Chapter 2 — Units And Measurement

Fig 1 — Educational Diagram: Introduction to Measurement

What is measurement?
Measurement is the process of comparing an unknown physical quantity with a chosen standard of the same kind to determine its magnitude. The result is expressed as a number followed by a unit (for example, 5 m).

Physical quantities
Physical quantities are properties that can be measured. They are classified as:

• Base (fundamental) quantities: Seven SI base quantities with their units and dimension symbols: mass (kilogram, kg, M), length (metre, m, L), time (second, s, T), electric current (ampere, A, I), thermodynamic temperature (kelvin, K, Θ), amount of substance (mole, mol, N), luminous intensity (candela, cd, J).
• Derived quantities: Formed from base quantities (e.g., speed = length/time with unit m s^-1, force = mass × acceleration with unit N = kg·m·s^-2).

Units and standards
A unit is a standard used to specify measurements. The International System of Units (SI) provides agreed standards so measurements are uniform worldwide. Example conversions: 1 km = 1000 m, 1 m = 100 cm, 1 hour = 3600 s.

Why measurement needs care
Real measurements are affected by limitations of instruments, methods and observers. Important concepts:

• Accuracy: How close a measured value is to the true value.
• Precision: How closely repeated measurements agree with each other (repeatability).
• Resolution: Smallest change an instrument can detect.
• Calibration: Adjusting an instrument to match a standard.

Errors and uncertainties
Measurements always have errors. Error = measured value − true value. We usually report uncertainty (an estimate of error range).

• Systematic errors: Biases that shift measurements consistently (e.g., zero error, calibration error). They affect accuracy and can often be corrected.
• Random errors: Scatter around the mean due to unpredictable fluctuations (e.g., reading fluctuations). They affect precision and are treated statistically.

Significant figures
Significant figures indicate the certainty of a measured number. Rules: non-zero digits are significant, zeros between non-zero digits are significant, leading zeros are not, trailing zeros are significant if there is a decimal point or indicated by scientific notation.

Dimensional analysis
Every physical quantity has a dimension expressed in terms of base dimensions (M, L, T, I, Θ, N, J). Dimensional analysis helps to:

• Check correctness of equations (both sides must have same dimensions).
• Derive possible forms of physical relations up to dimensionless constants.

Propagation of uncertainties (basic rules)
For independent uncertainties Δxi in measured quantities xi:

• For sum/difference: Δf ≈ sqrt(Σ (Δxi)²).
• For product/quotient: fractional uncertainty Δf/f ≈ sqrt(Σ (Δxi/xi)²).
• For power f = xⁿ: Δf/f ≈ |n| (Δx/x).

Practical measurement procedure

1. Choose proper instrument (sensitivity, range, resolution).
2. Calibrate the instrument if needed.
3. Take multiple readings to estimate random error.
4. Report result with uncertainty and correct significant figures.

Summary
Measurement connects the physical world to numbers through agreed units. Understanding units, errors, significant figures and dimensional analysis is essential to perform and interpret measurements correctly in physics.

Fig 2 — Educational Diagram: Systems of Units

What is a unit and a system of units?

A unit is a definite magnitude of a physical quantity adopted by convention to express values of that quantity. A system of units is a consistent set of units used to measure physical quantities, with specified base units from which derived units are obtained.

Historical systems

• FPS (Foot–Pound–Second): used in engineering in some countries.
• CGS (Centimetre–Gram–Second): common in older physics literature.
• MKS (Metre–Kilogram–Second): precursor to SI.

International System of Units (SI)

The SI is the modern globally accepted system. It is a coherent decimal-based system established to provide standardisation.

Seven SI base quantities and units

• Length: metre, symbol m
• Mass: kilogram, symbol kg
• Time: second, symbol s
• Electric current: ampere, symbol A
• Thermodynamic temperature: kelvin, symbol K
• Amount of substance: mole, symbol mol
• Luminous intensity: candela, symbol cd

Derived units

Derived units are obtained from base units by algebraic combinations. Examples: newton N for force, joule J for energy, pascal Pa for pressure, watt W for power. In a coherent system such as SI, derived units are directly formed from base units without additional numerical factors (for example 1 N = 1 kg m s-2).

Coherent system

A coherent system yields derived units that do not require extra numerical factors. SI is coherent: e.g. force F = m a gives unit kg m s-2 which is named newton.

Prefixes

SI uses decimal prefixes to represent multiples and submultiples: kilo (k, 10^3), centi (c, 10^-2), milli (m, 10^-3), micro (µ, 10^-6), nano (n, 10^-9), giga (G, 10^9), etc. These simplify representation of very large or small quantities.

Dimensional formula and dimensional analysis

Each physical quantity can be expressed in terms of base quantities using a dimensional formula. Example: velocity [v] = L T-1, acceleration [a] = L T-2, force [F] = M L T-2. Dimensional analysis helps check the correctness of equations and convert units.

Rules and conventions

• Unit names are written in lowercase (newton), symbols in standard letters (N).
• Symbols are case sensitive (m metre, M mega prefix different).
• Avoid mixing unit systems in one equation without conversion.

Why SI is preferred

It is universal, coherent, decimal-based and maintained by international agreement, which ensures reproducibility, compatibility and ease of conversion.

Fig 3 — Educational Diagram: Dimensions and Dimensional Formula

What are dimensions?
Dimensions of a physical quantity express its dependence on the chosen fundamental physical quantities (base quantities) such as mass, length and time. Dimensions are written using symbols for base quantities: M (mass), L (length), T (time), I (electric current), Θ (thermodynamic temperature), N (amount of substance) and J (luminous intensity).

Dimensional formula
The dimensional formula of a quantity Q is a compact representation showing powers of the base quantities:
[Q] = M^a L^b T^c I^d Θ^e N^f J^g, where a, b, c, ... are integers or fractions called the dimensions (or exponents) of Q.

How to find a dimensional formula
1. Express the quantity in terms of base quantities (mass, length, time, etc.).
2. Replace each derived quantity by its dimensional formula (e.g., velocity by L T⁻¹).
3. Collect powers of M, L, T, ... to get the exponents.

Important rules and points

• All terms added or equated in a physically meaningful equation must have the same dimensions (principle of dimensional homogeneity).
• Dimensions of fundamental constants can be determined by placing them into known equations (e.g., gravitational constant G from F = G m₁m₂/r²).
• Dimensional analysis cannot determine dimensionless numerical constants (like 2, π), nor the form of dimensionless functions (sin, exp). It also cannot distinguish sums of quantities of the same dimensions.
• Useful in checking formulas, deriving scaling laws (e.g., pendulum period), and designing experiments and models (similarity and scaling in wind tunnels).

Example of dimensional derivation (simple)
To check s = ut + ½ at²: [s] = L, [u t] = (L T⁻¹)(T) = L, [a t²] = (L T⁻²)(T²) = L. All terms have same dimension L, so the equation is dimensionally homogeneous.

Using dimensional analysis to get dependences (example: simple pendulum)
Assume period T depends on length l, mass m and gravitational acceleration g: T ∝ l^a m^b g^c. Dimensions: T = L^a M^b (L T⁻²)^c = M^b L^a+c T^−2c. Equate exponents to dimensions of T (M⁰ L⁰ T¹): b = 0, a + c = 0, −2c = 1 → c = −1/2, a = +1/2. So T ∝ sqrt(l/g). (A dimensionless constant 2π is not found by dimensional analysis.)

Limitations
Dimensional analysis cannot provide dimensionless multiplicative constants, cannot distinguish between addition of quantities unless they have identical dimensions, and fails for purely dimensionless relations. It also cannot predict the dependence on dimensionless parameters (e.g., Reynolds number) without additional reasoning.

Fig 4 — Educational Diagram: Principle of Homogeneity of Dimensions

Statement: In any physically meaningful equation, every term that is added or subtracted must have the same dimensional formula. This is called the Principle of Homogeneity of Dimensions.

What it means: If an equation contains several terms (for example, s = ut + (1/2)at²), each term must represent the same physical dimension (here, length [L]). You cannot add quantities of different dimensions (for example, you cannot add a length to a time or a velocity to an acceleration).

Why it is useful: Homogeneity is a quick check on whether an equation might be correct. If the dimensions do not match, the equation is definitely wrong. If they do match, the equation may be correct, but homogeneity alone cannot prove correctness (it cannot give dimensionless constants or detect some functional forms).

How to use it for dimensional analysis: Typical steps to find how one quantity depends on others using homogeneity:

• Assume the quantity Q depends on variables x, y, z as Q = k x^a y^b z^c, where k is dimensionless.
• Write the dimensional formula for each quantity (for example, [Q], [x], [y], [z]).
• Equate the powers of fundamental dimensions (M, L, T) on both sides to get linear equations for a, b, c.
• Solve for the exponents a, b, c and write the relation (up to a dimensionless constant k).

Worked examples (brief):

• Pendulum period: Assume T = k l^a g^b m^c. Dimensions: [T] = T, [l] = L, [g] = LT^-2, [m] = M. Equating dimensions gives M: 0 = c, L: 0 = a + b, T: 1 = -2b. Solving: b = -1/2, a = 1/2, c = 0. So T = k sqrt(l/g). (k = 2π from theory.)
• Mass–spring period: Assume T = k m^a k_s^b (k_s is spring constant with dimensions MT^-2). Doing dimensional matching gives T ∝ sqrt(m/k_s).

Limitations:

• Homogeneity cannot determine dimensionless constants (like 2π, 1/2, e, etc.).
• It cannot determine the form of dimensionless functions such as trigonometric, exponential or logarithmic dependences (arguments of these functions must be dimensionless).
• Different dimensionally-consistent relations may still be physically incorrect; additional physical reasoning or experiment is needed.

Practical tips for students: Always write dimensional formulae for each term before adding/subtracting them. When deriving relations, include only variables that could logically influence the quantity (ignore constant masses if they do not affect the dependence).

Fig 5 — Educational Diagram: Dimensional Analysis: Applications and Limitations

What is Dimensional Analysis?

Dimensional analysis is the method of studying the dimensions (fundamental physical quantities such as mass M, length L and time T) of physical quantities to check the consistency of equations, deduce possible forms of relations between quantities and perform scaling/estimation. The central idea is the principle of dimensional homogeneity: every physically meaningful equation must have the same dimensions on both sides.

Key concepts

• Dimensions and dimensional formula: the dimension of a quantity is expressed in terms of base dimensions (for Class 11 usually M, L, T). Example: velocity v has dimensional formula [L T⁻¹].
• Dimensional homogeneity: if A = B + C (or any relation) then dimensions(A) = dimensions(B) = dimensions(C). This rule applies to each term that is added or equated.
• Buckingham π-theorem (brief): if a physical problem involves n variables and k fundamental dimensions, it can be reduced to (n − k) independent dimensionless parameters (π-groups). This is the formal basis for finding scaling laws.

How to use dimensional analysis to find a relation (method)

1. List all relevant physical variables and write their dimensions.
2. Assume the desired dependent quantity is proportional to a product of powers of the variables: e.g. Q ∝ A^a B^b ...
3. Equate dimensions on both sides to find the exponents (solve linear equations in the exponents).
4. Note: the result gives the functional form up to a dimensionless constant (a pure number).

Typical applications

• Checking correctness of derived formulas or exam answers (dimensional consistency).
• Deriving the form of relationships between quantities when the exact constant is unknown (e.g., period of a simple pendulum).
• Scaling laws and order-of-magnitude estimates (how a quantity changes when system size or parameters scale).
• Reducing the number of variables via dimensionless groups (important in fluid mechanics, heat transfer, similarity experiments).
• Converting units and identifying errors in units.

Limitations — what dimensional analysis cannot do

• It cannot determine dimensionless numerical constants (for example, it cannot tell whether T = 2π√(l/g) or T = √(l/g); the 2π is dimensionless and not found by dimensional analysis).
• It cannot distinguish between quantities with the same dimensions but different physics (e.g., torque and energy have the same dimensions ML²T⁻²).
• It cannot give additive constants or find expressions where terms add (because homogeneity applies to each term separately).
• It cannot handle vector relationships (directional dependence) or sign information — it gives only magnitudes and scaling.
• It fails when important dimensionless parameters are omitted (e.g., Reynolds number in fluid flow) — results can be misleading if you leave out relevant variables that are dimensionless or hidden.
• It cannot determine functional forms involving non-algebraic functions (trigonometric, exponential, logarithmic) unless their arguments are dimensionless.

Practical advice for students

• Always check dimensional consistency after deriving a formula.
• List all potentially relevant variables; forgetting a variable (like viscosity or density) can give an incorrect scaling form.
• Remember that the method gives the dependence up to a multiplicative dimensionless constant — experiments or deeper theory are needed for that constant.

Fig 6 — Educational Diagram: Measurement Errors and Uncertainties

What is a measurement error?
Every measured value differs from the true value. The difference is called an error. Errors arise from limitations of instruments, observer, environment, and procedure.

Types of errors

• Systematic errors: Repeatable biases that shift all measurements in one direction (e.g., zero error in a scale, calibration offset). They affect accuracy. Can often be found and corrected.
• Random (statistical) errors: Unpredictable variations from measurement to measurement (e.g., fluctuations in reading, environment). They affect precision and are treated statistically.

Accuracy vs Precision
Accuracy = closeness to true value. Precision = reproducibility (scatter) of repeated measurements.

Uncertainty
Uncertainty is the quantitative estimate of the doubt in a measurement. We usually express a result as: measured value ± uncertainty (with units). Two common forms of uncertainty:

• Absolute uncertainty Δx: the amount by which the measured value x may differ from the true value (e.g., 5.00 ± 0.02 m).
• Relative uncertainty = Δx/|x| (dimensionless). Often multiplied by 100% to give percentage uncertainty.

Instrumental limits and least count
Least count is the smallest scale division of an instrument and provides a basic estimate of instrumental uncertainty (often taken as ±(1/2) least count for analog instruments).

Significant figures
The number of significant figures reported should reflect the uncertainty. Typically the uncertainty is given to one (or at most two) significant figures and the measured value rounded to the same decimal place.

Propagation of uncertainties
When a quantity z depends on measured quantities x, y, ... the uncertainty in z depends on uncertainties in x, y, ... Two commonly used methods:

• Worst-case (upper-bound) method: For addition/subtraction, absolute uncertainties add: if z = x ± y then Δz = Δx + Δy.
• Relative addition for products/quotients: For z = x * y or z = x / y, relative uncertainties add: (Δz)/|z| = (Δx)/|x| + (Δy)/|y|.
• Power rule: If z = x^n then (Δz)/|z| = |n| * (Δx)/|x|.
• General (quadrature) method for independent random errors: Combine contributions in quadrature using partial derivatives: Δz = sqrt( (∂z/∂x · Δx)^2 + (∂z/∂y · Δy)^2 + ... ). This gives a statistically better estimate when errors are uncorrelated and random.

Statistical treatment for repeated measurements
Make N repeated measurements x_i. The sample mean x̄ = (1/N) Σ x_i. The sample standard deviation (estimate of spread) is σ = sqrt( (1/(N-1)) Σ (x_i - x̄)^2 ). The standard error of the mean (uncertainty in x̄) is σ_x̄ = σ / sqrt(N).

Reporting a final result
Best practice: give value ± uncertainty and the unit, state whether uncertainty is 1 standard deviation, and round consistently. Example: length = (12.34 ± 0.05) m (1 s.d.).

Fig 7 — Educational Diagram: Propagation of Errors

What it is: Propagation of errors (uncertainties) tells how measurement uncertainties in independent variables affect the uncertainty in a quantity calculated from them. For small uncertainties, we can linearize the function and combine contributions from each variable.

Types of errors:

• Systematic errors: biases that shift measurements in one direction (calibration, zero error).
• Random errors: statistical fluctuations around a mean (reduced by repeated measurement).
• Absolute error (Δx): the uncertainty in the same units as x (e.g. 2.00 ± 0.01 m).
• Relative (fractional) error: Δx/x (often expressed as a percent).

Basic rules (for small independent uncertainties):

• Sum/Difference: If y = a ± b ± c..., absolute errors add (worst-case): Δy = Δa + Δb + Δc + ... . For random independent errors one uses root-sum-square (RSS): σ_y = sqrt(σ_a^2 + σ_b^2 + ...).
• Product/Quotient: If y = a·b / c..., relative (fractional) uncertainties add (worst-case): Δy/y ≈ Δa/a + Δb/b + Δc/c + ... . For independent random errors: (σ_y / y) = sqrt((σ_a/a)^2 + (σ_b/b)^2 + ...).
• Power: If y = a^n then Δy/y ≈ |n|·(Δa/a).

General formula (linear approximation):

If y = f(x1, x2, ..., xn) and uncertainties Δxi are small, the first-order (worst-case) estimate is

Δy ≈ |∂f/∂x1|·Δx1 + |∂f/∂x2|·Δx2 + ... + |∂f/∂xn|·Δxn.

For independent random uncertainties (standard deviations σi) use the root-sum-square form

σ_y = sqrt((∂f/∂x1)^2 σ1^2 + (∂f/∂x2)^2 σ2^2 + ... + (∂f/∂xn)^2 σn^2).

If variables are correlated include covariance terms: σ_y^2 = Σ_i (∂f/∂xi)^2 σ_i^2 + 2 Σ_{i

When to use which: Use the additive (worst-case) form for guaranteed bounds (conservative). Use RSS (statistical) when uncertainties are independent and random to estimate typical uncertainty (standard deviation).

Practical notes: (1) Propagation formulas rely on linearization, so they are accurate when relative uncertainties << 1. (2) Keep significant figures consistent: report uncertainty with 1–2 significant digits and round the result accordingly.

Fig 8 — Educational Diagram: Significant Figures and Rounding Off

What are significant figures?
Significant figures (sig figs) are the digits in a number that carry meaningful information about its precision — including all certain digits and the first uncertain digit. They communicate how precisely a quantity is known.

Why they matter
In experiments and calculations we must not imply greater precision than measurements justify. Using sig figs and correct rounding preserves proper uncertainty in results.

Basic rules to count significant figures

• All non‑zero digits are significant (e.g., 123 has 3 sig figs).
• Leading zeros are not significant (e.g., 0.0045 has 2 sig figs).
• Captive (embedded) zeros are significant (e.g., 1002 has 4 sig figs).
• Trailing zeros are significant only if the number contains a decimal point (e.g., 1500 has 2 sig figs; 1500. has 4; 15.00 has 4).
• Exact numbers (counting numbers or defined constants) have infinite sig figs (e.g., 1 dozen = 12 exactly).

Rounding off rules

• To round to a given digit, look at the next digit to the right:
  - If it is 0–4, leave the digit unchanged (round down).
  - If it is 5–9, increase the digit by 1 (round up).
• When rounding a 5 exactly, many scientific practices use "round half to even" (banker’s rounding) to avoid bias: if the digit to be kept is even, leave it; if odd, increase it. In simple lab practice, 5 is often rounded up.

Rules for propagation of significant figures in calculations

• Multiplication and division: The result should have as many sig figs as the operand with the fewest sig figs. Example: 3.142 (4 sf) × 2.1 (2 sf) ≈ 6.6 (2 sf).
• Addition and subtraction: The result should have the same number of decimal places as the value with the least number of decimal places. Example: 12.11 + 0.023 = 12.133 → 12.13 (2 decimal places).

Practical note
Apply rounding only at the final step of a multi‑step calculation. Rounding intermediate results can increase cumulative rounding error.

Relationship to measurement uncertainty
Significant figures reflect the precision of a measurement: the last significant digit is the first uncertain digit. For example, a ruler reading 12.3 cm implies uncertainty about the last digit (±0.1 cm if ruler scale is 0.1 cm).

Fig 9 — Educational Diagram: Measurement Instruments: Vernier Calipers and Screw Gauge

Overview
Vernier calipers and the screw gauge (micrometer) are precision length-measuring instruments used to measure external dimensions, internal dimensions and depths (calipers) and very small thicknesses or diameters (screw gauge). Both work by using a main scale and a finer secondary scale to read fractions of the smallest main-scale division.

Vernier Calipers — Construction & Principle
A vernier caliper consists of a main (fixed) scale and a movable jaw carrying the vernier scale, plus a fixed jaw, inside jaws (for internal diameters) and a depth rod. The vernier works on the principle of comparing two scales: a set number of divisions on the vernier equals a different number on the main scale, producing a small least count (LC) equal to the difference between one main-scale division and one vernier division.

Least Count (Vernier)
LC = value of one main-scale division − value of one vernier-scale division
If N vernier divisions = (N − 1) main-scale divisions (common design), then LC = main scale division / N.
Typical example: 1 main-scale division = 1 mm and 10 vernier divisions = 9 mm → LC = 0.1 mm.

Reading a Vernier Caliper

1. Close the jaws and check zero error (see below).
2. Open jaws to fit the object. Read the main-scale reading just left of the vernier zero (MSR).
3. Find the vernier division that exactly coincides with a main-scale division; its number × LC = vernier reading (VSR).
4. Observed reading = MSR + VSR.
5. Corrected reading = Observed reading − zero error (if any).

Zero Error (Vernier)
When jaws are fully closed, the vernier zero may not coincide with the main-scale zero. If a vernier division k (k > 0) coincides, zero error = k × LC. Convention: if vernier zero is to the right of main zero, zero error is positive; if to the left, it is negative. Corrected measurement = observed reading − zero error.

Example (Vernier)
Main scale reading left of vernier zero = 12 mm. Vernier division 7 coincides. LC = 0.1 mm → Observed = 12 + 7×0.1 = 12.7 mm. If when closed vernier division 3 coincides (zero error = +0.3 mm), corrected = 12.7 − 0.3 = 12.4 mm.

Screw Gauge (Micrometer) — Construction & Principle
A screw gauge has an anvil and a spindle moved by a calibrated screw (with a pitch), a main (sleeve) scale marked in mm (and often half-mm) and a circular (thimble) scale divided into equal parts. One full rotation of the thimble advances the spindle by the pitch (distance moved per revolution). The circular-scale divisions subdivide one pitch.

Least Count (Screw Gauge)
LC = pitch / number of divisions on the circular scale.
Common design: pitch = 0.5 mm and 50 divisions on circular scale → LC = 0.5/50 = 0.01 mm.

Reading a Screw Gauge

1. Close the spindle gently using the ratchet; check zero error.
2. Place the object and rotate the thimble until it lightly touches the object; use the ratchet for uniform pressure.
3. Read the main scale (sleeve) — this gives whole mm and sometimes half-mm.
4. Read the circular (thimble) scale division that aligns with the reference line; multiply by LC to get thimble reading.
5. Observed reading = main-scale reading + (circular division × LC).
6. Corrected reading = Observed reading − zero error.

Zero Error (Screw Gauge)
When spindle and anvil touch, if the reading is non-zero the zero error = (circular division coinciding with reference) × LC. Sign convention: if thimble zero lies ahead of sleeve zero (i.e., positive extra reading) it is positive zero error; otherwise negative. Always subtract zero error from observed value.

Example (Screw Gauge)
Main scale shows 5.5 mm (i.e., 5 mm + 0.5 mm), circular scale shows 21 divisions, LC = 0.01 mm → Observed = 5.5 + 21×0.01 = 5.71 mm. If zero error = +0.02 mm, corrected = 5.71 − 0.02 = 5.69 mm.

Precision & Uncertainty
Least count is the instrument's precision. Typical instrumental uncertainty is often taken as ±(LC/2) for random reading uncertainty and systematic uncertainty must be corrected via zero-error correction. Repeat measurements and take mean; report result with appropriate significant figures.

Practical Tips to Minimize Error

• Always check and correct for zero error.
• Keep the eye perpendicular to the scale to avoid parallax.
• Use the ratchet on a screw gauge for uniform contact pressure; do not overtighten.
• Take multiple readings at different orientations (for wires) and average.
• Handle instrument carefully and keep it clean; temperature affects metal dimensions.

Fig 10 — Educational Diagram: Direct and Indirect Measurement

Direct measurement means obtaining the value of a physical quantity by comparing it directly with a standard using a suitable instrument. Examples: measuring length with a ruler or vernier caliper, mass with a balance, time with a stopwatch. Direct measurements give a reading (plus an uncertainty) without the need for further calculation.

Indirect measurement means finding a desired quantity by calculating it from other quantities that are measured directly. If Q is not measured directly but computed as a function Q = f(x, y, ...), where x, y ... are measured, the measurement of Q is indirect. Examples: computing speed v = s/t (distance/time), finding resistance R = V/I from measured voltage and current, or determining density ρ = m/V from measured mass and volume.

Why use indirect measurement? Many important quantities cannot be measured directly (e.g., gravitational constant, refractive index from angles), or direct measurement is impractical, destructive, or less accurate. Indirect methods allow determination using relationships and laws of physics.

Errors and uncertainties: Every direct measurement has an uncertainty (instrumental, systematic, random). When you compute a quantity indirectly from measured values, those uncertainties combine and affect the uncertainty in the result. There are simple rules for propagating uncertainties for common operations and a general method using partial derivatives (for independent errors).

Classification of errors:

• Systematic errors: biases that shift all measurements similarly (e.g., zero error, calibration error).
• Random errors: scatter in repeated measurements due to unpredictable variations (reduced by averaging).

Repeated measurements and statistics: For direct measurements repeated N times, use the mean x̄ as the best estimate and the standard deviation σ (or standard error σ/√N) to quantify spread and uncertainty.

Fig 11 — Educational Diagram: Order of Magnitude and Estimation

Definition: Order of magnitude of a positive quantity is the power of ten nearest to that quantity. It gives a rough scale (size class) of a number by expressing it as 10^n. Order-of-magnitude estimates are simple, approximate calculations used to judge size, feasibility and compare physical quantities.

How to find it (standard method): express the number in scientific notation N = a × 10^n with 1 ≤ a < 10. Then either

• use logarithms: order ≈ round(log10 N) → 10^{round(log10 N)}, or
• use the coefficient a: if a < sqrt(10) (≈ 3.162) the order is 10^n, if a ≥ sqrt(10) the order is 10^{n+1}.

This rounding is done on a logarithmic (multiplicative) scale, not on a linear scale.

Why it is useful: it simplifies comparisons between very different-sized quantities, helps check plausibility of answers, and is the basis of Fermi or back-of-envelope estimates — fast, approximate solutions to complex questions.

Rules and shortcuts:

• Multiplication/division: multiply numbers in scientific form — add/subtract exponents. Orders add/subtract accordingly (order of AB ≈ order(A)+order(B)).
• Comparing sizes: two numbers are of the same order if their rounded exponents are equal (i.e., within a factor ≈ 3.16).
• Error handling: if a quantity is 10^n, a typical approximate uncertainty of one order corresponds to a factor of 10 up or down; small relative errors do not change the order unless they push a over the sqrt(10) threshold.

Fermi estimation approach: break a hard problem into simple factors you can guess, estimate each factor to ± a factor of a few, multiply them, and convert to scientific notation to get the order. Fermi problems emphasize reasonable assumptions, unit consistency, and robustness of final result to factor uncertainties.

Fig 12 — Educational Diagram: Practical Examples and Numerical Problems

Overview
"Practical Examples and Numerical Problems" covers how to apply the concepts of units, measurement, significant figures, errors and uncertainties, dimensional analysis and graphing to solve laboratory-style and textbook numerical problems in Class 11 Physics.

Key steps to solve numerical problems

• Identify physical quantities, their units and the required result.
• Convert all quantities to SI units before calculation.
• Note the precision and record significant figures of measurements.
• Estimate absolute and fractional (or percentage) uncertainties.
• Use appropriate error-propagation rules when combining measured quantities.
• Check dimensional consistency of the result (dimensional analysis).
• If using experimental data, plot graphs with error bars and extract slope/intercept with uncertainties.

Significant-figure rules (brief)

• Non-zero digits are significant (e.g., 123 has 3 s.f.).
• Zeros between non-zero digits are significant (e.g., 1002 has 4 s.f.).
• Leading zeros are not significant (0.0025 has 2 s.f.).
• Trailing zeros after decimal are significant (2.500 has 4 s.f.).
• In multiplication/division, round final result to the same number of s.f. as the least precise input. In addition/subtraction round to the least precise decimal place.

Errors and uncertainties — common formulas

• Absolute error: Δx (same unit as x).
• Fractional (relative) error: Δx/x.
• Percentage error: (Δx/x)×100%.
• Addition/subtraction (worst-case): If R = A ± B, then ΔR = ΔA + ΔB (upper bound). For random independent errors, probable error: ΔR ≈ sqrt((ΔA)^2 + (ΔB)^2).
• Multiplication/division: If R = A·B or R = A/B, fractional errors add: ΔR/R ≈ ΔA/A + ΔB/B (worst-case). For independent random errors, use quadrature on fractional errors.
• Powers: If R = A^n then ΔR/R ≈ |n| (ΔA/A).

Dimensional analysis
Use dimensions to check formula correctness and to deduce relationships. Example: for a simple pendulum period T ∝ sqrt(l/g) because [T] = sqrt([L]/[L T^{-2}]) = T.

Worked numerical examples

1. Simple unit conversion and s.f.
   Convert 5.60 km to meters and state the value with correct significant figures.
   Solution: 5.60 km = 5.60 × 1000 m = 5600 m. The given value 5.60 has 3 s.f., so write as 5.60×10^3 m or 5.60 km → 5600 m (use scientific notation to show 3 s.f.): 5.60×10^3 m.
2. Propagation of uncertainty — speed from measured distance and time
   Given: distance s = 12.3 ± 0.1 m and time t = 2.00 ± 0.05 s. Find v = s/t and its uncertainty.
   Solution:
   v = 12.3/2.00 = 6.15 m/s.
   Fractional uncertainties: Δs/s = 0.1/12.3 = 0.00813, Δt/t = 0.05/2.00 = 0.0250.
   For division (worst-case): Δv/v ≈ Δs/s + Δt/t = 0.00813 + 0.0250 = 0.03313.
   So Δv ≈ v × 0.03313 = 6.15 × 0.03313 ≈ 0.20 m/s.
   Result: v = 6.15 ± 0.20 m/s.
3. Pendulum: finding g from T and l with uncertainty
   Formula: T = 2π sqrt(l/g) ⇒ g = 4π^2 l / T^2.
   Suppose l = 1.000 ± 0.001 m and measured period T = 2.006 ± 0.002 s. Compute g and its uncertainty (approximate propagation).
   g = 4π^2 × 1.000 / (2.006)^2 ≈ 9.8079 m/s^2.
   Fractional uncertainties: Δl/l = 0.001/1.000 = 0.001; ΔT/T = 0.002/2.006 ≈ 0.001.
   For g = (const) × l × T^{-2}, fractional error: Δg/g ≈ Δl/l + 2(ΔT/T) ≈ 0.001 + 2×0.001 = 0.003.
   So Δg ≈ 0.003 × 9.808 ≈ 0.03 m/s^2.
   Result: g = 9.81 ± 0.03 m/s^2 (rounded appropriately).

Practical tips for experimental numerical problems

• When timing oscillations, measure time for many oscillations (N) and divide: T = t_total / N to reduce random error.
• Always include units with numerical answers and round uncertainties to one significant figure (or two if the leading digit is 1) and report the measured value to the same decimal place.
• Use dimensional analysis to catch algebraic mistakes before finalizing the answer.