In short

Acceleration is the rate of change of velocity with respect to time. Average acceleration over an interval is a_{\text{avg}} = \Delta v / \Delta t. Instantaneous acceleration is the derivative a = dv/dt = d^2x/dt^2. Acceleration is a vector — it can be positive (velocity increasing in the positive direction) or negative (velocity changing in the negative direction). Uniform acceleration means a is constant, the simplest and most common special case in introductory kinematics.

An autorickshaw is stopped at a red light on MG Road. The signal turns green. The driver floors the throttle and the auto lurches forward — slowly at first, then faster. After 5 seconds, the speedometer reads 30 km/h. After 10 seconds, it reads 50 km/h.

You already know the auto is speeding up. But how quickly is it speeding up? Is the velocity increasing at a steady rate, or is the auto gaining speed faster in the first few seconds than in the last few? To answer this, you need a quantity that measures not the velocity itself, but how fast the velocity is changing. That quantity is acceleration.

Here is the key idea, stated plainly: velocity tells you how fast position changes. Acceleration tells you how fast velocity changes. If velocity is the speedometer reading, acceleration is how quickly that reading is climbing (or falling).

Average acceleration — the big-picture rate

Start with what you can measure directly. The auto's velocity changed from v_i = 0 to v_f = 50 km/h in a time interval of \Delta t = 10 s. The average rate of change of velocity over that interval is:

a_{\text{avg}} = \frac{\Delta v}{\Delta t} = \frac{v_f - v_i}{t_f - t_i}

Why: this is the same structure as the definition of average velocity — v_{\text{avg}} = \Delta x / \Delta t — but one level up. Instead of position change per unit time, you have velocity change per unit time.

Convert 50 km/h to SI: 50 \times \frac{1000}{3600} \approx 13.9 m/s. Then:

a_{\text{avg}} = \frac{13.9 - 0}{10} = 1.39 \text{ m/s}^2

The unit m/s² reads "metres per second per second" — every second, the velocity changes by 1.39 m/s. After 1 second, the auto moves at about 1.4 m/s. After 2 seconds, about 2.8 m/s. The velocity accumulates at a steady rate (assuming uniform acceleration, which you will formalise later in this article).

What average acceleration hides

Average acceleration gives you the overall trend but hides the details. The auto might have accelerated sharply in the first 3 seconds and then gently for the remaining 7. Or it might have accelerated at a perfectly steady rate the whole time. The average is the same in both cases — 1.39 m/s² — but the ride feels very different.

This is exactly the same problem you face with average velocity: a car that drives 100 km in 2 hours has an average speed of 50 km/h, whether it cruised at a constant 50 or sprinted at 100 and then sat in traffic. To capture what is happening at each moment, you need the instantaneous version.

Instantaneous acceleration — the derivative of velocity

Instantaneous acceleration is the acceleration at a single instant. You get it the same way you get instantaneous velocity from average velocity: shrink the time interval to zero and take the limit.

a = \lim_{\Delta t \to 0} \frac{\Delta v}{\Delta t} = \frac{dv}{dt}

Why: as \Delta t shrinks, the average acceleration over that tiny interval approaches the true acceleration at that instant. The limit is the formal definition — the derivative of velocity with respect to time.

Since velocity is itself the derivative of position — v = dx/dt — acceleration is the second derivative of position:

a = \frac{dv}{dt} = \frac{d}{dt}\left(\frac{dx}{dt}\right) = \frac{d^2x}{dt^2}

Why: differentiating position once gives velocity. Differentiating velocity (which is already a first derivative of position) gives the second derivative of position. Acceleration is how the "how fast" itself is changing — a rate of a rate.

This is the hierarchy of motion, clean and complete:

Quantity Symbol Derivative relation
Position x —
Velocity v v = dx/dt
Acceleration a a = dv/dt = d^2x/dt^2

Each row is the derivative of the row above it. And each row is the integral of the row below it — a connection you will use in Uniformly Accelerated Motion to derive the equations of motion.

Positive and negative acceleration — it is not about speeding up or slowing down

This is where most textbooks create a misconception. They say "positive acceleration means speeding up" and "negative acceleration means slowing down." That is wrong. Or rather, it is only half right, and the half that is missing causes real confusion.

Here is the correct statement: the sign of acceleration tells you the direction of the velocity change, not whether the object is speeding up or slowing down.

Consider two scenarios on a straight road:

Scenario A: A car is moving to the right (positive direction) at 20 m/s and speeds up to 30 m/s in 5 seconds.

a = \frac{30 - 20}{5} = +2 \text{ m/s}^2

Positive acceleration, speeding up. So far, so good.

Scenario B: A car is moving to the left (negative direction) at -20 m/s and speeds up to -30 m/s in 5 seconds. It is going faster — the speedometer reads a higher number — but the velocity is becoming more negative.

a = \frac{-30 - (-20)}{5} = \frac{-10}{5} = -2 \text{ m/s}^2

Negative acceleration — but the car is speeding up! The magnitude of the velocity increased from 20 m/s to 30 m/s.

Sign of acceleration vs speeding up or slowing down A 2x2 grid showing four cases: positive velocity with positive acceleration means speeding up; positive velocity with negative acceleration means slowing down; negative velocity with positive acceleration means slowing down; negative velocity with negative acceleration means speeding up. Sign of acceleration a > 0 a < 0 v > 0 v < 0 v and a same direction SPEEDING UP v and a opposite SLOWING DOWN v and a opposite SLOWING DOWN v and a same direction SPEEDING UP
The real rule: if velocity and acceleration point in the same direction (both positive or both negative), the object speeds up. If they point in opposite directions, the object slows down. The sign of acceleration alone does not tell you whether the object is speeding up.

The real rule is simpler and always correct:

Think of the Delhi Metro braking as it approaches Rajiv Chowk station. The train is moving forward (positive velocity) and the brakes apply a backward force (negative acceleration). Velocity and acceleration are in opposite directions, so the train slows down. This is the familiar "deceleration" — but notice that deceleration is just negative acceleration when velocity is positive. It is not a fundamentally different quantity.

The word "deceleration"

You will see the word "deceleration" in textbooks. It means the magnitude of acceleration when the object is slowing down. It is always a positive number. If a car's acceleration is -3 m/s² and the car is moving in the positive direction (so it is slowing down), the deceleration is 3 m/s².

Be careful: "deceleration" is a description of what is happening (the object is getting slower), not a direction. Acceleration is the physics — it has a sign that tells you direction. Deceleration is a convenience word.

Visualising acceleration — spacing between positions

Before looking at graphs, build a physical picture of what acceleration looks like in the real world.

Imagine dropping ink dots on the road from the back of a moving car, one dot every second. If the car moves at constant velocity, the dots are equally spaced — same distance covered each second.

Now imagine the car is accelerating. Each second, it covers more distance than the second before. The dots spread apart. The increasing gaps between dots are the acceleration — you can see it.

If the car is decelerating (slowing down), the dots bunch together. Each second, the car covers less distance. The shrinking gaps tell you the velocity is decreasing.

Animated: three rows of dots — constant velocity, positive acceleration, and negative acceleration Top row: equally spaced dots for constant velocity. Middle row: dots spreading apart for positive acceleration. Bottom row: dots bunching together for negative acceleration (object slowing down).
Watch the trail of ghost dots each body leaves behind. Top (dark): constant velocity — equal spacing. Middle (red): positive acceleration — gaps grow wider as the object speeds up. Bottom (grey): negative acceleration — gaps shrink as the object slows down. Click replay to watch again.

This is the most physical way to "see" acceleration: not as a number, but as the pattern of spacing between positions at equal time intervals. Equal spacing means zero acceleration. Expanding spacing means positive acceleration. Contracting spacing means the object is slowing down.

Acceleration–time graphs

Reading a velocity–time graph for acceleration

The velocity–time (v–t) graph is the natural home for acceleration. On a v–t graph, acceleration is the slope of the curve at any point.

Velocity vs time for the Delhi Metro accelerating from rest. The straight line tells you the acceleration is constant — the slope is $\Delta v / \Delta t = 22.2/10 = 2.22$ m/s² everywhere. Every second, the velocity increases by the same amount.

The acceleration–time graph itself

The acceleration–time (a–t) graph plots acceleration on the vertical axis against time. For uniform acceleration, this is just a horizontal line.

The key insight: the area under the a–t graph between two times equals the change in velocity over that interval.

\Delta v = \int_{t_1}^{t_2} a \, dt

Why: acceleration is dv/dt, so a \, dt = dv. Summing (integrating) both sides over a time interval gives total velocity change on the right and the area under the a–t curve on the left. This is the fundamental theorem of calculus at work.

This means:

Feature of the a–t graph Physical meaning
Height of the curve at time t Instantaneous acceleration at t
Area under the curve from t_1 to t_2 Change in velocity from t_1 to t_2
A horizontal line at a = a_0 Uniform (constant) acceleration
The curve crosses the time axis Acceleration changes sign — the object switches from speeding up to slowing down (or vice versa)
An $a$–$t$ graph where acceleration decreases linearly from $+3$ m/s² to $-3$ m/s². The area above the time axis (from $t = 0$ to $t = 3$ s) is $\frac{1}{2} \times 3 \times 3 = 4.5$ m/s — the velocity gained. The area below the axis (from $t = 3$ to $t = 6$ s) is $4.5$ m/s — the velocity lost. The total velocity change from $t = 0$ to $t = 6$ is zero; the object has returned to its original speed.

Uniform (constant) acceleration — the special case that governs free fall

Most of introductory kinematics focuses on one special case: uniform acceleration, where a is constant. This is not an artificial simplification — it is the reality whenever a constant net force acts on an object of constant mass. The most important example is free fall: near Earth's surface, gravity pulls every object downward with a constant acceleration of g \approx 9.8 m/s², regardless of mass.

When a is constant:

Here is a preview of those equations, derived quickly from the definition of constant acceleration.

Starting from a = dv/dt = \text{constant}:

dv = a \, dt

Why: rearranging the definition to isolate dv. Since a is constant, you can integrate both sides directly.

\int_{v_0}^{v} dv = \int_{0}^{t} a \, dt
v - v_0 = at
\boxed{v = v_0 + at}

Why: integrating a constant a over time t gives at. This is the velocity at any time t: start with v_0 and add a multiplied by how much time has passed.

This single equation — v = v_0 + at — is the foundation of uniformly accelerated motion. It says velocity changes linearly with time when acceleration is constant, which is exactly what "constant acceleration" means.

Free fall as uniform acceleration

Drop a cricket ball from the top of a building. Neglect air resistance. The ball accelerates downward at g = 9.8 m/s². Taking downward as positive:

Every second, the velocity increases by exactly 9.8 m/s. The acceleration is uniform — the same 9.8 m/s² at every instant. A dropped cricket ball and a dropped coin gain speed at exactly the same rate (in the absence of air resistance), because g does not depend on mass.

Worked examples

Example 1: Delhi Metro accelerating from a station

A Delhi Metro train starts from rest at Hauz Khas station and reaches a speed of 80 km/h in 20 seconds, accelerating uniformly. Find the average acceleration in m/s².

Velocity vs time for the Metro train. The straight line from the origin confirms uniform acceleration. The slope of this line — rise over run — is the acceleration.

Step 1. Identify the knowns.

v_i = 0 (starts from rest), v_f = 80 km/h, \Delta t = 20 s.

Why: "starts from rest" means the initial velocity is zero. The final velocity is given in km/h, so you need to convert to m/s before computing acceleration in SI units.

Step 2. Convert 80 km/h to m/s.

v_f = 80 \times \frac{1000}{3600} = 80 \times \frac{5}{18} = \frac{400}{18} \approx 22.2 \text{ m/s}

Why: to convert km/h to m/s, multiply by 1000/3600 = 5/18. This is worth memorising — you will use it constantly in kinematics.

Step 3. Apply the definition of average acceleration.

a_{\text{avg}} = \frac{\Delta v}{\Delta t} = \frac{v_f - v_i}{\Delta t} = \frac{22.2 - 0}{20}
\boxed{a_{\text{avg}} = 1.11 \text{ m/s}^2}

Why: the velocity changed by 22.2 m/s over 20 seconds. Dividing gives 1.11 m/s² — every second, the train's speed increased by about 1.1 m/s, or roughly 4 km/h.

Result: The average acceleration is 1.11 m/s². On the v–t graph above, this is the slope of the straight line from the origin to the point (20, 22.2).

What this shows: Uniform acceleration from rest produces a straight line through the origin on a v–t graph. The slope of that line is the acceleration. A slope of 1.11 m/s² means the train gains about 1 metre per second of speed every second — a comfortable acceleration that does not throw standing passengers off balance.

Example 2: A braking car — when does it stop?

A car's velocity as a function of time is given by v(t) = 12 - 3t^2, where v is in m/s and t is in seconds. The car is braking (not uniformly — the deceleration increases with time). Find: (a) when the car stops, and (b) the acceleration at the moment it stops.

Velocity vs time for the braking car. The velocity starts at 12 m/s and drops to zero at $t = 2$ s. The curve is a downward parabola — the rate of deceleration increases as time goes on. Notice that $v$ goes negative after $t = 2$ s, meaning the car would reverse direction if the brakes released and the equation continued to hold.

(a) When does the car stop?

The car stops when v = 0.

12 - 3t^2 = 0
3t^2 = 12
t^2 = 4
\boxed{t = 2 \text{ s}}

Why: "stops" means velocity equals zero. Solving v(t) = 0 gives t = 2 s (take the positive root since t \geq 0).

(b) Acceleration at t = 2 s.

The acceleration is the derivative of velocity:

a(t) = \frac{dv}{dt} = \frac{d}{dt}(12 - 3t^2) = -6t

Why: differentiating term by term. The derivative of the constant 12 is 0. The derivative of -3t^2 is -6t (bring down the power, reduce the exponent by one).

At t = 2 s:

\boxed{a(2) = -6 \times 2 = -12 \text{ m/s}^2}

Why: substitute t = 2 into the expression for a(t). The negative sign tells you the acceleration is in the direction opposite to the initial motion — the car is decelerating.

Acceleration vs time for the same car. The acceleration becomes increasingly negative — the braking force grows with time. At $t = 2$ s (the moment the car stops), the acceleration is $-12$ m/s². This is strong braking — roughly 1.2 times the acceleration due to gravity.

Result: The car stops at t = 2 s with an acceleration of -12 m/s² at that instant. The magnitude of this deceleration is 12 m/s², which is about 1.2g — close to the maximum braking capability of a car with good tyres on dry asphalt.

What this shows: When acceleration is not constant, you cannot use a_{\text{avg}} = \Delta v / \Delta t to find the acceleration at a specific moment. You need the derivative a = dv/dt. The v–t graph is a curve (not a straight line), and the slope of that curve changes at every point. The acceleration at any instant is the slope of the v–t curve at that instant.

Common confusions

If you came here to understand what acceleration is, compute it in basic problems, and read a–t graphs, you have everything you need. What follows is for readers who want to see how acceleration connects to force, explore non-uniform acceleration, and understand what the a–t graph really tells you about the motion.

Acceleration and Newton's second law

Acceleration is not just a kinematic quantity — it is the bridge between kinematics (describing motion) and dynamics (explaining motion). Newton's second law says:

\vec{F}_{\text{net}} = m\vec{a}

Why: the net force on an object equals its mass times its acceleration. This is the most important equation in classical mechanics. It tells you that acceleration is not something that just happens — it is caused by a net force.

This means that whenever you calculate an acceleration, you can immediately ask: what force is producing it? The Delhi Metro accelerating at 1.11 m/s² in Example 1 — if a loaded Metro train has a mass of about 300,000 kg, the net force must be:

F = ma = 300{,}000 \times 1.11 \approx 333{,}000 \text{ N} \approx 333 \text{ kN}

That is the net force from the electric motors minus all resistive forces (friction, air drag). A third of a meganewton, quietly applied every time the train leaves a station.

Jerk — the derivative of acceleration

If acceleration is the rate of change of velocity, what is the rate of change of acceleration? It has a name: jerk (symbol j).

j = \frac{da}{dt} = \frac{d^2v}{dt^2} = \frac{d^3x}{dt^3}

Why: the chain of derivatives continues. Position \to velocity \to acceleration \to jerk. Each is the derivative of the one before it.

Jerk matters in engineering. When a lift accelerates, you feel the acceleration as a force pressing you into the floor. If the acceleration changes suddenly (high jerk), you feel a sudden lurch — the transition from standing still to being pushed is abrupt. Smooth-ride lifts are designed to minimise jerk: they ramp the acceleration up gradually, hold it steady, then ramp it down.

In Example 2, the acceleration was a(t) = -6t, so the jerk is:

j = \frac{da}{dt} = -6 \text{ m/s}^3

The jerk is constant and negative — the braking force increases at a steady rate. A passenger in that car would feel the braking pressure build smoothly over the 2-second stop.

Reconstructing velocity from an a–t graph

You know that the area under the a–t curve gives the velocity change. This lets you reconstruct the entire velocity history from an a–t graph, if you know the initial velocity.

Suppose a(t) = 4 - 2t (acceleration starts at 4 m/s² and drops linearly to 0 at t = 2 s, then goes negative). If v(0) = 0:

v(t) = v(0) + \int_0^t a \, dt' = 0 + \int_0^t (4 - 2t') \, dt' = 4t - t^2

At t = 2 s: v = 8 - 4 = 4 m/s. At t = 4 s: v = 16 - 16 = 0 m/s — the object stops again. The area under a from 0 to 2 (a triangle of area 4) is the velocity gained. The area from 2 to 4 is -4 (area below the axis), which takes the velocity back to zero.

This is the inverse problem: given the acceleration, find the velocity. Together with x = \int v \, dt, you can reconstruct the entire motion from the acceleration function alone — which is exactly what Newton's second law gives you once you know the forces.

Non-uniform acceleration in real life

Uniform acceleration is the textbook case, but most real accelerations are not constant. A car engine produces different forces at different speeds. Air resistance increases with speed, reducing the net force (and hence the acceleration) as the car goes faster. A rocket's acceleration increases as it burns fuel, because the mass is decreasing while the thrust stays roughly constant.

The mathematical tools remain the same — a = dv/dt and \Delta v = \int a \, dt — but the integrals become more interesting. In Free Fall, you will see the simplest case of truly constant acceleration. In later articles on drag and terminal velocity, you will see what happens when acceleration depends on velocity — a differential equation that changes the character of the motion entirely.

Where this leads next