for thermal design (temperature rise, heatsinking, airflow).

Exam Focus

• Typical question patterns:
  • Compute $I$, $V$, $R$, or $P$ from given circuit values and choose the correct formula.
  • Identify the correct way to connect a DMM for voltage vs current.
  • Reason about whether a measurement will “load” a circuit.
• Common mistakes:
  • Treating current as “consumed” by a component; confusing energy use with current flow.
  • Using the DMM in current mode across a voltage source (creating a short).
  • Mixing units (mA vs A, kΩ vs Ω) without converting.

Ohm’s Law and Real Resistors

Ohm’s law is the first tool for converting between voltage and current in resistive parts of a circuit. But mechanical engineers also need to know where it stops being accurate—because real parts have tolerances, temperature dependence, and power limits.

Ohm’s law (what it means physically)

Ohm’s law for an ohmic element is:

$V = IR$

This is not a universal law of nature for all components; it’s a model that fits many materials and resistor components over a certain range. It matters because it allows you to reduce many circuit behaviors to solvable linear equations.

Resistor ratings: tolerance, power, temperature coefficient

A real resistor has:

• Nominal resistance (e.g., $10\,\text{k}\Omega$).
• Tolerance (e.g., $\pm 1\%$), meaning the actual value can vary.
• Power rating (e.g., $0.25\,\text{W}$), limiting how much heat it can safely dissipate.
• Temperature coefficient (often specified in $\text{ppm}/\text{°C}$), meaning resistance changes with temperature.

Why this matters: if you design a voltage divider assuming an exact resistance but use $\pm 5\%$ parts, your sensor scaling can drift enough to break calibration.

Conductance and resistivity (bridging to materials/mechanics)

Sometimes it’s easier to think in conductance:

$G = \frac{1}{R}$

And for a uniform conductor of length $L$ and cross-sectional area $A$, resistance relates to **resistivity** $\rho$:

$R = \rho\frac{L}{A}$

This connects directly to mechanical design because wire sizing, connector selection, and temperature rise depend on geometry and material.

Loading: when a “simple” measurement changes the circuit

Even though DMM voltage inputs are high impedance, they’re not infinite. If you measure a high-resistance node (like the midpoint of a large resistor divider), the meter acts like another resistor to ground, changing the divider ratio. The concept you need is: any measurement device becomes part of the circuit model.

Worked example: choosing a resistor power rating

You need a resistor to drop $5\,\text{V}$ at $50\,\text{mA}$.

1) Required resistance:

$R = \frac{V}{I} = \frac{5\,\text{V}}{0.05\,\text{A}} = 100\,\Omega$

2) Power dissipated:

$P = VI = 5\,\text{V} \times 0.05\,\text{A} = 0.25\,\text{W}$

In practice you don’t pick a resistor exactly at its rating. You’d usually choose a higher power rating (e.g., $0.5\,\text{W}$ or more) to reduce temperature rise and improve reliability.

Exam Focus

• Typical question patterns:
  • Compute a resistor value to meet a current or voltage drop requirement.
  • Determine resistor power dissipation and select an appropriate rating.
  • Conceptual questions about why a measured voltage differs from the “ideal” value (loading).
• Common mistakes:
  • Forgetting that $P = \frac{V^2}{R}$ uses the voltage across the resistor, not the supply.
  • Ignoring tolerance and assuming exact values in dividers/sensor scaling.
  • Confusing resistivity $\rho$ (material property) with resistance $R$ (component property).

Circuit Laws: Nodes, Loops, and Systematic Analysis

As soon as circuits stop being single-resistor problems, you need a method that always works. That method is built on conservation laws: conservation of charge and conservation of energy.

Kirchhoff’s Current Law (KCL)

Kirchhoff’s Current Law says the algebraic sum of currents entering a node is zero:

$\sum I = 0$

This is conservation of charge: charge can’t accumulate at an ideal node in steady-state (and even in dynamic cases, KCL still holds when you properly model capacitors).

Kirchhoff’s Voltage Law (KVL)

Kirchhoff’s Voltage Law says the algebraic sum of voltage rises and drops around a closed loop is zero:

$\sum V = 0$

This is conservation of energy: if you start at a point in a circuit and return to it, the net change in potential energy per unit charge must be zero.

Series and parallel (what’s actually the same?)

Students often memorize rules without understanding what “series” and “parallel” mean.

• Series components share the same current (only one path).
• Parallel components share the same voltage (connected between the same two nodes).

From that definition you can derive everything else.

Equivalent resistance:

• Series:

$R_{\text{eq}} = R_1 + R_2 + \cdots$

• Parallel:

$\frac{1}{R_{\text{eq}}} = \frac{1}{R_1} + \frac{1}{R_2} + \cdots$

Voltage divider (a controlled fraction of a voltage)

A voltage divider uses two series resistors across a source. The midpoint voltage (relative to the bottom node) is:

$V_{\text{out}} = V_{\text{in}}\frac{R_2}{R_1 + R_2}$

where $R_2$ is the resistor from the midpoint to the bottom node. This matters in sensors (scaling a signal) and reference generation.

A critical caveat: the formula assumes the output is unloaded (no significant current drawn from the midpoint). If you connect a load, it effectively places a resistor in parallel with $R_2$ and changes the ratio.

Current divider (how current splits)

For two parallel resistors $R_1$ and $R_2$ carrying total current $I_{\text{in}}$, the current in branch 1 is:

$I_1 = I_{\text{in}}\frac{R_2}{R_1 + R_2}$

and similarly:

$I_2 = I_{\text{in}}\frac{R_1}{R_1 + R_2}$

It looks “backward” at first: the branch current is proportional to the other resistance. That’s because current prefers the lower resistance path.

A systematic method: nodal analysis (quick and scalable)

In nodal analysis, you:

1) Choose a reference node (ground).
2) Assign node voltages to other nodes.
3) Write KCL at each non-reference node in terms of node voltages and resistances.
4) Solve the resulting linear system.

This method scales to real circuits (sensor networks, biasing circuits) and is especially useful when you later add dependent sources or op-amp models.

Worked example: voltage divider with load

Let $V_{\text{in}} = 12\,\text{V}$, $R_1 = 2\,\text{k}\Omega$, $R_2 = 1\,\text{k}\Omega$. A load $R_L = 1\,\text{k}\Omega$ is connected from output to ground.

1) Combine $R_2$ and $R_L$ in parallel:

$R_{\text{down}} = \left(\frac{1}{R_2} + \frac{1}{R_L}\right)^{-1} = \left(\frac{1}{1000} + \frac{1}{1000}\right)^{-1}\,\Omega = 500\,\Omega$

2) Now the divider is $R_1$ in series with $R_{\text{down}}$:

$V_{\text{out}} = 12\,\text{V}\frac{500\,\Omega}{2000\,\Omega + 500\,\Omega} = 12\,\text{V}\times 0.2 = 2.4\,\text{V}$

Without the load, you’d get:

$V_{\text{out,ideal}} = 12\,\text{V}\frac{1000}{3000} = 4\,\text{V}$

So the load dramatically changes the output—this is exactly why buffering (using an op-amp follower, for example) matters in instrumentation.

Exam Focus

• Typical question patterns:
  • Identify whether elements are in series or parallel from a schematic.
  • Use KCL/KVL to find unknown currents/voltages.
  • Analyze a divider with and without load.
• Common mistakes:
  • Declaring components “in series” just because they’re drawn next to each other (instead of checking nodes).
  • Using the voltage divider formula on a loaded divider without adjusting $R_2$.
  • Sign errors in KVL loop equations (mixing rises and drops inconsistently).

Sources and Equivalent Circuits (Thévenin and Norton Thinking)

Mechanical systems are often modeled as equivalents (a spring-damper mass model, a transfer function, a lumped thermal network). Circuits have the same philosophy: replace complicated subcircuits with simple equivalents at a pair of terminals.

Ideal vs real voltage sources

An ideal voltage source maintains a fixed voltage regardless of current. A real supply behaves like an ideal source with a small internal resistance in series. Under load, that resistance causes voltage sag and heat.

Similarly, a current source has a compliance voltage limit and non-ideal behavior.

Thévenin equivalent (voltage source form)

Any linear two-terminal network can be represented as a Thévenin equivalent:

• A voltage source $V_{\text{th}}$
• In series with a resistance $R_{\text{th}}$

To find it (for resistor/source networks):

1) $V_{\text{th}}$ is the open-circuit terminal voltage.
2) $R_{\text{th}}$ is the resistance “seen” looking into the terminals with independent sources turned off:

• Voltage sources replaced by shorts.
• Current sources replaced by opens.

Why it matters: it simplifies load calculations, helps explain divider loading, and is foundational for sensor interfaces.

Norton equivalent (current source form)

The Norton equivalent is:

• A current source $I_{\text{n}}$
• In parallel with a resistance $R_{\text{n}}$

with the relationship:

$R_{\text{n}} = R_{\text{th}}$

$I_{\text{n}} = \frac{V_{\text{th}}}{R_{\text{th}}}$

Maximum power transfer (and why you usually don’t want it)

For a Thévenin source with resistance $R_{\text{th}}$ driving a load $R_L$, maximum power to the load occurs when:

$R_L = R_{\text{th}}$

This is a classic result, but in many mechanical/electronic systems you do not want maximum power transfer—you want high efficiency, meaning $R_L \gg R_{\text{th}}$ for a voltage source so most power is delivered to the load and little is wasted internally.

Worked example: Thévenin seen by a load

A divider has $V_{\text{in}} = 10\,\text{V}$, $R_1 = 1\,\text{k}\Omega$ (top), $R_2 = 1\,\text{k}\Omega$ (bottom). Find the Thévenin equivalent at the midpoint relative to ground.

1) Open-circuit voltage:

$V_{\text{th}} = 10\,\text{V}\frac{R_2}{R_1 + R_2} = 10\,\text{V}\times \frac{1}{2} = 5\,\text{V}$

2) Turn off the source: replace the 10 V source with a short. Then the midpoint sees $R_1$ to ground and $R_2$ to ground in parallel:

$R_{\text{th}} = R_1 \parallel R_2 = \left(\frac{1}{1000} + \frac{1}{1000}\right)^{-1}\,\Omega = 500\,\Omega$

So the network behaves like $5\,\text{V}$ in series with $500\,\Omega$.

Exam Focus

• Typical question patterns:
  • Reduce a network to a Thévenin/Norton equivalent at given terminals.
  • Predict loaded output voltage from $V_{\text{th}}$ and $R_{\text{th}}$.
  • Conceptual: when is maximum power transfer relevant vs undesirable?
• Common mistakes:
  • Forgetting to “turn off” sources correctly when finding $R_{\text{th}}$.
  • Mixing up open-circuit voltage with loaded voltage.
  • Applying maximum power transfer as a design goal in efficient power delivery systems.

Capacitors, Inductors, and First-Order Transients

Mechanical engineers often think in time constants—thermal warm-up, pressure equilibration, mechanical damping. RC and RL circuits are the electrical version of first-order dynamics, and they show up everywhere: sensor filtering, debouncing, motor drive protection, and power supply smoothing.

Capacitors (energy in an electric field)

A capacitor stores energy in an electric field. Its defining relationship is:

$i(t) = C\frac{dv(t)}{dt}$

• $C$ is capacitance in farads $\text{F}$.

The stored energy is:

$E = \frac{1}{2}CV^2$

Key intuition:

• A capacitor resists sudden changes in voltage (because that would require infinite current).
• In steady-state DC, an ideal capacitor is an open circuit (no DC current).

Inductors (energy in a magnetic field)

An inductor stores energy in a magnetic field. Its defining relationship is:

$v(t) = L\frac{di(t)}{dt}$

• $L$ is inductance in henries $\text{H}$.

Stored energy:

$E = \frac{1}{2}LI^2$

Key intuition:

• An inductor resists sudden changes in current (because that would require infinite voltage).
• In steady-state DC, an ideal inductor is a short circuit.

RC time constant and step response

For a resistor $R$ and capacitor $C$ in a simple first-order configuration, the time constant is:

$\tau = RC$

If a capacitor charges toward a final voltage $V_f$ from initial voltage $V_0$, a common form is:

$v_C(t) = V_f + (V_0 - V_f)e^{-t/\tau}$

You don’t need to memorize every configuration if you remember the behavior: after about $5\tau$, the response is very close to the final value.

RL time constant

For a resistor $R$ and inductor $L$:

$\tau = \frac{L}{R}$

Why transients matter in electromechanical systems

• Switching a motor (inductive load) creates voltage spikes—this is why flyback diodes or snubbers are used.
• Filtering sensor noise often uses RC low-pass filters.
• Power supplies use capacitance to handle load steps (sudden current demand changes).

Worked example: RC charging time

A sensor signal is filtered with $R = 10\,\text{k}\Omega$ and $C = 1\,\mu\text{F}$.

1) Time constant:

$\tau = RC = 10\,000\,\Omega \times 1 \times 10^{-6}\,\text{F} = 0.01\,\text{s}$

2) Approximate settling time to near-final value:

$t \approx 5\tau = 0.05\,\text{s}$

Interpretation: the filter will noticeably slow changes faster than roughly a few tens of milliseconds.

Exam Focus

• Typical question patterns:
  • Compute $\tau$ for RC/RL networks and estimate rise/decay times.
  • Determine initial and final conditions for capacitor/inductor in switching events.
  • Conceptual: explain why inductive loads generate voltage spikes.
• Common mistakes:
  • Treating capacitors as shorts in DC steady-state (it’s the opposite: open circuit).
  • Forgetting units: $\mu\text{F}$ and $\text{k}\Omega$ conversions can shift $\tau$ by orders of magnitude.
  • Assuming the voltage across an inductor can’t change (it’s the current that can’t change instantly).

AC Fundamentals: Sinusoids, RMS, and Impedance

Once you move beyond batteries and simple DC, you’ll encounter AC in mains power, transformers, and motor drives. Even when your device is “DC powered,” the ripple and switching behavior often require AC-like thinking.

Sinusoidal signals and frequency

A sinusoidal voltage can be written as:

$v(t) = V_p\sin(\omega t + \phi)$

• $V_p$ is peak voltage.
• $\omega$ is angular frequency in $\text{rad/s}$.
• $\phi$ is phase.

Frequency $f$ relates to $\omega$ by:

$\omega = 2\pi f$

RMS (why it exists)

For power calculations, you often care about an “equivalent DC” value that would deliver the same heating in a resistor. That value is the root mean square (RMS).

For a sinusoid:

$V_{\text{rms}} = \frac{V_p}{\sqrt{2}}$

$I_{\text{rms}} = \frac{I_p}{\sqrt{2}}$

In resistive loads, average power is:

$P = V_{\text{rms}}I_{\text{rms}}$

A classic misconception is using peak values directly in $P = VI$ for AC; you generally use RMS values for power in linear resistive loads.

Reactance and impedance

In AC steady-state, capacitors and inductors behave like frequency-dependent resistances.

Capacitive reactance magnitude:

$X_C = \frac{1}{\omega C}$

Inductive reactance magnitude:

$X_L = \omega L$

Because these effects are phase-shifted, we use impedance $Z$ to represent opposition to AC including phase. In many introductory problems you may only need magnitudes or the idea that capacitors pass high-frequency signals more easily (lower $X_C$ at higher $\omega$) while inductors oppose high-frequency current more strongly (higher $X_L$ at higher $\omega$).

Mechanical analogy (useful intuition)

• Capacitor: like a spring (stores energy, relates force to displacement)
• Inductor: like a mass (stores kinetic energy, relates force to acceleration)
• Resistor: like a damper (dissipates energy)

The analogy isn’t perfect, but it helps you predict qualitative behavior: filters, resonance, and phase shifts.

Worked example: RMS conversion

A mains-like source provides $V_p = 170\,\text{V}$ sinusoidal. Find $V_{\text{rms}}$.

$V_{\text{rms}} = \frac{170\,\text{V}}{\sqrt{2}} \approx 120\,\text{V}$

This is why a “$120\,\text{V}$” supply has about $170\,\text{V}$ peak.

Exam Focus

• Typical question patterns:
  • Convert between peak and RMS values for sinusoidal signals.
  • Compute $X_C$ or $X_L$ at a given frequency and interpret the result.
  • Qualitative questions about whether a capacitor/inductor passes or blocks high-frequency signals.
• Common mistakes:
  • Using peak instead of RMS in power calculations.
  • Forgetting $\omega = 2\pi f$ and plugging $f$ directly into reactance formulas.
  • Assuming capacitors “block AC” (they block DC in steady-state; AC behavior depends on frequency).

Diodes and Power Rectification Essentials

Many electromechanical systems need DC rails derived from AC, and many circuits need protection against reverse polarity or inductive kick. Diodes are the simplest semiconductor device used for one-way conduction and protection.

What a diode does (ideal and real)

An ideal diode conducts perfectly in one direction and blocks perfectly in the other. Real diodes have:

• A forward voltage drop (for silicon diodes, often around $0.7\,\text{V}$ at moderate currents, but it depends on diode type and current).
• Reverse leakage (small reverse current).
• Breakdown voltage (beyond which reverse conduction occurs, sometimes destructively unless designed for it).

Why it matters: in low-voltage systems (e.g., $3.3\,\text{V}$ logic), a diode drop can be a significant fraction of the supply. In power systems, diode losses mean heat.

Common diode uses in mechanical/electromechanical systems

• Reverse polarity protection on DC inputs.
• Flyback (freewheel) diode across a relay coil or DC motor winding to clamp the voltage spike when switching off.
• Rectification: converting AC to DC.

Rectifiers: half-wave and full-wave

• Half-wave rectifier uses one diode, passes only one half-cycle.
• Full-wave rectifier (often a bridge rectifier) uses four diodes to flip the negative half-cycle so both halves contribute to the DC output.

After rectification, a capacitor is often used to smooth the ripple. The capacitor charges near the waveform peaks and discharges between them.

A subtle but common mistake is thinking the rectifier “creates” a constant DC voltage. Without filtering and regulation, rectified output is a pulsating waveform with ripple.

Worked example: bridge rectifier peak output estimate

A transformer provides $12\,\text{V}$ RMS AC. Approximate the peak DC after a bridge rectifier with a smoothing capacitor, ignoring ripple, and assuming $0.7\,\text{V}$ drop per diode.

1) Convert RMS to peak:

$V_p = V_{\text{rms}}\sqrt{2} = 12\,\text{V}\sqrt{2} \approx 17.0\,\text{V}$

2) Bridge rectifier conducts through two diodes at a time:

$V_{\text{DC,peak}} \approx 17.0\,\text{V} - 2\times 0.7\,\text{V} = 15.6\,\text{V}$

In a real design you’d also consider transformer regulation, diode type, load current, and ripple.

Exam Focus

• Typical question patterns:
  • Identify diode orientation and determine whether it is forward- or reverse-biased.
  • Estimate rectified DC peak from an AC RMS input, including diode drops.
  • Explain why flyback diodes are used with coils/motors.
• Common mistakes:
  • Forgetting that a bridge has two diode drops in the conduction path.
  • Assuming a diode drop is always exactly $0.7\,\text{V}$ regardless of current/type.
  • Omitting flyback protection when switching inductive loads.

Transistors as Switches: Relays, MOSFETs, and Motor/Actuator Driving

Mechanical engineers frequently need to control actuators—solenoids, DC motors, valves, pumps, heaters—from low-power control signals. A microcontroller pin cannot directly drive these loads, so you use transistors and driver circuits.

Why a transistor is needed

A control signal (from a PLC output or microcontroller GPIO) might supply only a few milliamps at $3.3\,\text{V}$ or $5\,\text{V}$. But a solenoid might require $1\,\text{A}$ at $24\,\text{V}$. A transistor allows a small input signal to control a larger current.

BJT vs MOSFET (high-level comparison)

A BJT (bipolar junction transistor) is current-controlled: base current controls collector current.

A MOSFET (metal-oxide-semiconductor field-effect transistor) is voltage-controlled: gate voltage controls drain-source conduction, and ideally the gate draws negligible DC current.

In modern switching applications (motor control, power conversion), MOSFETs are often preferred due to lower on-state loss and easier drive for many cases—though gate drive requirements and protection still matter.

Low-side switching (common and intuitive)

A very common pattern is low-side switching, where the load connects to the positive supply and the transistor connects the other end of the load to ground when turned on.

Key design elements:

• Flyback diode across inductive loads.
• Gate/base resistor to limit drive current and reduce ringing.
• Common ground between control electronics and power stage (unless isolation is used).

On-state loss and heating

When a MOSFET is on, it behaves roughly like a small resistance $R_{\text{DS(on)}}$. Conduction loss is approximately:

$P \approx I^2R_{\text{DS(on)}}$

This links directly to thermal design: even “small” resistances matter at high current.

Relays vs transistor switching

A relay is an electromechanical switch. Advantages: galvanic isolation and ability to switch AC or high voltages easily. Disadvantages: slower, wears out mechanically, coil power consumption, contact arcing.

A transistor switch: fast, no mechanical wear, supports PWM control. Disadvantages: requires correct polarity, voltage ratings, and protection; no inherent isolation.

Worked example: MOSFET conduction loss

A DC motor draws $I = 8\,\text{A}$. You select a MOSFET with $R_{\text{DS(on)}} = 20\,\text{m}\Omega$ at your gate drive voltage.

Conduction loss:

$P \approx I^2R = (8\,\text{A})^2 \times 0.02\,\Omega = 64 \times 0.02\,\text{W} = 1.28\,\text{W}$

That $1.28\,\text{W}$ becomes heat in the transistor package—small enough that it might be manageable, but not ignorable.

Exam Focus

• Typical question patterns:
  • Choose between relay and transistor switching for a given load and control requirement.
  • Compute approximate conduction loss from $I$ and $R_{\text{DS(on)}}$.
  • Identify where the flyback diode goes in an inductive load circuit.
• Common mistakes:
  • Placing the flyback diode backwards (it should be reverse-biased during normal operation).
  • Assuming any MOSFET will fully turn on with a logic pin; gate drive voltage affects $R_{\text{DS(on)}}$.
  • Forgetting that the control and power grounds must share a reference in non-isolated low-side switching.

Operational Amplifiers and Signal Conditioning for Sensors

Sensors are where mechanical and electrical worlds meet. A strain gauge, thermocouple, pressure transducer, or accelerometer often produces small signals that must be amplified, filtered, and shifted into the input range of an ADC or control system.

What an op-amp is (model first)

An operational amplifier (op-amp) is a high-gain differential amplifier. The idealized model (used in many analyses) assumes:

• Infinite open-loop gain.
• Infinite input impedance (no input current).
• Zero output impedance.

Real op-amps deviate from this, but the ideal model is powerful because feedback makes circuits predictable.

Negative feedback (the reason op-amps are useful)

With negative feedback, the op-amp adjusts its output to make the input difference small. In many linear circuits you can use the “virtual short” idea: the op-amp drives its output so that $V_- \approx V_+$ (assuming it is not saturated and feedback is present).

A common misunderstanding is believing $V_- = V_+$ is “always true.” It’s only approximately true when the op-amp is operating in its linear region with negative feedback.

Non-inverting and inverting amplifiers

Non-inverting amplifier: input goes to $V_+$, feedback network sets gain.

Ideal gain:

$A_v = 1 + \frac{R_f}{R_g}$

Inverting amplifier: input goes through $R_{\text{in}}$ to $V_-$, $V_+$ at reference (often ground).

Ideal gain:

$A_v = -\frac{R_f}{R_{\text{in}}}$

Buffer (voltage follower)

A voltage follower has gain near 1 but high input impedance and low output impedance. This is the “fix” for loaded voltage dividers and high-impedance sensors.

Filtering and anti-aliasing (why mechanical engineers care)

If you sample a sensor with an ADC, high-frequency noise can alias into your band of interest. A simple RC or active low-pass filter can reduce this. In vibration measurement or motor control, getting filtering wrong can destabilize control loops or hide resonances.

Wheatstone bridge and strain gauges (common instrumentation pattern)

Strain gauges often form part of a Wheatstone bridge, converting small resistance changes into a differential voltage. Because the signal can be millivolts, you often need an instrumentation amplifier (a specialized differential amplifier with high input impedance and high common-mode rejection).

Even if you don’t derive the full bridge equation, you should understand the purpose: reject supply variation and amplify only the small imbalance caused by strain.

Worked example: non-inverting amplifier gain

Design a non-inverting op-amp gain of $11$ using $R_g = 10\,\text{k}\Omega$.

From:

$A_v = 1 + \frac{R_f}{R_g}$

Set $11 = 1 + \frac{R_f}{10\,000\,\Omega}$, so:

$\frac{R_f}{10\,000\,\Omega} = 10$

$R_f = 100\,000\,\Omega = 100\,\text{k}\Omega$

Exam Focus

• Typical question patterns:
  • Compute op-amp gain for inverting/non-inverting configurations.
  • Identify when a buffer is needed to prevent loading.
  • Conceptual questions about negative feedback and saturation.
• Common mistakes:
  • Applying the “virtual short” rule when the op-amp has no negative feedback or is saturated.
  • Ignoring input/output range limits (real op-amps may not swing to the rails).
  • Forgetting that sensor signals may need offsetting to fit within ADC input range.

Digital Electronics Interfaces: Logic Levels, Pull-Ups, and PWM

Modern mechanical systems are mechatronic: sensors feed microcontrollers/PLCs, which compute control outputs, which drive actuators. Understanding digital interfaces prevents the classic failures: floating inputs, noise-triggered signals, and burnt I/O pins.

Logic levels and the idea of noise margin

Digital electronics represent signals as two ranges: “low” and “high.” Exact threshold voltages depend on the logic family and supply voltage. The key concept is noise margin: how much unwanted voltage disturbance a signal can tolerate without flipping interpretation.

Floating inputs and pull-up/pull-down resistors

A microcontroller input pin left unconnected is floating—it can pick up noise and read randomly. A pull-up resistor weakly connects the input to a high voltage; a pull-down resistor weakly connects it to ground. The resistor is sized to dominate leakage/noise without wasting too much power.

This shows up constantly with switches, limit sensors, and open-collector/open-drain outputs.

Debouncing switches

Mechanical switches bounce: when they close, the contact can rapidly make/break for a few milliseconds. Without debouncing, a single press can look like many presses.

Debouncing can be done:

• In hardware: RC filter and a Schmitt-trigger input.
• In software: require a stable state for a minimum time.

PWM (pulse-width modulation) for power control

PWM is a way to control average power by rapidly switching a load on and off.

• Duty cycle $D$ is the fraction of time the signal is “on.”
• For a PWM voltage switching between $0$ and $V$, the average (DC) component is:

$V_{\text{avg}} = DV$

Why mechanical engineers care: PWM is used for DC motor speed control, heater control, solenoid force modulation, and LED dimming. The load’s electrical and mechanical time constants often “average” the switching into a smooth effect.

A common mistake is assuming PWM creates a lower voltage level. Electrically, the instantaneous voltage is still full supply—what changes is average power and effective behavior.

Worked example: PWM average voltage

A motor driver applies $24\,\text{V}$ PWM with duty cycle $D = 0.35$. The average voltage component is:

$V_{\text{avg}} = 0.35 \times 24\,\text{V} = 8.4\,\text{V}$

The motor still sees $24\,\text{V}$ pulses; current ripple depends on motor inductance and switching frequency.

Exam Focus

• Typical question patterns:
  • Explain why floating inputs are unreliable and how pull-ups/pull-downs fix it.
  • Compute PWM average output from duty cycle and supply.
  • Diagnose switch bounce and propose hardware/software debouncing.
• Common mistakes:
  • Leaving digital inputs floating and expecting stable readings.
  • Using too-small pull resistors, wasting power or overloading outputs.
  • Treating PWM as “analog voltage reduction” rather than switched full-scale voltage.

Practical Wiring, Grounding, and Noise in Electromechanical Systems

In real machines, electrical problems are often not about the schematic—they’re about layout, wiring impedance, grounding, and electromagnetic interference (EMI). Mechanical engineers who understand these issues design systems that work reliably on the factory floor, not just on a bench.

Grounds: reference, return path, and safety earth

The word “ground” is overloaded:

• Signal ground: the voltage reference for measurements and logic.
• Power return: the conductor that carries load current back to the source.
• Protective earth: a safety connection to exposed conductive parts.

A common misconception is that ground is a magical zero-voltage sink. In reality, ground conductors have resistance and inductance, so current flowing through them causes voltage differences (ground bounce), which can corrupt sensor readings.

Star grounding vs daisy chaining

A practical guideline is star grounding for mixed-signal systems: route high currents (motors, heaters) separately from sensitive sensor returns, meeting at a common point near the supply return. If you daisy-chain grounds, the motor current can induce voltage drops that appear in the sensor reference, creating measurement error.

Shielding and twisted pairs

• Twisted pair wiring reduces magnetic pickup by making the loop area small.
• Shielded cable reduces capacitive coupling from noisy sources.

For low-level signals (strain gauges, encoders), cable choice and routing matter as much as the sensor.

Inductive loads and EMI

Switching currents rapidly (high $di/dt$) creates voltage spikes across inductance:

$v = L\frac{di}{dt}$

This equation is the reason you add:

• Flyback diodes for DC coils.
• Snubbers (RC networks) for contact arcing.
• Proper decoupling capacitors near IC power pins.

Connectors, wire gauge, and voltage drop

Long cable runs with significant current cause voltage drop and heating. Even if you don’t do detailed ampacity calculations, you should be comfortable with the idea:

$V_{\text{drop}} = IR_{\text{wire}}$

and that $R_{\text{wire}}$ increases with length and decreases with cross-sectional area.

Worked example: voltage drop in a supply cable

A $24\,\text{V}$ actuator draws $3\,\text{A}$. The round-trip cable resistance (out and back combined) is $0.6\,\Omega$.

Voltage drop:

$V_{\text{drop}} = IR = 3\,\text{A}\times 0.6\,\Omega = 1.8\,\text{V}$

Actuator voltage:

$V_{\text{load}} = 24\,\text{V} - 1.8\,\text{V} = 22.2\,\text{V}$

Power wasted in cable:

$P_{\text{cable}} = I^2R = (3\,\text{A})^2 \times 0.6\,\Omega = 5.4\,\text{W}$

That $5.4\,\text{W}$ becomes heat along the cable—an enclosure thermal concern and sometimes a safety concern.

Exam Focus

• Typical question patterns:
  • Diagnose noise problems based on grounding and wiring topology.
  • Compute voltage drop and cable power dissipation from current and wire resistance.
  • Conceptual: explain how shielding/twisting reduce interference.
• Common mistakes:
  • Sharing sensor ground returns with high-current motor returns (creating measurement offset).
  • Adding shielding but grounding it incorrectly (creating ground loops or leaving it ineffective).
  • Ignoring return path geometry—current loops matter for EMI.

Sensors and Actuators in Mechatronic Systems (Electrical View)

“Electrical/electronics” in mechanical engineering is ultimately about interfacing with the physical world: converting physical quantities to signals (sensors) and signals to physical action (actuators). This section ties together circuits, filtering, and drivers in the way you’ll see them in real systems.

Sensor output types you must recognize

1) Resistive sensors: output is resistance change.

• Examples: thermistors, RTDs, strain gauges, potentiometers.
• Common interface: voltage divider or Wheatstone bridge.

2) Voltage-output sensors: provide a voltage proportional to the measured quantity.

• Often need buffering, filtering, and scaling.

3) Current-loop sensors (common in industry): frequently $4\,\text{mA}$ to $20\,\text{mA}$ loops.

• Advantage: less sensitive to voltage drop and noise over long cables.
• Interface: sense resistor to convert current to voltage.

4) Digital sensors: output pulses (encoders) or communication (I2C/SPI/CAN in some systems).

The key skill is to identify the sensor type and choose the correct conditioning method.

ADC basics: resolution and quantization

An analog-to-digital converter (ADC) maps an input voltage range into discrete counts.

For an $N$-bit ADC with reference voltage $V_{\text{ref}}$, the ideal step size (least significant bit, LSB) is approximately:

$\Delta V \approx \frac{V_{\text{ref}}}{2^N}$

This matters because you can’t measure changes smaller than about one LSB without additional techniques (averaging, oversampling, gain).

Actuator electrical models

• DC motor: electrically it looks like resistance and inductance plus a back-emf proportional to speed. Practically, starting (stall) current can be much higher than running current—important for driver sizing.
• Solenoid: mostly inductive; force depends on current and magnetic circuit geometry.
• Heater: mostly resistive; power and thermal time constant dominate behavior.

Bringing it together: a typical control chain

A common mechatronic chain looks like:

1) Sensor produces small/noisy signal.
2) Signal conditioning: bridge excitation, amplifier, filter.
3) ADC samples into controller.
4) Control algorithm computes command.
5) Output stage: PWM, transistor driver, or relay.
6) Protection: fuses, flyback diodes, snubbers, thermal cutoffs.

When troubleshooting, you should test each stage separately: confirm sensor excitation, verify analog scaling at the ADC pin, check PWM duty, confirm driver voltage at the actuator, and validate grounding.

Worked example: ADC step size

A microcontroller ADC is $12$-bit with $V_{\text{ref}} = 3.3\,\text{V}$.

Step size:

$\Delta V \approx \frac{3.3\,\text{V}}{2^{12}} = \frac{3.3\,\text{V}}{4096} \approx 8.06 \times 10^{-4}\,\text{V} = 0.806\,\text{mV}$

So a sensor change of $0.2\,\text{mV}$ will be difficult to resolve without amplification or filtering/averaging.

Exam Focus

• Typical question patterns:
  • Choose an interface circuit for a resistive sensor (divider vs bridge) and explain why.
  • Compute ADC resolution (LSB size) and relate it to sensor output range.
  • Conceptual actuator questions: why is motor start current high, why do solenoids need flyback diodes?
• Common mistakes:
  • Feeding a high-impedance sensor directly into an ADC without buffering, causing measurement error.
  • Forgetting that ADC input range is limited to $0$ to $V_{\text{ref}}$ (overvoltage can damage inputs).
  • Sizing drivers for “typical” current instead of worst-case (stall/inrush) current.