←
Mathematics — Grade 12, Part II
VECTOR ALGEBRA · 4 / 7
⌕
Chapter 4 of 7

VECTOR ALGEBRA

28 sections · 9,968 words

Chapter 10: Vector Algebra

36 words · 1 min read

Chapter 10

Vector Algebra

Big idea: some quantities need a direction as well as a size. Vectors give them an algebra, so that pushes, journeys and velocities can be added, scaled and compared with simple arithmetic.

10.1 Introduction

302 words · 1 min read

Think of two kinds of information. "The room is 44 metres long" needs only a number. "Walk 44 metres" is incomplete: which way? Some quantities are complete with a number and a unit. Others also need a direction.

  • A scalar has only a size (magnitude). Examples: mass, time, temperature, distance, speed, area, energy.
  • A vector has a size and a direction. Examples: displacement, velocity, acceleration, force, momentum.

A person walks 33 blocks east and then 44 blocks north. The distance walked is 3+4=73 + 4 = 7 blocks: a scalar, which only counts how far the feet moved. The displacement is the straight line from the start to the end. It is a vector. By Pythagoras its size is 32+42=25=5\sqrt{3^2 + 4^2} = \sqrt{25} = 5 blocks, and its direction is north of east.

The walk (solid) is 7 blocks long. The displacement (dashed) is the straight arrow from start to end, 5 blocks long.

Vectors are used wherever direction matters: in physics (forces, velocities), engineering (structures, circuits), computer graphics (movement, lighting) and navigation.

Scalar (size only)Vector (size and direction)
distancedisplacement
speedvelocity
massforce, weight
energy, workmomentum, acceleration
In this chapter you will learn: how to describe a vector by its components; how to add vectors, scale them and find a point between two others; and the two ways to multiply vectors: the dot product (gives a number) and the cross product (gives a vector).

10.2 Some Basic Concepts

277 words · 1 min read

Every point in space is described by three coordinates (x,y,z)(x, y, z), measured along three perpendicular axes that meet at the origin O=(0,0,0)O = (0, 0, 0). The zz-axis points up. The distance between two points comes from Pythagoras' theorem, used twice:

PQ=(x2−x1)2+(y2−y1)2+(z2−z1)2PQ = \sqrt{(x_2 - x_1)^2 + (y_2 - y_1)^2 + (z_2 - z_1)^2}

Example: for P=(1,2,3)P = (1, 2, 3) and Q=(4,−2,5)Q = (4, -2, 5), the differences are 33, −4-4, 22. The squares add to 2929, so PQ=29PQ = \sqrt{29}.

A vector is a directed line segment: a straight line with a start and an end, drawn with an arrowhead at the end. If it starts at AA and ends at BB, we write AB→\overrightarrow{AB}. The start AA is the initial point and the end BB is the terminal point. A vector is also written with a single letter and an arrow, such as a⃗\vec{a}.

The magnitude of AB→\overrightarrow{AB} is the length of the segment, written ∣AB→∣|\overrightarrow{AB}|. It equals the distance ABAB. Two things describe a vector completely: its magnitude and its direction. It does not matter where on the page it is drawn.

The vector AB starts at A = (1, 1, 0) and ends at B = (4, 3, 2).

Here ∣AB→∣=17≈4.1231|\overrightarrow{AB}| = \sqrt{17} \approx 4.1231.

Common mistake: thinking a vector is a point. A point is a place. A vector is a movement. The same vector can start anywhere.

Position Vector

245 words · 1 min read

If you fix the origin OO, every point PP gets its own vector OP→\overrightarrow{OP} from the origin to PP. This is the position vector of PP. If P=(x,y,z)P = (x, y, z), then

OP→=x i^+y j^+z k^.\overrightarrow{OP} = x\,\hat{i} + y\,\hat{j} + z\,\hat{k}.

Here i^\hat{i}, j^\hat{j}, k^\hat{k} are unit vectors (length 11) along the xx, yy and zz axes. The numbers xx, yy, zz are the components of the vector. The magnitude comes from the distance formula:

∣OP→∣=x2+y2+z2.|\overrightarrow{OP}| = \sqrt{x^2 + y^2 + z^2}.

Example 1. For P=(2,3,6)P = (2, 3, 6): OP→=2i^+3j^+6k^\overrightarrow{OP} = 2\hat{i} + 3\hat{j} + 6\hat{k} and ∣OP→∣=4+9+36=49=7|\overrightarrow{OP}| = \sqrt{4 + 9 + 36} = \sqrt{49} = 7.

Example 2. For P=(1,−2,2)P = (1, -2, 2): OP→=i^−2j^+2k^\overrightarrow{OP} = \hat{i} - 2\hat{j} + 2\hat{k} and ∣OP→∣=1+4+4=3|\overrightarrow{OP}| = \sqrt{1 + 4 + 4} = 3.

The position vector of P = (2, 3, 6). Go 2 along x, 3 along y and 6 up in z. Its length is 7.
In simple words: the position vector is the recipe for getting from the origin to the point: so many steps along xx, so many along yy, so many up.

Direction Cosines

393 words · 2 min read

A vector r⃗\vec{r} makes three angles with the positive xx, yy, zz axes. Call them α\alpha, β\beta, γ\gamma (the direction angles). The cosines of these angles are the direction cosines:

l=cos⁡α,m=cos⁡β,n=cos⁡γ.l = \cos\alpha, \qquad m = \cos\beta, \qquad n = \cos\gamma.

For r⃗=xi^+yj^+zk^\vec{r} = x\hat{i} + y\hat{j} + z\hat{k} of length rr, the xx-component is the adjacent side of the angle with the xx-axis, so

l=xr,m=yr,n=zr.l = \frac{x}{r}, \qquad m = \frac{y}{r}, \qquad n = \frac{z}{r}.

Square and add: l2+m2+n2=x2+y2+z2r2=1l^2 + m^2 + n^2 = \dfrac{x^2 + y^2 + z^2}{r^2} = 1. This always holds. The direction cosines are also the components of the unit vector r^=li^+mj^+nk^\hat{r} = l\hat{i} + m\hat{j} + n\hat{k}.

For r⃗=2i^+3j^+6k^\vec{r} = 2\hat{i} + 3\hat{j} + 6\hat{k} we have r=7r = 7. The direction cosines are 27\frac{2}{7}, 37\frac{3}{7}, 67\frac{6}{7}. Their squares are 449,949,3649\frac{4}{49}, \frac{9}{49}, \frac{36}{49}, which add to 4949=1\frac{49}{49} = 1. The direction angles are α≈73.4∘\alpha \approx 73.4^{\circ}, β≈64.62∘\beta \approx 64.62^{\circ}, γ≈31∘\gamma \approx 31^{\circ}.

For r⃗=i^−2j^+2k^\vec{r} = \hat{i} - 2\hat{j} + 2\hat{k} we have r=3r = 3. The direction cosines are 13\frac{1}{3}, −23-\frac{2}{3}, 23\frac{2}{3}. The negative sign tells us the vector leans away from the positive yy-axis, so β≈131.81∘\beta \approx 131.81^{\circ} is more than 90∘90^{\circ}. The other angles are α≈70.53∘\alpha \approx 70.53^{\circ} and γ≈48.19∘\gamma \approx 48.19^{\circ}.

Any numbers proportional to the direction cosines are called direction ratios. For r⃗\vec{r} above, (2,3,6)(2, 3, 6) and (4,6,12)(4, 6, 12) are both direction ratios, because the second is the first times 22. Direction cosines are the special direction ratios whose squares add to 11. To get them from direction ratios a,b,ca, b, c, divide each by a2+b2+c2\sqrt{a^2 + b^2 + c^2}.

If a vector makes equal angles with all three axes, then l=m=nl = m = n. From l2+m2+n2=1l^2 + m^2 + n^2 = 1 we get 3l2=13l^2 = 1, so l=±13l = \pm\dfrac{1}{\sqrt{3}}. The vector i^+j^+k^\hat{i} + \hat{j} + \hat{k} is one such vector, and its angle with each axis is cos⁡−113≈54.74∘\cos^{-1}\dfrac{1}{\sqrt{3}} \approx 54.74^{\circ}.

Common mistake: using the components as direction cosines without dividing by the length. The cosines of (2,3,6)(2, 3, 6) are 27\frac{2}{7}, 37\frac{3}{7}, 67\frac{6}{7}, not 22, 33, 66. A cosine can never be larger than 11.

10.3 Types of Vectors

330 words · 2 min read
TypeWhat it meansExample
Zero vector 0⃗\vec{0}size 00, no direction (start and end point are the same)0⃗=0i^+0j^+0k^\vec{0} = 0\hat{i} + 0\hat{j} + 0\hat{k}
Unit vectorsize exactly 11i^\hat{i}, j^\hat{j}, k^\hat{k}
Co-initial vectorsstart at the same pointtwo arrows from the origin
Collinear (parallel) vectorslie along the same line or parallel lines, whatever their sizesa⃗\vec{a} and 3a⃗3\vec{a}
Equal vectorssame size and same direction (the starting point does not matter)two arrows that match exactly
Negative of a⃗\vec{a}same size, opposite direction, written −a⃗-\vec{a}a⃗+(−a⃗)=0⃗\vec{a} + (-\vec{a}) = \vec{0}

Dividing a vector by its own length gives a unit vector in the same direction. We write it a^=a⃗∣a⃗∣\hat{a} = \dfrac{\vec{a}}{|\vec{a}|}. For a⃗=3i^+4j^+12k^\vec{a} = 3\hat{i} + 4\hat{j} + 12\hat{k} the length is 9+16+144=13\sqrt{9 + 16 + 144} = 13. So

a^=313i^+413j^+1213k^.\hat{a} = \frac{3}{13}\hat{i} + \frac{4}{13}\hat{j} + \frac{12}{13}\hat{k}.

As a check, 9+16+144169=1\frac{9 + 16 + 144}{169} = 1, and the decimals are 0.2308,0.3077,0.92310.2308, 0.3077, 0.9231.

Two vectors are collinear if one is a number times the other. For a⃗=2i^−4j^+6k^\vec{a} = 2\hat{i} - 4\hat{j} + 6\hat{k} and b⃗=−i^+2j^−3k^\vec{b} = -\hat{i} + 2\hat{j} - 3\hat{k}, we see that a⃗=−2b⃗\vec{a} = -2\vec{b}. So they are collinear. The minus sign means they point in opposite directions.

Take P=(0,1,1)P = (0, 1, 1) and Q=(1,3,3)Q = (1, 3, 3). The vector PQ→\overrightarrow{PQ} goes 11 along xx, 22 along yy, 22 along zz, so PQ→=i^+2j^+2k^\overrightarrow{PQ} = \hat{i} + 2\hat{j} + 2\hat{k}. It is equal to the vector i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k} drawn from the origin, even though it starts somewhere else.

In simple words: two arrows are equal if you can slide one onto the other without turning or stretching it.

EXERCISE 10.1

472 words · 2 min read
  1. Say whether each quantity is a scalar or a vector: temperature, displacement, area, force, volume, acceleration, kinetic energy, momentum.
  2. Find the magnitude of each position vector: 3i^+4j^+12k^3\hat{i} + 4\hat{j} + 12\hat{k}, −2i^+6j^+9k^-2\hat{i} + 6\hat{j} + 9\hat{k}, i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k}, −5j^+12k^-5\hat{j} + 12\hat{k}.
  3. Find the direction cosines of r⃗=6i^−2j^+3k^\vec{r} = 6\hat{i} - 2\hat{j} + 3\hat{k} and the angle it makes with the xx-axis.
  4. Find the unit vector in the direction of 4i^−3j^+12k^4\hat{i} - 3\hat{j} + 12\hat{k}.
  5. Which of a⃗=i^−2j^+3k^\vec{a} = \hat{i} - 2\hat{j} + 3\hat{k}, b⃗=−2i^+4j^−6k^\vec{b} = -2\hat{i} + 4\hat{j} - 6\hat{k}, c⃗=3i^−6j^+9k^\vec{c} = 3\hat{i} - 6\hat{j} + 9\hat{k} are collinear? Which pairs point the same way?
  6. Are i^+2j^+3k^\hat{i} + 2\hat{j} + 3\hat{k} and 2i^+4j^+5k^2\hat{i} + 4\hat{j} + 5\hat{k} collinear?
  7. Find xx so that xi^−2j^+4k^x\hat{i} - 2\hat{j} + 4\hat{k} is collinear with 3i^−6j^+12k^3\hat{i} - 6\hat{j} + 12\hat{k}.
  8. Write the vector from A=(1,2,3)A = (1, 2, 3) to B=(4,6,3)B = (4, 6, 3) and find its magnitude.
  9. Find the magnitude of 3a⃗−2b⃗3\vec{a} - 2\vec{b} for a⃗=i^+2j^−k^\vec{a} = \hat{i} + 2\hat{j} - \hat{k} and b⃗=2i^−j^+3k^\vec{b} = 2\hat{i} - \hat{j} + 3\hat{k}.
  10. Explain why a⃗=i^+j^\vec{a} = \hat{i} + \hat{j} and b⃗=i^+j^+0k^\vec{b} = \hat{i} + \hat{j} + 0\hat{k} are the same vector, and give its direction cosines.

Answers

  1. Scalars: temperature, area, volume, kinetic energy. Vectors: displacement, force, acceleration, momentum.
  2. 1313, 1111, 33, 1313. For example 9+16+144=169=13\sqrt{9 + 16 + 144} = \sqrt{169} = 13.
  3. ∣r⃗∣=7|\vec{r}| = 7, so the direction cosines are 67\frac{6}{7}, −27-\frac{2}{7}, 37\frac{3}{7}. The angle with the xx-axis is cos⁡−167≈31∘\cos^{-1}\frac{6}{7} \approx 31^{\circ}.
  4. The length is 16+9+144=13\sqrt{16 + 9 + 144} = 13, so the unit vector is 413i^−313j^+1213k^\dfrac{4}{13}\hat{i} - \dfrac{3}{13}\hat{j} + \dfrac{12}{13}\hat{k}.
  5. All three are collinear: b⃗=−2a⃗\vec{b} = -2\vec{a} and c⃗=3a⃗\vec{c} = 3\vec{a}. The vectors a⃗\vec{a} and c⃗\vec{c} point the same way (positive multiple). b⃗\vec{b} points opposite to both.
  6. No. The first two components of the second are twice those of the first, but the third is not (5≠65 \ne 6), so it is not a multiple of the first.
  7. The second vector is 33 times i^−2j^+4k^\hat{i} - 2\hat{j} + 4\hat{k}, so x=1x = 1.
  8. AB→=3i^+4j^\overrightarrow{AB} = 3\hat{i} + 4\hat{j} and ∣AB→∣=9+16+0=5|\overrightarrow{AB}| = \sqrt{9 + 16 + 0} = 5.
  9. 3a⃗=3i^+6j^−3k^3\vec{a} = 3\hat{i} + 6\hat{j} - 3\hat{k} and 2b⃗=4i^−2j^+6k^2\vec{b} = 4\hat{i} - 2\hat{j} + 6\hat{k}. Their difference is −i^+8j^−9k^-\hat{i} + 8\hat{j} - 9\hat{k}, with magnitude 1+64+81=146\sqrt{1 + 64 + 81} = \sqrt{146}.
  10. The component of k^\hat{k} in a⃗\vec{a} is 00, so they are equal. The length is 2\sqrt{2}, so the direction cosines are 12\frac{1}{\sqrt{2}}, 12\frac{1}{\sqrt{2}}, 00 (angles 45∘45^{\circ}, 45∘45^{\circ}, 90∘90^{\circ}).

10.4 Addition of Vectors

418 words · 2 min read

A boat points straight east and its engine gives it a speed of 33 metres per second through still water. At the same time a river current pushes it north at 44 metres per second. Where does the boat actually go? Its real velocity is the sum of the two velocities. This is vector addition.

Put the second arrow so that it starts where the first one ends. The sum is the arrow from the very start to the very end. If a⃗=AB→\vec{a} = \overrightarrow{AB} and b⃗=BC→\vec{b} = \overrightarrow{BC}, then

AB→+BC→=AC→.\overrightarrow{AB} + \overrightarrow{BC} = \overrightarrow{AC}.

The triangle law: a goes 3 east, b goes 4 north, and a + b goes straight from the start to the end.

Here a⃗+b⃗=3i^+4j^\vec{a} + \vec{b} = 3\hat{i} + 4\hat{j}, with magnitude 9+16=5\sqrt{9 + 16} = 5. The boat moves at 55 m/s, in a direction between east and north. Notice that 5<3+45 < 3 + 4: the resultant is shorter than the two sides added, because they do not point the same way.

If two vectors start at the same point, draw the parallelogram they span. The sum is the diagonal from the common starting point. This is the same as the triangle law, with b⃗\vec{b} moved to the far side.

In practice, add component by component. For a⃗=2i^+3j^−k^\vec{a} = 2\hat{i} + 3\hat{j} - \hat{k} and b⃗=i^−2j^+4k^\vec{b} = \hat{i} - 2\hat{j} + 4\hat{k}:

a⃗+b⃗=3i^+j^+3k^.\vec{a} + \vec{b} = 3\hat{i} + \hat{j} + 3\hat{k}.

Subtraction means adding the negative: a⃗−b⃗=a⃗+(−b⃗)=i^+5j^−5k^\vec{a} - \vec{b} = \vec{a} + (-\vec{b}) = \hat{i} + 5\hat{j} - 5\hat{k}.

Any side of a triangle is shorter than the other two together, so for all vectors ∣a⃗+b⃗∣≤∣a⃗∣+∣b⃗∣|\vec{a} + \vec{b}| \le |\vec{a}| + |\vec{b}|. For the numbers above, ∣a⃗∣≈3.742|\vec{a}| \approx 3.742, ∣b⃗∣≈4.583|\vec{b}| \approx 4.583 and ∣a⃗+b⃗∣≈4.359|\vec{a} + \vec{b}| \approx 4.359, and indeed 4.359<8.3244.359 < 8.324. Equality happens only when both vectors point the same way.

Common mistake: adding the lengths of vectors instead of the vectors. In the boat example the lengths add to 77, but the speed is 55.

Properties of vector addition

285 words · 1 min read

Vector addition follows the same rules as adding numbers. Each rule was checked with numbers by computer, using a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k}, b⃗=4i^−j^+2k^\vec{b} = 4\hat{i} - \hat{j} + 2\hat{k} and c⃗=−2i^+5j^\vec{c} = -2\hat{i} + 5\hat{j}.

  • Commutative: a⃗+b⃗=b⃗+a⃗\vec{a} + \vec{b} = \vec{b} + \vec{a}. Both give 5i^+j^+5k^5\hat{i} + \hat{j} + 5\hat{k}. The order does not matter: the two routes of the parallelogram end at the same corner.
  • Associative: (a⃗+b⃗)+c⃗=a⃗+(b⃗+c⃗)(\vec{a} + \vec{b}) + \vec{c} = \vec{a} + (\vec{b} + \vec{c}). Both give 3i^+6j^+5k^3\hat{i} + 6\hat{j} + 5\hat{k}. You can group in any way, so we may write a⃗+b⃗+c⃗\vec{a} + \vec{b} + \vec{c}.
  • Zero vector: a⃗+0⃗=a⃗\vec{a} + \vec{0} = \vec{a}. Adding the zero vector changes nothing.
  • Negative: a⃗+(−a⃗)=0⃗\vec{a} + (-\vec{a}) = \vec{0}. Going somewhere and coming straight back means no displacement.
The parallelogram law. Going along a then b, or along b then a, reaches the same far corner (5, 3, 0). That is why a + b = b + a.

Closed polygon. If you walk around a closed shape (a triangle, a square, any polygon) and come back to the start, the displacements add to the zero vector. For the triangle AA, BB, CC: AB→+BC→+CA→=0⃗\overrightarrow{AB} + \overrightarrow{BC} + \overrightarrow{CA} = \vec{0}.

In simple words: you can add vectors in any order, group them as you like, and a vector plus its opposite cancels.

10.5 Multiplication of a Vector by a Scalar

299 words · 1 min read

Multiplying a vector a⃗\vec{a} by a number λ\lambda stretches or shrinks it. The result λa⃗\lambda\vec{a} has:

  • length ∣λa⃗∣=∣λ∣ ∣a⃗∣|\lambda\vec{a}| = |\lambda|\,|\vec{a}|;
  • the same direction as a⃗\vec{a} if λ>0\lambda > 0, the opposite direction if λ<0\lambda < 0;
  • the zero vector if λ=0\lambda = 0.
a, 2a and −a lie on one line. Doubling doubles the length. A minus sign reverses the direction.

In components, multiply every component: λ(xi^+yj^+zk^)=λx i^+λy j^+λz k^\lambda(x\hat{i} + y\hat{j} + z\hat{k}) = \lambda x\,\hat{i} + \lambda y\,\hat{j} + \lambda z\,\hat{k}.

The unit vector a^=a⃗∣a⃗∣\hat{a} = \dfrac{\vec{a}}{|\vec{a}|} has length 11. So m a^m\,\hat{a} has length mm and the direction of a⃗\vec{a}. Example: find a vector of length 1010 in the direction of a⃗=2i^−3j^+6k^\vec{a} = 2\hat{i} - 3\hat{j} + 6\hat{k}.

The length of a⃗\vec{a} is 4+9+36=7\sqrt{4 + 9 + 36} = 7. So the vector is 107a⃗=17(20i^−30j^+60k^)\dfrac{10}{7}\vec{a} = \dfrac{1}{7}\left(20\hat{i} - 30\hat{j} + 60\hat{k}\right). Check: 10a⃗=20i^−30j^+60k^10\vec{a} = 20\hat{i} - 30\hat{j} + 60\hat{k} has length 7070, and dividing by 77 gives length 1010.

So the answer is 207i^−307j^+607k^\dfrac{20}{7}\hat{i} - \dfrac{30}{7}\hat{j} + \dfrac{60}{7}\hat{k}.

Common mistake: multiplying the length of a vector by a negative number and getting a negative length. Length is never negative. The sign of λ\lambda changes the direction, and ∣λ∣|\lambda| changes the length.

10.5.1 Components of a vector

409 words · 2 min read

Every vector can be written in one way as a⃗=a1i^+a2j^+a3k^\vec{a} = a_1\hat{i} + a_2\hat{j} + a_3\hat{k}. The numbers a1a_1, a2a_2, a3a_3 are its components. This is the same as the position vector of the point (a1,a2,a3)(a_1, a_2, a_3). Working with components turns geometry into arithmetic.

  • Magnitude: ∣a⃗∣=a12+a22+a32|\vec{a}| = \sqrt{a_1^2 + a_2^2 + a_3^2}.
  • Equal vectors: a⃗=b⃗\vec{a} = \vec{b} exactly when all three components match.
  • Sum and difference: add or subtract the matching components.
  • Scalar multiple: multiply each component.
  • Collinear (parallel) vectors: a⃗∥b⃗\vec{a} \parallel \vec{b} exactly when b⃗=λa⃗\vec{b} = \lambda\vec{a} for some number λ\lambda, that is, the components are proportional: b1a1=b2a2=b3a3\dfrac{b_1}{a_1} = \dfrac{b_2}{a_2} = \dfrac{b_3}{a_3}.

A drone climbs with velocity 3i^+4j^+12k^3\hat{i} + 4\hat{j} + 12\hat{k} metres per second (33 east, 44 north, 1212 up). Its speed is 9+16+144=13\sqrt{9 + 16 + 144} = 13 m/s.

Let a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=2i^−3j^+k^\vec{b} = 2\hat{i} - 3\hat{j} + \hat{k}. Then

  • a⃗+b⃗=3i^−j^+4k^\vec{a} + \vec{b} = 3\hat{i} - \hat{j} + 4\hat{k}, with ∣a⃗+b⃗∣=26|\vec{a} + \vec{b}| = \sqrt{26}.
  • 2a⃗−3b⃗=2i^+4j^+6k^−(6i^−9j^+3k^)=−4i^+13j^+3k^2\vec{a} - 3\vec{b} = 2\hat{i} + 4\hat{j} + 6\hat{k} - \left(6\hat{i} - 9\hat{j} + 3\hat{k}\right) = -4\hat{i} + 13\hat{j} + 3\hat{k}, with ∣2a⃗−3b⃗∣=194|2\vec{a} - 3\vec{b}| = \sqrt{194}.

For which xx is p⃗=2i^−4j^+6k^\vec{p} = 2\hat{i} - 4\hat{j} + 6\hat{k} parallel to q⃗=xi^+2j^−3k^\vec{q} = x\hat{i} + 2\hat{j} - 3\hat{k}? The ratio of the jj-components is 2−4=−12\frac{2}{-4} = -\frac{1}{2}, and for the kk-components −36=−12\frac{-3}{6} = -\frac{1}{2}. They agree, so x=−12×2=−1x = -\frac{1}{2} \times 2 = -1. The computer confirms p⃗\vec{p} and −i^+2j^−3k^-\hat{i} + 2\hat{j} - 3\hat{k} are parallel.

Write 5i^+3j^5\hat{i} + 3\hat{j} as p (i^+2j^)+q (3i^−j^)p\,(\hat{i} + 2\hat{j}) + q\,(3\hat{i} - \hat{j}). Comparing components: p+3q=5p + 3q = 5 and 2p−q=32p - q = 3. Solving gives p=2p = 2 and q=1q = 1. Check: 2(i^+2j^)+(3i^−j^)=5i^+3j^2(\hat{i} + 2\hat{j}) + (3\hat{i} - \hat{j}) = 5\hat{i} + 3\hat{j}.

In simple words: the components say how far to go along each axis. Vector algebra is just ordinary algebra done three times at once.

10.5.2 Vector joining two points

262 words · 1 min read

Suppose the points PP and QQ have position vectors p⃗=OP→\vec{p} = \overrightarrow{OP} and q⃗=OQ→\vec{q} = \overrightarrow{OQ}. To walk from PP to QQ you can go from PP back to OO and then on to QQ. In the triangle OPQOPQ the triangle law gives OP→+PQ→=OQ→\overrightarrow{OP} + \overrightarrow{PQ} = \overrightarrow{OQ}, so

PQ→=OQ→−OP→=q⃗−p⃗.\overrightarrow{PQ} = \overrightarrow{OQ} - \overrightarrow{OP} = \vec{q} - \vec{p}.

Rule: the vector from PP to QQ is "end minus start", component by component.

Example 1. For P=(1,2,3)P = (1, 2, 3) and Q=(4,−2,5)Q = (4, -2, 5): PQ→=(4−1)i^+(−2−2)j^+(5−3)k^=3i^−4j^+2k^\overrightarrow{PQ} = (4 - 1)\hat{i} + (-2 - 2)\hat{j} + (5 - 3)\hat{k} = 3\hat{i} - 4\hat{j} + 2\hat{k}, and ∣PQ→∣=9+16+4=29|\overrightarrow{PQ}| = \sqrt{9 + 16 + 4} = \sqrt{29}. The vector QP→\overrightarrow{QP} is the negative: −3i^+4j^−2k^-3\hat{i} + 4\hat{j} - 2\hat{k}.

The vector PQ is the position vector of Q minus the position vector of P.

Example 2: a triangle. Take A=(1,2,3)A = (1, 2, 3), B=(4,6,3)B = (4, 6, 3), C=(2,−1,5)C = (2, -1, 5). Then AB→=3i^+4j^\overrightarrow{AB} = 3\hat{i} + 4\hat{j}, BC→=−2i^−7j^+2k^\overrightarrow{BC} = -2\hat{i} - 7\hat{j} + 2\hat{k} and CA→=−i^+3j^−2k^\overrightarrow{CA} = -\hat{i} + 3\hat{j} - 2\hat{k}. Adding the three gives 0⃗\vec{0}, as it must for a closed triangle.

Common mistake: "start minus end". The vector from PP to QQ is q⃗−p⃗\vec{q} - \vec{p}. Reversing the order gives the opposite vector.

10.5.3 Section formula

475 words · 2 min read

A point RR lies on the line through AA and BB and divides the segment ABAB in the ratio m:nm : n. This means AR:RB=m:nAR : RB = m : n. If AA and BB have position vectors a⃗\vec{a} and b⃗\vec{b}, then RR has position vector

r⃗=n a⃗+m b⃗m+n(internal division).\vec{r} = \frac{n\,\vec{a} + m\,\vec{b}}{m + n} \qquad (\text{internal division}).

Notice the crossing: the far vector gets the larger share when its number mm is bigger. The mid-point is the special case m=nm = n: r⃗=a⃗+b⃗2\vec{r} = \dfrac{\vec{a} + \vec{b}}{2}.

External division. If RR lies on the line outside the segment, with AR:RB=m:nAR : RB = m : n (and m≠nm \ne n), then

r⃗=m b⃗−n a⃗m−n.\vec{r} = \frac{m\,\vec{b} - n\,\vec{a}}{m - n}.

Let A=(1,2,3)A = (1, 2, 3) and B=(6,−3,8)B = (6, -3, 8). The point dividing ABAB internally in the ratio 2:32 : 3 is 3a⃗+2b⃗5=15i^+25k^5=(3,0,5)\dfrac{3\vec{a} + 2\vec{b}}{5} = \dfrac{15\hat{i} + 25\hat{k}}{5} = (3, 0, 5). Externally in the ratio 3:23 : 2 it is 3b⃗−2a⃗1=(16,−13,18)\dfrac{3\vec{b} - 2\vec{a}}{1} = (16, -13, 18). The mid-point is (72,−12,112)(\frac{7}{2}, -\frac{1}{2}, \frac{11}{2}).

R = (3, 0, 5) divides AB internally in the ratio 2 : 3.

The centroid GG of a triangle is where the three medians meet. It lies 23\frac{2}{3} of the way from each vertex to the opposite mid-point. For A=(1,2,3)A = (1, 2, 3), B=(4,6,3)B = (4, 6, 3), C=(7,1,0)C = (7, 1, 0): the mid-point of BCBC is M=(112,72,32)M = (\frac{11}{2}, \frac{7}{2}, \frac{3}{2}). Doubling MM gives b⃗+c⃗=11i^+7j^+3k^\vec{b} + \vec{c} = 11\hat{i} + 7\hat{j} + 3\hat{k}. Dividing AMAM internally in the ratio 2:12 : 1 gives G=a⃗+2m⃗3=a⃗+b⃗+c⃗3=12i^+9j^+6k^3=(4,3,2)G = \dfrac{\vec{a} + 2\vec{m}}{3} = \dfrac{\vec{a} + \vec{b} + \vec{c}}{3} = \dfrac{12\hat{i} + 9\hat{j} + 6\hat{k}}{3} = (4, 3, 2). In general, the centroid has position vector a⃗+b⃗+c⃗3\dfrac{\vec{a} + \vec{b} + \vec{c}}{3}.

Take A=(1,2,3)A = (1, 2, 3), B=(3,5,7)B = (3, 5, 7), C=(5,8,11)C = (5, 8, 11). Then AB→=2i^+3j^+4k^\overrightarrow{AB} = 2\hat{i} + 3\hat{j} + 4\hat{k} and AC→=4i^+6j^+8k^\overrightarrow{AC} = 4\hat{i} + 6\hat{j} + 8\hat{k}. Since AC→=2AB→\overrightarrow{AC} = 2\overrightarrow{AB}, the vectors are parallel and share the point AA, so AA, BB, CC are on one line. Indeed B=(3,5,7)B = (3, 5, 7) is the mid-point of ACAC.

A quick way to remember: the point that divides ABAB in the ratio m:nm : n is the starting point plus mm+n\frac{m}{m + n} of the vector AB→\overrightarrow{AB}.

EXERCISE 10.2

456 words · 2 min read
  1. Find PQ→\overrightarrow{PQ} and ∣PQ→∣|\overrightarrow{PQ}| for P=(2,−1,3)P = (2, -1, 3) and Q=(−1,4,5)Q = (-1, 4, 5).
  2. Find the mid-point of the line segment joining A=(2,4,−6)A = (2, 4, -6) and B=(8,−2,2)B = (8, -2, 2).
  3. Find the point that divides the segment from P=(−1,3,2)P = (-1, 3, 2) to Q=(4,−2,7)Q = (4, -2, 7) internally in the ratio 2:32 : 3.
  4. Find the point that divides the same segment externally in the ratio 2:12 : 1.
  5. ABCDABCD is a parallelogram with A=(1,1,1)A = (1, 1, 1), B=(3,2,0)B = (3, 2, 0), C=(6,5,2)C = (6, 5, 2). Find DD.
  6. Show that A=(1,−1,2)A = (1, -1, 2), B=(3,1,6)B = (3, 1, 6), C=(7,5,14)C = (7, 5, 14) are on one line.
  7. For a⃗=2i^−j^+3k^\vec{a} = 2\hat{i} - \hat{j} + 3\hat{k} and b⃗=i^+4j^−2k^\vec{b} = \hat{i} + 4\hat{j} - 2\hat{k}, find 2a⃗+3b⃗2\vec{a} + 3\vec{b} and its magnitude.
  8. Write 5i^+3j^5\hat{i} + 3\hat{j} as p(i^+2j^)+q(3i^−j^)p(\hat{i} + 2\hat{j}) + q(3\hat{i} - \hat{j}).
  9. In what ratio does the plane y=0y = 0 (the xzxz-plane) divide the segment joining A=(1,2,3)A = (1, 2, 3) and B=(6,−3,8)B = (6, -3, 8)? Where does it meet the segment?
  10. Find the centroid of the triangle with vertices (1,2,3)(1, 2, 3), (4,6,3)(4, 6, 3), (7,1,0)(7, 1, 0).

Answers

  1. PQ→=−3i^+5j^+2k^\overrightarrow{PQ} = -3\hat{i} + 5\hat{j} + 2\hat{k} and ∣PQ→∣=9+25+4=38|\overrightarrow{PQ}| = \sqrt{9 + 25 + 4} = \sqrt{38}.
  2. a⃗+b⃗2=10i^+2j^−4k^2=(5,1,−2)\dfrac{\vec{a} + \vec{b}}{2} = \dfrac{10\hat{i} + 2\hat{j} - 4\hat{k}}{2} = (5, 1, -2).
  3. 3p⃗+2q⃗5=5i^+5j^+20k^5=(1,1,4)\dfrac{3\vec{p} + 2\vec{q}}{5} = \dfrac{5\hat{i} + 5\hat{j} + 20\hat{k}}{5} = (1, 1, 4).
  4. 2q⃗−p⃗2−1=9i^−7j^+12k^=(9,−7,12)\dfrac{2\vec{q} - \vec{p}}{2 - 1} = 9\hat{i} - 7\hat{j} + 12\hat{k} = (9, -7, 12).
  5. The diagonals of a parallelogram bisect each other, so A+C=B+DA + C = B + D, so D=A+C−B=(4,4,3)D = A + C - B = (4, 4, 3). Check: the mid-point of ACAC and the mid-point of BDBD are both (72,3,32)(\frac{7}{2}, 3, \frac{3}{2}).
  6. AB→=2i^+2j^+4k^\overrightarrow{AB} = 2\hat{i} + 2\hat{j} + 4\hat{k} and AC→=6i^+6j^+12k^=3 AB→\overrightarrow{AC} = 6\hat{i} + 6\hat{j} + 12\hat{k} = 3\,\overrightarrow{AB}. The vectors are parallel and share AA, so the three points are collinear.
  7. 2a⃗+3b⃗=7i^+10j^2\vec{a} + 3\vec{b} = 7\hat{i} + 10\hat{j}, with magnitude 49+100=149\sqrt{49 + 100} = \sqrt{149}.
  8. p+3q=5p + 3q = 5 and 2p−q=32p - q = 3 give p=2p = 2, q=1q = 1. Check: 2(i^+2j^)+(3i^−j^)=5i^+3j^2(\hat{i} + 2\hat{j}) + (3\hat{i} - \hat{j}) = 5\hat{i} + 3\hat{j}.
  9. Let the ratio be m:nm : n. The yy-coordinate is n⋅2+m⋅(−3)m+n=0\dfrac{n \cdot 2 + m \cdot (-3)}{m + n} = 0, so 2n=3m2n = 3m and m:n=2:3m : n = 2 : 3. The meeting point is (3,0,5)(3, 0, 5).
  10. a⃗+b⃗+c⃗3=12i^+9j^+6k^3=(4,3,2)\dfrac{\vec{a} + \vec{b} + \vec{c}}{3} = \dfrac{12\hat{i} + 9\hat{j} + 6\hat{k}}{3} = (4, 3, 2).

10.6 Product of Two Vectors

141 words · 1 min read

You can add two vectors and you can scale one. Can you multiply two vectors? There is more than one sensible answer, because two arrows can be combined in two useful ways:

  • The scalar product (or dot product), written a⃗⋅b⃗\vec{a} \cdot \vec{b}. The answer is a number. It measures how much the two vectors point the same way. Example: work $= $ force ⋅\cdot displacement.
  • The vector product (or cross product), written a⃗×b⃗\vec{a} \times \vec{b}. The answer is a vector that is perpendicular to both. It measures how much they point different ways. Examples: the area of a parallelogram, and the turning effect (torque) of a force.
In simple words: the dot product asks "how much do these two arrows agree?". The cross product asks "how much do they spread apart, and in which direction does the spread point?".

10.6.1 Scalar (or dot) product of two vectors

326 words · 1 min read

Let θ\theta be the angle between two non-zero vectors a⃗\vec{a} and b⃗\vec{b} when they are drawn from the same point, with 0≤θ≤π0 \le \theta \le \pi. The scalar product is

a⃗⋅b⃗=∣a⃗∣ ∣b⃗∣cos⁡θ.\vec{a} \cdot \vec{b} = |\vec{a}|\,|\vec{b}|\cos\theta.

If either vector is the zero vector, the product is 00. The answer is a plain number, which can be positive, zero or negative, depending on whether θ\theta is less than, equal to, or more than 90∘90^{\circ}.

A force F⃗\vec{F} moves an object through a displacement d⃗\vec{d}. Only the part of the force along the motion does work. So the work done is W=∣F⃗∣ ∣d⃗∣cos⁡θ=F⃗⋅d⃗W = |\vec{F}|\,|\vec{d}|\cos\theta = \vec{F} \cdot \vec{d}. For example, a force F⃗=6i^+8j^\vec{F} = 6\hat{i} + 8\hat{j} newtons moves a box by d⃗=5i^\vec{d} = 5\hat{i} metres. The work is 3030 joules: the 88 in the j^\hat{j}-direction is sideways to the motion and contributes nothing.

Writing both vectors in components, the dot product is simply

a⃗⋅b⃗=a1b1+a2b2+a3b3.\vec{a} \cdot \vec{b} = a_1b_1 + a_2b_2 + a_3b_3.

Example. For a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=4i^−j^+2k^\vec{b} = 4\hat{i} - \hat{j} + 2\hat{k}: a⃗⋅b⃗=1⋅4+2⋅(−1)+3⋅2=8\vec{a} \cdot \vec{b} = 1 \cdot 4 + 2 \cdot (-1) + 3 \cdot 2 = 8. The lengths are ∣a⃗∣=14|\vec{a}| = \sqrt{14} and ∣b⃗∣=21|\vec{b}| = \sqrt{21}. So

cos⁡θ=a⃗⋅b⃗∣a⃗∣ ∣b⃗∣=8294\cos\theta = \frac{\vec{a} \cdot \vec{b}}{|\vec{a}|\,|\vec{b}|} = \frac{8}{\sqrt{294}}

So θ≈62.19∘\theta \approx 62.19^{\circ}.

Two vectors from the origin. Their dot product is 8 and the angle between them is about 62.2°.
Common mistake: writing a⃗⋅b⃗\vec{a} \cdot \vec{b} as a vector. It is a number. Also, never write a⃗×b⃗\vec{a} \times \vec{b} when you mean the dot product. The dot and the cross are different operations.

Observations on the dot product

237 words · 1 min read
  • Perpendicular vectors. If a⃗\vec{a} and b⃗\vec{b} are non-zero, then a⃗⋅b⃗=0\vec{a} \cdot \vec{b} = 0 exactly when they are perpendicular (because cos⁡90∘=0\cos 90^{\circ} = 0). Example: i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k} and 2i^+j^−2k^2\hat{i} + \hat{j} - 2\hat{k} give 1⋅2+2⋅1+2⋅(−2)=01 \cdot 2 + 2 \cdot 1 + 2 \cdot (-2) = 0, so they are perpendicular.
  • The sign tells the angle. Positive means θ<90∘\theta < 90^{\circ} (acute). Zero means 90∘90^{\circ}. Negative means θ>90∘\theta > 90^{\circ} (obtuse). Example: i^\hat{i} and −i^+j^-\hat{i} + \hat{j} give −1-1, and the angle is 135∘135^{\circ}.
  • A vector with itself. a⃗⋅a⃗=∣a⃗∣2\vec{a} \cdot \vec{a} = |\vec{a}|^2, since θ=0\theta = 0 and cos⁡0=1\cos 0 = 1. For a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} this is 1+4+9=141 + 4 + 9 = 14. This is how we get the length: ∣a⃗∣=a⃗⋅a⃗|\vec{a}| = \sqrt{\vec{a} \cdot \vec{a}}.
  • Order does not matter. a⃗⋅b⃗=b⃗⋅a⃗\vec{a} \cdot \vec{b} = \vec{b} \cdot \vec{a}.
  • The unit vectors. Each of i^\hat{i}, j^\hat{j}, k^\hat{k} has length 11, and the three are mutually perpendicular. So:
i^\hat{i}j^\hat{j}k^\hat{k}
i^\hat{i}110000
j^\hat{j}001100
k^\hat{k}000011

This table is why the component formula a1b1+a2b2+a3b3a_1b_1 + a_2b_2 + a_3b_3 works: all the cross-terms like i^⋅j^\hat{i} \cdot \hat{j} are zero, and all the square terms are one.

Useful test: to check whether two vectors are perpendicular, compute their dot product. If it is 00, they are.

Two important properties of scalar product

317 words · 1 min read

Property 1 (distributive law). a⃗⋅(b⃗+c⃗)=a⃗⋅b⃗+a⃗⋅c⃗\vec{a} \cdot (\vec{b} + \vec{c}) = \vec{a} \cdot \vec{b} + \vec{a} \cdot \vec{c}. The dot product spreads over addition, like ordinary multiplication.

Check with a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k}, b⃗=4i^−j^+2k^\vec{b} = 4\hat{i} - \hat{j} + 2\hat{k}, c⃗=−2i^+5j^\vec{c} = -2\hat{i} + 5\hat{j}. Then a⃗⋅b⃗=8\vec{a} \cdot \vec{b} = 8 and a⃗⋅c⃗=8\vec{a} \cdot \vec{c} = 8. On the other side, b⃗+c⃗=2i^+4j^+2k^\vec{b} + \vec{c} = 2\hat{i} + 4\hat{j} + 2\hat{k} and a⃗⋅(b⃗+c⃗)=16\vec{a} \cdot (\vec{b} + \vec{c}) = 16, which equals 8+88 + 8.

Property 2 (scalars come out). (λa⃗)⋅b⃗=λ (a⃗⋅b⃗)(\lambda\vec{a}) \cdot \vec{b} = \lambda\,(\vec{a} \cdot \vec{b}). With λ=3\lambda = 3: (3a⃗)⋅b⃗=3i^+6j^+9k^⋅4i^−j^+2k^=24(3\vec{a}) \cdot \vec{b} = 3\hat{i} + 6\hat{j} + 9\hat{k} \cdot 4\hat{i} - \hat{j} + 2\hat{k} = 24, which is 3×83 \times 8.

The component formula. Expand (a1i^+a2j^+a3k^)⋅(b1i^+b2j^+b3k^)(a_1\hat{i} + a_2\hat{j} + a_3\hat{k}) \cdot (b_1\hat{i} + b_2\hat{j} + b_3\hat{k}) with Property 1 into nine terms. The table in the last section kills the six mixed terms and leaves a1b1+a2b2+a3b3a_1b_1 + a_2b_2 + a_3b_3.

Squares of sums. Since a⃗⋅a⃗=∣a⃗∣2\vec{a} \cdot \vec{a} = |\vec{a}|^2, we get

∣a⃗+b⃗∣2=∣a⃗∣2+2 a⃗⋅b⃗+∣b⃗∣2.|\vec{a} + \vec{b}|^2 = |\vec{a}|^2 + 2\,\vec{a} \cdot \vec{b} + |\vec{b}|^2.

Check: a⃗+b⃗=5i^+j^+5k^\vec{a} + \vec{b} = 5\hat{i} + \hat{j} + 5\hat{k} has ∣a⃗+b⃗∣2=51|\vec{a} + \vec{b}|^2 = 51. And ∣a⃗∣2+2a⃗⋅b⃗+∣b⃗∣2=14+16+21=51|\vec{a}|^2 + 2\vec{a} \cdot \vec{b} + |\vec{b}|^2 = 14 + 16 + 21 = 51. In the same way ∣a⃗−b⃗∣2=∣a⃗∣2−2 a⃗⋅b⃗+∣b⃗∣2=19|\vec{a} - \vec{b}|^2 = |\vec{a}|^2 - 2\,\vec{a} \cdot \vec{b} + |\vec{b}|^2 = 19.

Common mistake: writing (a⃗+b⃗)2=a⃗2+b⃗2(\vec{a} + \vec{b})^2 = \vec{a}^2 + \vec{b}^2. The middle term 2 a⃗⋅b⃗2\,\vec{a} \cdot \vec{b} is missing. It is zero only when the vectors are perpendicular.

10.6.2 Projection of a vector on a line

366 words · 2 min read

Shine a light straight down onto a line from the tip of a⃗\vec{a}. The shadow of a⃗\vec{a} on the line is its projection. Suppose the line has the direction of b⃗\vec{b}, and θ\theta is the angle between a⃗\vec{a} and b⃗\vec{b}. Then the (signed) length of the shadow is ∣a⃗∣cos⁡θ|\vec{a}|\cos\theta. This number is called the projection of a⃗\vec{a} on b⃗\vec{b}. It equals

a⃗⋅b⃗∣b⃗∣\frac{\vec{a} \cdot \vec{b}}{|\vec{b}|}

This is a number. It is negative if θ\theta is obtuse (the shadow points backwards). As a vector, the shadow is this length times the unit vector b^\hat{b}:

vector projection=a⃗⋅b⃗∣b⃗∣2 b⃗.\text{vector projection} = \frac{\vec{a} \cdot \vec{b}}{|\vec{b}|^2}\,\vec{b}.

For a⃗=3i^+4j^\vec{a} = 3\hat{i} + 4\hat{j} and b⃗=2i^\vec{b} = 2\hat{i} (along the xx-axis): a⃗⋅b⃗=6\vec{a} \cdot \vec{b} = 6 and ∣b⃗∣=2|\vec{b}| = 2, so the projection is 62=3\dfrac{6}{2} = 3. The vector projection is 3i^3\hat{i}. This simply keeps the xx-part of a⃗\vec{a}.

For a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=4i^+4j^+2k^\vec{b} = 4\hat{i} + 4\hat{j} + 2\hat{k}: a⃗⋅b⃗=18\vec{a} \cdot \vec{b} = 18 and ∣b⃗∣=36=6|\vec{b}| = \sqrt{36} = 6. So the projection is 186=3\dfrac{18}{6} = 3, and the vector projection is 1836 b⃗=2i^+2j^+k^\dfrac{18}{36}\,\vec{b} = 2\hat{i} + 2\hat{j} + \hat{k}.

What is left over, a⃗−2i^+2j^+k^=−i^+2k^\vec{a} - 2\hat{i} + 2\hat{j} + \hat{k} = -\hat{i} + 2\hat{k}, is perpendicular to b⃗\vec{b}: its dot product with b⃗\vec{b} is 00. So a⃗\vec{a} splits into a part along b⃗\vec{b} and a part at right angles to it.

The orange arrow is the projection of a on b. The dashed line from the tip of a meets b at a right angle.
In simple words: the projection tells you how much of a⃗\vec{a} points along b⃗\vec{b}. If the two are perpendicular, the projection is zero.

Observations on projection, angles and inequalities

384 words · 2 min read

The unit vector i^\hat{i} points along the xx-axis, so the angle α\alpha between a⃗=a1i^+a2j^+a3k^\vec{a} = a_1\hat{i} + a_2\hat{j} + a_3\hat{k} and the xx-axis satisfies cos⁡α=a⃗⋅i^∣a⃗∣=a1∣a⃗∣\cos\alpha = \dfrac{\vec{a} \cdot \hat{i}}{|\vec{a}|} = \dfrac{a_1}{|\vec{a}|}. This is the direction cosine from Section 10.2. In the same way cos⁡β=a2∣a⃗∣\cos\beta = \dfrac{a_2}{|\vec{a}|} and cos⁡γ=a3∣a⃗∣\cos\gamma = \dfrac{a_3}{|\vec{a}|}.

If ∣a⃗∣=2|\vec{a}| = 2, ∣b⃗∣=3|\vec{b}| = 3 and a⃗⋅b⃗=3\vec{a} \cdot \vec{b} = 3, then cos⁡θ=32×3=12\cos\theta = \dfrac{3}{2 \times 3} = \dfrac{1}{2}, so θ=60∘\theta = 60^{\circ}.

Since ∣cos⁡θ∣≤1|\cos\theta| \le 1, we always have ∣a⃗⋅b⃗∣≤∣a⃗∣ ∣b⃗∣|\vec{a} \cdot \vec{b}| \le |\vec{a}|\,|\vec{b}|. This is the Cauchy–Schwarz inequality. For a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=4i^−j^+2k^\vec{b} = 4\hat{i} - \hat{j} + 2\hat{k}: ∣a⃗⋅b⃗∣=8|\vec{a} \cdot \vec{b}| = 8, while ∣a⃗∣ ∣b⃗∣=294≈17.146|\vec{a}|\,|\vec{b}| = \sqrt{294} \approx 17.146.

Equality holds exactly when the vectors are parallel. For a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=2i^+4j^+6k^\vec{b} = 2\hat{i} + 4\hat{j} + 6\hat{k}: a⃗⋅b⃗=28\vec{a} \cdot \vec{b} = 28 and ∣a⃗∣2∣b⃗∣2=14×56=784=282|\vec{a}|^2|\vec{b}|^2 = 14 \times 56 = 784 = 28^2.

We know ∣a⃗+b⃗∣≤∣a⃗∣+∣b⃗∣|\vec{a} + \vec{b}| \le |\vec{a}| + |\vec{b}|. Equality needs the two vectors to point the same way. With the vectors just above, a⃗+b⃗=3i^+6j^+9k^\vec{a} + \vec{b} = 3\hat{i} + 6\hat{j} + 9\hat{k}, and ∣a⃗+b⃗∣2=126=9×14|\vec{a} + \vec{b}|^2 = 126 = 9 \times 14, so ∣a⃗+b⃗∣=314=∣a⃗∣+∣b⃗∣|\vec{a} + \vec{b}| = 3\sqrt{14} = |\vec{a}| + |\vec{b}| (because ∣b⃗∣=2∣a⃗∣|\vec{b}| = 2|\vec{a}|).

For any vectors, (a⃗+b⃗)⋅(a⃗−b⃗)=∣a⃗∣2−∣b⃗∣2(\vec{a} + \vec{b}) \cdot (\vec{a} - \vec{b}) = |\vec{a}|^2 - |\vec{b}|^2. So the sum and difference are perpendicular exactly when ∣a⃗∣=∣b⃗∣|\vec{a}| = |\vec{b}|. Example: a⃗=i^+2j^+2k^\vec{a} = \hat{i} + 2\hat{j} + 2\hat{k} and b⃗=2i^+j^−2k^\vec{b} = 2\hat{i} + \hat{j} - 2\hat{k} both have length 33. Then a⃗+b⃗=3i^+3j^\vec{a} + \vec{b} = 3\hat{i} + 3\hat{j}, a⃗−b⃗=−i^+j^+4k^\vec{a} - \vec{b} = -\hat{i} + \hat{j} + 4\hat{k}, and their dot product is 00. This is the fact that the diagonals of a rhombus cross at right angles.

EXERCISE 10.3

607 words · 3 min read
  1. Find a⃗⋅b⃗\vec{a} \cdot \vec{b} and the angle between them for a⃗=i^−j^+2k^\vec{a} = \hat{i} - \hat{j} + 2\hat{k} and b⃗=3i^+2j^−k^\vec{b} = 3\hat{i} + 2\hat{j} - \hat{k}.
  2. Find λ\lambda so that λi^+3j^−2k^\lambda\hat{i} + 3\hat{j} - 2\hat{k} and 2i^−4j^+4k^2\hat{i} - 4\hat{j} + 4\hat{k} are perpendicular.
  3. Show that 3i^−2j^+k^3\hat{i} - 2\hat{j} + \hat{k} and i^+2j^+k^\hat{i} + 2\hat{j} + \hat{k} are perpendicular.
  4. Find the angle between i^+j^\hat{i} + \hat{j} and j^+k^\hat{j} + \hat{k}.
  5. Find the projection of 2i^+3j^+6k^2\hat{i} + 3\hat{j} + 6\hat{k} on i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k}, and the vector projection.
  6. A force 4i^−2j^+6k^4\hat{i} - 2\hat{j} + 6\hat{k} newtons moves an object through 3i^+j^+2k^3\hat{i} + \hat{j} + 2\hat{k} metres. Find the work done.
  7. a⃗=3i^\vec{a} = 3\hat{i} and b⃗=4j^\vec{b} = 4\hat{j}. Show that ∣a⃗+b⃗∣=∣a⃗−b⃗∣|\vec{a} + \vec{b}| = |\vec{a} - \vec{b}|. Why does this happen?
  8. Show that a⃗=2i^+3j^+6k^\vec{a} = 2\hat{i} + 3\hat{j} + 6\hat{k} and b⃗=6i^−3j^+2k^\vec{b} = 6\hat{i} - 3\hat{j} + 2\hat{k} have the same length, and that a⃗+b⃗\vec{a} + \vec{b} is perpendicular to a⃗−b⃗\vec{a} - \vec{b}.
  9. Find the angle between a space diagonal of a cube and one of its edges (take the cube with vertices at 00 and 11).
  10. Check the inequality ∣a⃗⋅b⃗∣≤∣a⃗∣ ∣b⃗∣|\vec{a} \cdot \vec{b}| \le |\vec{a}|\,|\vec{b}| for a⃗=2i^−j^+2k^\vec{a} = 2\hat{i} - \hat{j} + 2\hat{k} and b⃗=i^+4j^−2k^\vec{b} = \hat{i} + 4\hat{j} - 2\hat{k}.
  11. For a⃗=2i^\vec{a} = 2\hat{i} and b⃗=i^+j^\vec{b} = \hat{i} + \hat{j}, find the projection of a⃗\vec{a} on b⃗\vec{b} and of b⃗\vec{b} on a⃗\vec{a}. Are they equal?
  12. The points A=(0,1,1)A = (0, 1, 1), B=(1,3,3)B = (1, 3, 3), C=(2,2,−1)C = (2, 2, -1) form a triangle. Show that it has a right angle at AA.

Answers

  1. a⃗⋅b⃗=3−2−2=−1\vec{a} \cdot \vec{b} = 3 - 2 - 2 = -1. The dot product is negative, so the angle is obtuse: θ≈96.26∘\theta \approx 96.26^{\circ}.
  2. The dot product is 2λ−12−8=02\lambda - 12 - 8 = 0, so λ=10\lambda = 10. The computer confirmed that 10i^+3j^−2k^10\hat{i} + 3\hat{j} - 2\hat{k} and 2i^−4j^+4k^2\hat{i} - 4\hat{j} + 4\hat{k} have dot product 00.
  3. 3⋅1+(−2)⋅2+1⋅1=3−4+1=03 \cdot 1 + (-2) \cdot 2 + 1 \cdot 1 = 3 - 4 + 1 = 0.
  4. The dot product is 11 and each length is 2\sqrt{2}, so cos⁡θ=12\cos\theta = \frac{1}{2} and θ=60∘\theta = 60^{\circ}.
  5. a⃗⋅b⃗=20\vec{a} \cdot \vec{b} = 20 and ∣b⃗∣=3|\vec{b}| = 3, so the projection is 203≈6.6667\frac{20}{3} \approx 6.6667. The vector projection is 209b⃗=209i^+409j^+409k^\frac{20}{9}\vec{b} = \frac{20}{9}\hat{i} + \frac{40}{9}\hat{j} + \frac{40}{9}\hat{k}.
  6. W=F⃗⋅d⃗=12−2+12=22W = \vec{F} \cdot \vec{d} = 12 - 2 + 12 = 22 joules.
  7. Both are 9+16=5\sqrt{9 + 16} = 5. The two vectors are perpendicular, so 2 a⃗⋅b⃗=02\,\vec{a} \cdot \vec{b} = 0 and the cross-terms in ∣a⃗±b⃗∣2|\vec{a} \pm \vec{b}|^2 vanish.
  8. ∣a⃗∣=49=7|\vec{a}| = \sqrt{49} = 7 and ∣b⃗∣=36+9+4=7|\vec{b}| = \sqrt{36 + 9 + 4} = 7. Then a⃗+b⃗=8i^+8k^\vec{a} + \vec{b} = 8\hat{i} + 8\hat{k} and a⃗−b⃗=−4i^+6j^+4k^\vec{a} - \vec{b} = -4\hat{i} + 6\hat{j} + 4\hat{k}, with dot product −32+0+32=0-32 + 0 + 32 = 0.
  9. The diagonal is i^+j^+k^\hat{i} + \hat{j} + \hat{k} and an edge is i^\hat{i}. Then cos⁡θ=13\cos\theta = \frac{1}{\sqrt{3}}, so θ≈54.74∘\theta \approx 54.74^{\circ}.
  10. a⃗⋅b⃗=2−4−4=−6\vec{a} \cdot \vec{b} = 2 - 4 - 4 = -6, so ∣a⃗⋅b⃗∣=6|\vec{a} \cdot \vec{b}| = 6. Also ∣a⃗∣=3|\vec{a}| = 3 and ∣b⃗∣=21|\vec{b}| = \sqrt{21}, so ∣a⃗∣ ∣b⃗∣≈13.748|\vec{a}|\,|\vec{b}| \approx 13.748. Indeed 6≤13.7486 \le 13.748.
  11. Projection of a⃗\vec{a} on b⃗\vec{b}: 22≈1.4142\frac{2}{\sqrt{2}} \approx 1.4142. Projection of b⃗\vec{b} on a⃗\vec{a}: 22=1\frac{2}{2} = 1. They are not equal, because the two lengths in the denominators differ.
  12. AB→=i^+2j^+2k^\overrightarrow{AB} = \hat{i} + 2\hat{j} + 2\hat{k} and AC→=2i^+j^−2k^\overrightarrow{AC} = 2\hat{i} + \hat{j} - 2\hat{k}. Their dot product is 2+2−4=02 + 2 - 4 = 0, so the sides at AA are perpendicular.

10.6.3 Vector (or cross) product of two vectors

417 words · 2 min read

The vector product of two non-zero vectors a⃗\vec{a} and b⃗\vec{b}, with angle θ\theta between them (0≤θ≤π0 \le \theta \le \pi), is

a⃗×b⃗=∣a⃗∣ ∣b⃗∣sin⁡θ  n^.\vec{a} \times \vec{b} = |\vec{a}|\,|\vec{b}|\sin\theta\;\hat{n}.

Here n^\hat{n} is a unit vector perpendicular to both a⃗\vec{a} and b⃗\vec{b}. Its direction follows the right-hand rule: curl the fingers of your right hand from a⃗\vec{a} towards b⃗\vec{b}. Your thumb points along n^\hat{n}. If a⃗\vec{a} and b⃗\vec{b} are parallel, or one is zero, then a⃗×b⃗=0⃗\vec{a} \times \vec{b} = \vec{0}.

The size is an area. ∣a⃗×b⃗∣=∣a⃗∣ ∣b⃗∣sin⁡θ|\vec{a} \times \vec{b}| = |\vec{a}|\,|\vec{b}|\sin\theta is exactly the area of the parallelogram with sides a⃗\vec{a} and b⃗\vec{b} (base ∣a⃗∣|\vec{a}| times height ∣b⃗∣sin⁡θ|\vec{b}|\sin\theta).

a = (2, 0, 0) and b = (1, 2, 0) lie on the floor. Their cross product points straight up and has length 4, which is the area of the parallelogram (base 2 × height 2).

Using the products i^×j^=k^\hat{i} \times \hat{j} = \hat{k}, j^×k^=i^\hat{j} \times \hat{k} = \hat{i}, k^×i^=j^\hat{k} \times \hat{i} = \hat{j} (and the reverse orders giving the negatives), one finds

a⃗×b⃗=(a2b3−a3b2)i^+(a3b1−a1b3)j^+(a1b2−a2b1)k^\begin{aligned}\vec{a} \times \vec{b} = {} & (a_2b_3 - a_3b_2)\hat{i} \\ & + (a_3b_1 - a_1b_3)\hat{j} \\ & + (a_1b_2 - a_2b_1)\hat{k}\end{aligned}

This is the expansion of the determinant ∣i^j^k^a1a2a3b1b2b3∣\begin{vmatrix} \hat{i} & \hat{j} & \hat{k} \\ a_1 & a_2 & a_3 \\ b_1 & b_2 & b_3 \end{vmatrix}.

Let a⃗=2i^+j^+3k^\vec{a} = 2\hat{i} + \hat{j} + 3\hat{k} and b⃗=3i^+5j^−2k^\vec{b} = 3\hat{i} + 5\hat{j} - 2\hat{k}. Then

  • i^\hat{i}-part: 1⋅(−2)−3⋅5=−171 \cdot (-2) - 3 \cdot 5 = -17.
  • j^\hat{j}-part: 3⋅3−2⋅(−2)=133 \cdot 3 - 2 \cdot (-2) = 13.
  • k^\hat{k}-part: 2⋅5−1⋅3=72 \cdot 5 - 1 \cdot 3 = 7.

So a⃗×b⃗=−17i^+13j^+7k^\vec{a} \times \vec{b} = -17\hat{i} + 13\hat{j} + 7\hat{k}, with magnitude 289+169+49=507=133\sqrt{289 + 169 + 49} = \sqrt{507} = 13\sqrt{3}. As a check, its dot product with a⃗\vec{a} is 00 and with b⃗\vec{b} is 00, so it is perpendicular to both.

Order matters! a⃗×b⃗\vec{a} \times \vec{b} and b⃗×a⃗\vec{b} \times \vec{a} point in opposite directions. Swapping the factors changes the sign. This is very different from numbers, and from the dot product.

Observations on the cross product

432 words · 2 min read
  • Swapping flips the sign. b⃗×a⃗=−(a⃗×b⃗)\vec{b} \times \vec{a} = -(\vec{a} \times \vec{b}). For the vectors in the last section, b⃗×a⃗=17i^−13j^−7k^\vec{b} \times \vec{a} = 17\hat{i} - 13\hat{j} - 7\hat{k}.
  • A vector crossed with itself is zero. a⃗×a⃗=0⃗\vec{a} \times \vec{a} = \vec{0}, because sin⁡0=0\sin 0 = 0.
  • Parallel vectors have cross product zero. For non-zero vectors, a⃗×b⃗=0⃗\vec{a} \times \vec{b} = \vec{0} exactly when a⃗∥b⃗\vec{a} \parallel \vec{b}. Example: 2i^−4j^+6k^2\hat{i} - 4\hat{j} + 6\hat{k} and −i^+2j^−3k^-\hat{i} + 2\hat{j} - 3\hat{k} give 0⃗\vec{0}.
  • Scalars come out. (λa⃗)×b⃗=λ(a⃗×b⃗)(\lambda\vec{a}) \times \vec{b} = \lambda(\vec{a} \times \vec{b}). Check with λ=3\lambda = 3: both sides are −51i^+39j^+21k^-51\hat{i} + 39\hat{j} + 21\hat{k}.
  • Distributive law. a⃗×(b⃗+c⃗)=a⃗×b⃗+a⃗×c⃗\vec{a} \times (\vec{b} + \vec{c}) = \vec{a} \times \vec{b} + \vec{a} \times \vec{c}. Check with a⃗=i^+2j^\vec{a} = \hat{i} + 2\hat{j}, b⃗=j^+3k^\vec{b} = \hat{j} + 3\hat{k}, c⃗=2i^+k^\vec{c} = 2\hat{i} + \hat{k}: both sides equal 8i^−4j^−3k^8\hat{i} - 4\hat{j} - 3\hat{k}.

  • Area of a parallelogram with adjacent sides a⃗\vec{a}, b⃗\vec{b}: ∣a⃗×b⃗∣|\vec{a} \times \vec{b}|.
  • Area of a triangle with sides a⃗\vec{a}, b⃗\vec{b} from one vertex: 12∣a⃗×b⃗∣\frac{1}{2}|\vec{a} \times \vec{b}|.
  • A perpendicular direction. a⃗×b⃗\vec{a} \times \vec{b} is perpendicular to both. Dividing by its length gives a unit vector perpendicular to both.
  • The sine of the angle. sin⁡θ=∣a⃗×b⃗∣∣a⃗∣ ∣b⃗∣\sin\theta = \dfrac{|\vec{a} \times \vec{b}|}{|\vec{a}|\,|\vec{b}|}.

Example: area of a triangle. The triangle has corners A=(1,0,0)A = (1, 0, 0), B=(2,3,1)B = (2, 3, 1), C=(0,1,4)C = (0, 1, 4). Then AB→=i^+3j^+k^\overrightarrow{AB} = \hat{i} + 3\hat{j} + \hat{k} and AC→=−i^+j^+4k^\overrightarrow{AC} = -\hat{i} + \hat{j} + 4\hat{k}, so AB→×AC→=11i^−5j^+4k^\overrightarrow{AB} \times \overrightarrow{AC} = 11\hat{i} - 5\hat{j} + 4\hat{k}. Its length is 162=92\sqrt{162} = 9\sqrt{2}. The area of the triangle is half of this: 922≈6.364\dfrac{9\sqrt{2}}{2} \approx 6.364.

Lagrange's identity. Adding the dot and cross products gives a neat link: ∣a⃗×b⃗∣2=∣a⃗∣2∣b⃗∣2−(a⃗⋅b⃗)2|\vec{a} \times \vec{b}|^2 = |\vec{a}|^2|\vec{b}|^2 - (\vec{a} \cdot \vec{b})^2. It comes from sin⁡2θ=1−cos⁡2θ\sin^2\theta = 1 - \cos^2\theta. Check with a⃗=i^+2j^+2k^\vec{a} = \hat{i} + 2\hat{j} + 2\hat{k} and b⃗=3i^+4k^\vec{b} = 3\hat{i} + 4\hat{k}: a⃗×b⃗=8i^+2j^−6k^\vec{a} \times \vec{b} = 8\hat{i} + 2\hat{j} - 6\hat{k}, so ∣a⃗×b⃗∣2=104|\vec{a} \times \vec{b}|^2 = 104. On the other side, 9⋅25−112=225−121=1049 \cdot 25 - 11^2 = 225 - 121 = 104.

Example: the angle. For a⃗=i^\vec{a} = \hat{i} and b⃗=i^+j^\vec{b} = \hat{i} + \hat{j}: ∣a⃗×b⃗∣2=1|\vec{a} \times \vec{b}|^2 = 1 and ∣a⃗∣∣b⃗∣=2|\vec{a}||\vec{b}| = \sqrt{2}, so sin⁡θ=12\sin\theta = \frac{1}{\sqrt{2}} and θ=45∘\theta = 45^{\circ}.

In simple words: the dot product measures how much two arrows line up. The cross product measures how much they fail to line up. The size is an area and the direction is a right angle to both.

EXERCISE 10.4

664 words · 3 min read
  1. Find a⃗×b⃗\vec{a} \times \vec{b} for a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k} and b⃗=4i^−j^+2k^\vec{b} = 4\hat{i} - \hat{j} + 2\hat{k}.
  2. Find ∣a⃗×b⃗∣|\vec{a} \times \vec{b}| for a⃗=2i^+j^\vec{a} = 2\hat{i} + \hat{j} and b⃗=3j^+4k^\vec{b} = 3\hat{j} + 4\hat{k}.
  3. Find a unit vector perpendicular to both i^+j^\hat{i} + \hat{j} and j^+k^\hat{j} + \hat{k}.
  4. Find the area of the parallelogram with adjacent sides 3i^+k^3\hat{i} + \hat{k} and i^+2j^\hat{i} + 2\hat{j}.
  5. Find the area of the triangle with vertices (1,1,1)(1, 1, 1), (3,2,4)(3, 2, 4), (2,5,3)(2, 5, 3).
  6. Show that 2i^−4j^+6k^2\hat{i} - 4\hat{j} + 6\hat{k} and −i^+2j^−3k^-\hat{i} + 2\hat{j} - 3\hat{k} are parallel by computing their cross product.
  7. Show that a⃗×b⃗\vec{a} \times \vec{b} from Question 1 is perpendicular to both a⃗\vec{a} and b⃗\vec{b}.
  8. Check Lagrange's identity ∣a⃗×b⃗∣2=∣a⃗∣2∣b⃗∣2−(a⃗⋅b⃗)2|\vec{a} \times \vec{b}|^2 = |\vec{a}|^2|\vec{b}|^2 - (\vec{a} \cdot \vec{b})^2 for a⃗=i^+2j^+2k^\vec{a} = \hat{i} + 2\hat{j} + 2\hat{k} and b⃗=3i^+4k^\vec{b} = 3\hat{i} + 4\hat{k}.
  9. Find sin⁡θ\sin\theta for the angle θ\theta between i^\hat{i} and i^+j^\hat{i} + \hat{j}, using the cross product.
  10. A wrench has the lever arm r⃗=2i^\vec{r} = 2\hat{i} metres. A force F⃗=3j^\vec{F} = 3\hat{j} newtons is applied at its end. The torque (turning effect) is r⃗×F⃗\vec{r} \times \vec{F}. Find it.
  11. A parallelogram has sides a⃗=2i^+j^\vec{a} = 2\hat{i} + \hat{j} and b⃗=3j^+k^\vec{b} = 3\hat{j} + \hat{k}. Find its area two ways: from a⃗×b⃗\vec{a} \times \vec{b}, and from half the cross product of the diagonals.
  12. Find xx so that i^+xj^+2k^\hat{i} + x\hat{j} + 2\hat{k} and 2i^+4j^+4k^2\hat{i} + 4\hat{j} + 4\hat{k} have cross product 0⃗\vec{0}.

Answers

  1. i^\hat{i}: 2⋅2−3⋅(−1)=72 \cdot 2 - 3 \cdot (-1) = 7. j^\hat{j}: 3⋅4−1⋅2=103 \cdot 4 - 1 \cdot 2 = 10. k^\hat{k}: 1⋅(−1)−2⋅4=−91 \cdot (-1) - 2 \cdot 4 = -9. So a⃗×b⃗=7i^+10j^−9k^\vec{a} \times \vec{b} = 7\hat{i} + 10\hat{j} - 9\hat{k}.
  2. a⃗×b⃗=4i^−8j^+6k^\vec{a} \times \vec{b} = 4\hat{i} - 8\hat{j} + 6\hat{k}, so ∣a⃗×b⃗∣=16+64+36=116=229|\vec{a} \times \vec{b}| = \sqrt{16 + 64 + 36} = \sqrt{116} = 2\sqrt{29}.
  3. a⃗×b⃗=i^−j^+k^\vec{a} \times \vec{b} = \hat{i} - \hat{j} + \hat{k}, with length 3\sqrt{3}. A unit vector is 13(i^−j^+k^)\dfrac{1}{\sqrt{3}}\left(\hat{i} - \hat{j} + \hat{k}\right). (The negative of this vector also works.)
  4. a⃗×b⃗=−2i^+j^+6k^\vec{a} \times \vec{b} = -2\hat{i} + \hat{j} + 6\hat{k}, so the area is 4+1+36=41≈6.403\sqrt{4 + 1 + 36} = \sqrt{41} \approx 6.403 square units.
  5. AB→=2i^+j^+3k^\overrightarrow{AB} = 2\hat{i} + \hat{j} + 3\hat{k}, AC→=i^+4j^+2k^\overrightarrow{AC} = \hat{i} + 4\hat{j} + 2\hat{k}, and AB→×AC→=−10i^−j^+7k^\overrightarrow{AB} \times \overrightarrow{AC} = -10\hat{i} - \hat{j} + 7\hat{k} with length 150=56\sqrt{150} = 5\sqrt{6}. The area is 562≈6.124\dfrac{5\sqrt{6}}{2} \approx 6.124.
  6. The cross product is 0⃗\vec{0}, the zero vector, so they are parallel.
  7. 7i^+10j^−9k^⋅i^+2j^+3k^=7+20−27=07\hat{i} + 10\hat{j} - 9\hat{k} \cdot \hat{i} + 2\hat{j} + 3\hat{k} = 7 + 20 - 27 = 0 and 7i^+10j^−9k^⋅4i^−j^+2k^=28−10−18=07\hat{i} + 10\hat{j} - 9\hat{k} \cdot 4\hat{i} - \hat{j} + 2\hat{k} = 28 - 10 - 18 = 0.
  8. ∣a⃗∣2=9|\vec{a}|^2 = 9, ∣b⃗∣2=25|\vec{b}|^2 = 25 and a⃗⋅b⃗=11\vec{a} \cdot \vec{b} = 11. Right side: 225−121=104225 - 121 = 104. Left side: a⃗×b⃗=8i^+2j^−6k^\vec{a} \times \vec{b} = 8\hat{i} + 2\hat{j} - 6\hat{k} and ∣a⃗×b⃗∣2=104|\vec{a} \times \vec{b}|^2 = 104. They agree.
  9. ∣i^×(i^+j^)∣=∣k^∣=1|\hat{i} \times (\hat{i} + \hat{j})| = |\hat{k}| = 1 and ∣i^∣ ∣i^+j^∣=2|\hat{i}|\,|\hat{i} + \hat{j}| = \sqrt{2}, so sin⁡θ=12\sin\theta = \frac{1}{\sqrt{2}} and θ=45∘\theta = 45^{\circ}.
  10. r⃗×F⃗=6k^\vec{r} \times \vec{F} = 6\hat{k} newton-metres. It points along the zz-axis, the axis the wrench turns about, and has size 66 (lever arm 22 times force 33, since they are perpendicular).
  11. a⃗×b⃗=i^−2j^+6k^\vec{a} \times \vec{b} = \hat{i} - 2\hat{j} + 6\hat{k}, so the area is 41\sqrt{41}. The diagonals are 2i^+4j^+k^2\hat{i} + 4\hat{j} + \hat{k} and 2i^−2j^−k^2\hat{i} - 2\hat{j} - \hat{k}, with cross product −2i^+4j^−12k^-2\hat{i} + 4\hat{j} - 12\hat{k} and length 164=241\sqrt{164} = 2\sqrt{41}. Half of it is 41\sqrt{41}, the same.
  12. The second vector is twice i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k}, so the first must be this vector too: x=2x = 2. (The computer confirmed that i^+2j^+2k^\hat{i} + 2\hat{j} + 2\hat{k} and 2i^+4j^+4k^2\hat{i} + 4\hat{j} + 4\hat{k} are parallel.)

Miscellaneous Examples

376 words · 2 min read

Show that (a⃗−b⃗)×(a⃗+b⃗)=2 (a⃗×b⃗)(\vec{a} - \vec{b}) \times (\vec{a} + \vec{b}) = 2\,(\vec{a} \times \vec{b}). Expand with the distributive law: a⃗×a⃗+a⃗×b⃗−b⃗×a⃗−b⃗×b⃗\vec{a} \times \vec{a} + \vec{a} \times \vec{b} - \vec{b} \times \vec{a} - \vec{b} \times \vec{b}. Since a⃗×a⃗=b⃗×b⃗=0⃗\vec{a} \times \vec{a} = \vec{b} \times \vec{b} = \vec{0} and −b⃗×a⃗=a⃗×b⃗-\vec{b} \times \vec{a} = \vec{a} \times \vec{b}, the result is 2(a⃗×b⃗)2(\vec{a} \times \vec{b}). Check with a⃗=i^+2j^\vec{a} = \hat{i} + 2\hat{j}, b⃗=j^+3k^\vec{b} = \hat{j} + 3\hat{k}: both sides equal 12i^−6j^+2k^12\hat{i} - 6\hat{j} + 2\hat{k}.

Find a unit vector perpendicular to both 2i^−j^+k^2\hat{i} - \hat{j} + \hat{k} and i^+2j^−k^\hat{i} + 2\hat{j} - \hat{k}. Their cross product is −i^+3j^+5k^-\hat{i} + 3\hat{j} + 5\hat{k} and its length is 35\sqrt{35}. So the unit vector is 135(−i^+3j^+5k^)\dfrac{1}{\sqrt{35}}\left(-\hat{i} + 3\hat{j} + 5\hat{k}\right).

Suppose a⃗+b⃗+c⃗=0⃗\vec{a} + \vec{b} + \vec{c} = \vec{0} and the lengths are 33, 44, 55. Find a⃗⋅b⃗+b⃗⋅c⃗+c⃗⋅a⃗\vec{a} \cdot \vec{b} + \vec{b} \cdot \vec{c} + \vec{c} \cdot \vec{a}. Square the sum: 0=∣a⃗+b⃗+c⃗∣2=∣a⃗∣2+∣b⃗∣2+∣c⃗∣2+2(a⃗⋅b⃗+b⃗⋅c⃗+c⃗⋅a⃗)0 = |\vec{a} + \vec{b} + \vec{c}|^2 = |\vec{a}|^2 + |\vec{b}|^2 + |\vec{c}|^2 + 2(\vec{a} \cdot \vec{b} + \vec{b} \cdot \vec{c} + \vec{c} \cdot \vec{a}). So the sum of dot products is −9+16+252=−25-\dfrac{9 + 16 + 25}{2} = -25. Example: a⃗=3i^\vec{a} = 3\hat{i}, b⃗=4j^\vec{b} = 4\hat{j}, c⃗=−3i^−4j^\vec{c} = -3\hat{i} - 4\hat{j} add to 0⃗\vec{0} and the dot products total −25-25.

Take A=(0,0,0)A = (0, 0, 0), B=(2,1,0)B = (2, 1, 0), C=(3,3,0)C = (3, 3, 0), D=(1,2,0)D = (1, 2, 0). Then AB→=2i^+j^\overrightarrow{AB} = 2\hat{i} + \hat{j} and DC→=2i^+j^\overrightarrow{DC} = 2\hat{i} + \hat{j} are equal, so ABCDABCD is a parallelogram. Its area is ∣AB→×AD→∣=∣3k^∣=3|\overrightarrow{AB} \times \overrightarrow{AD}| = |3\hat{k}| = 3.

A box of weight W⃗=−10k^\vec{W} = -10\hat{k} units rests on a ramp that points along d⃗=4i^+3k^\vec{d} = 4\hat{i} + 3\hat{k} (44 along, 33 up). The part of the weight that pulls the box along the ramp is the projection of W⃗\vec{W} on d⃗\vec{d}: W⃗⋅d⃗∣d⃗∣=−305=−6\dfrac{\vec{W} \cdot \vec{d}}{|\vec{d}|} = \dfrac{-30}{5} = -6. The minus sign means it pulls down the slope, with strength 66.

Miscellaneous Exercise on Chapter 10

481 words · 2 min read
  1. Find a⃗⋅b⃗\vec{a} \cdot \vec{b} and a⃗×b⃗\vec{a} \times \vec{b} for a⃗=2i^−3j^+k^\vec{a} = 2\hat{i} - 3\hat{j} + \hat{k} and b⃗=−i^+j^+2k^\vec{b} = -\hat{i} + \hat{j} + 2\hat{k}.
  2. Find λ\lambda so that a⃗+λb⃗\vec{a} + \lambda\vec{b} is perpendicular to c⃗\vec{c}, where a⃗=i^+2j^+3k^\vec{a} = \hat{i} + 2\hat{j} + 3\hat{k}, b⃗=i^−j^\vec{b} = \hat{i} - \hat{j}, c⃗=2i^+j^\vec{c} = 2\hat{i} + \hat{j}.
  3. Show that the points (1,0,0)(1, 0, 0), (0,1,0)(0, 1, 0), (0,0,1)(0, 0, 1) are the corners of an equilateral triangle, and find its area.
  4. Two unit vectors make an angle of 60∘60^{\circ}. Find the lengths of their sum and difference.
  5. If a⃗⋅b⃗=a⃗⋅c⃗\vec{a} \cdot \vec{b} = \vec{a} \cdot \vec{c} and a⃗×b⃗=a⃗×c⃗\vec{a} \times \vec{b} = \vec{a} \times \vec{c} with a⃗≠0⃗\vec{a} \ne \vec{0}, show that b⃗=c⃗\vec{b} = \vec{c}.
  6. Find the unit vector along the bisector of the angle between a⃗=i^+2j^+2k^\vec{a} = \hat{i} + 2\hat{j} + 2\hat{k} and b⃗=2i^+j^−2k^\vec{b} = 2\hat{i} + \hat{j} - 2\hat{k}, which have equal lengths.
  7. A parallelogram has sides 4i^4\hat{i} and i^+3j^\hat{i} + 3\hat{j}. Show that the cross product gives the same area as base ×\times height.
  8. Show that ∣a⃗×b⃗∣2+(a⃗⋅b⃗)2=∣a⃗∣2∣b⃗∣2|\vec{a} \times \vec{b}|^2 + (\vec{a} \cdot \vec{b})^2 = |\vec{a}|^2|\vec{b}|^2 in general.

Answers

  1. a⃗⋅b⃗=−2−3+2=−3\vec{a} \cdot \vec{b} = -2 - 3 + 2 = -3. a⃗×b⃗=−7i^−5j^−k^\vec{a} \times \vec{b} = -7\hat{i} - 5\hat{j} - \hat{k}.
  2. a⃗+λb⃗=(1+λ)i^+(2−λ)j^+3k^\vec{a} + \lambda\vec{b} = (1 + \lambda)\hat{i} + (2 - \lambda)\hat{j} + 3\hat{k}. Its dot product with c⃗\vec{c} is 2+2λ+2−λ=4+λ2 + 2\lambda + 2 - \lambda = 4 + \lambda. This is 00 for λ=−4\lambda = -4, and then a⃗+λb⃗=−3i^+6j^+3k^\vec{a} + \lambda\vec{b} = -3\hat{i} + 6\hat{j} + 3\hat{k}.
  3. Each side vector, such as −i^+j^-\hat{i} + \hat{j}, has length 2\sqrt{2}, so all three sides are equal. The area is 12∣AB→×AC→∣=12∣i^+j^+k^∣=32\frac{1}{2}|\overrightarrow{AB} \times \overrightarrow{AC}| = \frac{1}{2}|\hat{i} + \hat{j} + \hat{k}| = \dfrac{\sqrt{3}}{2}.
  4. ∣a⃗+b⃗∣2=1+1+2cos⁡60∘=3|\vec{a} + \vec{b}|^2 = 1 + 1 + 2\cos 60^{\circ} = 3, so ∣a⃗+b⃗∣=3≈1.7321|\vec{a} + \vec{b}| = \sqrt{3} \approx 1.7321. ∣a⃗−b⃗∣2=2−2cos⁡60∘=1|\vec{a} - \vec{b}|^2 = 2 - 2\cos 60^{\circ} = 1, so ∣a⃗−b⃗∣=1|\vec{a} - \vec{b}| = 1.
  5. Let d⃗=b⃗−c⃗\vec{d} = \vec{b} - \vec{c}. Then a⃗⋅d⃗=0\vec{a} \cdot \vec{d} = 0 and a⃗×d⃗=0⃗\vec{a} \times \vec{d} = \vec{0}. The first says d⃗\vec{d} is perpendicular to a⃗\vec{a} (or zero). The second says d⃗\vec{d} is parallel to a⃗\vec{a} (or zero). Only the zero vector is both, so d⃗=0⃗\vec{d} = \vec{0}.
  6. Both have length 33. For equal lengths, a⃗+b⃗\vec{a} + \vec{b} is along the bisector (the diagonal of a rhombus). a⃗+b⃗=3i^+3j^\vec{a} + \vec{b} = 3\hat{i} + 3\hat{j} has length 18=32\sqrt{18} = 3\sqrt{2}. The unit vector is 12(i^+j^)\dfrac{1}{\sqrt{2}}(\hat{i} + \hat{j}).
  7. a⃗×b⃗=12k^\vec{a} \times \vec{b} = 12\hat{k}, so the area is 1212. The base is 44 and the height (the yy-component of b⃗\vec{b}) is 33, and 4×3=124 \times 3 = 12.
  8. ∣a⃗×b⃗∣2=∣a⃗∣2∣b⃗∣2sin⁡2θ|\vec{a} \times \vec{b}|^2 = |\vec{a}|^2|\vec{b}|^2\sin^2\theta and (a⃗⋅b⃗)2=∣a⃗∣2∣b⃗∣2cos⁡2θ(\vec{a} \cdot \vec{b})^2 = |\vec{a}|^2|\vec{b}|^2\cos^2\theta. Adding and using sin⁡2θ+cos⁡2θ=1\sin^2\theta + \cos^2\theta = 1 gives the result.

Summary

320 words · 1 min read
  • A vector has magnitude and direction. A scalar has magnitude only.
  • Position vector of P=(x,y,z)P = (x, y, z): OP→=xi^+yj^+zk^\overrightarrow{OP} = x\hat{i} + y\hat{j} + z\hat{k}, with ∣OP→∣=x2+y2+z2|\overrightarrow{OP}| = \sqrt{x^2 + y^2 + z^2}.
  • Direction cosines l=xrl = \frac{x}{r}, m=yrm = \frac{y}{r}, n=zrn = \frac{z}{r} satisfy l2+m2+n2=1l^2 + m^2 + n^2 = 1. The unit vector is a^=a⃗∣a⃗∣\hat{a} = \frac{\vec{a}}{|\vec{a}|}.
  • Vector joining two points: from PP to QQ is PQ→=q⃗−p⃗\overrightarrow{PQ} = \vec{q} - \vec{p} (end minus start).
  • Addition: triangle law and parallelogram law; add components. It is commutative and associative. ∣a⃗+b⃗∣≤∣a⃗∣+∣b⃗∣|\vec{a} + \vec{b}| \le |\vec{a}| + |\vec{b}|.
  • Scalar multiple: ∣λa⃗∣=∣λ∣∣a⃗∣|\lambda\vec{a}| = |\lambda||\vec{a}|. Vectors are parallel exactly when one is a multiple of the other.
  • Section formula: the point dividing ABAB internally in the ratio m:nm : n is na⃗+mb⃗m+n\frac{n\vec{a} + m\vec{b}}{m + n}. Externally: mb⃗−na⃗m−n\frac{m\vec{b} - n\vec{a}}{m - n}. Centroid of a triangle: a⃗+b⃗+c⃗3\frac{\vec{a} + \vec{b} + \vec{c}}{3}.
  • Dot product: a⃗⋅b⃗=∣a⃗∣∣b⃗∣cos⁡θ=a1b1+a2b2+a3b3\vec{a} \cdot \vec{b} = |\vec{a}||\vec{b}|\cos\theta = a_1b_1 + a_2b_2 + a_3b_3. It is a number. Zero exactly when the vectors are perpendicular. a⃗⋅a⃗=∣a⃗∣2\vec{a} \cdot \vec{a} = |\vec{a}|^2.
  • Projection of one vector on another: a⃗⋅b⃗∣b⃗∣\frac{\vec{a} \cdot \vec{b}}{|\vec{b}|}. The vector projection of a⃗\vec{a} on b⃗\vec{b} is a⃗⋅b⃗∣b⃗∣2b⃗\frac{\vec{a} \cdot \vec{b}}{|\vec{b}|^2}\vec{b}.
  • Cross product: a⃗×b⃗=∣a⃗∣∣b⃗∣sin⁡θ n^\vec{a} \times \vec{b} = |\vec{a}||\vec{b}|\sin\theta\,\hat{n}, perpendicular to both (right-hand rule). In components it is the determinant with rows i^ j^ k^\hat{i}\,\hat{j}\,\hat{k}; a1 a2 a3a_1\,a_2\,a_3; b1 b2 b3b_1\,b_2\,b_3. Swapping reverses the sign. It is zero exactly when the vectors are parallel.
  • Areas: parallelogram =∣a⃗×b⃗∣= |\vec{a} \times \vec{b}|; triangle =12∣a⃗×b⃗∣= \frac{1}{2}|\vec{a} \times \vec{b}|.
  • Lagrange's identity: ∣a⃗×b⃗∣2+(a⃗⋅b⃗)2=∣a⃗∣2∣b⃗∣2|\vec{a} \times \vec{b}|^2 + (\vec{a} \cdot \vec{b})^2 = |\vec{a}|^2|\vec{b}|^2.

Historical Note

241 words · 1 min read

The word vector comes from the Latin vectus, which means "carried": a vector carries you from one point to another.

  • Around 1800. Caspar Wessel (1745–1818) and Jean-Robert Argand (1768–1822) showed that a complex number a+bia + bi can be drawn as an arrow in a plane. This was one of the first geometric uses of directed line segments.
  • Forces. The idea that two pushes on an object combine by the parallelogram rule is very old. It goes back to ancient Greek thinkers, and in the 16th century Simon Stevin showed how to add forces correctly.
  • 1840s and 1850s. William Rowan Hamilton (1805–1865), an Irish mathematician, invented quaternions: numbers of the form a+bi^+cj^+dk^a + b\hat{i} + c\hat{j} + d\hat{k} that follow special rules. He wanted a way to multiply vectors in three dimensions. He was the first to use the word "vector" for a directed line segment, in his Lectures on Quaternions (1853).
  • 1880s. Josiah Willard Gibbs (1839–1903), an American scientist, and Oliver Heaviside (1850–1925), an English engineer, separated the "scalar part" of a quaternion from its "vector part". This created the vector analysis used today, with the dot product and cross product that you learned in this chapter. Gibbs printed his Elements of Vector Analysis in 1881 and 1884.

Today vectors are everywhere: in physics, in engineering, and in every computer game or animation, where a character's position, speed and direction are all stored as vectors.

Up next
Continue to the next chapter