Why Dimensional Analysis? Dimensional Homogeneity
The first rule of any physical equation: every term must have the same dimensions. Primary dimensions, unit-system traps, and why homogeneity catches errors before numbers do.
Suppose you are asked how the drag on a submarine depends on its speed, size, and the water it moves through. You could set up a massive test matrix: dozens of speeds, dozens of hull lengths, fresh water and salt water, every viscosity imaginable. Or you could notice something deeper — the physics does not care whether you measure in metres or feet, seconds or hours. That indifference to units is not a curiosity; it is a <strong>constraint</strong> that collapses the problem from many variables to a handful of dimensionless groups. This lesson builds the foundation: <em>dimensional homogeneity</em>, the requirement that every term in a physical equation carry the same dimensions, and the primary dimensions M, L, and T from which every other unit is built.
The indifference of physics to units
Water flowing past a ship at 10 m/s behaves identically to the same ship moving at 32.8 ft/s through the same water — the physics does not change when you swap rulers. That indifference is the starting point of dimensional analysis: because nature cares only about ratios of lengths, masses, and times, the form of a physical law must be independent of the units used to measure it. A law that only works in SI is not a law at all.
The practical consequence is dimensional homogeneity. If you add two quantities in a physical equation, they must have the same dimensions — you cannot add a length to a mass any more than you can add apples to seconds. This sounds obvious, but it is a surprisingly powerful tool. An equation that fails the homogeneity test is wrong, full stop, before a single number is plugged in. An equation that passes is not automatically right, but it has cleared a basic structural check that eliminates a large class of common errors.
Homogeneity also tells you what must be in an equation. If a result has dimensions of pressure M L⁻¹ T⁻², and the only variables available are density M L⁻³ and velocity L T⁻¹, then pressure must scale as ρ V² times some dimensionless function — because ρ V² is the only combination of those variables that produces the right dimensions. That single insight, expanded systematically, is the Buckingham Pi theorem we develop in the next lesson.
The primary dimensions are mass M, length L, time T, and temperature Θ. Every other quantity is a derived combination: force is M L T⁻², pressure is M L⁻¹ T⁻², energy is M L² T⁻². The SI unit newton is just a name for kg·m·s⁻²; the British unit lbf is a name for slug·ft·s⁻². Working in primary dimensions lets you convert between any unit system without memorising conversion factors, because you are tracking what the quantity is, not what it is called.
Checking equations by dimensions
Dimensional homogeneity is most useful as a filter. Take Bernoulli's equation, which we met in Module 5: each term must have the dimensions of pressure, M L⁻¹ T⁻². The static-pressure term p obviously does. The dynamic term ½ρV² has dimensions (M L⁻³)(L T⁻¹)² = M L⁻¹ T⁻². The hydrostatic term ρgh has dimensions (M L⁻³)(L T⁻²)(L) = M L⁻¹ T⁻². All three match, so the equation is at least dimensionally plausible. If someone handed you p = ρV² + ρg, the second term would fail — ρg is M L⁻² T⁻², not pressure — and you would know instantly that the equation is wrong, without knowing any numbers at all.
The same check catches errors in derived formulas. A student who forgets the square on velocity and writes kinetic energy as ½mv rather than ½mv² produces a term with dimensions M L T⁻¹ — momentum, not energy — and the homogeneity test flags it immediately. Dimensional checking is not a substitute for physical reasoning, but it is an excellent safety net, and it costs nothing.
The most common dimensional mistake in engineering is confusing mass with force in the British system. In SI the kilogram is a mass and the newton is the corresponding force (F = ma gives 1 N = 1 kg·m/s²). In the British system the pound-mass (lbm) is a mass, but the pound-force (lbf) is a force. They are not equal and opposite — they are different physical quantities with different dimensions. To use F = ma consistently in British units you must either convert lbm to slugs (1 slug = 32.174 lbm) or introduce g_c = 32.174 lbm·ft/(lbf·s²). A similar trap exists in older metric literature with the kilogram-force (kgf), which is a force, not a mass. The safest defence is to work in primary dimensions (M, L, T) until the final numerical step, then convert once.
- Dimensions of R: [J/(kg·K)] = [energy]/([mass]·[temperature]) = (M L² T⁻²)/(M·Θ) = L² T⁻² Θ⁻¹.
- Dimensions of T: Θ.
- Product RT: (L² T⁻² Θ⁻¹)(Θ) = L² T⁻².
- Because γ is dimensionless, [γRT] = L² T⁻².
- Square root: [√(γRT)] = (L² T⁻²)^½ = L T⁻¹, which is the dimension of velocity. The equation is dimensionally homogeneous.
- 1 lbf/ft² = 4.448 N / (0.3048 m)² = 4.448 / 0.09290 ≈ 47.88 Pa.
- 85 lbf/ft² = 85 × 47.88 ≈ 4070 Pa.
- (Working in primary dimensions: [force/area] = (M L T⁻²)/L² = M L⁻¹ T⁻², the same as pressure in any system.)
- c = √(1.4 × 287 × 288) = √(115 718) ≈ 340.2 m/s.
- This is the standard sea-level speed of sound (~340 m/s).
Check your understanding
- Dimensional homogeneity requires every term in a physical equation to carry the same primary dimensions (M, L, T, Θ).
- Common fluid quantities in M-L-T: density M L⁻³, velocity L T⁻¹, pressure M L⁻¹ T⁻², viscosity M L⁻¹ T⁻¹.
- Checking equations by dimensions catches errors before any numbers are inserted and predicts how variables must combine.
- Unit-system traps (lbm vs lbf, kgf vs N) are avoided by working in primary dimensions until the final numerical step.
🎓 Go deeper: external courses & trusted references
External sites are listed for reference only. This course is independent and has no affiliation with, or endorsement from, the institutions named.