Scalar (dot) product of vectors

Two forces can point in wildly different directions, yet physics keeps asking a single number out of them: how much do they act together. A tow rope pulls a sledge forward and slightly upward; only the forward-leaning part of that pull actually drags the sledge across the snow. A satellite feels gravity pulling straight down toward Earth while it coasts sideways along its orbit — and because those two directions are perpendicular, gravity does no work on it at all, orbit after orbit. Both stories are answered by the same piece of vector algebra: the scalar (dot) product, a rule that takes two vectors and, using nothing but their sizes and the angle between them, hands back one plain number. That number turns out to be exactly the currency of Work, Energy and Power. Work done by a constant force is defined as $W = \vec F\cdot\vec d$ ; instantaneous power is defined as $P = \vec F\cdot\vec v$ . Neither definition is really new physics once you know what a dot product is — they are the dot product, applied to force-and-displacement and force-and-velocity respectively. Get comfortable multiplying two vectors this way — from magnitudes and an angle, or from components — and half of the sign-reading and angle-hunting in energy problems becomes mechanical rather than mysterious. This article rebuilds the dot product from its definition, shows exactly why it must produce a scalar and not a vector, derives the component formula that makes it fast to compute, and then puts it to work on the same two numerical examples used earlier in the chapter — extended a little further to squeeze out every angle and sign the CBSE board loves to test.

Key takeaways

Definition. For vectors $\vec A$ and $\vec B$ with angle $\theta$ between them (placed tail to tail), $\vec A\cdot\vec B = AB\cos\theta$ — a pure number, never a vector.
Component formula. In three dimensions, $\vec A\cdot\vec B = A_xB_x + A_yB_y + A_zB_z$ ; this follows from $\hat i\cdot\hat i=\hat j\cdot\hat j=\hat k\cdot\hat k=1$ and all mixed unit-vector products being zero.
Commutative, and self-squaring. $\vec A\cdot\vec B=\vec B\cdot\vec A$ ; and $\vec A\cdot\vec A=A^2$ , a shortcut for a vector's squared magnitude.
Sign carries the physics. $\theta<90^\circ$ gives a positive dot product, $\theta=90^\circ$ gives exactly zero, and $\theta>90^\circ$ gives a negative dot product — the algebraic test for acute, perpendicular, and obtuse.
Work and power are dot products. $W=\vec F\cdot\vec d=Fd\cos\theta$ and $P=\vec F\cdot\vec v=Fv\cos\theta$ ; a force perpendicular to the motion, like the normal force under gravity-free horizontal sliding or the centripetal pull on a satellite, does zero work no matter how large it is.
Angle backed out from a known dot product. Rearranging gives $\cos\theta=\dfrac{\vec A\cdot\vec B}{AB}$ , the standard way to find the angle between two force or velocity vectors given in component form.
Key idea: the scalar product turns two vectors and the angle between them into one signed number, $\vec A\cdot\vec B=AB\cos\theta=A_xB_x+A_yB_y+A_zB_z$ , and this is precisely the operation the syllabus uses to define work and power.

The core insight

The scalar product $\vec A\cdot\vec B=AB\cos\theta=A_xB_x+A_yB_y+A_zB_z$ compresses two vectors into one signed number whose sign tells you acute, perpendicular, or obtuse — and this exact operation is what defines work, $W=\vec F\cdot\vec d$ , and power, $P=\vec F\cdot\vec v$ .

Where it fits in the syllabus

The scalar product is formally introduced in the Motion in a Plane unit, alongside vector addition, resolution into rectangular components, and its counterpart the vector (cross) product. It resurfaces here in Work, Energy and Power as the operation hiding inside two of the chapter's opening definitions: work by a constant force, $W=\vec F\cdot\vec d=Fd\cos\theta$ , and instantaneous power, $P=\vec F\cdot\vec v=Fv\cos\theta$ . The CBSE Class 11 board expects you to define the dot product, compute it both from magnitudes-and-angle and from rectangular components, use it to extract the angle between two given vectors, and apply it fluently to work and power problems — including reading the sign of the result as positive, negative, or zero work.

The physical picture

Picture two vectors $\vec A$ and $\vec B$ drawn from a common starting point, with an angle $\theta$ between them. The scalar product asks: how much of $\vec B$ lies along the direction of $\vec A$ , and then multiplies that projected length by the size of $\vec A$ .

\vec A\cdot\vec B = AB\cos\theta

where $A=|\vec A|$ and $B=|\vec B|$ are the magnitudes of the two vectors and $\theta$ is the angle between them measured with both vectors tail to tail, $0^\circ\le\theta\le180^\circ$ . Because $B\cos\theta$ is exactly the length of the projection of $\vec B$ onto the line of $\vec A$ , the dot product is nothing more than base times projected height — a plain number, carrying whatever units the two vectors carry between them, with no leftover direction attached. That is precisely why it is called the scalar product: however you rotate the picture in space, the number $AB\cos\theta$ does not change, because both $A$ , $B$ , and the angle between them are rotation-independent.

Figure

Two vectors from a common point with the projection of one vector onto the other shown as a dashed drop line, illustrating that the dot product is the base length times the projected length

Two properties fall out immediately. First, the dot product is commutative: $\vec A\cdot\vec B=\vec B\cdot\vec A$ , since swapping which vector you call $\vec A$ and which you call $\vec B$ does not change either magnitude or the angle between them. Second, dotting a vector with itself gives $\vec A\cdot\vec A=A\cdot A\cdot\cos0^\circ=A^2$ , an instant way to recover a vector's squared magnitude — this is the trick behind writing $v^2=\vec v\cdot\vec v$ in kinetic-energy work. For everyday calculation, though, the magnitude-and-angle form is rarely the fastest route, because the angle between two vectors given in component form is not obvious by eye. The component formula solves that. If $\vec A=A_x\hat i+A_y\hat j+A_z\hat k$ and $\vec B=B_x\hat i+B_y\hat j+B_z\hat k$ , then

\vec A\cdot\vec B = A_xB_x + A_yB_y + A_zB_z

where $A_x,A_y,A_z$ and $B_x,B_y,B_z$ are the rectangular components of the two vectors along the mutually perpendicular unit vectors $\hat i,\hat j,\hat k$ . This formula follows directly from the dot products of the unit vectors with each other: $\hat i\cdot\hat i=\hat j\cdot\hat j=\hat k\cdot\hat k=1$ (each unit vector dotted with itself, angle $0^\circ$ ), and $\hat i\cdot\hat j=\hat j\cdot\hat k=\hat k\cdot\hat i=0$ (any two distinct unit vectors are perpendicular, angle $90^\circ$ ). Multiplying out $(A_x\hat i+A_y\hat j+A_z\hat k)\cdot(B_x\hat i+B_y\hat j+B_z\hat k)$ term by term and dropping every mixed term that involves $\hat i\cdot\hat j$ or similar leaves exactly $A_xB_x+A_yB_y+A_zB_z$ . Three special cases are worth memorising because they turn up constantly when judging the sign of work: at $\theta=0^\circ$ (parallel vectors), $\cos\theta=1$ and $\vec A\cdot\vec B=AB$ , the largest possible positive value; at $\theta=90^\circ$ (perpendicular vectors), $\cos\theta=0$ and $\vec A\cdot\vec B=0$ regardless of how large $A$ and $B$ are; at $\theta=180^\circ$ (antiparallel vectors), $\cos\theta=-1$ and $\vec A\cdot\vec B=-AB$ , the largest possible negative value.

Derivation AWork done by a constant force is a scalar product

The goal here is to show, starting only from the general definition $\vec A\cdot\vec B=AB\cos\theta$ and the component identity $\vec A\cdot\vec B=A_xB_x+A_yB_y+A_zB_z$ , that the syllabus formula for work done by a constant force, $W=\vec F\cdot\vec d=Fd\cos\theta$ , is not an independent physical law bolted onto vector algebra but simply that same scalar product with $\vec A$ renamed $\vec F$ (the constant force) and $\vec B$ renamed $\vec d$ (the displacement), and to confirm the two ways of computing it — magnitude-angle and component-wise — agree on a concrete pair of vectors.

Step 1. Write work as the definition given in the syllabus: for a constant force $\vec F$ producing a displacement $\vec d$ , $W=\vec F\cdot\vec d$ . By the general definition of the scalar product this immediately equals $W=Fd\cos\theta$ , where $F=|\vec F|$ , $d=|\vec d|$ , and $\theta$ is the angle between the force vector and the displacement vector, tail to tail.

Step 2. Confirm the same result appears from components. Take $\vec F=F_x\hat i+F_y\hat j$ and $\vec d=d_x\hat i+d_y\hat j$ in a plane. By the component formula, $W=\vec F\cdot\vec d=F_xd_x+F_yd_y$ . This is the version you will actually compute with whenever the force and displacement are handed to you as $\hat i,\hat j$ components rather than as a magnitude and a compass angle.

Step 3. Verify the two forms agree numerically. Let $\vec F=(30\hat i+40\hat j)\ \text{N}$ and $\vec d=(5\hat i+12\hat j)\ \text{m}$ — the Pass-1 worked pair. Components give $W=(30)(5)+(40)(12)=150+480=630\ \text{J}$ . Magnitudes give $F=\sqrt{30^2+40^2}=50\ \text{N}$ and $d=\sqrt{5^2+12^2}=13\ \text{m}$ , so $\cos\theta=\dfrac{630}{(50)(13)}=\dfrac{630}{650}=0.969$ , giving $\theta\approx14.4^\circ$ and $W=Fd\cos\theta=(50)(13)(0.969)\approx630\ \text{J}$ — the same number both ways, as the algebra in Step 1 and Step 2 guarantees for any $\vec F$ and $\vec d$ .

Step 4. Read off the sign rule this buys you for free. Because $\cos\theta$ is positive for $\theta<90^\circ$ , zero at $\theta=90^\circ$ , and negative for $\theta>90^\circ$ , the work done by a constant force is automatically positive when the force has a component along the motion, exactly zero when the force is perpendicular to the motion (a normal reaction on a horizontal floor, or the centripetal force on a particle moving in a circle at constant speed), and negative when the force opposes the motion — no separate rule to memorise, only the one dot-product definition applied consistently. The identical reasoning, with $\vec d$ replaced by $\vec v$ , gives instantaneous power $P=\vec F\cdot\vec v=Fv\cos\theta$ .

Worked examples

Board level - work done by a force given in components

A block is pulled with a force $\vec F=(30\hat i+40\hat j)\ \text{N}$ through a displacement $\vec d=(5\hat i+12\hat j)\ \text{m}$ . Find the work done, the angle between $\vec F$ and $\vec d$ , and the instantaneous power delivered if this displacement takes $t=3\ \text{s}$ at constant velocity.

Step 1. Compute the work using components. $W=\vec F\cdot\vec d=F_xd_x+F_yd_y=(30)(5)+(40)(12)=150+480=630\ \text{J}$ .

Step 2. Compute the magnitudes for the angle check. $F=\sqrt{30^2+40^2}=\sqrt{900+1600}=\sqrt{2500}=50\ \text{N}$ and $d=\sqrt{5^2+12^2}=\sqrt{25+144}=\sqrt{169}=13\ \text{m}$ .

Step 3. Find the angle between the vectors. $\cos\theta=\dfrac{\vec F\cdot\vec d}{Fd}=\dfrac{630}{(50)(13)}=\dfrac{630}{650}=0.969$ , so $\theta\approx14.4^\circ$ — the force is nearly aligned with the displacement, consistent with the large positive work.

Step 4. Find the average velocity and use it for power. $\vec v=\dfrac{\vec d}{t}=\dfrac{(5\hat i+12\hat j)}{3}=(1.67\hat i+4\hat j)\ \text{m/s}$ , so $P=\vec F\cdot\vec v=(30)(1.67)+(40)(4)=50+160=210\ \text{W}$ . This checks against $P=W/t=630/3=210\ \text{W}$ , exactly as expected since the force is constant.

The work done is $\mathbf{630\ J}$ , the angle between force and displacement is $\mathbf{\approx14.4^\circ}$ , and the average power delivered is $\mathbf{210\ W}$ .

Competitive level - work by tension and gravity on an incline, with a component cross-check

A crate of mass $m=10\ \text{kg}$ ( $mg=98\ \text{N}$ , taking $g=9.8\ \text{m/s}^2$ ) is dragged $d=4\ \text{m}$ up a frictionless incline that rises at $\theta_i=37^\circ$ to the horizontal, by a rope tension $T=150\ \text{N}$ acting $20^\circ$ above the incline surface. Find the work done by tension and by gravity using the dot-product definition directly in components, and check the two results are consistent with the incline geometry.

Step 1. Set up axes with $\hat i$ along the incline surface (the direction of motion) and $\hat j$ perpendicular to it. Then $\vec d=4\hat i\ \text{m}$ .

Step 2. Resolve the tension into these axes. Since the tension is $20^\circ$ above the incline surface, $\vec T=T\cos20^\circ\,\hat i+T\sin20^\circ\,\hat j=(150)(0.940)\hat i+(150)(0.342)\hat j=(141\hat i+51.3\hat j)\ \text{N}$ .

Step 3. Dot the tension into the displacement. $W_T=\vec T\cdot\vec d=(141)(4)+(51.3)(0)=564\ \text{J}$ , matching the direct magnitude-angle computation $W_T=Td\cos20^\circ=(150)(4)(0.940)=564\ \text{J}$ — the perpendicular component of tension does zero work since it has no $\hat i$ part to dot with $\vec d$ .

Step 4. Resolve gravity into the same incline axes. Gravity acts vertically down; along the incline direction (up-slope taken positive) its component is $-mg\sin\theta_i=-(98)(0.602)=-59.0\ \text{N}$ , and perpendicular to the incline it is $-mg\cos\theta_i$ , which again cannot do work since $\vec d$ has no perpendicular component.

Step 5. Dot gravity into the displacement. $W_g=(-59.0)(4)+(\text{perp. component})(0)=-236\ \text{J}$ , matching the earlier magnitude-angle route $W_g=mgd\cos127^\circ=(98)(4)(-0.602)=-236\ \text{J}$ , since $127^\circ$ is the angle between the straight-down gravity vector and the up-the-incline displacement.

The work done by tension is $\mathbf{+564\ J}$ and the work done by gravity is $\mathbf{-236\ J}$ ; the net work by these two forces alone is $564-236=328\ \text{J}$ , which (with the incline frictionless and normal force doing zero work) equals the crate's gain in kinetic energy by the work-energy theorem. Timing note: in an exam, resolving forces along and perpendicular to the direction of motion before dotting — as in Steps 2-5 — is faster and safer than computing a raw angle between two arbitrarily oriented vectors; reserve the direct $\cos\theta$ route for when the angle between force and displacement is already stated in the question, as in the board-level example above.

Common traps

Writing the dot product as a vector. The scalar product always returns a pure number; if a "dot-product answer" has a direction attached to it, the cross product has been used by mistake, or a stray unit vector has been left in the final line.
Using the wrong angle. $\theta$ must be the angle between the two vectors placed tail to tail — not the angle a force makes with the horizontal, unless the displacement itself happens to be horizontal. On an incline, the angle between force and displacement is usually not the same as the incline angle; resolve into incline-aligned axes first, as in the competitive example, to avoid this slip.
Forgetting $\cos\theta$ can be negative. For $90^\circ<\theta<180^\circ$ the work comes out negative — a genuine physical result (a retarding force), never a sign to drop or "correct."
Dropping the cosine altogether. $\vec A\cdot\vec B=AB\cos\theta$ , not $AB$ ; multiplying two magnitudes and forgetting the cosine factor is one of the most common numerical slips, particularly when the vectors are given in component form and a student jumps straight to multiplying magnitudes.
Treating a component sum as automatically correct without checking axes. $A_xB_x+A_yB_y+A_zB_z$ only equals $\vec A\cdot\vec B$ when both vectors are resolved along the same set of mutually perpendicular axes; mixing an incline-aligned $x$ -component of one vector with a horizontal-aligned $x$ -component of the other silently produces a wrong answer.

Common trapPerpendicular does not mean small — it means exactly zero, however large the force

The single misconception that costs the most marks is assuming a "big force" must do "big work." A force perpendicular to the displacement does exactly zero work, no matter how large its magnitude is — the normal reaction from a floor on a sliding block, the tension in a conical pendulum's string relative to its instantaneous velocity, and the centripetal force holding a satellite in circular orbit are all textbook-sized examples of forces that do zero work purely because $\theta=90^\circ$ makes $\cos\theta=0$ in $W=Fd\cos\theta$ . Students who reason from magnitude alone, without checking the angle, routinely credit these forces with work they never do.

You should be able to…

State the definition $\vec A\cdot\vec B=AB\cos\theta$ , explain in one sentence why the result is a scalar, and identify the angle $\theta$ correctly as the angle between the two vectors tail to tail.
Compute a dot product from rectangular components using $\vec A\cdot\vec B=A_xB_x+A_yB_y+A_zB_z$ , and know why this formula follows from $\hat i\cdot\hat i=1$ and $\hat i\cdot\hat j=0$ .
Use a known dot product to find the angle between two given vectors via $\cos\theta=\dfrac{\vec A\cdot\vec B}{AB}$ .
Recognise instantly when a dot product is zero (perpendicular vectors), maximum positive (parallel), or maximum negative (antiparallel), without computing a decimal.
Apply the dot product to compute work done by a constant force, $W=Fd\cos\theta$ or $W=F_xd_x+F_yd_y+F_zd_z$ , and correctly interpret a positive, negative, or zero result physically.
Apply the identical operation to instantaneous power, $P=\vec F\cdot\vec v$ , and connect it to $W=Pt$ for constant power.

Ask the tutor: Still mixing up when to use $AB\cos\theta$ versus adding components, or want more practice spotting when a force does zero work by the perpendicularity test — ask the AI tutor to work through a few more dot-product and work-power problems with you, step by step.

Why does the dot product use cosine and not sine?

Cosine measures how much of one vector lies along the direction of the other — exactly the "useful" part of a force that actually pushes something along its displacement. When two vectors are perpendicular ( $\theta=90^\circ$ ), $\cos\theta=0$ , matching the physical fact that a sideways force does not help (or hinder) motion along a different direction at all. Sine, by contrast, is largest at $90^\circ$ and would give the wrong physical answer — that is the signature of the vector (cross) product instead, which measures the "twisting" or perpendicular part of two vectors, not the "along" part.

If work is a dot product, why do we sometimes just write W equals Fd without any cosine?

That simplified form is only valid in the special case where the force and the displacement point in exactly the same direction, so $\theta=0^\circ$ and $\cos\theta=1$ . It is common in simple problems — lifting a box straight up with a force straight up, or pushing a cart forward with a force that points forward — but it is a special case of $W=Fd\cos\theta$ , not a separate formula. The moment the force is applied at an angle, as with the tension in the incline example, you must bring the cosine back in, or resolve into components and dot directly.

How is the dot product different from just multiplying the x-components and adding, without worrying about the angle at all?

They give the same answer, but the component method and the magnitude-angle method are two routes to the identical number, not two different quantities. $A_xB_x+A_yB_y+A_zB_z$ is derived directly from $AB\cos\theta$ by writing both vectors in unit-vector form and using $\hat i\cdot\hat i=1$ , $\hat i\cdot\hat j=0$ , and so on. Use the component form when vectors are already given as components (it is usually faster and avoids computing an angle at all); use the magnitude-angle form when a diagram gives you lengths and an angle directly.

Is the dot product going to show up outside of work and power in this chapter?

Within Work, Energy and Power specifically, its two headline uses are the definitions of work, $W=\vec F\cdot\vec d$ , and instantaneous power, $P=\vec F\cdot\vec v$ — both are dot products applied to different vector pairs. You will also lean on the self-dot-product trick $\vec v\cdot\vec v=v^2$ whenever kinetic energy $K=\tfrac12mv^2$ is written in vector form. The vector (cross) product, by contrast, belongs to torque and angular momentum in the rotational-motion chapter, not here.