Dimensional Analysis and Its Applications

Dimensional analysis is the payoff of everything so far: using the dimensions of physical quantities as a machine that checks equations, derives new relations and converts values between unit systems. It is the single most productive topic of this chapter in JEE Main — numerical-value questions on unit conversion and derivation steps appear almost every year — and NEET tests it through Assertion–Reason and consistency checks. This page works all three applications as step-by-step pipelines, then closes with the five limitations that every honest examiner asks about. You will need the dimensional catalogue from the Dimensional Formulae and Equations page.

What is Dimensional Analysis? — Complete Theory

Application A — Checking the consistency of an equation

Principle of homogeneity: every term of a valid equation must carry the same dimensions.

Check 1: v² = u² + 2as. Each term must be [v²] = [L T⁻¹]² = [L² T⁻²]. [u²] = [L² T⁻²] ✔. [as] = [L T⁻²][L] = [L² T⁻²] ✔. All three terms identical → dimensionally consistent.

Check 2: T = 2π√(l/g). [l/g] = [L]/[L T⁻²] = [T²]; square root → [T]. LHS [T] ✔ consistent (2π is dimensionless and invisible to the check).

Catch 3: v = at². RHS = [L T⁻²][T²] = [L] ≠ [L T⁻¹] → condemned. And T = 2π√(g/l): [g/l] = [T⁻²], √ → [T⁻¹] ≠ [T] → condemned. The method detects structural errors instantly.

Application B — Deriving a relation (up to a dimensionless constant)

Derivation 1 — Period of a simple pendulum. Assume T depends on the string length l and gravity g: T = k·lᵃ·gᵇ with k dimensionless.

Step 1. Write dimensions: [T]¹ = [L]ᵃ[L T⁻²]ᵇ = [L]ᵃ⁺ᵇ[T⁻²ᵇ].

Step 2. Match exponents: L: a + b = 0; T: −2b = 1.

Step 3. Solve: b = −1/2, a = +1/2 → T = k√(l/g). Experiment supplies k = 2π. (Note: mass never appears — a small-angle pendulum is mass-independent, which dimensional analysis "knew" for free.)

Derivation 2 — Speed of a transverse wave on a string. Assume v depends on tension F and linear mass density μ = m/l: v = k·Fᵃ·μᵇ.

Step 1. [L T⁻¹] = [M L T⁻²]ᵃ[M L⁻¹]ᵇ = [M]ᵃ⁺ᵇ[L]ᵃ⁻ᵇ[T⁻²ᵃ].

Step 2–3. T: −2a = −1 → a = 1/2; M: a + b = 0 → b = −1/2 → v = k√(F/μ), and the exact solution of the wave equation gives k = 1. ✔

Application C — Converting between unit systems

If a quantity Q has dimensional formula [Mᵃ Lᵇ Tᶜ] and its numerical value is n₁ in system 1 (base units M₁, L₁, T₁) and n₂ in system 2, the quantity itself is invariant: n₁u₁ = n₂u₂. Since u ∝ MᵃLᵇTᶜ, dividing gives the master conversion formula:

n₂ = n₁ (M₁/M₂)ᵃ (L₁/L₂)ᵇ (T₁/T₂)ᶜ

Worked conversion 1: 1 joule → erg. Energy: a = 1, b = 2, c = 0. SI bases: M₁ = 1 kg, L₁ = 1 m; CGS bases: M₂ = 1 g, L₂ = 1 cm. n₂ = 1 × (10³ g/1 g)¹ × (10² cm/1 cm)² × 1 = 10³ × 10⁴ = 10⁷ erg. ✔ (1 J = 10⁷ erg — memorised bank confirmed.)

Worked conversion 2: G = 6.67 × 10⁻¹¹ N m² kg⁻² → CGS. [G] = [M⁻¹ L³ T⁻²]: a = −1, b = 3, c = 0. n₂ = 6.67 × 10⁻¹¹ × (10³)⁻¹ × (10²)³ = 6.67 × 10⁻¹¹ × 10⁻³ × 10⁶ = 6.67 × 10⁻⁸ dyne cm² g⁻². ✔

The five limitations (where the machine goes blind)

Visualising Dimensional Analysis and Its Applications

Hold the two pipelines of this page as one mental diagram: the derivation pipeline (assume → dimension → equate exponents → solve) and the conversion pipeline (identify (a, b, c) → apply base-unit ratios → multiply). Both are exam-scriptable in four lines each. When a question says "show that", draw the pipeline; when it says "express in CGS", draw the map. The five limitations form the exit ramp — quote them by number and example.

Solved Examples on Dimensional Analysis (Step-by-Step)

Solved Example 1 — Value of g in CGS

Q. g = 9.8 m s⁻². Find its value in CGS units.

Step 1. [g] = [L T⁻²]: (a, b, c) = (0, 1, 2).

Step 2. n₂ = 9.8 × (M-ratio)⁰ × (10²)¹ × 1 = 9.8 × 10².

Step 3. g = 980 cm s⁻². ✔ (time identical in both systems → T-ratio 1).

Solved Example 2 — Consistency check with a twist

Q. Is the equation x = A(1 − cos ωt) dimensionally consistent, where x is displacement and ω angular frequency?

Step 1. The argument of cos must be dimensionless: [ωt] = [T⁻¹][T] = 1 ✔.

Step 2. (1 − cos ωt) is a pure number (1 minus a ratio), so [A] must equal [x] = [L].

Step 3. Both sides [L] → consistent. Note this equation passes only because the trig function's argument is dimensionless — limitation #2 means DA could not have derived it.

Practice Questions on Dimensional Analysis (With Solutions)

Practice Q1 (Numerical-value type)

Q. 1 kWh = 3.6 × 10⁶ J = n × 10¹³ erg. Find n.

Solution. 1 J = 10⁷ erg → 3.6 × 10⁶ J = 3.6 × 10⁶ × 10⁷ = 3.6 × 10¹³ erg. So n = 3.6.

Practice Q2 (Single-correct MCQ)

Q. Which relation can be obtained completely (including all constants) by dimensional analysis? (a) T = 2π√(l/g) (b) s = ut + ½at² (c) v = k√(F/μ), k unknown (d) none of these

Solution. (a) and (b) contain dimensionless constants (2π, ½) that DA cannot produce; (c) is only obtained up to k. Hence none — the honest answer NTA rewards.

Practice Q3 (Assertion–Reason)

Q. Assertion (A): Dimensional analysis cannot distinguish between work and torque. Reason (R): Both have the dimensional formula [M L² T⁻²].

Solution. Both statements are true, and the identical dimensions are precisely why the method — which sees only dimensions — cannot separate them. R explains A.

Key Formulas & Takeaways

FAQs on Dimensional Analysis and Its Applications

What are the three uses of dimensional analysis?

Checking the dimensional consistency of an equation, deriving a relation among quantities (up to a dimensionless constant), and converting a value from one system of units to another using n₂ = n₁(M₁/M₂)ᵃ(L₁/L₂)ᵇ(T₁/T₂)ᶜ.

Why can dimensional analysis not give the constant 2π?

Pure numbers such as 2, π and 1/2 carry no dimensions, so the method is blind to them; the constant must come from experiment or a full derivation.

Can we derive v = u + at by dimensional analysis?

No. The method can confirm each term has dimension [L T⁻¹] but cannot handle the additive structure or fix the coefficient 2 in 'at' — derivation needs the kinematic definitions.

What fails when a quantity depends on more than three variables?

In mechanics only M, L and T are available, giving at most three independent exponent equations; with four or more unknown powers the system cannot be solved uniquely.