• Foundation
  • About 22 min
  • 5 practice questions

Dimensional analysis: checking equations and deducing relations

Use the principle of homogeneity to test equations, deduce a formula from the quantities it depends on, convert units between systems, and state the limits of the method.

If something looks wrong, please tell us.

By the end of this lesson you can

  • State the principle of homogeneity of dimensions and use it to test an equation.
  • Explain why a dimensionally correct equation may still be wrong.
  • Deduce the form of a relation by the method of dimensions.
  • Convert a quantity from one system of units to another using its dimensions.
  • List the limitations of dimensional analysis.

Dimensions are not just labels. Because only like quantities can be added or equated, they give us a quick test for any equation and, sometimes, a way to guess a formula before we can derive it. This is dimensional analysis.

The principle of homogeneity

Quantities can be added, subtracted or equated only if they have the same dimensions. You cannot add a velocity to a force or subtract a current from a temperature. Hence in a correct equation every term on both sides must have the same dimensions. This is the principle of homogeneity of dimensions. In the same way, units can be cancelled like algebraic symbols, and so can dimensions.

Test x = x₀ + v₀t + ½at², the position of a body moving with uniform acceleration. [x] = [L]; [x₀] = [L]; [v₀t] = [L T⁻¹][T] = [L]; [½at²] = [L T⁻²][T²] = [L]. Every term is a length, so the equation is dimensionally consistent.

The test works in one direction only. If an equation fails the test, it is certainly wrong. If it passes, it is not proved right, because a pure number (such as ½ or 2π) or a dimensionless function could be wrong without changing any dimensions. Also, the argument of any trigonometric, logarithmic or exponential function must be dimensionless: in y = A sin(ωt − kx), ωt and kx must be pure numbers, so ω has dimensions [T⁻¹] and k has dimensions [L⁻¹].

A dimensionally wrong equation must be wrong; a dimensionally correct equation need not be right.

Deducing a relation: the method of dimensions

If we know (from experiment or insight) which quantities a result depends on, we can often find the form of the relation. Write the result as a product of powers of those quantities times a dimensionless constant k, equate the powers of M, L and T on both sides, and solve. The method works when the result depends on at most three independent quantities in mechanics, because there are only three equations (for M, L and T).

Method of dimensions, step by step

  1. List the dependence Assume Q = k aˣ bʸ cᶻ with k a dimensionless constant.
  2. Substitute dimensions Write [Q] and each of [a], [b], [c] in M, L, T form.
  3. Equate powers Powers of M, of L and of T must match on both sides.
  4. Solve for x, y, z Three equations give three unknown powers.
  5. Write the formula Q = k aˣ bʸ cᶻ; the value of k must come from experiment or theory.

Worked example 1: speed of a wave on a stretched string

The speed v of a wave on a stretched string depends on the tension F and the mass per unit length μ. Find the form of the relation. Step 1: v = k Fᵃ μᵇ. Step 2: [v] = [L T⁻¹], [F] = [M L T⁻²], [μ] = [M L⁻¹]. So [L T⁻¹] = [M L T⁻²]ᵃ [M L⁻¹]ᵇ = [Mᵃ⁺ᵇ Lᵃ⁻ᵇ T⁻²ᵃ]. Step 3: equate powers. M: a + b = 0. L: a − b = 1. T: −2a = −1. Step 4: a = ½, so b = −½ (and a − b = ½ + ½ = 1 checks). Answer: v = k √(F/μ). Dimensions cannot give k; experiment and theory show k = 1.

Worked example 2: force needed for circular motion

The force F needed to keep a body of mass m moving at speed v in a circle of radius r is assumed to be F = k mᵃ vᵇ rᶜ. Step 1: [M L T⁻²] = [M]ᵃ [L T⁻¹]ᵇ [L]ᶜ. Step 2: M: a = 1. T: −b = −2, so b = 2. L: b + c = 1, so c = −1. Answer: F = k mv²/r. (You will meet this as the centripetal force, with k = 1.)

Worked example 3: changing systems of units

A quantity has the same physical size in any system, so its numerical value n and unit u satisfy n₁u₁ = n₂u₂. If its dimensional formula is [Mᵃ Lᵇ Tᶜ], then n₂ = n₁ (M₁/M₂)ᵃ (L₁/L₂)ᵇ (T₁/T₂)ᶜ. Convert 1 joule into CGS units (ergs). Energy is [M L² T⁻²], so a = 1, b = 2, c = −2. M₁/M₂ = 1 kg/1 g = 1000; L₁/L₂ = 1 m/1 cm = 100; T₁/T₂ = 1. Then n₂ = 1 × 1000¹ × 100² × 1 = 10³ × 10⁴ = 10⁷. Answer: 1 J = 10⁷ erg.

Limitations of dimensional analysis

What dimensional analysis cannot do

  • Constants
    • Cannot find dimensionless constants such as ½ or 2π
  • Some functions
    • Cannot deduce relations involving sin, cos, log or exponentials
    • Cannot handle a sum of terms, such as ut + ½at², by the product method
  • Look-alike quantities
    • Cannot tell apart quantities with the same dimensions, such as work and torque
  • Too many variables
    • In mechanics, fails if the result depends on more than three quantities

Key takeaways

  • Every term of a correct equation has the same dimensions (homogeneity).
  • Failing the test proves an equation wrong; passing it does not prove it right.
  • The method of dimensions gives the form of a relation but never the dimensionless constant.
  • To convert units, use n₂ = n₁ (M₁/M₂)ᵃ (L₁/L₂)ᵇ (T₁/T₂)ᶜ.

Check your understanding

  1. 1. Which equation is dimensionally INCORRECT? (v, u: speeds; a: acceleration; s: distance; t: time)

    1. v = u + at
    2. v² = u² + 2as
    3. s = ut + ½at²
    4. s = ut + at³
    Show answer

    Answer: D. In s = ut + at³, the term at³ has dimensions [L T⁻²][T³] = [L T], not [L], so it cannot be added to a distance. Each term of the other three equations has matching dimensions: [L T⁻¹] for the first, [L² T⁻²] for the second and [L] for the third.

  2. 2. In y = A sin(ωt), where y and A are displacements and t is time, the dimensions of ω are:

    1. [T⁻¹]
    2. [L T⁻¹]
    3. [T]
    4. dimensionless
    Show answer

    Answer: A. The argument of a sine must be dimensionless, so ωt has no dimensions. Hence [ω] = 1/[T] = [T⁻¹].

  3. 3. The time period T of a simple pendulum is assumed to depend on its length l and on g. Dimensional analysis gives:

    1. T = k l g
    2. T = k √(l/g)
    3. T = k √(g/l)
    4. T = k l²/g
    Show answer

    Answer: B. Let T = k lˣ gʸ. Then [T] = [L]ˣ [L T⁻²]ʸ. T: −2y = 1, so y = −½. L: x + y = 0, so x = ½. Hence T = k √(l/g). (Experiment and theory give k = 2π.)

  4. 4. The value of 1 newton in dynes (CGS unit of force, g cm s⁻²), found from the dimensions [M L T⁻²], is:

    1. 10³
    2. 10⁷
    3. 10⁵
    4. 10²
    Show answer

    Answer: C. n₂ = n₁ (M₁/M₂)¹ (L₁/L₂)¹ (T₁/T₂)⁻² = 1 × (1000 g/1 g) × (100 cm/1 cm) × 1 = 1000 × 100 = 10⁵. So 1 N = 10⁵ dyn.

  5. 5. Assertion (A): The method of dimensions cannot tell whether kinetic energy is ½mv² or ¾mv². Reason (R): Dimensionless constants do not affect the dimensions of an expression.

    1. Both A and R are true, and R is the correct explanation of A.
    2. Both A and R are true, but R is not the correct explanation of A.
    3. A is true, but R is false.
    4. A is false, but R is true.
    Show answer

    Answer: A. Both ½mv² and ¾mv² have dimensions [M L² T⁻²], so dimensional analysis cannot choose between them; only the definition of kinetic energy can. That is because pure numbers have no dimensions, which is exactly what R says. Both are true and R explains A.

Privacy · Terms · Refunds · Delivery