In short
Every physical quantity can be expressed as a combination of seven base dimensions: mass (M), length (L), time (T), electric current (A), temperature (K), amount of substance (mol), and luminous intensity (cd). Dimensional analysis lets you check whether an equation is physically consistent, derive formulas you have never seen before, and catch algebraic mistakes — all by tracking the dimensions through each step. If the dimensions on the left side of an equation do not match the right, the equation is wrong, guaranteed.
Someone writes on the board: "the speed of a falling ball is v = \sqrt{2gh^3}." You have never solved this problem. You do not know the derivation. But you know instantly that the formula is wrong — because the dimensions do not work out. Speed has dimensions of length per time. The right side has g (acceleration, length per time squared) multiplied by h^3 (length cubed). Inside the square root, that gives length to the fourth power divided by time squared. Take the square root: length squared per time. That is not speed. The formula is wrong, and you caught it without lifting a pencil.
This is the power of dimensional analysis. It is not a calculation technique — it is a consistency check that runs on every equation you will ever write. And sometimes, it does more than check: it can derive formulas from scratch, using nothing but the requirement that the dimensions must balance.
What dimensions actually are
A dimension is the type of physical quantity — stripped of its specific unit. Length is a dimension. Whether you measure it in metres, kilometres, miles, or the wingspan of a Garuda — those are units. The dimension is still "length."
The SI system recognises seven base dimensions. Every measurable quantity in physics is built from some combination of these seven:
| Base dimension | Symbol | SI unit | Example |
|---|---|---|---|
| Mass | M | kilogram (kg) | A cricket ball has mass 160 g |
| Length | L | metre (m) | A cricket pitch is 20.12 m |
| Time | T | second (s) | A fast bowler's delivery takes about 0.5 s |
| Electric current | A | ampere (A) | A phone charger draws about 2 A |
| Temperature | K | kelvin (K) | Water boils at 373 K |
| Amount of substance | mol | mole (mol) | 1 mol of water is about 18 g |
| Luminous intensity | cd | candela (cd) | A Diwali diya emits roughly 1 cd |
For most of mechanics — the first year of physics — you need only three: M, L, and T. Electric current enters with electromagnetism. Temperature enters with thermodynamics. The last two (mol and cd) are specialised and rarely appear in dimensional analysis problems at the JEE level.
The notation is: square brackets around a quantity mean "the dimensions of." So [v] means "the dimensions of velocity," and you write it as [\text{L}\,\text{T}^{-1}] — length divided by time.
Dimensional formulae of common quantities
Every derived quantity has a dimensional formula built from the base dimensions. Here is a table you will use constantly:
| Quantity | Formula | Dimensions | How to read it |
|---|---|---|---|
| Velocity | v = \text{distance}/\text{time} | [\text{L}\,\text{T}^{-1}] | Length per time |
| Acceleration | a = \text{velocity}/\text{time} | [\text{L}\,\text{T}^{-2}] | Length per time squared |
| Force | F = ma | [\text{M}\,\text{L}\,\text{T}^{-2}] | Mass times acceleration |
| Work / Energy | W = F \cdot d | [\text{M}\,\text{L}^2\,\text{T}^{-2}] | Force times distance |
| Power | P = W/t | [\text{M}\,\text{L}^2\,\text{T}^{-3}] | Energy per time |
| Pressure | p = F/A | [\text{M}\,\text{L}^{-1}\,\text{T}^{-2}] | Force per area |
| Momentum | p = mv | [\text{M}\,\text{L}\,\text{T}^{-1}] | Mass times velocity |
| Frequency | f = 1/T | [\text{T}^{-1}] | Inverse time |
| Angular velocity | \omega = \theta/t | [\text{T}^{-1}] | Angle (dimensionless) per time |
| Gravitational constant | G | [\text{M}^{-1}\,\text{L}^3\,\text{T}^{-2}] | From F = Gm_1m_2/r^2 |
Notice the pattern: you never memorise dimensional formulae. You derive them. Force is mass times acceleration, so its dimensions are [\text{M}] \times [\text{L}\,\text{T}^{-2}] = [\text{M}\,\text{L}\,\text{T}^{-2}]. Energy is force times distance, so [\text{M}\,\text{L}\,\text{T}^{-2}] \times [\text{L}] = [\text{M}\,\text{L}^2\,\text{T}^{-2}]. Pressure is force per area, so [\text{M}\,\text{L}\,\text{T}^{-2}] / [\text{L}^2] = [\text{M}\,\text{L}^{-1}\,\text{T}^{-2}]. Each formula follows from the definition. The table above is not something to memorise — it is something you can reconstruct in thirty seconds.
Checking the dimensional consistency of equations
The most immediate use of dimensional analysis: verifying that an equation makes physical sense. The rule is absolute — every term that is added or subtracted in an equation must have the same dimensions. You cannot add a velocity to a force. You cannot subtract an energy from a momentum. And if an equation violates this, it is wrong, no matter how impressive the derivation looked.
Here is the method, applied step by step.
Is v^2 = u^2 + 2as dimensionally consistent?
This is one of the kinematic equations. Check each term.
Step 1. Find the dimensions of v^2.
Why: squaring a quantity squares its dimensions. Velocity is \text{L}\,\text{T}^{-1}, so velocity squared is \text{L}^2\,\text{T}^{-2}.
Step 2. Find the dimensions of u^2.
Why: u is also a velocity, so u^2 has the same dimensions as v^2. Good — the first two terms match.
Step 3. Find the dimensions of 2as.
Why: a is acceleration (\text{L}\,\text{T}^{-2}) and s is displacement (\text{L}). The factor 2 is dimensionless — pure numbers carry no dimensions. The product as gives \text{L}^2\,\text{T}^{-2}, which matches the other terms.
Step 4. Compare.
All three terms have dimensions [\text{L}^2\,\text{T}^{-2}]. The equation is dimensionally consistent.
Why: dimensional consistency does not prove the equation is correct — only that it is not obviously wrong. The equation v^2 = u^2 + 3as would also pass a dimensional check, even though the correct coefficient is 2, not 3. Dimensional analysis catches type errors (adding metres to seconds), not numerical errors (wrong coefficient).
Catching an error: is T = 2\pi\sqrt{g/L} correct?
Someone claims the time period of a simple pendulum is T = 2\pi\sqrt{g/L}.
Step 1. Dimensions of the left side: [T] = [\text{T}] (time).
Step 2. Dimensions of the right side: 2\pi is dimensionless. Check inside the square root:
Why: g has dimensions \text{L}\,\text{T}^{-2}, and L has dimensions \text{L}. Dividing cancels the length, leaving \text{T}^{-2}.
Step 3. Take the square root: \sqrt{[\text{T}^{-2}]} = [\text{T}^{-1}].
Why: the square root halves the exponents. \text{T}^{-2} becomes \text{T}^{-1}, which is inverse time — a frequency, not a time period.
Step 4. The right side has dimensions [\text{T}^{-1}], but the left side has [\text{T}]. These do not match. The formula is dimensionally inconsistent — it must be wrong.
The correct formula is T = 2\pi\sqrt{L/g}. Swapping g and L inside the square root fixes the dimensions: \sqrt{[\text{L}] / [\text{L}\,\text{T}^{-2}]} = \sqrt{[\text{T}^2]} = [\text{T}].
Deriving formulas using dimensional analysis
Dimensional analysis does not just check equations — it can build them. If you know which physical quantities a result depends on, the requirement that dimensions must balance often pins down the formula up to a dimensionless constant.
This is the most powerful application. Here is the classic example.
The time period of a simple pendulum
You have a pendulum: a mass m hanging from a string of length L, swinging under gravity g. What does the time period T depend on?
Physically, T could depend on three things: the string length L, the mass of the bob m, and the gravitational acceleration g. Assume the relationship is a power law:
where k is a dimensionless constant and a, b, c are unknown exponents you need to find.
Why: this is the key assumption — that T is a product of powers of the relevant quantities. This works because dimensions multiply (there are no "dimension sums"). The constant k absorbs any pure number (2\pi, \frac{1}{3}, etc.) that dimensional analysis cannot determine.
Step 1. Write the dimensions of both sides.
Left side: [T] = [\text{T}], which is [\text{M}^0\,\text{L}^0\,\text{T}^1].
Right side:
Why: L has dimension L, m has dimension M, and g has dimensions \text{L}\,\text{T}^{-2}. When you raise g to the power c, every exponent in its dimensional formula gets multiplied by c: \text{L}^c \cdot \text{T}^{-2c}.
Step 2. Equate the exponents of each base dimension.
For M: b = 0
For L: a + c = 0
For T: -2c = 1
Why: the left side has \text{M}^0\,\text{L}^0\,\text{T}^1. The right side has \text{M}^b\,\text{L}^{a+c}\,\text{T}^{-2c}. For the dimensions to match, the exponent of each base dimension must be equal on both sides. This gives three equations in three unknowns.
Step 3. Solve.
From the T equation: -2c = 1, so c = -\frac{1}{2}.
From the L equation: a + c = 0, so a = -c = \frac{1}{2}.
From the M equation: b = 0.
Why: the mass exponent is zero — the time period does not depend on the mass of the bob. This is a real physical fact, not just a mathematical convenience. A heavier pendulum swings at the same rate as a lighter one, as long as the string length and gravity are the same. You derived this purely from dimensions.
Step 4. Write the result.
Why: L^{1/2} \cdot g^{-1/2} = \sqrt{L/g}. The mass has dropped out entirely. Dimensional analysis has determined the functional form. The only thing it cannot tell you is the value of k.
The full derivation from Newton's second law (which you will meet in the Simple Pendulum article) gives k = 2\pi, so:
Dimensional analysis got you the entire structure of the formula. The only information it could not provide was the factor 2\pi — a pure number that carries no dimensions.
Worked examples
Example 1: Checking a formula for the range of a projectile
A student writes the formula for the horizontal range of a projectile launched at speed u and angle \theta as R = u^2 \sin(2\theta) / g. Is this dimensionally consistent?
Step 1. Dimensions of the left side.
R is a distance: [R] = [\text{L}].
Step 2. Dimensions of the right side.
[u^2] = [\text{L}\,\text{T}^{-1}]^2 = [\text{L}^2\,\text{T}^{-2}]
Why: u is a velocity (\text{L}\,\text{T}^{-1}), and squaring doubles all exponents.
[\sin(2\theta)] = dimensionless. Trigonometric functions take dimensionless arguments (angles in radians are dimensionless) and return dimensionless results.
[g] = [\text{L}\,\text{T}^{-2}]
Why: gravitational acceleration has dimensions of acceleration.
Step 3. Combine.
Why: the \text{T}^{-2} cancels between numerator and denominator. One power of L in the denominator cancels one power from the numerator, leaving \text{L}^1. The sine factor is dimensionless and does not affect the result.
Step 4. Compare: both sides are [\text{L}].
Result: The formula R = u^2 \sin(2\theta) / g is dimensionally consistent.
What this shows: The dimensional check confirms the formula is plausible. It cannot confirm the coefficient or the specific trigonometric function — \sin(\theta), \cos(\theta), or \sin(2\theta) all have the same dimensions (none). But if the formula had been R = u \sin(2\theta)/g, the dimensional check would catch it immediately: [\text{L}\,\text{T}^{-1}]/[\text{L}\,\text{T}^{-2}] = [\text{T}], which is time, not length.
Example 2: Deriving the formula for the speed of waves on a string
The speed v of a transverse wave on a stretched string depends on the tension F in the string and the mass per unit length \mu (linear mass density). Derive the formula for v using dimensional analysis.
Step 1. Assume a power-law relationship.
Why: v depends on tension F and linear density \mu. These are the only two relevant physical quantities. The constant k is dimensionless.
Step 2. Write dimensions.
[v] = [\text{L}\,\text{T}^{-1}]
[F] = [\text{M}\,\text{L}\,\text{T}^{-2}]
[\mu] = [\text{M}\,\text{L}^{-1}]
Why: \mu is mass per unit length, so its dimensions are \text{M}/\text{L} = \text{M}\,\text{L}^{-1}.
Step 3. Equate dimensions.
For M: a + b = 0
For L: a - b = 1
For T: -2a = -1
Step 4. Solve.
From T: a = \frac{1}{2}.
From M: b = -a = -\frac{1}{2}.
Check L: a - b = \frac{1}{2} - (-\frac{1}{2}) = 1. Consistent.
Step 5. Write the result.
The exact derivation (from Newton's second law applied to a small element of the string) gives k = 1, so v = \sqrt{F/\mu}.
Result: v = \sqrt{F/\mu}. A tighter string (larger F) carries waves faster. A heavier string (larger \mu) carries waves slower.
What this shows: Dimensional analysis correctly determined the functional form. For this particular problem, the dimensionless constant turns out to be exactly 1, but dimensional analysis could not have predicted that — it could just as easily have been \frac{1}{2} or \pi.
Common confusions
-
"If two quantities have the same dimensions, they are the same thing." No. Work and torque both have dimensions [\text{M}\,\text{L}^2\,\text{T}^{-2}], but they are completely different physical concepts — work is a scalar (\vec{F} \cdot \vec{d}), torque is a vector (\vec{r} \times \vec{F}). Same dimensions, different physics.
-
"Dimensional analysis can find the exact formula." It cannot. Dimensional analysis determines the structure of a formula — which quantities appear and with which exponents — but it cannot determine dimensionless constants like 2\pi, \frac{1}{2}, or \frac{4}{3}. The pendulum formula is T = 2\pi\sqrt{L/g}, not T = \sqrt{L/g}.
-
"Angles have dimensions." They do not. An angle in radians is defined as arc length divided by radius — both lengths — so the dimensions cancel: [\text{L}]/[\text{L}] = \text{dimensionless}. This is why \sin(\theta), \cos(\theta), and e^\theta are all dimensionally valid only when \theta is dimensionless.
-
"You can take the logarithm of a quantity with dimensions." You cannot. The expression \ln(5 \text{ m}) is physically meaningless. Logarithms, exponentials, and trigonometric functions require dimensionless arguments. If you see \ln(x) in a physics equation, x must be a ratio or a pure number. For example, in the decibel formula \beta = 10 \log_{10}(I/I_0), the argument is the ratio I/I_0, which is dimensionless.
-
"If the dimensions match, the equation is correct." Dimensional consistency is necessary but not sufficient. The equation F = 2ma is dimensionally consistent but physically wrong (the correct equation is F = ma). Dimensional analysis catches structural errors — wrong powers, missing quantities — but not numerical errors.
The method of dimensional analysis — a recipe
Here is the systematic procedure, summarised so you can apply it to any problem:
- Identify the dependent variable — the quantity you want to express (e.g., time period T).
- List all relevant independent variables — the quantities T could depend on (e.g., L, m, g).
- Assume a power-law form: T = k \cdot L^a \cdot m^b \cdot g^c.
- Write the dimensions of every quantity.
- Equate the exponents of each base dimension (M, L, T) on both sides.
- Solve the system of linear equations for the unknown exponents.
- Write the result — the formula is determined up to the dimensionless constant k.
This works cleanly when the number of independent variables equals the number of base dimensions involved (typically three for mechanics). When there are more variables than base dimensions, the system is underdetermined — you get relationships between the exponents but not unique values. That is where the Buckingham Pi theorem (below) becomes necessary.
If you are here to use dimensional analysis for checking equations and deriving formulas, you have what you need. What follows is for readers preparing for JEE Advanced or those curious about the formal foundations and limitations.
The Buckingham Pi theorem
When the number of relevant variables exceeds the number of base dimensions, simple dimensional analysis cannot pin down all the exponents. The Buckingham Pi theorem provides a systematic way to handle this.
The theorem states: if a physical relationship involves n variables built from k independent base dimensions, the relationship can be rewritten in terms of n - k independent dimensionless groups, called Pi groups (\Pi_1, \Pi_2, \ldots).
For example, the drag force F_D on a sphere moving through a fluid depends on:
- the sphere's diameter d (dimension L)
- its speed v (dimension LT^{-1})
- the fluid density \rho (dimension ML^{-3})
- the fluid viscosity \eta (dimension ML^{-1}T^{-1})
That is n = 4 variables involving k = 3 base dimensions (M, L, T). The Buckingham Pi theorem tells you there is 4 - 3 = 1 independent dimensionless group. You can verify that:
is dimensionless. The theorem says that \Pi can only depend on other dimensionless groups formed from the variables — and since there is only one such group, \Pi must be a function of that group alone. In this case, that group is the Reynolds number \text{Re} = \rho v d / \eta, so:
The function f cannot be determined from dimensional analysis — it requires experiment or a full fluid-dynamics calculation. But dimensional analysis has reduced a problem with four variables to one with a single dimensionless parameter. That is an enormous simplification.
Limitations of dimensional analysis
Dimensional analysis is powerful, but it has sharp boundaries:
-
Cannot determine dimensionless constants. The factor 2\pi in the pendulum formula, the factor \frac{1}{2} in KE = \frac{1}{2}mv^2, the factor \frac{4}{3} in the volume of a sphere — all invisible to dimensional analysis.
-
Cannot distinguish between functions of the same dimension. \sin(\theta), \cos(\theta), \tan(\theta), e^{\theta}, \ln(1 + \theta), and \theta itself all have the same dimensions (none). Dimensional analysis cannot tell you which one appears in a formula.
-
Requires correct identification of relevant variables. If you omit a variable that matters, or include one that does not, the analysis gives wrong results. For the pendulum, if you forgot g or incorrectly included the bob's colour, the exponents would come out wrong or the system would be inconsistent.
-
Works only for power-law relationships. The assumption y = k \cdot x_1^a \cdot x_2^b \cdot \ldots fails when the true relationship involves sums, differences, or transcendental functions. The equation x(t) = A\sin(\omega t + \phi) for simple harmonic motion cannot be derived by dimensional analysis — you can find the dimensions of A and \omega, but not the sine.
-
Cannot handle relationships with more than one term. The equation v^2 = u^2 + 2as has three terms. Dimensional analysis can check that all three terms have the same dimensions, but it cannot derive the equation from scratch — it does not know there are three terms or how they relate.
Despite these limitations, dimensional analysis remains one of the most powerful tools in a physicist's kit. ISRO engineers use it during the early design phases of rocket engines — when exact equations are not yet available, dimensional analysis constrains what the answer can look like, eliminating entire families of wrong guesses before any detailed calculation begins.
Dimensionless numbers in physics
Certain dimensionless combinations appear so often that they have names. Here are the most important ones:
| Dimensionless number | Definition | Physical meaning |
|---|---|---|
| Reynolds number Re | \rho v L / \eta | Ratio of inertial to viscous forces |
| Mach number Ma | v / v_s | Speed relative to the speed of sound |
| Froude number Fr | v / \sqrt{gL} | Inertial force vs gravitational force |
| Coefficient of restitution e | v_{\text{sep}} / v_{\text{app}} | Energy retained in a collision |
These numbers are the output of the Buckingham Pi analysis. They encode the physics that dimensional analysis cannot resolve into a single formula — the physics that depends on the ratios of effects, not their absolute magnitudes.
Where this leads next
- Units and the SI System — the foundation this article builds on: what SI units are, how they are defined, and why the seven base units exist.
- Errors in Measurement — how to quantify uncertainty, propagate errors, and distinguish systematic from random errors.
- Significant Figures and Rounding — the rules for reporting measured quantities with the right number of digits.
- Estimation and Order of Magnitude — another tool for quick-checking: can you estimate the answer to within a factor of 10 before solving?
- Simple Pendulum — the full derivation of T = 2\pi\sqrt{L/g} from Newton's second law, including the small-angle approximation that dimensional analysis cannot see.