Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
18 views

Solution - Problem Set 3 Multivariable Calculus

This document contains solutions to problems from a multivariable calculus problem set. It includes determining the parametric equations for points on moving objects, finding parameterizations of curves on surfaces, and calculating distortions caused by shadows.

Uploaded by

Sharan Balasubramanian
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
18 views

Solution - Problem Set 3 Multivariable Calculus

This document contains solutions to problems from a multivariable calculus problem set. It includes determining the parametric equations for points on moving objects, finding parameterizations of curves on surfaces, and calculating distortions caused by shadows.

Uploaded by

Sharan Balasubramanian
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

18.

02 Problem Set 3, Part II Solutions

1. (b) To get to the point P , start at the origin, add the vector �0, 2� to go
up to the center of the disk at t = 0; then add the vector �t, 0� to get to the
center of the disk at time t; and finally (do what it takes to) shift over by one
unit at an angle θ = t/2, measured clockwise from the vertical (see below).
How to do that last move? Just add the vector −�sin(t/2), cos(t/2)�, as can
be seen by from a sketch.
−→
Combining, we get OP (t) = �0, 2� + �t, 0� + �− sin(t/2), − cos(t/2)� or
r = r(t) = �t − sin(t/2), 2 − cos(t/2)�.
The reason that the angle is θ = t/2 after time t is as follows: the center of
the disk has moved t units to the right. Since the disk rolls without slipping,
the horizontal distance it travels is equal to r θ = the circular arclength
“unrolled” by the disk, where r is the radius and θ the central angle in
radians. Since r = 2 and the horizontal distance is t, we get t = 2θ.
(c) See applet. The shape of of the curve is an elongated periodic ‘wave’
created by the up-and-down movement of the point as the wheel moves to
the right.

2. (a) r(t) is clearly in the plane through O defined by u and v. But | r(t) |2
= (u cos(t) + v sin(t)) · (u cos(t) + v sin(t)) = cos2 (t) + sin2 (t) = 1 since
u · u = v · v = 1 and u · v = 0 by assumption. So r(t) sweeps out the unit
circle centered at O in P.
b) We need to find two orthogonal unit vectors u and v which lie in the plane
given by x + 2y + z = 0; then we can use the result of part(a). To get the
first one, we can use take any vector orthogonal to n = �1, 2, 1�, i.e. any
vector u1 = �p, q, r� with p + 2q + r = 0, and then make it a unit vector.
We’ll pick �1, −1, 1� to get u = √13 �1, −1, 1�. To get v, we can just take the
cross- product of u1 × n, and then also make it a unit vector. This gives v1
= �1, −1, 1� × �1, 2, 1� = 3�−1, 0, 1� and then v = √12 �−1, 0, 1�. With these
choices we get
1 1
r(t) = √ �1, −1, 1� cos(t) + √ �−1, 0, 1� sin(t),
3 2
1 1 1 1 1
or x = √ cos(t) − √ sin(t), y = −√ cos(t), z = √ cos(t) + √ sin(t).
3 2 3 3 2
Note: other choices for u1 will give other vectors v and the resulting equa­
tions will look different, but they will be equivalent (i.e. give the same curve
C).

1
3. a) Any such line L� must have direction vector �a, b, c� which is normal to
�1, 2, 1�. So we must have
a + 2b + c = 0.
Since the line L� passes though O, we can write the equation of L � in the
point-direction form: x = a t, y = b t, z = c t, and then use the equation
a + 2b + c = 0 above, we get

x = a t, y = b t, z = (−a − 2b) t.

But L� is described redundantly: scaling a and b equally gives an equivalent


equation. To remove this redundancy, we need to look at two cases: either
b = 0 in which case we get the line x = t, y = 0, z = −t, or b =
� 0 in which
case we may divide through by b to get

x = α t, y = t, z = (−α − 2) t

where the parameter α corresponds to a/b.


b) Since these lines all pass through O, they can be described by their unit
direction vectors. The fact that the lines all lie in P means that their direction
vectors all lie in P. The tips of these unit vectors (considered as origin
vectors) form a circle. So we have a circle’s worth of lines through O lying
in P.
Note that the family of lines described algebraically in part(a) is a one-
parameter family, with α as the parameter, and that this is consistent with
the geometric fact that we have 1-dimensional circle of lines.

4. (a) Let Pt be the shadow of the point (cos t, sin t, 2) on the plane Pα :
my + z = 0 for an arbitrary but temporarily fixed value of t. See figure.

2
(b) Let Pt be the intersection of the line Lt joining (0,0,4) to (cos t, sin t, 2)
with the plane Pα . Let u be the parameter for Lt , then the vector-parametric
equation for Lt is given by
� t (u) = �0, 0, 4� + u�cos t, sin t, −2�
Lt : R

(since direction vector for Lt is �cos t, sin t, 2� − �0, 0, 4�).


Pt = Lt ∩ Pα : substitute x = u cos t, y = u sin t,�z = 4 − 2u into my + z =�0 →
mu sin t+(4−2u) = 0 → u∗ = 2−m4sin t → Pt = 2−m 4 cos t
, 4 sin t , −4m sin t (af­
sin t 2−m sin t 2−m sin t
� t = the answer given (with m = tan α).
ter simplification). Thus �rα (t) = OP
(c) α = 0, so �r0 (t) = �2 cos t, 2 sin t, 0� is indeed an enlarged circle in the x-y plane.
√ sin t
|�rα (t) − �r0 (t)| = 2 5m| 2−m sin t
| (for m > 0, after simplification). That gives
the following differences:

t = 0 t =√ π/2 t = π t =√3π/2
2 5m 2 5m
0 2−m
0 2+m

√
2 5m
So the largest value is 2−m
.
(The following paragraph is extra) To put this in perspective, if α = .1 rad (so
m = .1003 . . . ), then this distortion is 0.236 . . . linear units. More generally,
for small α, m √ is pretty close
√ to α, αso weα can approximate the maximum
α/2 2
distortion by 2 5 1− α = 5α(1 + 2 + ( 2 ) + · · · ) (geometric series), or
√ 2
5α + O(α2 ).

3
MIT OpenCourseWare
http://ocw.mit.edu

18.02SC Multivariable Calculus


Fall 2010

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

You might also like