| Francis
Woodhouse |
I'm having problems getting the correct answer for Question 3 in exercise 13.1 of "Teach Yourself Calculus". The question is to find ![]() The book gives the answer as: ![]() My answer, however, is: ![]() I got my answer by first performing polynomial long division, which gave two fractions which were easy to integrate. However, it appears that either I keep going wrong somewhere, or the book is wrong. Can anyone get the answer the book has, and if so, how? Thanks! |
||
| Alice
Thompson |
I agree with them, but don't get the a/b2 term, but it can be part of the constant. You can rewrite the fraction as: .When you integrate , you must divide by b,
giving their answer.
|
||
| Kerwin
Hui |
To get the a/b2 term: Let u=a+bx, then we have du/dx=b, x=(u-a)/b, so
Kerwin |
||
| Francis
Woodhouse |
Aha, substitution. Performing the substitution myself, I got the same answer as the book. It's a bit odd that substitution is required to get the exact same answer in a chapter all about integration of algebraic fractions... haven't had to use a substitution in all of the other questions in the exercise. |
||
| Mark
Durkee |
You can do it without substitution using Alice's method above: which can be
integrated.
|
||
| Francis
Woodhouse |
Turns out that my initial answer was different from the answer I was getting the rest of the time - should've been 1 / b2 , not 1 / b, outside the brackets. And yes, you can get the answer with Alice's method - however, as she says, it's missing the a/b2 term. Although this could be said to be in the constant of integration, a second integration would be wrong if it was missing the a/b2 term. I think, anyway. Thanks, you three - the help is appreciated. |
||
| Philip
Ellison |
The second integration would not be affected, as ò(a/b2 + C1)dx=(a/b2 + C1)x and òC2 dx=C2 x, i.e. in both cases a second integration results in an arbitrary constant multiplied by x. |
||
| Francis
Woodhouse |
Agh, silly me. Forgot about integrating the first constant. |
||
| Chris
Tynan |
I think we also have to consider the case when b=0? The integral is then trivial, as long as a is not 0, but i think it should still be dealt with. |