Physical Quantities, Measurement and Experimental Methods

Introduction

Engineering physics begins with a disciplined answer to a simple question: How much? Electrical engineering quantities such as voltage, current, resistance, charge, energy, frequency and temperature acquire scientific meaning only when they are defined, measured and reported consistently. Measurement is therefore not merely the reading of an instrument; it is a process that connects a physical phenomenon to a numerical value, a unit, an uncertainty and a traceable method.

This chapter develops the foundations required for laboratory work and engineering analysis. It introduces SI units, dimensions, vectors, measurement systems, calibration, uncertainty, error analysis, significant figures, technical reporting, reproducibility and laboratory safety.

Learning Objectives

After studying this chapter, you should be able to:

• use SI base and derived units correctly;
• test equations using dimensional analysis;
• resolve and combine vectors in Cartesian form;
• distinguish accuracy, precision, error and uncertainty;
• select suitable instruments and calibration procedures;
• propagate uncertainty through sums, products and powers;
• use repeated measurements and statistical analysis appropriately;
• report results with suitable significant figures and units;
• explain traceability, reproducibility and laboratory safety requirements.

1. Physical Quantities and the SI System

A physical quantity consists of a numerical value and a unit. For example, 12 V describes an electric potential difference whose numerical value is 12 when expressed in volts. The number alone is incomplete because changing the unit changes its numerical representation.

The International System of Units (SI) is founded on seven base quantities:

Derived units are formed from base units. Examples include:

SI prefixes provide powers of ten. Common examples are kilo (k, 10³), mega (M, 10⁶), giga (G, 10⁹), milli (m, 10⁻³), micro (μ, 10⁻⁶) and nano (n, 10⁻⁹). Prefix symbols are case-sensitive: 1 mA is 10⁻³ A, whereas 1 MA is 10⁶ A.

A unit symbol is not pluralized and is separated from its numerical value by a space: 5 V, 20 Ω and 3.0 m. The degree Celsius is written °C; temperature differences of one degree Celsius and one kelvin have the same magnitude, although their zero points differ.

2. Dimensions and Dimensional Analysis

Dimensions describe the physical nature of a quantity independently of the particular units used. The fundamental dimensions used in mechanics and electrical engineering include mass [M], length [L], time [T] and current [I]. For example,

and

where v is velocity and R is electrical resistance.

An equation is dimensionally homogeneous if every term added or equated has the same dimensions. Dimensional analysis is useful for checking equations, converting units and anticipating functional relationships. It cannot determine dimensionless numerical constants such as 2 or π, and it cannot distinguish quantities having identical dimensions, such as torque and energy.

Worked Example 1: Checking an Equation

Suppose a proposed expression for the period of a simple pendulum is T = 2π√(L/g), where L is length and g is acceleration due to gravity. Since [L] = L and [g] = L T⁻²,

Thus √(L/g) has dimension T, and 2π is dimensionless. The expression is dimensionally consistent. Dimensional consistency does not by itself prove that the model is physically correct; experimental assumptions, such as small oscillation amplitude, are also required.

Unit Conversion

Unit conversion uses ratios equal to unity. For example,

Units should be carried through every calculation. A result without units is incomplete unless the quantity is explicitly dimensionless.

3. Scalars, Vectors and Coordinate Methods

A scalar has magnitude only; examples include temperature, resistance, mass and energy. A vector has magnitude and direction; examples include displacement, velocity, acceleration, force and electric field.

In a Cartesian coordinate system, a vector A may be written as

where Aₓ, Aᵧ and A_z are components and i, j and k are unit vectors along the x-, y- and z-axes. Its magnitude is

For a two-dimensional vector with magnitude A and angle θ measured counterclockwise from the positive x-axis,

The direction can be recovered using a quadrant-aware inverse tangent, conventionally written as θ = atan2(Aᵧ,Aₓ). Ordinary inverse tangent may give an incorrect quadrant if the signs of the components are not considered.

Vector addition is performed component by component:

The scalar product is

and produces a scalar. The vector product produces a vector perpendicular to both factors and is defined in a right-handed coordinate system by

Vector methods are essential in circuit field theory, electromagnetic force calculations, mechanical loading and three-dimensional instrumentation.

4. Measurement as a System

A measurement system normally contains four functional stages:

1. Sensing: a sensor responds to the measurand, the quantity intended to be measured.
2. Conversion and conditioning: the signal is converted, amplified, filtered or linearized.
3. Transmission and processing: the signal may be digitized, stored or mathematically processed.
4. Indication or recording: the result is displayed or documented with units and uncertainty.

Important instrument characteristics include:

• Range: interval over which the instrument can operate safely and meaningfully.
• Resolution: smallest change that can be distinguished.
• Sensitivity: change in output per unit change in input.
• Linearity: closeness of the response to a straight-line relationship.
• Hysteresis: dependence of output on the direction of input change.
• Response time: time required to respond to a change in input.
• Loading: alteration of the measured system caused by connecting the instrument.

A digital display may show many digits without providing equivalent accuracy. Resolution is a property of indication; accuracy depends on calibration, instrument design and operating conditions.

Calibration and Traceability

Calibration compares an instrument or measurement system with a reference of known value. A calibration may establish correction values, verify performance or determine whether the instrument meets a specification. Calibration does not necessarily mean adjustment.

Metrological traceability is the property by which a measurement result can be related to a reference through a documented, unbroken chain of calibrations, each contributing to measurement uncertainty. A traceability statement should identify the reference, calibration date, method, environmental conditions and uncertainty.

A typical calibration procedure is:

1. define the measurand and acceptable range;
2. select a reference with substantially better performance;
3. control temperature, humidity, supply voltage and other influences;
4. apply known values across the operating range;
5. record indication, reference value and deviations;
6. estimate uncertainty and document corrections;
7. label the instrument with status and next review date.

Calibration intervals should be based on stability, usage, risk and historical performance rather than an arbitrary calendar rule.

5. Error, Accuracy, Precision and Uncertainty

An error is the difference between a measured value and a reference or true value. Because the true value is generally not known exactly, practical work often uses a certified reference or accepted value. Error may be systematic or random.

• Systematic error produces a consistent bias, such as zero offset, incorrect calibration or thermal drift. Repeating the measurement does not remove it.
• Random error varies unpredictably between observations. Repetition and statistical analysis can reduce uncertainty associated with random variation.
• Gross error results from mistakes such as incorrect range selection, transcription or wiring. It must be prevented or identified, not statistically averaged away.

Accuracy describes closeness to the accepted value. Precision describes agreement among repeated values. A result can be precise but inaccurate if a systematic bias is present.

If x is a measured value and x_ref is a reference value, the absolute error is

and the relative error, when x_ref ≠ 0, is

The percentage error is 100e_rel %. The sign is meaningful: positive indicates that the measurement is above the reference.

Uncertainty is a parameter characterizing the dispersion of values reasonably attributable to the measurand. It is not identical to error, because the unknown error cannot generally be determined exactly.

6. Statistical Treatment of Repeated Measurements

For n repeated observations x₁, x₂, …, xₙ, the arithmetic mean is

where x̄ is the sample mean. The sample standard deviation is

The denominator n − 1 provides an unbiased estimate of population variance when the observations are independent samples from a stable population. The standard uncertainty of the mean is

This reduction applies to independent random variation, not to systematic bias.

Worked Example 2: Repeated Voltage Measurements

Five readings are 5.02, 5.00, 5.01, 5.03 and 4.99 V. Their mean is

The deviations from the mean are 0.010, −0.010, 0, 0.020 and −0.020 V. The sum of squared deviations is 0.0010 V², giving

The standard uncertainty of the mean is

A suitable report, subject to instrument and model uncertainties, is approximately 5.010 ± 0.007 V. The uncertainty should not be reported with unjustified precision.

7. Uncertainty Propagation

A measured result often depends on several input quantities. If y = f(x₁, x₂, …, xₙ), and the input uncertainties are independent and small, the combined standard uncertainty is approximated by

where u(xᵢ) is the standard uncertainty of input xᵢ. If inputs are correlated, covariance terms must be included:

The simpler expression is valid only when covariance is negligible.

Useful special cases are:

For addition or subtraction, y = a ± b, with independent uncertainties,

For multiplication or division, y = ab or y = a/b,

For y = axᵖ, where a is a constant and p is a known power,

These relative forms assume nonzero quantities and independent inputs.

Worked Example 3: Resistance from Voltage and Current

Let V = 12.0 ± 0.1 V and I = 2.00 ± 0.02 A. Since R = V/I,

The relative uncertainty is

Therefore u(R) = 6.00 × 0.0120 = 0.072 Ω, and the result is R = 6.00 ± 0.07 Ω, assuming independent standard uncertainties and negligible loading effects.

For expanded uncertainty, a coverage factor k may be used: U = ku_c. A statement such as R = 6.00 ± 0.14 Ω, k = 2 communicates an interval intended to provide approximately 95% coverage when the underlying assumptions are appropriate. The coverage probability must not be implied without stating the method and assumptions.

8. Significant Figures and Technical Reporting

Significant figures communicate the resolution justified by the measurement. Leading zeros are not significant; zeros between nonzero digits are significant; trailing zeros after a decimal point are significant. Thus 0.0042 has two significant figures, while 4.200 has four.

Do not retain calculator digits that exceed the information in the inputs. In intermediate calculations, retain guard digits and round the final result. An uncertainty is commonly reported with one or two significant figures, and the measured value is rounded to the same decimal place.

A technically complete result should include:

• the measurand and method;
• numerical value and SI unit;
• uncertainty and coverage information where relevant;
• environmental and operating conditions;
• instrument identity and calibration status;
• date, operator and sample identification;
• observations, assumptions and deviations from procedure.

Graphs should label both axes with quantity and unit, use an informative scale and distinguish data from fitted models. A linear fit should not be presented as proof of causation. Residuals, range limitations and outliers should be examined before accepting a model.

9. Reproducibility, Repeatability and Experimental Design

Repeatability concerns agreement when the same method, operator, instrument, location and short time interval are used. Reproducibility concerns agreement under changed conditions, such as different laboratories, operators or instruments. A result may be repeatable but not reproducible if a hidden systematic effect is present.

Good experimental design begins by defining the question and measurand. The investigator should identify variables, controls, expected scales, sources of uncertainty and criteria for rejecting invalid data before collecting observations. Randomizing the order of trials can reduce the influence of drift. Replication estimates random variation; it does not correct bias.

Outliers must be investigated rather than deleted merely because they are inconvenient. A reading may be retained, excluded with documented justification, or repeated after identifying a procedural fault. Data integrity requires preservation of raw records, audit trails and transparent correction procedures.

10. Laboratory Safety and Professional Responsibility

Electrical laboratories combine shock, arc-flash, fire, stored-energy, mechanical and chemical hazards. Safety is an engineering requirement, not an administrative formality.

Essential practices include:

• perform a risk assessment before energizing a circuit;
• use appropriately rated probes, leads, fuses and personal protective equipment;
• de-energize, isolate and verify absence of voltage before modifying a circuit;
• discharge capacitors and control inductive stored energy;
• use current-limited supplies where possible;
• keep the work area dry and uncluttered;
• use one-hand practices where required by the local procedure;
• never defeat interlocks or protective earth connections;
• inspect leads and insulation before use;
• know emergency shutdown, first-aid and fire-response procedures.

A multimeter must be connected correctly: an ammeter is placed in series and has low resistance, whereas a voltmeter is placed in parallel and has high resistance. Incorrect connection can create a short circuit or damage both instrument and circuit.

Ethically, engineers must report unfavorable results, preserve raw data, disclose limitations and refuse to certify measurements beyond their competence or calibration scope. Safety decisions should be based on credible risk rather than convenience or schedule pressure.

11. Illustrative Case: Calibrating a Current Measurement System

Background

An undergraduate laboratory uses a data-acquisition system to measure current in a motor-drive experiment. The displayed current is suspected to be biased because the sensor output is nonzero when no current flows.

Problem

The system must be evaluated over 0–5 A without exposing students or equipment to unnecessary risk. A calibrated reference ammeter has a stated standard uncertainty of 0.010 A.

Analysis

The team identifies zero offset, gain error, temperature drift, reference uncertainty and display resolution as contributors. The measurand is defined as the true steady-state current at the sensor terminals under specified temperature and supply conditions. The reference ammeter is selected with a lower uncertainty than the system being tested.

Approach and Solution

The circuit is tested at 0, 1, 2, 3, 4 and 5 A, with three repeated readings at each point. The reference and test readings are recorded simultaneously. At zero current, the test device reads 0.06 A, establishing a measurable offset. A plot of test reading against reference reading shows an approximately linear response. A correction model is fitted only over the validated range, and residuals are inspected. The uncertainty budget combines repeatability, reference uncertainty, resolution and calibration-model uncertainty.

The corrected instrument is labelled with its valid range, correction equation, uncertainty and calibration date. A later check at 2.5 A confirms that the correction remains within the acceptance criterion.

Outcome and Learning Points

The original system was precise but inaccurate because of zero bias. Calibration improved the measurement only after the measurand, reference, conditions and uncertainty were explicitly defined. The case demonstrates that repeated readings cannot remove a systematic offset, traceability requires documentation, and a calibration result is valid only within its stated scope.

12. Contemporary Tools and Best Practices

Modern laboratories use digital oscilloscopes, automated calibration software, programmable supplies and networked data acquisition. These tools improve speed and repeatability but introduce new risks: software configuration errors, incorrect metadata, clock synchronization problems, firmware changes and cybersecurity threats.

Artificial intelligence can assist with anomaly detection, signal classification and uncertainty budgeting, but it cannot replace physical calibration, engineering judgment or traceable raw data. Any automated decision should be validated against known data, monitored for drift and documented sufficiently for independent review.

Best practice is to maintain a measurement plan, use calibrated equipment, record raw data, separate correction from uncertainty, check dimensional consistency, repeat critical observations, protect personnel and report limitations honestly.

Chapter Summary

Physical quantities require both numerical values and units, and the SI system provides a coherent international foundation. Dimensions allow equations and unit conversions to be checked, although dimensional consistency alone cannot establish physical validity. Scalars and vectors require different mathematical treatments, with vector components providing a practical method for engineering calculations.

Measurement is a complete system involving sensing, conditioning, processing and reporting. Calibration establishes performance and traceability, while accuracy, precision, error and uncertainty describe different aspects of measurement quality. Repeated observations support estimates of mean and random variation, but they do not remove systematic bias. Uncertainty propagation connects uncertainties in input quantities to the reported result.

Professional reporting requires appropriate significant figures, units, conditions, methods and uncertainty statements. Reproducibility depends on controlled procedures, transparent records and investigation of outliers. Finally, electrical laboratory work requires systematic risk control, correct instrument connections, respect for stored energy and ethical communication of results.

Key Terms

Physical quantity: A property expressed by a numerical value and a unit.

SI: The International System of Units.

Dimension: The physical nature of a quantity expressed in fundamental dimensions.

Scalar: A quantity described by magnitude only.

Vector: A quantity described by magnitude and direction.

Measurand: The quantity intended to be measured.

Resolution: The smallest change distinguishable by an instrument.

Sensitivity: Change in output per unit change in input.

Calibration: Comparison with a reference to establish measurement performance.

Traceability: Relation of a result to a reference through a documented calibration chain.

Accuracy: Closeness to an accepted reference value.

Precision: Agreement among repeated measurements.

Systematic error: Consistent bias that repetition does not remove.

Random error: Unpredictable variation among observations.

Uncertainty: Parameter describing dispersion reasonably attributed to a measurand.

Standard deviation: A measure of spread in a set of observations.

Repeatability: Agreement under substantially unchanged measurement conditions.

Reproducibility: Agreement under changed measurement conditions.

Significant figures: Digits justified by the precision of a measurement.

Measuring instrument loading: Change in a system caused by connecting the instrument.

Review Questions

Short Answer Questions

1. Why must a numerical measurement normally be reported with a unit?
2. Distinguish accuracy from precision.
3. What is the purpose of calibration?
4. Why does repeating a measurement not remove systematic error?
5. State the difference between repeatability and reproducibility.

Long Answer Questions

1. Explain the SI system and show how voltage and resistance are expressed in base units.
2. Describe a complete calibration procedure for a digital voltage instrument.
3. Explain the statistical treatment of repeated measurements, including the sample standard deviation and standard uncertainty of the mean.
4. Derive the uncertainty expressions for addition and multiplication of independent quantities.
5. Discuss the requirements of a scientifically defensible technical report.

Analytical / Application Questions

1. A sensor output is consistently 0.15 V high over its operating range. Identify the likely error type and propose a correction and validation procedure.
2. A resistance is calculated from voltage and current measurements with specified uncertainties. Formulate an uncertainty budget and explain which assumptions must be checked.
3. Design a safe and traceable experiment to verify a current sensor from 0 to 5 A, including equipment, data collection, calibration, uncertainty and reporting requirements.

Multiple Choice Questions with Answers

Question 1

Which SI base unit is used for electric current?

A. Volt

B. Ampere

C. Coulomb

D. Ohm

Correct Answer: B. Ampere

Question 2

Which expression has the dimensions of velocity?

A. L T⁻¹

B. L T⁻²

C. M L T⁻²

D. M L² T⁻²

Correct Answer: A. L T⁻¹

Question 3

A vector has components Aₓ = 3 and Aᵧ = 4. What is its magnitude?

A. 1

B. 5

C. 7

D. 12

Correct Answer: B. 5

Question 4

Which statement best describes precision?

A. Closeness to an accepted reference value

B. Agreement among repeated measurements

C. The smallest instrument range

D. The calibration interval

Correct Answer: B. Agreement among repeated measurements

Question 5

Which error is not reduced simply by repeating measurements?

A. Random error

B. Sampling variation

C. Systematic error

D. Readout fluctuation

Correct Answer: C. Systematic error

Question 6

For five independent measurements, the standard uncertainty of the mean is calculated as:

A. s × √5

B. s/5

C. s/√5

D. 5s

Correct Answer: C. s/√5

Question 7

If R = V/I, V = 12.0 V and I = 2.00 A, the resistance is:

A. 0.167 Ω

B. 6.00 Ω

C. 10.0 Ω

D. 24.0 Ω

Correct Answer: B. 6.00 Ω

Question 8

Which connection is correct for measuring voltage across a component?

A. A voltmeter in series

B. An ammeter in parallel

C. A voltmeter in parallel

D. An ohmmeter across an energized circuit

Correct Answer: C. A voltmeter in parallel

Question 9

Metrological traceability requires:

A. A highly resolved display only

B. An undocumented comparison with any instrument

C. An unbroken documented calibration chain to a reference

D. Repeating a measurement exactly twice

Correct Answer: C. An unbroken documented calibration chain to a reference

Question 10

For independent quantities y = a + b, the combined standard uncertainty is:

A. u(a) + u(b)

B. √[u²(a) + u²(b)]

C. u(a)u(b)

D. |u(a) − u(b)|

Correct Answer: B. √[u²(a) + u²(b)]