Integration involving Substitution

hungryhippo
Messages
10
Reaction score
0

Homework Statement


Find the arc length of the function f(x) = x(sqrt(x/2-x) from 0 to x

Integral forms involving a+bu
--> integral [sqrt(a+bu)]/u^2du

Homework Equations


Arc Length = integral sqrt(1+(f'(x))^2)dx

The Attempt at a Solution


First, I took the derrivate
f'(x) = sqrt(x/2-x) + x/[(2-x)^2(sqrt(x/2-x))]
= [x(2-x)+x]/[(2-x)^2(sqrt(x/2-x))]
=[x(3-x)]/[(2-x)^2(sqrt(x/2-x)]

Then, the square of f'(x)
= [x^2(3-x)^2]/[x(2-x)^3]
=[x(3-x)^2]/[(2-x)^3]

Then I added 1 to the square of the derivative, which simplified to
=[8-3x]/[2-x]^3

Then, the square root of the above
=sqrt([8-3x]/[2-x]^3)

I substituted 2-x = t
=sqrt[(3t+2)/(t^3)] <--multiplied top and bottom by sqrt[t]
=sqrt(3t^2+2t)/t^2

I'm stuck here..I'm allowed to use formulas from a table of integrals (i.e. the one specified above), but I can't seem to get it into that form ..
 
Physics news on Phys.org
Congratulations on getting that far. That's one nasty integral. I don't have a good integral table so I put sqrt(3t^2+2t)/t^2 into Wolfram Alpha and said "Show steps". It says, i) complete the square in the numerator, ii) do a sec trig substitution and then iii) use the universal tan(x/2) Weierstrass substitution and finally iv) get something you can solve by partial fractions. The result is less than pleasant, but it looks simpler than the steps involved in getting there. There may be a shortcut. But who cares? Life is short. Can you use WA or a computer algebra system instead of an integral table? A system like maxima also seems to handle it ok. And it's free.
 
Last edited:
Hey, I don't think we're suppose to use the Wolfram alpha application. I asked my prof, and he said there was an "easy" way to simplify, although I have failed to find one yet.
 
hungryhippo said:
Hey, I don't think we're suppose to use the Wolfram alpha application. I asked my prof, and he said there was an "easy" way to simplify, although I have failed to find one yet.

Ok, so maybe there is an easier way. I guess I'm not seeing one right now either.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top