ENGI 5432 Advanced Calculus

    Faculty of Engineering and Applied Science
    2009 Winter

    Problem Set 4   -   Solutions

    1. For the vector field

      F  =  < x / sqrt{x^2 + y^2} , y / sqrt{x^2 + y^2} , 0 >

      1. Convert   F   from Cartesian coordinates to spherical polar coordinates.


        Let   c1 = cos q   ,       s1 = sin q
                c2 = cos f   and   s2 = sin f ,   then
        x   =   r s1c2     and     y   =   r s1s2
        Þ     x2 + y2   =   r2s12 (c22 + s22)   =   r2s12
        Þ     sqrt{x^2 + y^2}   =   r s1
        [Note:   r >= 0 and sin(theta) >= 0   for all values in the spherical polar (r, q, f) system.
        There is therefore no ambiguity in the square root.]
        Fcart = < c2, s2, 0 >
        Fsph = A Fcart
        ==> [ Fr ; Ftheta ; Fphi ] = [ s1 ; c1 ; 0 ]

        F = sin(theta)(r-hat) + cos(theta)(theta-hat)


      2. Relative to the coordinate axes, in what direction is   F   pointing?


        [diagram of F in horizontal plane] zero z component ==> vector F is parallel to the x-y plane everywhere in real 3 space
        (except on the z-axis, where F is undefined)

        In the horizontal plane   z = r cos q :
        Fy = sin f
        Fx = cos f       (from part (a) above)
        psi = phi
        Þ     vector F points directly away from the z-axis

        OR

        In the vertical plane containing the point P (r, q, f) and the z-axis
        (which is the plane   f = constant):

        [vertical slice showing r and theta components] The vertical component of F is
        Fr cos q   -   Fq sin q
        =   sin q cos q   -   cos q sin q   =   0
        Therefore F is parallel to the x-y plane.

        The azimuthal component (perpendicular into the paper) is
        -Fx sin f   +   Fy cos f
        =   - cos f sin f   +   sin f cos f   =   0

        F is therefore in the same vertical plane as the point P and the z-axis.

        Þ     vector F points directly away from the z-axis


    1. The coordinate conversion matrix   A   for Cartesian coordinates to spherical polar coordinates is

               [ s1*c2  s1*s2   c1 ]
A  =  [ c1*c2  c1*s2  -s1 ]  c1 = cos theta   s1 = sin theta
         [  -s2     c2    0  ]       c2 = cos phi     s2 = sin phi

      so that   Fsph   =   A Fcart   where

                 [ F_r     ]
F    =  [ F_theta ] , 
 sph    [ F_phi   ]

  F  =  F_r r^  +  F_theta theta^  +  F_phi phi^
                  [ F_x ]
F     =  [ F_y ]    and 
 cart   [ F_z ]

   F  =  F_x i  +  F_y j  +  F_z k

      Show that the conversion matrix   B   for the inverse conversion from spherical polar back to Cartesian coordinates, such that   Fcart = B Fsph,   is simply

      B = AT       (the transpose of matrix   A).


      B = A-1   (because the two transformations are inverses of each other)
      In order to show that A-1 = AT, it is sufficient to show that ATA = I.
      [This is much faster than the adjoint/determinant method or the row reduction method for finding A-1.]

      AT A = [ s1c2  c1c2  (-s2) : s1s2  c1s2  c2 : c1  (-s1)  0 ]
    [ s1c2  s1s2  c1 : c1c2  c1s2  (-s1) ; (-s2)  c2  0]
      =[ details of matrix multiplication ]
      matrix product = Identity matrix
      B = A^(-1) = A^T   [and the coordinate transformation matrix A is orthogonal.]


    1. Convert from spherical polar to Cartesian coordinates the vector field

      F  =  cos theta r^  -  sin theta theta^


      From Question 2, B = AT.
      Use the same abbreviations as in questions 1 and 2 :
      Let   c1 = cos q   ,       s1 = sin q
              c2 = cos f   and   s2 = sin f ,   then
      Fr   =   c1   ,     Fq   =   -s1   ,       Ff   =   0
      ==> F(cart) = [ Fx ; Fy ; Fx ] 
= [ s1c2  c1c2  (-s2) ; s1s2 c1s2  c2 ; c1  (-s1)  0 ] [ c1 ; (-s1) ; 0 ]
      =[matrix product] = k-hat

      cos theta rHat - sin theta thetaHat = kHat


    1. Find the divergence and curl of each of the following:

      1. F(r)  =  < x*y^2,  y*z^2,  x^2*z >

        div F = (d/dx) (xy^2) + (d/dy)(yz^2) + (d/dz)((x^2)z)
                      =   y2 + z2 + x2         Therefore

        div F-arrow = x^2 + y^2 + z^2

        curl F = = < (0-2yz), (0-2zx), (0-2xy) >

        curl F = -2 < yz, zx, xy >


      2. F(r)  =  < x/y,  y/z,  z/x >

        div F = d/dx(x/y) + d/dy(y/z) + d/dz(z/x)

        div F = 1/x + 1/y + 1/z

        curl F = < yz^(-2), zx^(-2), xz^(-2) >

        curl F = < yz^(-2), zx^(-2), xz^(-2) >

        [Note:   the cyclic symmetry between x, y, z   in parts (a) and (b) above allows any two parts of the calculations for divergence and curl to be deduced from the other part.]


      3. F(r)  =  < y*sqrt{1+x^2+y^2},  x*sqrt{1+x^2+y^2},  z^4 >

        ==> div F = d/dx( y(1 + x^2 + y^2)^(-.5))
         + d/dy( x(1 + x^2 + y^2)^(-.5) + d/dz(z^4)
        (-.5)((1 + x^2 + y^2)^(-.5))(2xy) 
         - (.5)((1 + x^2 + y^2)^(-1.5))(2xy) + 4z^3

        div F = 4z^3 - (2xy)/(((1 + x^2 + y^2)^(3/2))

        curl F = |  i  j  k ; d/dx  d/dy  d/dz ;
         y(1 + x^2 + y^2)^(-.5)  x(1 + x^2 + y^2)^(-.5)  z^4
        < (0-0), (0-0), 
         ( 1(1 + x^2 + y^2)^(-.5) + x(-.5)(1 + x^2 + y^2)^(-1.5)(2x) -
         1(1 + x^2 + y^2)(-.5) - y(-.5)(1 + x^2 + y^2)^(-1.5)(2y)) >
        < 0, 0, (y^2 - x^2) / (1+ x^2 + y^2)^(1.5) >

        curl F = (y^2 - x^2) / (1 + x^2 + y^2)^(3/2) k-Hat


    1. For the vector field defined in spherical polar coordinates by

      u = (sin theta rHat + cos theta thetaHat) sin phi
     + cos phi phiHat

      find   du/dt.


      Recall, from page 1-38 of the lecture notes, that
              [derivative of unit vector rHat]
              [derivative of unit vector thetaHat]
      and   [derivative of unit vector phiHat]
      Use the abbreviations
      [abbreviations]
      [abbreviated form of vector u]
      Using the product and chain rules of differentiation,
      [applying the product and chain rules]
      [gathering up components]
      Therefore, everywhere in space and at all values of t,

      du/dt = 0

      OR

      Convert the vector u from spherical polar coordinates to Cartesian coordinates:

      converting, u = jHat
      and the derivative of any absolutely constant vector is the zero vector!


    1. Consider the purely radial vector field   F (r, q, f) = f (r) r Hat, where r Hat is the unit radial vector in the spherical polar coordinate system and   f (r)   is any function of r that is differentiable everywhere in real 3 space (except possibly at the origin).

      1. Find an expression, in terms of   r,   f (r)   and   f ' (r), for the divergence of F.

        First note that   f (r)   is a function of   r   only, so that
        partial df/dr = total df/dr = f'(r)
        div F = d/dr { r^2 sin theta f(r) } / (r^2 sin theta)
        Therefore, everywhere except possibly on the z axis,

        div F = d/dr { r^2 f(r) } / (r^2)


      2. Find an expression, in terms of   r,   f (r)   and   f ' (r), for the curl of F.

        [curl in determinant form for spherical polar]
        curl F = zero vector
        limit curl F = vector 0  as  r sin theta --> 0

        curl F identically = zero vector

        [All purely radial differentiable vector fields are therefore automatically irrotational.]


      3. Of particular interest is the central force law

        F  =  (k/r^n) rHat
        (k, r > 0)

        Show that the divergence of   F   vanishes everywhere in real 3 space (except possibly at the origin) if and only if   n = 2 .
        [Two of the four fundamental forces of nature, electromagnetism and gravity, both obey this inverse square law.]


        [Substitute f(r) into part (a)]
        div F = (k/r^2) d/dr r^(2-n)
        div F   therefore vanishes for all   r > 0   if and only if   n = 2 .

        [Note:   n = 2   recovers the familiar inverse square law,
                        F = k/r^2 r-Hat  for which
    div F = 0  and curl F = 0  for all r > 0]


    Return to Index of Problem Set Solutions   [Return to the Index of Solutions.]
    Back to previous page   [Return to your previous page]
    Created 2008 01 04 and most recently modified 2008 12 27 by Dr. G.H. George