Karl's Calculus Forum: November/December 2004

© 2004 by Karl Hahn
KCT logo (click for Table of Contents)

Home Page Table of Contents How to email Math Notation Send Email to Karl





    

Archived Forum Discussion for November/December 2004

return to current forum

Reply to Gordon: You didn't give any initial condition, so I'll state one here: y(0) = A. This one is easy enough to find a closed solution for. The function multiplying y is x2. You integrate that and take the exponential of the negative of the result. When you apply the initial condition you find

   y(x)  =  Ae-x3/3
which you can expand using a Maclaurin series. When you do you will find that you will have a series in which the powers are all multiples of 3.

You can also use the method of finding the powers one by one. Your equation is equivalent to y' = x2y. So you have

   y(0)  =  A     initial condition given
   y'(0) =  x2y  =  0  because x=0
   y"(0) =  2xy + x2y'  =  0  for the same reason
   y'"(0) = 2y + 4xy' + x2y"  =  2A
and so on, using the product rule to take higher derivatives as you go. You can evaluate each derivative at x = 0 using previous derivatives and the initial condition. When you have them, you can set up the Maclaurin series. You should get the same thing as doing it the first way.
Karl <Click to Send Email to Karl>
USA - Saturday, December 18, 2004 at 17:33:22 (EST)
Reply to Sarah: Figure out the surface area of the water at the top of the trough as a function of the height of water, h. That area will be precisely the same as dV/dh. The problem gives you dV/dt = -2 ft3/sec. To get the rate at which height is changing, the chain rule requires that
  dV dh    dV
        =    
  dh dt    dt
Solve for dh/dt.
Karl <Click to Send Email to Karl>
USA - Saturday, December 18, 2004 at 17:12:37 (EST)
Hi, I'm having trouble solving y'-x^2*y=0 using series. To be exact, I'm having trouble equatin the indexes(where the series start) and making the exponents of x the same. Thanks a lot.
Gordon
USA - Thursday, December 16, 2004 at 21:33:08 (EST)
I need help with this problem of the week. A trough is 5 feet long, and its vertical cross sections are inverted isosceles trianges with base 2 and height 3 feet. Water is being siphoned out of the trough at the rate of 2 cubic feet per minute. At any time t, let h be the depth and V be the volume of water in the trough. a) Find the volume of water in the trough when it is full. (I got 10 cubic feet) b) What is the rate of change in h at the instant when the trough is 1/4 full by volume? c) What is the rate of change in the area of the surface of the water at the instant when the trough is 1/4 full by volume? PLEASE HELP.
Sarah


USA - Thursday, December 16, 2004 at 14:49:09 (EST)
Reply to Gerald Farley: Problem 1:-
equation 1:     2x2y + x2 = c       (where c is a constant)
equation 2:     x + y = 1

hence y = 1 - x
Substitute for y into the first equation to get:
  2x2(1 - x) + x2  =  c

  3x2 - 2x3  =  c

  3x2 - 2x3 - c  =  0
which is a cubic. Use the cubic formula to solve this for x, then back-substitute the solutions into the second equation to get y. Then check each (x,y) solution to make sure it is not extraneous.

Problem 2:

equation 1:     6x3y2 + 3x3y + x3 = c       (where c is a constant)
equation 2:     x + y = 1
Same method except this time you'll get a fifth degree polynomial in x. Sorry, there's no formula for solving that. You have to use an analytic method of successive approximation like Newton-Raphson to find the solution set for each value of c.
Karl <Click to Send Email to Karl>
USA - Wednesday, December 15, 2004 at 15:50:49 (EST)
Hi


A simple real-life situation has led me to two separate problems.

In each problem I wish to get two equations simplified into one equation of the form x=...


Solutions to either/both of these problems much appreciated:-


Problem 1:-
equation 1:     2x2y + x2 = c       (where c is a constant)
equation 2:     x + y = 1


Problem 2:
equation 1:     6x3y2 + 3x3y + x3 = c       (where c is a constant)
equation 2:     x + y = 1


For some reason these problems seem resistant to the simple techniques I learnt at school - apologies if they are in fact trivially easy !


Gerald Farley


UK - Wednesday, December 15, 2004 at 08:41:40 (EST)
Reply to Vinay: The definition of ln(x) is

  x
  
  1
  dt
    
   t

The log of 2x will be that same integral taken from 1 to 2x. Now break that integral into two pieces:

   ln(2x)  =
  2
  
  1
  dt
      +
   t
  2x
  
  2
  dt
    
   t
Now look at the 1/x curve from 2 to 2x. Isn't it just half of the same curve taken from 1 to x, except stretched out? Use that fact to evaluate the second integral in the sum in terms of ln(x). If you let u = 2t, and du = 2dt as a substitution in the second integral and substitute the limits as well, it all works out. Observe that the first integral in the sum is, by definition, ln(2). See if you can finish it now.
Karl <Click to Send Email to Karl>
USA - Tuesday, December 07, 2004 at 19:09:39 (EST)


Need some helps on Logs.... I need to prove that ln(2x) = ln(2) + ln (x) and I have to use the definition of ln which is integral from X to 1 of 1 / x DX thanks for the help!
Vinay


union City, ca USA - Monday, December 06, 2004 at 18:20:16 (EST)
Reply to Gordon: First, this is not a homogeneous equation. You'd have to have zero to the right of the equal for it to be a homogeneous equation. Nevertheless, the first thing to do with the nonhomogeneous equation that you have is to solve the corresponding homogeneous equation, y" - y' = 0. The solution for this is yh = Aex + B, where both A and B are undetermined constants (and the h subscript indicates that this is the solution to the homogeneous equation).

From here you can proceed two ways. One is to recognize that the solution will be in the form of y = (ax2 + bx + c)ex + B, then plug this back into the original equation to solve for a, b, and c. Actually both c and B will be undetermined, and you determine them from the initial conditions given in the problem.

The other (more general) method is by variation of parameters. To do this you replace A and B in the homogeneous solution with two unknown functions, u(x) and v(x). So you have the solution, y = u(x)ex + v(x). The trick now is to find u and v. To do this we solve the following two equations simultaneously:

   u'(x)ex + v'(x)  =  0

   u'(x)(d/dx)ex + v'(x)(d/dx)1  =  xex
The first equation is using the components of the homogeneous solution. The second is using the derivatives of the homogeneous solution, and on the right, the forcing function, xex. The second equation is equivalent to:
   u'(x)ex  =  xex

   u'(x)  =  x

   u(x)  =  (1/2)x2 + C
From that you can easily find v'(x) and integrate it (don't forget a constant of integration) to get v(x). The resulting solution will have two undetermined constants, C, and the one you get by integrating v'. Use the initial conditions given in the problem to determine them.
Karl <Click to Send Email to Karl>
USA - Sunday, December 05, 2004 at 16:15:07 (EST)
I'm asked to solve y"-y'=xex, with initial conditions y(0)=2 and y'(0)=1. This is  a second order homogeneous differential equation, so I try to subsitute y=(Ax+B)*ex. After substituting the second derivative and first derivative of my y, I found that I'm left with Aex=xex. Can someone tell me what I did wrong?

Gordon


Union City, CA USA - Saturday, December 04, 2004 at 22:13:13 (EST)
Reply to Will Pittman: If a and b are the lengths of the legs and c is the length of the hypotenuse, then
   a2 + b2 = c2
according to the Pythagorean rule. All three quantities are functions of time, t. Taking the derivative of the above with respect to t:
  2a da/dt  +  2b db/dt  =  2c dc/dt
The problem gives you a, b, as well as da/dt and db/dt. You can find c from a and b. Solve for dc/dt.
Karl <Click to Send Email to Karl>
USA - Wednesday, December 01, 2004 at 19:53:16 (EST)
Hi After doing a set of operations, I am left with the following expression:
                     (ax1 + bx2)(x1 dc + x2 de)
            G  =     ---------------------
                     (a + c)x1 + (b + e)x2

where x1 + x2 = X.

Is it possibly to simplify? Would be useful, for example, to have an expression with dc/c and de/e. Many thanks
Lars


London, UK - Monday, November 29, 2004 at 06:47:33 (EST)
At a given instant the legs of a right triangle are 5 cm and 12 cm long. If the short leg is increasing at the rate of 1 cm/sec and the long leg is decreasing at the rate of 2 cm/sec, how fast is the hypotenuse changing?
Will Pittman


Charlotte, NC USA - Saturday, November 27, 2004 at 12:21:39 (EST)
Reply to Jason: If g is the identity function,
   g(x)  =  x
then it works for you. Also if g is the inverse function of f(x) = 3x+2, then it also works. So you need to solve for
   f(g(x))  =  3g(x) + 2  =  x
The way you find an inverse function is:
  3g + 2  =  x
Solve for g in terms of x and you'll have a g that is the inverse function of f(x).
Karl <Click to Send Email to Karl>
USA - Monday, November 22, 2004 at 00:23:31 (EST)
Reply to Tamara: The volume of a cone is given by
   V  =  (1/3)πr2h
where V, r, and h are all functions of time, t. The area of the base is given by:
   A  =  πr2
Taking the derivative of the area equation with respect to time, t:
   dA/dt  =  2πr dr/dt
The problem gives you r and dr/dt. For part a), solve for dA/dt.

For part b), you need to take the time-derivative of the volume equation using the product and chain rules:

   dV/dt  =  (1/3)π(2rh dr/dt  +  r2 dh/dt)
The problem also gives you V and dV/dt. Use the original volume equation to find h. Then plug all the information you know into the derivative-of-the-volume equation above and solve for dh/dt to get the answer for part b).

For part c), divide dA/dt by dh/dt
Karl <Click to Send Email to Karl>
USA - Monday, November 22, 2004 at 00:16:53 (EST)


Reply to Steve: Let the three dimensions of the box be L, H, and W. The problem stipulates that the base is square, so W = L. So the volume equation is
   V  =  32000 cm  =  L2H
There are four sides whose dimensions are each L × H. There is also the base whose dimensions are L × L. So the area of material needed is
   A  =  4LH + L2
From the first equation you have
  H  =  V/L2
Replacing H in the area equation:
   A  =  4L(V/L2) + L2  =  4V/L + L2
Taking the derivative, dA/dL, and setting it to zero you get:
  dA
      =  0  =  -4V/L2 + 2L
  dL
From this, you can solve for L in terms of the volume, V. You can back-substitute then to find H.
Karl <Click to Send Email to Karl>
USA - Sunday, November 21, 2004 at 23:49:52 (EST)
i need a little help! suppose that f(x)= 3x+2. find the two different functions g such that g(f(x))=f(g(x))
jason


USA - Wednesday, November 17, 2004 at 13:00:14 (EST)
the volume V of a cone is increasing at the rate of 28 (pie)cubic units per second. At the instant when the radius r of the Cone is 3 units, its volume is 12pie cubic units and the radius is increasing at 1/2 unit per second a) at the instant when the radius of the cone is 3 units, what is the rate of change of the area of its base? b) at the instant when the radius of the cone is 3 units, what is th erate of change of its height h? c) at the instant when the radius of the ocne is 3 units, what is the instantaneous rate of change of the area of its base with respect to its height h?
Tamara


PA USA - Monday, November 15, 2004 at 20:22:51 (EST)
    I cant figure out this problem.
    a box with a square base and open top must have a volume of 32,000cm^2. find the demensions of the
box that minimize the amount of material used.

    is this a square box or what, i am having trouble with it.


steve


USA - Monday, November 15, 2004 at 16:50:26 (EST)
Need help with this problem: If the voltage in a circuit varied according to v=-20t^2, what is the exact rate of change dv/dt of the voltage when t=3 seconds? Thanks.
Betty Clark


USA - Sunday, November 14, 2004 at 01:22:23 (EST)
Reply to Gorden: Observe that 2n/3n+1 = (2/3)n × (1/3). So if you can show that (2/3)n converges, you're done.
Karl <Click to Send Email to Karl>
USA - Monday, November 08, 2004 at 14:24:59 (EST)
Hi, I'm given this sequence 2^n/3^(n+1), and I have to determine if it converges or diverges. I've tried taking L'Hopital's rule of its corresponding function but I can't get rid of the 2^x/3^(x+1). Thank you in advance.
Gorden


Union City, CA USA - Sunday, November 07, 2004 at 00:46:29 (EST)

Home Page Table of Contents How to email Math Notation



    

You can email me by clicking this button:

  Use your own emailer     Use form