Checking dimensional correctness

An equation in physics is a claim, and claims can be tested before you ever plug in a number. Dimensional analysis gives the cheapest test there is: write out the dimensional formula of every quantity in the equation, simplify, and see whether the two sides actually match. If they don't, the equation is wrong — full stop, no further work needed. If they do match, the equation has passed a necessary check, though not a complete one: constants like 12\tfrac12 or 2π2\pi are invisible to this test, and two different physical quantities can share a dimensional formula. This article works through the check on three familiar equations — a pendulum's time period, kinetic energy, and uniformly accelerated motion — so that the method becomes a habit you reach for whenever you're unsure a remembered formula is right.

The core insight

The principle of homogeneity of dimensions says a physical equation can only be correct if every additive term on both sides reduces to the same dimensional formula, which gives a quick pass/fail test for catching algebraic errors — though it cannot verify dimensionless constants or distinguish quantities that happen to share a dimensional formula.

The core idea

Every physical quantity has a dimensional formula built from the seven base dimensions MM, LL, TT, and so on. The principle of homogeneity of dimensions states that in a correct equation, every term that is added or subtracted anywhere — not just the two overall sides — must carry the identical dimensional formula. You cannot add a length to a time any more than you can add three apples to five oranges and call the answer "eight." This gives a three-step recipe for checking whether a remembered or derived equation could possibly be right:

First, write the dimensional formula of every physical quantity that appears in the equation. Second, simplify each side (or each additive term) using the ordinary rules of exponents — dimensions multiply and divide just like algebraic symbols. Third, compare. If

[LHS]=[RHS][\text{LHS}] = [\text{RHS}]

where: LHS and RHS are the dimensional formulae obtained by simplifying the left- and right-hand sides — then the equation is dimensionally consistent: it may be correct, though the check cannot confirm the value of any dimensionless numerical constant. If the two sides do not reduce to the same dimensional formula, the equation is definitely wrong and can be discarded without further checking. The test has two well-known blind spots. It says nothing about pure numbers such as 12\tfrac12 or 2π2\pi, since numbers are dimensionless and vanish from the comparison. And it cannot tell apart two different quantities that happen to share a dimensional formula — energy and torque are both [ML2T−2][ML^2T^{-2}], so a dimensionally correct equation might still have swapped one for the other. The check also assumes the equation contains no trigonometric, exponential, or logarithmic function of a dimensional quantity, since the argument of such a function must itself be dimensionless before the check can even begin.

Figure
Flow diagram of the three step check write dimensions then simplify then compare with a tick for matching sides and a cross for mismatched sides
Checking dimensional correctness — the three step test Step 1 write dimensions of each quantity Step 2 simplify each side using exponent rules Step 3 compare LHS and RHS of every added term match dimensionally consistent may be correct mismatch definitely wrong A match cannot confirm a dimensionless constant such as one half or two pi
Derivation ATesting the simple pendulum's time period

The goal of this derivation is to apply the three-step homogeneity check to the standard formula for the time period of a simple pendulum, T=2πL/gT = 2\pi\sqrt{L/g}, confirming that the powers of length and gravitational acceleration under the square root combine to give exactly a time, so that the equation passes the dimensional-consistency test even though the check cannot vouch for the constant 2π2\pi itself.

Step 1. Write the dimensional formula of each quantity appearing in the equation: the time period is [T]=[T][T] = [T], the pendulum's length is [L]=[L][L] = [L], and the acceleration due to gravity is [g]=[LT−2][g] = [LT^{-2}]. The factor 2π2\pi is a pure number and carries no dimensions at all, so it is dropped from the check.

Step 2. Simplify the right-hand side. Substituting the dimensional formulae under the square root,

[Lg]=[L][LT−2]=[T2]=[T]\left[\sqrt{\frac{L}{g}}\right] = \sqrt{\frac{[L]}{[LT^{-2}]}} = \sqrt{[T^{2}]} = [T]

The LL in the numerator and denominator cancels completely, leaving only the T−2T^{-2} inside the root, and taking the square root of [T2][T^2] gives [T][T].

Step 3. Compare the two sides: LHS =[T]= [T] and RHS =[T]= [T]. Since they match, T=2πL/gT = 2\pi\sqrt{L/g} is dimensionally consistent. This confirms the form of the equation — that a time period should scale with the square root of length over acceleration — but it says nothing about whether 2π2\pi is the right constant; that number only falls out of the full derivation from simple harmonic motion, done elsewhere. Result: dimensions match, [T]=[T][T] = [T], so the pendulum formula passes the check.

Worked example

Problem: A student remembers two possible formulae for kinetic energy, K=mv2K = mv^2 (forgetting the factor 12\tfrac12) and K=mvK = mv (an error in the power of vv). Use dimensional analysis to test both against the correct dimensional formula of energy, [ML2T−2][ML^2T^{-2}].

Step 1. Test K=mv2K = mv^2. Since [m]=[M][m] = [M] and [v]=[LT−1][v] = [LT^{-1}],

[m][v]2=[M][LT−1]2=[ML2T−2][m][v]^2 = [M][LT^{-1}]^2 = [ML^2T^{-2}]

This matches the dimensional formula of energy exactly, so K=mv2K = mv^2 is dimensionally consistent. Dimensional analysis cannot catch the missing 12\tfrac12, because a pure number leaves no dimensional trace — the check is blind to it by design.

Step 2. Test K=mvK = mv. Combining the same base dimensions,

[m][v]=[M][LT−1]=[MLT−1][m][v] = [M][LT^{-1}] = [MLT^{-1}]

This is the dimensional formula of momentum, not energy, and it does not match [ML2T−2][ML^2T^{-2}].

Step 3. Conclude. Since K=mvK = mv fails the comparison, it is dimensionally incorrect and must be wrong — no further checking needed. K=mv2K = mv^2 passes the check (though the true formula also needs the factor 12\tfrac12), while K=mvK = mv is dimensionally inconsistent and definitely wrong.

Common traps

  • Assuming a dimensionally correct equation must also be numerically or physically correct — dimensional consistency is necessary, but never sufficient.
  • Checking only the overall two sides of an equation and skipping the individual additive terms, when every term added or subtracted (as in s=ut+12at2s = ut + \tfrac12at^2) must independently match the dimensions of the other side.
  • Forgetting that arguments of trigonometric, logarithmic, or exponential functions must be dimensionless before the check even applies — an expression like a sine of a length has no meaning.
  • Making an algebra slip while simplifying exponents or square roots of dimensional formulae, producing a false "mismatch" or a false "match" that isn't really there.

Ask the tutor: Want to test one of your own remembered formulae for dimensional consistency, or work through another example where the check catches a hidden error?

If an equation passes the dimensional check, can I be confident it is the correct formula?

Not fully. Passing the check tells you the equation is dimensionally consistent, meaning it is not automatically ruled out — but it cannot confirm the value of any dimensionless constant like 12\tfrac12 or 2π2\pi, and it cannot tell apart two different physical quantities that happen to share the same dimensional formula, such as energy and torque. Treat a pass as "possible, not disproven," not as a guarantee. The full derivation, not dimensional analysis alone, is what fixes the exact formula.

Why does the constant 2 pi in the pendulum formula not show up anywhere in the dimensional check?

Because 2π2\pi is a pure number with no physical dimension attached to it — it has no length, mass, or time associated with it, so its dimensional formula is simply 11, the same as any other dimensionless number. When you write out dimensions, dimensionless constants contribute nothing and effectively disappear from the comparison. This is exactly why the check can never catch a missing or wrong numerical factor: the test only ever compares the dimensional parts of an equation, never its pure numbers.

How is checking dimensional correctness different from deriving a dimensional formula?

They use the same dimensional formulae but run in opposite directions. Deriving a dimensional formula starts from a known relationship between quantities and works out what dimensions an unknown quantity must have. Checking correctness starts from a complete, already-written equation and simply verifies that both sides reduce to the same dimensions using rules you already know. One builds a new dimensional formula from scratch; the other tests an existing equation for a specific kind of error.

Before this

Where this leads