Navigation

Vectors and Geometry in Two and Three Dimensions

1.4 Equations of Planes in 3d

20 problems · hints, answers and solutions shown beside each one

1Stage 1Conceptual

Do you understand the idea? Usually little or no calculation.

Q1Stage 1

The vector k^\hk is a normal vector (i.e. is perpendicular) to the plane z=0z=0. Find another nonzero vector that is normal to z=0z=0.

Hint

You are looking for a vector that is perpendicular to z=0z=0 and hence is parallel to k^\hk.

Answer

Any vector of the form ck^c\,\hk with c0c\ne 0 and c1c\ne 1 works. Three possible choices are k^-\hk, 2k^2\,\hk, 7.12345k^7.12345\,\hk.

Full solution

We are looking for a vector that is perpendicular to z=0z=0 and hence is parallel to k^\hk. To be parallel of k^\hk, the vector has to be of the form ck^c\,\hk for some real number cc. For the vector to be nonzero, we need c0c\ne 0 and for the vector to be different from k^\hk, we need c1c\ne 1. So three possible choices are k^-\hk, 2k^2\,\hk, 7.12345k^7.12345\,\hk.

Q2Stage 1

Consider the plane PP with equation 3x+12y+z=43x+\frac{1}{2}y+z=4.

  1. Find the intersection of PP with the yy-axis.

  2. Find the intersection of PP with the zz-axis.

  3. Sketch the part of the intersection of PP with the yzyz-plane that is in the first octant. (That is, with x,y,z0x,y,z\ge 0.)

Hint

A point (x,y,z)(x,y,z) is on the yy-axis if and only if x=z=0x=z=0. Similarly, a point (x,y,z)(x,y,z) is on the zz-axis if and only if x=y=0x=y=0.

Answer

(a) (0,8,0)(0,8,0) (b) (0,0,4)(0,0,4)

(c)

Figure from prob_s1.4, line 52

Figure from prob_s1.4, line 52

Full solution
  1. Each point on the yy-axis is of the form (0,y,0)(0,y,0). Such a point is on the plane PP if

    3(0)+12y+0=4    y=8\begin{equation*} 3(0)+\frac{1}{2}y+0=4 \iff y=8 \end{equation*}

    So the intersection of PP with the yy-axis is the single point (0,8,0)(0,8,0).

  2. Each point on the zz-axis is of the form (0,0,z)(0,0,z). Such a point is on the plane PP if

    3(0)+12(0)+z=4    z=4\begin{equation*} 3(0)+\frac{1}{2}(0)+z=4 \iff z=4 \end{equation*}

    So the intersection of PP with the zz-axis is the single point (0,0,4)(0,0,4).

  3. The intersection of the plane PP with the yzyz-plane is a line. We have shown in parts (a) and (b) that the points (0,8,0)(0,8,0) and (0,0,4)(0,0,4) are on that line. Here is a sketch of the part of that line that is in the first octant.

    Figure from prob_s1.4, line 52

    Figure from prob_s1.4, line 52

Q3Stage 1
  1. Find the equation of the plane that passes through the origin and has normal vector <1,2,3>\llt 1,2,3\rgt.

  2. Find the equation of the plane that passes through the point (0,0,1)(0,0,1) and has normal vector <1,1,3>\llt 1,1,3\rgt.

  3. Find, if possible, the equation of a plane that passes through both (1,2,3)(1,2,3) and (1,0,0)(1,0,0) and has normal vector <4,5,6>\llt 4,5,6\rgt.

  4. Find, if possible, the equation of a plane that passes through both (1,2,3)(1,2,3) and (0,3,4)(0,3,4) and has normal vector <2,1,1>\llt 2,1,1\rgt.

Answer

(a) x+2y+3z=0x+2y+3z=0 (b) x+y+3z=3x+y+3z=3

(c) There is no plane that passes through both (1,2,3)(1,2,3) and (1,0,0)(1,0,0) and has normal vector <4,5,6>\llt 4,5,6\rgt.

(d) 2x+y+z=72x+y+z=7

Full solution

(a) If (x,y,z)(x,y,z) is any point on the plane, then both the head and the tail of the vector from (x,y,z)(x,y,z) to (0,0,0)(0,0,0), namely <x,y,z>\llt x,y,z\rgt, lie in the plane. So the vector <x,y,z>\llt x,y,z\rgt must be perpendicular to <1,2,3>\llt 1,2,3\rgt and

0=<x,y,z><1,2,3>=x+2y+3z\begin{equation*} 0=\llt x,y,z\rgt\cdot \llt 1,2,3\rgt =x+2y+3z \end{equation*}

(b) If (x,y,z)(x,y,z) is any point on the plane, then both the head and the tail of the vector from (x,y,z)(x,y,z) to (0,0,1)(0,0,1), namely <x,y,z1>\llt x,y,z-1\rgt, lie in the plane. So the vector <x,y,z1>\llt x,y,z-1\rgt must be perpendicular to <1,1,3>\llt 1,1,3\rgt and

0=<x,y,z1><1,1,3>=x+y+3(z1)    x+y+3z=3\begin{equation*} 0=\llt x,y,z-1\rgt\cdot \llt 1,1,3\rgt =x+y+3(z-1) \iff x+y+3z=3 \end{equation*}

(c) If both (1,2,3)(1,2,3) and (1,0,0)(1,0,0) are on the plane, then both the head and the tail of the vector from (1,2,3)(1,2,3) to (1,0,0)(1,0,0), namely <0,2,3>\llt 0,2,3\rgt, lie in the plane. So the vector <0,2,3>\llt 0,2,3\rgt must be perpendicular to <4,5,6>\llt 4,5,6\rgt. As

<0,2,3><4,5,6>=280\begin{equation*} \llt 0,2,3\rgt \cdot \llt 4,5,6\rgt = 28\ne 0 \end{equation*}

the vector <0,2,3>\llt 0,2,3\rgt is not perpendicular to <4,5,6>\llt 4,5,6\rgt. So there is no plane that passes through both (1,2,3)(1,2,3) and (1,0,0)(1,0,0) and has normal vector <4,5,6>\llt 4,5,6\rgt.

(d) If both (1,2,3)(1,2,3) and (0,3,4)(0,3,4) are on the plane, then both the head and the tail of the vector from (1,2,3)(1,2,3) to (0,3,4)(0,3,4), namely <1,1,1>\llt 1,-1,-1\rgt, lie in the plane. So the vector <1,1,1>\llt 1,-1,-1\rgt must be perpendicular to <2,1,1>\llt 2,1,1\rgt. As

<1,1,1><2,1,1>=0\begin{equation*} \llt 1,-1,-1\rgt \cdot \llt 2,1,1\rgt = 0 \end{equation*}

the vector <1,1,1>\llt 1,-1,-1\rgt is indeed perpendicular to <2,1,1>\llt 2,1,1\rgt. So there is a plane that passes through both (1,2,3)(1,2,3) and (0,3,4)(0,3,4) and has normal vector <2,1,1>\llt 2,1,1\rgt. We now just have to build its equation.

If (x,y,z)(x,y,z) is any point on the plane, then both the head and the tail of the vector from (x,y,z)(x,y,z) to (1,2,3)(1,2,3), namely <x1,y2,z3>\llt x-1,y-2,z-3\rgt, lie in the plane. So the vector <x1,y2,z3>\llt x-1,y-2,z-3\rgt must be perpendicular to <2,1,1>\llt 2,1,1\rgt and

0=<x1,y2,z3><2,1,1>=2(x1)+(y2)+(z3)    2x+y+z=7\begin{equation*} 0=\llt x-1,y-2,z-3\rgt\cdot \llt 2,1,1\rgt =2(x-1)+(y-2)+(z-3) \iff 2x+y+z=7 \end{equation*}

As a check, note that both (x,y,z)=(1,2,3)(x,y,z)=(1,2,3) and (x,y,z)=(0,3,4)(x,y,z)=(0,3,4) obey the equation 2x+y+z=72x+y+z=7.

Q4Stage 1Past exam · M200 2008A

Find the equation of the plane that contains (1,0,0)(1,0,0), (0,1,0)(0,1,0) and (0,0,1)(0,0,1).

Hint

Guess.

Answer

x+y+z=1x+y+z=1

Full solution

Solution 1: That's too easy. We just guess. The plane x+y+z=1x+y+z=1 contains all three given points.

Solution 2: The plane does not pass through the origin. (You can see this by just making a quick sketch.) So the plane has an equation of the form ax+by+cz=1ax+by+cz=1.

  • For (1,0,0)(1,0,0) to be on the plane we need that

    a(1)+b(0)+c(0)=1    a=1\begin{equation*} a(1) +b(0) +c(0) =1 \implies a=1 \end{equation*}
  • For (0,1,0)(0,1,0) to be on the plane we need that

    a(0)+b(1)+c(0)=1    b=1\begin{equation*} a(0) +b(1) +c(0) =1 \implies b=1 \end{equation*}
  • For (0,0,1)(0,0,1) to be on the plane we need that

    a(0)+b(0)+c(1)=1    c=1\begin{equation*} a(0) +b(0) +c(1) =1 \implies c=1 \end{equation*}

So the plane is x+y+z=1x+y+z=1.

Solution 3:
Both the head and the tail of the vector from (1,0,0)(1,0,0) to (0,1,0)(0,1,0), namely <1,1,0>\llt -1,1,0\rgt, lie in the plane. Similarly, both the head and the tail of the vector from (1,0,0)(1,0,0) to (0,0,1)(0,0,1), namely <1,0,1>\llt -1,0,1\rgt, lie in the plane. So the vector

<1,1,0>×<1,0,1>=det[ı^ȷ^k^110101]=<1,1,1>\begin{align*} \llt -1,1,0\rgt \times \llt -1,0,1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ -1 & 1 & 0 \\ -1 & 0 & 1 \end{matrix}\right] =\llt 1 , 1 ,1 \rgt \end{align*}

is a normal vector for the plane. As (1,0,0)(1,0,0) is a point in the plane,

<1,1,1><x1,y0,z0>=0orx+y+z=1\begin{align*} \llt 1 , 1 ,1 \rgt \cdot \llt x-1\,,\,y-0\,,\, z-0 \rgt = 0\qquad \text{or}\quad x+y+z=1 \end{align*}

is an equation for the plane.

Q5Stage 1
  1. Find the equation of the plane containing the points (1,0,1)(1,0,1), (1,1,0)(1,1,0) and (0,1,1)(0,1,1).

  2. Is the point (1,1,1)(1,1,1) on the plane?

  3. Is the origin on the plane?

  4. Is the point (4,1,1)(4,-1,-1) on the plane?

Hint

(a) See Question 4 — or just have a guess!

Answer

(a) x+y+z=2x+y+z=2 (b) No. (c) No. (d) Yes.

Full solution

(a) Solution 1: That's too easy. We just guess. The plane x+y+z=2x+y+z=2 contains all three given points.

Solution 2:
Both the head and the tail of the vector from (1,0,1)(1,0,1) to (0,1,1)(0,1,1), namely <1,1,0>\llt 1,-1,0\rgt, lie in the plane. Similarly, both the head and the tail of the vector from (1,1,0)(1,1,0) to (0,1,1)(0,1,1), namely <1,0,1>\llt 1,0,-1\rgt, lie in the plane. So the vector

<1,1,0>×<1,0,1>=det[ı^ȷ^k^110101]=<1,1,1>\begin{align*} \llt 1,-1,0\rgt \times \llt 1,0,-1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & -1 & 0 \\ 1 & 0 & -1 \end{matrix}\right] =\llt 1 , 1 ,1 \rgt \end{align*}

is a normal vector for the plane. As (0,1,1)(0,1,1) is a point in the plane,

<1,1,1><x0,y1,z1>=0orx+y+z=2\begin{align*} \llt 1 , 1 ,1 \rgt \cdot \llt x-0\,,\,y-1\,,\, z-1 \rgt = 0\qquad \text{or}\quad x+y+z=2 \end{align*}

is an equation for the plane.

(b) Since

[x+y+z](x,y,z)=(1,1,1)=32\begin{equation*} \Big[x+y+z\Big]_{(x,y,z)=(1,1,1)} = 3\ne 2 \end{equation*}

the point (1,1,1)(1,1,1) is not on x+y+z=2x+y+z=2.

(c) Since

[x+y+z](x,y,z)=(0,0,0)=02\begin{equation*} \Big[x+y+z\Big]_{(x,y,z)=(0,0,0)} = 0\ne 2 \end{equation*}

the origin is not on x+y+z=2x+y+z=2.

(d) Since

[x+y+z](x,y,z)=(4,1,1)=2\begin{equation*} \Big[x+y+z\Big]_{(x,y,z)=(4,-1,-1)} = 2 \end{equation*}

the point (4,1,1)(4,-1,-1) is on x+y+z=2x+y+z=2.

Q6Stage 1

What's wrong with the following exercise? “Find the equation of the plane containing (1,2,3)(1,2,3), (2,3,4)(2,3,4) and (3,4,5)(3,4,5).”

Hint

Three points don't always determine a plane — why?

Answer

All three points (1,2,3), (2,3,4)(1,2,3),\ (2,3,4) and (3,4,5)(3,4,5) are on the line x(t)=(1,2,3)+t(1,1,1)\vx(t)=(1,2,3)+t(1,1,1). There are many planes through that line.

Full solution

The vector from (1,2,3)(1,2,3) to (2,3,4)(2,3,4), namely <1,1,1>\llt 1,1,1\rgt is parallel to the vector from (1,2,3)(1,2,3) to (3,4,5)(3,4,5), namely <2,2,2>\llt 2,2,2\rgt. So the three given points are collinear. Precisely, all three points (1,2,3), (2,3,4)(1,2,3),\ (2,3,4) and (3,4,5)(3,4,5) are on the line <x1,y2,z3>=t<1,1,1>\llt x-1,y-2,z-3\rgt=t\llt 1,1,1\rgt. There are many planes through that line.

2Stage 2Procedural

Practising the skill itself, until applying it is automatic.

Q7Stage 2

Find the plane containing the given three points.

  1. (1,0,1), (2,4,6), (1,2,1)(1,0,1),\ (2,4,6),\ (1,2,-1)

  2. (1,2,3), (4,4,4), (3,2,3)(1,-2,-3),\ (4,-4,4),\ (3,2,-3)

  3. (1,2,3), (5,2,1), (1,4,5)(1,-2,-3),\ (5,2,1),\ (-1,-4,-5)

Answer

(a) 9xyz=89x-y-z=8 (b) 14x7y8z=5214x-7y-8z=52

(c) For any real numbers aa and bb, the plane
ax+by(a+b)z=4a+bax+by-(a+b)z=4a+b contains the three given points.

Full solution

(a) The plane must be parallel to <2,4,6><1,0,1>=<1,4,5>\llt 2,4,6\rgt -\llt 1,0,1\rgt =\llt 1,4,5\rgt and to <1,2,1><1,0,1>=<0,2,2>\llt 1,2,-1\rgt -\llt 1,0,1\rgt =\llt 0,2,-2\rgt. So its normal vector must be perpendicular to both <1,4,5>\llt 1,4,5\rgt and <0,2,2>\llt 0,2,-2\rgt and hence parallel to

<1,4,5>×<0,2,2>=det[ı^ȷ^k^145022]=<18,2,2>\begin{equation*} \llt 1,4,5\rgt \times\llt 0,2,-2\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & 4 & 5 \\ 0 & 2 & -2 \end{matrix}\right] =\llt -18,2,2\rgt \end{equation*}

The plane is 9(x1)y(z1)=09(x-1)-y-(z-1)=0 or 9xyz=89x-y-z=8.

We can check this by observing that (1,0,1), (2,4,6)(1,0,1),\ (2,4,6) and (1,2,1)(1,2,-1) all satisfy 9xyz=89x-y-z=8.

(b) The plane must be parallel to <4,4,4><1,2,3>=<3,2,7>\llt 4,-4,4\rgt -\llt 1,-2,-3\rgt =\llt 3,-2,7\rgt and to <3,2,3><1,2,3>=<2,4,0>\llt 3,2,-3\rgt -\llt 1,-2,-3\rgt =\llt 2,4,0\rgt. So its normal vector must be perpendicular to both <3,2,7>\llt 3,-2,7\rgt and <2,4,0>\llt 2,4,0\rgt and hence parallel to

<3,2,7>×<2,4,0>=det[ı^ȷ^k^327240]=<28,14,16>\begin{equation*} \llt 3,-2,7\rgt \times\llt 2,4,0\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 3 & -2 & 7 \\ 2 & 4 & 0 \end{matrix}\right] =\llt -28,14,16\rgt \end{equation*}

The plane is 14(x1)7(y+2)8(z+3)=014(x-1)-7(y+2)-8(z+3)=0 or 14x7y8z=5214x-7y-8z=52.

We can check this by observing that (1,2,3), (4,4,4)(1,-2,-3),\ (4,-4,4) and (3,2,3)(3,2,-3) all satisfy 14x7y8z=5214x-7y-8z=52.

(c) The plane must be parallel to <5,2,1><1,2,3>=<4,4,4>\llt 5,2,1\rgt -\llt 1,-2,-3\rgt =\llt 4,4,4\rgt and to <1,4,5><1,2,3>=<2,2,2>\llt -1,-4,-5\rgt -\llt 1,-2,-3\rgt =\llt -2,-2,-2\rgt. My, my. These two vectors are parallel. So the three points are all on the same straight line. Any plane containing the line contains all three points. If <a,b,c>\llt a,b,c\rgt is any vector perpendicular to <1,1,1>\llt 1,1,1\rgt (i.e. which obeys a+b+c=0a+b+c=0) then the plane
a(x1)+b(y+2)+c(z+3)=0a(x-1)+b(y+2)+c(z+3)=0 or a(x1)+b(y+2)(a+b)(z+3)=0a(x-1)+b(y+2)-(a+b)(z+3)=0 or ax+by(a+b)z=4a+bax+by-(a+b)z=4a+b contains the three given points.

We can check this by observing that (1,2,3), (5,2,1)(1,-2,-3),\ (5,2,1) and (1,4,5)(-1,-4,-5) all satisfy the equation ax+by(a+b)z=4a+bax+by-(a+b)z=4a+b for all aa and bb.

Q8Stage 2

Find the distance from the given point to the given plane.

  1. point (1,2,3)(-1,2,3), plane x+y+z=7x+y+z=7

  2. point (1,4,3)(1,-4,3), plane x2y+z=5x-2y+z=5

Answer

(a) 3\sqrt{3} (b) 7/67/\sqrt{6}

Full solution

(a) One point on the plane is (0,0,7)(0,0,7). The vector from (1,2,3)(-1,2,3) to (0,0,7)(0,0,7) is <0,0,7><1,2,3>=<1,2,4>\llt 0,0,7\rgt -\llt -1,2,3\rgt =\llt 1,-2,4\rgt. A unit vector perpendicular to the plane is 13<1,1,1>\frac{1}{\sqrt{3}}\llt 1,1,1\rgt. The distance from
(1,2,3)(-1,2,3) to the plane is the length of the projection of <1,2,4>\llt 1,-2,4\rgt on 13<1,1,1>\frac{1}{\sqrt{3}}\llt 1,1,1\rgt which is

13<1,1,1><1,2,4>=33=3\begin{align*} \frac{1}{\sqrt{3}}\llt 1,1,1\rgt \cdot\llt 1,-2,4\rgt =\frac{3}{\sqrt{3}} =\sqrt{3} \end{align*}

(b) One point on the plane is (0,0,5)(0,0,5). The vector from (1,4,3)(1,-4,3) to (0,0,5)(0,0,5) is <0,0,5><1,4,3>=<1,4,2>\llt 0,0,5\rgt -\llt 1,-4,3\rgt =\llt -1,4,2\rgt. A unit vector perpendicular to the plane is 16<1,2,1>\frac{1}{\sqrt{6}}\llt 1,-2,1\rgt. The distance from
(1,4,3)(1,-4,3) to the plane is the length of the projection of <1,4,2>\llt -1,4,2\rgt on 16<1,2,1>\frac{1}{\sqrt{6}}\llt 1,-2,1\rgt which is the absolute value of

16<1,2,1><1,4,2>=76\begin{equation*} \frac{1}{\sqrt{6}}\llt 1,-2,1\rgt \cdot \llt -1,4,2\rgt =\frac{-7}{\sqrt{6}} \end{equation*}

or 7/67/\sqrt{6}.

Q9Stage 2Past exam · M200 2007A

A plane Π\Pi passes through the points A=(1,1,3)A = (1, 1, 3), B=(2,0,2)B = (2, 0, 2) and C=(2,1,0)C = (2, 1, 0) in R3\bbbr^3.

  1. Find an equation for the plane Π\Pi.

  2. Find the point EE in the plane Π\Pi such that the line LL through D=(6,1,2)D = (6, 1, 2) and EE is perpendicular to Π\Pi.

Answer

(a) 3x+2y+z=83x+2y+z = 8 (b) (3,1,1)\big(3\,,\,-1\,,\,1\big)

Full solution

(a) The vector from CC to AA, namely <12,11,30>=<1,0,3>\llt 1-2\,,\,1-1\,,\,3-0\rgt = \llt -1\,,\,0\,,\,3\rgt lies entirely inside Π\Pi. The vector from CC to BB, namely <22,01,20>=<0,1,2>\llt 2-2\,,\,0-1\,,\,2-0\rgt = \llt 0\,,\,-1\,,\,2\rgt also lies entirely inside Π\Pi. Consequently, the vector

<1,0,3>×<0,1,2>=det[ı^ȷ^k^103012]=<3,2,1>\begin{align*} \llt -1\,,\,0\,,\,3\rgt\times \llt 0\,,\,-1\,,\,2\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ -1 & 0 & 3 \\ 0 & -1 & 2 \end{matrix}\right] =\llt 3 \,,\, 2 \,,\, 1 \rgt \end{align*}

is perpendicular to Π\Pi. The equation of Π\Pi is then

<3,2,1><x2,y1,z>=0or3x+2y+z=8\begin{align*} \llt 3 \,,\, 2 \,,\, 1 \rgt \cdot \llt x-2 \,,\, y-1 \,,\, z \rgt = 0 \qquad\text{or}\qquad 3x+2y+z = 8 \end{align*}

(b) Let EE be (x,y,z)(x,y,z). Then the vector from DD to EE, namely <x6,y1,z2>\llt x-6\,,\,y-1\,,\,z-2\rgt has to be parallel to the vector <3,2,1>\llt 3 \,,\, 2 \,,\, 1 \rgt, which is perpendicular to Π\Pi. That is, there must be a number tt such that

<x6,y1,z2>=t<3,2,1>or x=6+3t, y=1+2t, z=2+t\begin{align*} & \llt x-6\,,\,y-1\,,\,z-2\rgt = t \llt 3 \,,\, 2 \,,\, 1 \rgt \\ &\text{or }x=6+3t,\ y=1+2t,\ z=2+t \end{align*}

As (x,y,z)(x,y,z) must be in Π\Pi,

8=3x+2y+z=3(6+3t)+2(1+2t)+(2+t)=22+14t    t=1\begin{align*} 8 = 3x+2y+z = 3(6+3t) + 2(1+2t) +(2+t) = 22 +14 t \implies t=-1 \end{align*}

So (x,y,z)=(6+3(1),1+2(1),2+(1))=(3,1,1)(x,y,z) = \big(6+3(-1)\,,\,1+2(-1)\,,\,2+(-1)\big) = \big(3\,,\,-1\,,\,1\big).

Q10Stage 2Past exam · M200 2011A

Let A=(2,3,4)A = (2, 3, 4) and let LL be the line given by the equations x+y=1x + y = 1 and x+2y+z=3x + 2y + z = 3.

  1. Write an equation for the plane containing AA and perpendicular to LL.

  2. Write an equation for the plane containing AA and LL.

Answer

(a) xy+z=3x-y+z=3 (b) 5x+y4z=35x+y-4z=-3

Full solution

We are going to need a direction vector for LL in both parts (a) and (b). So we find one first.

  • The vector <1,1,0>\llt 1,1,0\rgt is perpendicular to x+y=1x + y = 1 and hence to LL.

  • The vector <1,2,1>\llt 1,2,1\rgt is perpendicular to x+2y+z=3x + 2y + z = 3 and hence to LL.

So the vector

<1,1,0>×<1,2,1>=det[ı^ȷ^k^110121]=<1,1,1>\begin{align*} \llt 1,1,0\rgt \times \llt 1,2,1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & 1 & 0 \\ 1 & 2 & 1 \end{matrix}\right] =\llt 1 \,,\, -1 \,,\, 1 \rgt \end{align*}

is a direction vector for LL.

(a) The plane is to contain the point (2,3,4)(2,3,4) and is to have <1,1,1>\llt-1,1,-1 \rgt as a normal vector. So

<1,1,1><x2,y3,z4>=0orxy+z=3\begin{align*} \llt-1,1,-1 \rgt\cdot \llt x-2,y-3,z-4 \rgt =0\qquad\text{or}\qquad x-y+z=3 \end{align*}

does the job.

(b) The plane is to contain the points A=(2,3,4)A=(2,3,4) and (1,0,2)(1,0,2) (which is on LL) so that the vector <21,30,42>=<1,3,2>\llt 2-1,3-0,4-2 \rgt =\llt 1,3,2 \rgt is to be parallel to the plane. The direction vector of LL, namely <1,1,1>\llt -1 , 1 , -1 \rgt, is also to be parallel to the plane. So the vector

<1,3,2>×<1,1,1>=det[ı^ȷ^k^132111]=<5,1,4>\begin{align*} \llt 1,3,2\rgt \times \llt -1,1,-1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & 3 & 2 \\ -1 & 1 & -1 \end{matrix}\right] =\llt -5 \,,\, -1 \,,\, 4 \rgt \end{align*}

is to be normal to the plane. So

<5,1,4><x2,y3,z4>=0or5x+y4z=3\begin{align*} \llt-5,-1,4 \rgt\cdot \llt x-2,y-3,z-4 \rgt =0\qquad\text{or}\qquad 5x+y-4z=-3 \end{align*}

does the job.

Q11Stage 2Past exam · M200 2015D

Consider the plane 4x+2y4z=34x + 2y - 4z = 3. Find all parallel planes that are distance 22 from the above plane. Your answers should be in the following form: 4x+2y4z=C4x + 2y - 4z = C.

Answer

4x+2y4z=154x + 2y - 4z=15 and 4x+2y4z=94x + 2y - 4z=-9

Full solution

All planes that are parallel to the plane 4x+2y4z=34x + 2y - 4z = 3 must have <4,2,4>\llt 4\,,\,2\,,\,-4\rgt as a normal vector and hence must have an equation of the form 4x+2y4z=C4x + 2y - 4z = C for some constant CC. We must find the CC's for which the distance from 4x+2y4z=34x + 2y - 4z = 3 to 4x+2y4z=C4x + 2y - 4z = C is 22. One point on 4x+2y4z=34x + 2y - 4z = 3 is (0,32,0)\big(0,\frac{3}{2},0\big). The two points (x,y,z)(x',y',z') with

<x0,y32,z0>vector from (0,32,0) to (x,y,z)=±2 <4,2,4>16+4+16unit vector=±26<4,2,4>=±<43,23,43>\begin{align*} \overbrace{\llt x'-0\,,\,y'-\frac{3}{2}\,,\,z'-0\rgt}^{\text{vector from }\big(0,\frac{3}{2},0\big)\text{ to }(x',y',z')} =\pm 2\ \overbrace{\frac{\llt 4\,,\,2\,,\,-4\rgt}{\sqrt{16+4+16}}}^{\text{unit vector}} =\pm \frac{2}{6} \llt 4\,,\,2\,,\,-4\rgt =\pm\llt \frac{4}{3}\,,\,\frac{2}{3}\,,\,-\frac{4}{3}\rgt \end{align*}

are the two points that are a distance 22 from (0,32,0)\big(0,\frac{3}{2},0\big) in the direction of the normal.The two points (x,y,z)(x',y',z') are

(0+43,32+23,043)=(43,136,43) and (043,3223,0+43)=(43,56,43)\begin{align*} \left(0+\frac{4}{3}\,,\, \frac{3}{2}+\frac{2}{3}\,,\,0-\frac{4}{3}\right) &=\left(\frac{4}{3}\,,\, \frac{13}{6}\,,\,-\frac{4}{3}\right) \\\text{ and } \left(0-\frac{4}{3}\,,\, \frac{3}{2}-\frac{2}{3}\,,\,0+\frac{4}{3}\right) &=\left(-\frac{4}{3}\,,\, \frac{5}{6}\,,\,\frac{4}{3}\right) \end{align*}

These two points lie on the desired planes, so the two desired planes are

4x+2y4z=4×43+2×1364×43=32+26+326=15\begin{align*} 4x + 2y - 4z =\frac{4\times 4}{3} +\frac{2\times 13}{6}-\frac{-4\times 4}{3} =\frac{32+26+32}{6} =15 \end{align*}

and

4x+2y4z=4×(4)3+2×564×43=32+10326=9\begin{align*} 4x + 2y - 4z =\frac{4\times(-4)}{3} +\frac{2\times 5}{6}-\frac{4\times 4}{3} =\frac{-32+10-32}{6} =-9 \end{align*}
Q12Stage 2Past exam · M200 2004A

Find the distance from the point (1,2,3)(1,2,3) to the plane that passes through the points (0,1,1)(0,1,1), (1,1,3)(1,-1,3) and (2,0,1)(2,0,-1).

Answer

22

Full solution

The two vectors

a=<1,1,3><0,1,1>=<1,2,2>b=<2,0,1><0,1,1>=<2,1,2>\begin{alignat*}{3} \va&=\llt 1,-1,3\rgt - \llt 0,1,1\rgt &&= \llt 1,-2, 2\rgt \\ \vb&=\llt 2,0,-1\rgt - \llt 0,1,1\rgt &&= \llt 2,-1,-2\rgt \end{alignat*}

both lie entirely inside the plane. So the vector

a×b=det[ı^ȷ^k^122212]=<6,6,3>\begin{equation*} \va\times\vb =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & -2 & 2 \\ 2 & -1 & -2 \end{matrix}\right] =\llt 6,6,3\rgt \end{equation*}

is perpendicular to the plane. The vector c=13<6,6,3>=<2,2,1>\vc=\frac{1}{3}\llt 6,6,3\rgt =\llt 2,2,1\rgt is also perpendicular to the plane. The vector

d=<1,2,3><0,1,1>=<1,1,2>\begin{equation*} \vd=\llt 1,2,3\rgt-\llt 0,1,1\rgt=\llt 1,1,2\rgt \end{equation*}

joins the point to the plane. So, if θ\theta is the angle between d\vd and c\vc, the distance is

dcosθ=cdc=69=2\begin{equation*} |\vd|\cos\theta=\frac{\vc\cdot\vd}{|\vc|}=\frac{6}{\sqrt{9}}=2 \end{equation*}

3Stage 3Application

Further than practice: several ideas at once, or an unfamiliar situation.

Q13Stage 3Past exam · M200 2014A

Consider two planes W1W_1, W2W_2, and a line MM defined by:

W1 : 2x+y+z=7,W2 : x+3y+3z=6,M : x2=2y44=z+5.\begin{equation*} W_1\ :\ -2x + y + z = 7,\qquad W_2\ :\ -x + 3y + 3z = 6,\qquad M\ :\ \frac{x}{2} = \frac{2y-4}{4} = z + 5. \end{equation*}
  1. Find a parametric equation of the line of intersection LL of W1W_1 and W2W_2.

  2. Find the distance from L to M .

Answer

(a) (x,y,z)=(3,1,0)+t<0,1,1>(x,y,z) = (-3,1,0) +t \llt 0,-1,1\rgt (b) 17\sqrt{17}

Full solution

(a) Let's use zz as the parameter and rename it to tt. That is, z=tz=t. Subtracting 22 times the W2W_2 equation from the W1W_1 equation gives

5y5z=5    y=1z=1t or y1=t\begin{equation*} -5y -5z = -5 \implies y = 1-z = 1-t\quad\text{ or }\quad y-1=-t \end{equation*}

Substituting the result into the equation for W2W_2 gives

x+3(1t)+3t=6    x=3 or x+3=0\begin{align*} -x +3(1-t) +3t =6 \implies x = -3\quad\text{ or }\quad x+3=0 \end{align*}

So a parametric equation is

<x+3,y1,z>=t<0,1,1>\begin{equation*} \llt x+3,y-1,z\rgt = t \llt 0,-1,1\rgt \end{equation*}

(b) Solution 1

We can also parametrize MM using z=tz=t:

x=2z+10=2t+10, y=2z+12=2t+12    <x,y,z>=<10,12,0>+t<2,2,1>\begin{equation*} x=2z+10=2t+10,\ y=2z+12=2t+12\quad\implies\quad \llt x,y,z\rgt = \llt 10,12,0\rgt +t \llt 2,2,1\rgt \end{equation*}

So one point on MM is (10,12,0)(10,12,0) and one point on LL is (3,1,0)(-3,1,0) and

v=<(3)10,112,00>=<13,11,0>\begin{equation*} \vv= \llt (-3)-10\,,\, 1 -12 \,,\, 0-0\rgt =\llt -13 \,,\, -11 \,,\, 0\rgt \end{equation*}

is one vector from a point on MM to a point on LL.

The direction vectors of LL and MM are <0,1,1>\llt 0,-1,1\rgt and <2,2,1>\llt 2,2,1\rgt, respectively. The vector

n=<0,1,1>×<2,2,1>=det[ı^ȷ^k^011221]=<3,2,2>\begin{align*} \vn = \llt 0,-1,1\rgt \times \llt 2,2,1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 0 & -1 & 1 \\ 2 & 2 & 1 \end{matrix}\right] =\llt -3 \,,\, 2 \,,\, 2 \rgt \end{align*}

is then perpendicular to both LL and MM.

The distance from LL to MM is then the length of the projection of v\vv on n\vn, which is

vnn=3922+09+4+4=17\begin{align*} \frac{|\vv\cdot\vn|}{|\vn|} =\frac{|39-22+0|}{\sqrt{9+4+4}} =\sqrt{17} \end{align*}

(b) Solution 2
We can also parametrize MM using z=sz=s:

x=2z+10=2s+10, y=2z+12=2s+12    <x,y,z>=<10,12,0>+s<2,2,1>\begin{equation*} x=2z+10=2s+10,\ y=2z+12=2s+12\quad\implies\quad \llt x,y,z\rgt = \llt 10,12,0\rgt +s \llt 2,2,1\rgt \end{equation*}

The vector from the point (3,1t,t)(-3\,,\,1-t\,,\,t) on LL to the point (10+2s,12+2s,s)(10+2s\,,\,12+2s\,,\,s) on MM is

<13+2s,11+2s+t,st>\begin{equation*} \llt 13+2s \,,\, 11+2s+t \,,\, s-t \rgt \end{equation*}

So the distance from the point (3,1t,t)(-3\,,\,1-t\,,\,t) on LL to the point (10+2s,12+2s,s)(10+2s\,,\,12+2s\,,\,s) on MM is the square root of

D(s,t)=(13+2s)2+(11+2s+t)2+(st)2\begin{equation*} D(s,t) = (13+2s)^2 +(11+2s+t)^2 +(s-t)^2 \end{equation*}

That distance is minimized when

0=Ds=4(13+2s)+4(11+2s+t)+2(st)0=Dt=2(11+2s+t)2(st)\begin{align*} 0 = \pdiff{D}{s} &=4(13+2s) +4(11+2s+t) +2(s-t) \\ 0 = \pdiff{D}{t} &=2(11+2s+t) -2(s-t) \end{align*}

Cleaning up those equations gives

18s+2t=962s+4t=22\begin{align*} 18s +2t &= -96 \\ 2s +4t &=-22 \end{align*}

or

9s+t=48s+2t=11\begin{align*} 9s +t &=-48 \tag{E1}\\ s+2t &=-11 \tag{E2} \end{align*}

Subtracting (E2) from twice (E1) gives

17s=85    s=5\begin{equation*} 17s = -85 \implies s=-5 \end{equation*}

Substituting that into (E2) gives

2t=11+5    t=3\begin{equation*} 2t= -11 +5 \implies t = -3 \end{equation*}

Note that

13+2s=311+2s+t=2st=2\begin{align*} 13+2s &= 3 \\ 11+2s+t &= -2 \\ s-t &= -2 \end{align*}

So the distance is

D(5,3)=32+(2)2+(2)2=17\begin{align*} \sqrt{D(-5,-3)} &=\sqrt{3^2 + (-2)^2 + (-2)^2} =\sqrt{17} \end{align*}
Q14Stage 3

Find the equation of the sphere which has the two planes x+y+z=3, x+y+z=9x+y+z=3,\ x+y+z=9 as tangent planes if the center of the sphere is on the planes 2xy=0, 3xz=02x-y=0,\ 3x-z=0.

Answer

(x1)2+(y2)2+(z3)2=3(x-1)^2+(y-2)^2+(z-3)^2=3

Full solution

The two planes x+y+z=3x+y+z=3 and x+y+z=9x+y+z=9 are parallel. The centre must be on the plane x+y+z=6x+y+z=6 half way between them. So, the center is on x+y+z=6x+y+z=6, 2xy=02x-y=0 and 3xz=03x-z=0. Solving these three equations, or equvalently,

y=2x, z=3x, x+y+z=6x=6\begin{equation*} y=2x,\ z=3x,\ x+y+z=6x=6 \end{equation*}

gives (1,2,3)(1,2,3) as the centre. (1,1,1)(1,1,1) is a point on x+y+z=3x+y+z=3. (3,3,3)(3,3,3) is a point on x+y+z=9x+y+z=9. So <2,2,2>\llt 2,2,2\rgt is a vector with tail on x+y+z=3x+y+z=3 and head on x+y+z=9x+y+z=9. Furthermore <2,2,2>\llt 2,2,2\rgt is perpendicular to the two planes. So the distance between the planes is <2,2,2>=23|\llt 2,2,2\rgt|=2\sqrt{3} and the radius of the sphere is 3\sqrt{3}. The sphere is

(x1)2+(y2)2+(z3)2=3\begin{equation*} (x-1)^2+(y-2)^2+(z-3)^2=3 \end{equation*}
Q15Stage 3

Find the equation of the plane that passes through the point (2,0,1)(-2,0,1) and through the line of intersection of 2x+3yz=0, x4y+2z=52x+3y-z=0,\ x-4y+2z=-5.

Answer

3xy+z=53x-y+z=-5

Full solution

Set y=0y=0 and then solve 2x+3yz=0, x4y+2z=52x+3y-z=0,\ x-4y+2z=-5, i.e. 2xz=02x-z=0, x+2z=5x+2z=-5, or equvalently

z=2x, x+2z=5x=5\begin{equation*} z=2x,\ x+2z=5x=-5 \end{equation*}

The result, (1,0,2)(-1,0,-2), is one point on the plane. Set y=5y=5 and then solve 2x+3yz=0, x4y+2z=52x+3y-z=0,\ x-4y+2z=-5, i.e. 2x+15z=02x+15-z=0, x20+2z=5x-20+2z=-5, or equivalently

z=2x+15, x20+4x+30=5\begin{equation*} z=2x+15,\ x-20+4x+30=-5 \end{equation*}

The result, (3,5,9)(-3,5,9), is another point on the plane. So three points on the plane are (2,0,1), (1,0,2)(-2,0,1), \ (-1,0,-2) and (3,5,9)(-3,5,9). <2+1,00,1+2>=<1,0,3>\llt -2+1\,,\,0-0\,,\, 1+2 \rgt=\llt -1,0,3\rgt and <2+3,05,19>=<1,5,8>\llt -2+3\,,\,0-5\,,\, 1-9 \rgt = \llt 1,-5,-8\rgt are two vectors having both head and tail in the plane.

<1,0,3>×<1,5,8>=det[ı^ȷ^k^103158]=<15,5,5>\begin{align*} \llt -1,0,3\rgt \times \llt 1,-5,-8\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ -1 & 0 & 3 \\ 1 & -5 & -8 \end{matrix}\right] =\llt 15 , -5 ,5 \rgt \end{align*}

is a vector perpendicular to the plane. 15<15,5,5>=<3,1,1>\frac{1}{5}\llt 15 , -5 ,5 \rgt=\llt 3,-1,1\rgt is also a vector perpendicular to the plane. The plane is

3(x+1)(y0)+(z+2)=0or3xy+z=5\begin{equation*} 3(x+1) -(y-0) + (z+2)=0\quad\text{or}\quad 3x-y+z=-5 \end{equation*}
Q16Stage 3

Find the distance from the point p\vp to the plane nx=c\vn\cdot \vx= c.

Answer

cnp/n|c-\vn\cdot\vp|/|\vn|

Full solution

The vector n\vn is perpendicular to the plane nx=c\vn\cdot\vx= c. So the line

x(t)=p+tn\begin{equation*} \vx(t)=\vp+t\vn \end{equation*}

passes through p\vp and is perpendicular to the plane.

Figure from prob_s1.4, line 937

Figure from prob_s1.4, line 937

It crosses the plane at the value of tt which obeys

nx(t)=corn[p+tn]=c\begin{equation*} \vn\cdot\vx(t)= c \hskip .25 in\text{or}\hskip .25 in \vn\cdot[\vp+t\vn]= c \end{equation*}

namely

t=[cnp]/n2\begin{equation*} t=[c-\vn\cdot\vp]/|\vn|^2 \end{equation*}

The vector

x(t)p=tn=n[cnp]/n2\begin{equation*} \vx(t)-\vp=t\,\vn=\vn\,[c-\vn\cdot\vp]/|\vn|^2 \end{equation*}

has head on the plane nx=c\vn\cdot\vx= c, tail at p\vp, and is perpendicular to the plane. So the distance is the length of that vector, which is

cnp/n\begin{equation*} |c-\vn\cdot\vp|/|\vn| \end{equation*}
Q17Stage 3

Describe the set of points equidistant from (1,2,3)(1,2,3) and (5,2,7)(5,2,7).

Answer

It is the plane x+z=8x+z=8, which is the plane through (3,2,5)=12(1,2,3)+12(5,2,7)(3,2,5)=\half(1,2,3)+\half(5,2,7) with normal <1,0,1>=14(<5,2,7><1,2,3>)\llt 1,0,1\rgt=\frac{1}{4}\big(\llt 5,2,7\rgt-\llt 1,2,3\rgt \big).

Full solution

The distance from the point (x,y,z)(x,y,z) to (1,2,3)(1,2,3) is (x1)2+(y2)2+(z3)2\sqrt{(x-1)^2+(y-2)^2+(z-3)^2} and the distance from (x,y,z)(x,y,z)to (5,2,7)(5,2,7) is (x5)2+(y2)2+(z7)2\sqrt{(x-5)^2+(y-2)^2+(z-7)^2}. Hence (x,y,z)(x,y,z) is equidistant from (1,2,3)(1,2,3) and (5,2,7)(5,2,7) if and only if

(x1)2+(y2)2+(z3)2=(x5)2+(y2)2+(z7)2    x22x+1+z26z+9=x210x+25+z214z+49    8x+8z=64    x+z=8\begin{align*} & & (x-1)^2+(y-2)^2+(z-3)^2&=(x-5)^2+(y-2)^2+(z-7)^2\\ &\iff & x^2-2x+1+z^2-6z+9&=x^2-10x+25+z^2-14z+49\\ &\iff & 8x+8z&=64\\ &\iff & x+z&=8 \end{align*}

This is the plane through (3,2,5)=12(1,2,3)+12(5,2,7)(3,2,5)=\half(1,2,3)+\half(5,2,7) with normal vector <1,0,1>=14(<5,2,7><1,2,3>)\llt 1,0,1\rgt=\frac{1}{4}\big(\llt 5,2,7\rgt-\llt 1,2,3\rgt \big).

Q18Stage 3

Describe the set of points equidistant from a\va and b\vb.

Answer

It is the plane 2(ba)x=b2a22(\vb-\va)\cdot\vx=|\vb|^2-|\va|^2, which is the plane through 12a+12b\half\va+\half\vb with normal vector ba\vb-\va.

Full solution

The distance from the point x\vx to a\va is (xa)(xa)\sqrt{(\vx-\va)\cdot(\vx-\va)} and the distance from x\vx to b\vb is (xb)(xb)\sqrt{(\vx-\vb)\cdot(\vx-\vb)}. Hence x\vx is equidistant from a\va and b\vb if and only if

(xa)(xa)=(xb)(xb)    x22ax+a2=x22bx+b2    2(ba)x=b2a2\begin{alignat*}{3} & & (\vx-\va)\cdot(\vx-\va)&=(\vx-\vb)\cdot(\vx-\vb) \\ &\iff\qquad & |\vx|^2-2\va\cdot\vx+|\va|^2&=|\vx|^2-2\vb\cdot\vx+|\vb|^2 \\ &\iff & 2(\vb-\va)\cdot\vx&=|\vb|^2-|\va|^2 \end{alignat*}

This is the plane through 12a+12b\half\va+\half\vb with normal vector ba\vb-\va.

Q19Stage 3Past exam · M200 2003D

Consider a point P(5,10,2)P(5,-10,2) and the triangle with vertices A(0,1,1)A(0,1,1), B(1,0,1)B(1,0,1) and C(1,3,0)C(1,3,0).

  1. Compute the area of the triangle ABCABC.

  2. Find the distance from the point PP to the plane containing the triangle.

Answer

(a) 12111.658\half\sqrt{11}\approx 1.658 (b) 3110.9045\frac{3}{\sqrt{11}}\approx 0.9045

Full solution

(a) One side of the triangle is AB=<1,0,1><0,1,1>=<1,1,0>\overrightarrow{AB}=\llt 1,0,1\rgt - \llt 0,1,1\rgt = \llt 1,-1,0\rgt. A second side of the triangle is AC=<1,3,0><0,1,1>=<1,2,1>\overrightarrow{AC}=\llt 1,3,0\rgt - \llt 0,1,1\rgt = \llt 1,2,-1\rgt. If the angle between AB\overrightarrow{AB} and AC\overrightarrow{AC} is θ\theta and if we take AB\overrightarrow{AB} as the base of the triangle, then the triangle has base length b=ABb=|\overrightarrow{AB}| and height h=ACsinθh=|\overrightarrow{AC}|\sin\theta and hence

area=12bh=12ABACsinθ=12AB×AC=12<1,1,0>×<1,2,1>\begin{equation*} \text{area}=\half bh =\half |\overrightarrow{AB}|\,|\overrightarrow{AC}|\sin\theta =\half|\overrightarrow{AB}\times\overrightarrow{AC}| =\half|\llt 1,-1,0\rgt \times \llt 1,2,-1\rgt| \end{equation*}

As

<1,1,0>×<1,2,1>=det[ı^ȷ^k^110121]=ı^+ȷ^+3k^\begin{equation*} \llt 1,-1,0\rgt\times\llt 1,2,-1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & -1 & 0 \\ 1 & 2 & -1 \end{matrix}\right] =\hi+\hj+3\hk \end{equation*}

we have

area=12<1,1,3>=12111.658\begin{equation*} \text{area}=\half|\llt 1,1,3\rgt|=\half\sqrt{11}\approx 1.658 \end{equation*}

(b) A unit vector perpendicular to the plane containing the triangle is

n^=AB×ACAB×AC=<1,1,3>11\begin{equation*} \hn=\frac{\overrightarrow{AB}\times\overrightarrow{AC}} {|\overrightarrow{AB}\times\overrightarrow{AC}|} =\frac{\llt 1,1,3\rgt}{\sqrt{11}} \end{equation*}

The distance from PP to the plane containing the triangle is the length of the projection of AP=<5,10,2><0,1,1>=<5,11,1>\overrightarrow{AP}=\llt 5,-10,2\rgt-\llt 0,1,1\rgt =\llt 5,-11,1\rgt on n^\hn. If θ\theta the angle between AP\overrightarrow{AP} and n^\hn, then this is

distance=APcosθ=APn^=<5,11,1><1,1,3>11=3110.9045\begin{equation*} \text{distance}= |\overrightarrow{AP}|\,|\cos\theta| =\big|\overrightarrow{AP}\cdot\hn\big| =\left|\llt 5,-11,1\rgt\cdot\frac{\llt 1,1,3\rgt}{\sqrt{11}}\right| =\frac{3}{\sqrt{11}}\approx 0.9045 \end{equation*}
Q20Stage 3Past exam · M200 2003A

Consider the sphere given by

(x1)2+(y2)2+(z+1)2=2(x-1)^2+(y-2)^2+(z+1)^2=2

Suppose that you are at the point (2,2,0)(2,2,0) on SS, and you plan to follow the shortest path on SS to (2,1,1)(2,1,-1). Express your initial direction as a cross product.

Answer

Any positive constant times <1,1,1>×<1,0,1>=<1,2,1>\llt 1,1,-1\rgt\times\llt 1,0,1\rgt =\llt 1, -2, -1\rgt

Full solution

Switch to a new coordinate system with

X=x1Y=y2Z=z+1X=x-1\qquad Y=y-2\qquad Z=z+1

In this new coordinate system, the sphere has equation X2+Y2+Z2=2X^2+Y^2+Z^2=2. So the sphere is centred at (X,Y,Z)=(0,0,0)(X,Y,Z)=(0,0,0) and has radius 2\sqrt{2}. In the new coordinate system, the initial point (x,y,z)=(2,2,0)(x,y,z)=(2,2,0) has (X,Y,Z)=(1,0,1)(X,Y,Z)=(1,0,1) and our final point (x,y,z)=(2,1,1)(x,y,z)=(2,1,-1) has (X,Y,Z)=(1,1,0)(X,Y,Z)=(1,-1,0). Call the initial point PP and the final point QQ. The shortest path will follow the great circle from PP to QQ.

Figure from prob_s1.4, line 1125

Figure from prob_s1.4, line 1125

A great circle on a sphere is the intersection of the sphere with a plane that contains the centre of the sphere. Our strategy for finding the initial direction will be based on two observations.

  • The shortest path lies on the plane Π\Pi that contains the origin and the points PP and QQ. Since the shortest path lies on Π\Pi, our direction vector must also lie on Π\Pi and hence must be perpendicular to the normal vector to Π\Pi.

  • The shortest path also remains on the sphere, so our initial direction must also be perpendicular to the normal vector to the sphere at our initial point PP.

As our initial direction is perpendicular to the two normal vectors, it is parallel to their cross product.

So our main job is to find normal vectors to the plane Π\Pi and to the sphere at PP.

  • One way to find a normal vector to Π\Pi is to guess an equation for Π\Pi. As (0,0,0)(0,0,0) is on Π\Pi, (0,0,0)(0,0,0) must obey Π\Pi's equation. So Π\Pi's equation must be of the form aX+bY+cZ=0aX+bY+cZ=0. That (X,Y,Z)=(1,0,1)(X,Y,Z)=(1,0,1) is on Π\Pi forces a+c=0a+c=0. That (X,Y,Z)=(1,1,0)(X,Y,Z)=(1,-1,0) is on Π\Pi forces ab=0a-b=0. So we may take a=1a=1, b=1b=1 and c=1c=-1. That is, Π\Pi is
    X+YZ=0X+Y-Z=0. (Check that all three points (0,0,0)(0,0,0), (1,0,1)(1,0,1) and (1,1,0)(1,-1,0) do indeed obey X+YZ=0X+Y-Z=0.) A normal vector to Π\Pi is <1,1,1>\llt 1,1,-1\rgt.

  • A second way to find a normal vector to Π\Pi is to observe that both

    • the vector from (0,0,0)(0,0,0) to (1,0,1)(1,0,1), that is <1,0,1>\llt 1,0,1\rgt, lies completely inside Π\Pi and

    • the vector from (0,0,0)(0,0,0) to (1,1,0)(1,-1,0), that is <1,1,0>\llt 1,-1,0\rgt, lies completely inside Π\Pi.

    So the vector

    <1,0,1>×<1,1,0>=det[ı^ȷ^k^101110]=ı^+ȷ^k^\begin{equation*} \llt 1,0,1\rgt\times\llt 1,-1,0\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & 0 & 1 \\ 1 & -1 & 0 \end{matrix}\right] =\hi + \hj -\hk \end{equation*}

    is perpendicular to Π\Pi.

  • The vector from the centre of the sphere to the point PP on the sphere is perpendicular to the sphere at PP. So a normal vector to the sphere at our initial point (X,Y,Z)=(1,0,1)(X,Y,Z)=(1,0,1) is <1,0,1>\llt 1,0,1\rgt.

Since our initial direction (Note that the change of coordinates X=x1X=x-1, Y=y2Y=y-2, Z=z+1Z=z+1 has absolutely no effect on any velocity or direction vector. If our position at time tt is (x(t),y(t),z(t))(x(t),y(t),z(t)) in the original coordinate system, then it is (X(t),Y(t),Z(t))=(x(t)1,y(t)2,z(t)+1)(X(t),Y(t),Z(t)) =(x(t)-1,y(t)-2,z(t)+1) in the new coordinate system. The velocity vectors in the two coordinate systems <x(t),y(t),z(t)>=<X(t),Y(t),Z(t)>\llt x'(t),y'(t),z'(t)\rgt =\llt X'(t),Y'(t),Z'(t)\rgt are identical.) must be perpendicular to both <1,1,1>\llt 1,1,-1\rgt and <1,0,1>\llt 1,0,1\rgt, it must be one of ±<1,1,1>×<1,0,1>\pm\llt 1,1,-1\rgt\times\llt 1,0,1\rgt. To get from (1,0,1)(1,0,1) to (1,1,0)(1,-1,0) by the shortest path, our ZZ coordinate should decrease from 11 to 00. So the ZZ coordinate of our initial direction should be negative. This is the case for

<1,1,1>×<1,0,1>=det[ı^ȷ^k^111101]=ı^2ȷ^k^\begin{equation*} \llt 1,1,-1\rgt\times\llt 1,0,1\rgt =\det\left[\begin{matrix} \hi & \hj & \hk \\ 1 & 1 & -1 \\ 1 & 0 & 1 \end{matrix}\right] =\hi - 2\,\hj -\hk \end{equation*}

My list

nothing marked yet

Loading…

Open the whole list →

Your tutor can open this list with you. It follows your account, so it is there on whichever device you study on.

From the UBC Math 200 open textbook project, © Joel Feldman, Andrew Rechnitzer, Elyse Yeager and others. Licensed CC BY-NC-SA 4.0; this HTML adaptation is released under the same licence. Full attribution.