Solving Troublesome Integral: 100 \int \frac{h^2-20h}{(h-10)^2} dh

  • Thread starter Thread starter Bashyboy
  • Start date Start date
  • Tags Tags
    Integral
Bashyboy
Messages
1,419
Reaction score
5
Hello,

I have tried computing the following integral three times now, and I can not figure out what I am doing wrong.

The integral is 100 \int \frac{h^2-20h}{(h-10)^2} dh, which wolfram calculates as 100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 (-h - \frac{100}{h-10} + c ).

Here is one sample of my work:

Using integrating by parts, let u = h^2 - 20h, which becomes \frac{du}{dh} = 2h - 20. Multiplying both sides by the differential u, we get \frac{du}{dh} dh = (2h-20)dh. The definition of the differential of u is du = \frac{du}{dh} dh. Making this substitution, du = (2h - 20) dh. Let dv = \frac{dh}{(h-10)^2}. Integrating both sides, v = \frac{1}{h-10}.

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[ (h^2-20h) \left(\frac{1}{h-10} \right) - \int \frac{2h-20}{h-10} dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2 \int \frac{h-10}{h-10} dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2 \int dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2h + c \right]

which is not the same...

Can anyone see what I might have done wrong?
 
Last edited:
Physics news on Phys.org
Bashyboy said:
Hello,

I have tried computing the following integral three times now, and I can not figure out what I am doing wrong.

The integral is 100 \int \frac{h^2-20h}{(h-10)^2} dh, which wolfram calculates as 100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 (-h - \frac{100}{h-10} + c.

Here is one sample of my work:

Using integrating by parts, let u = h^2 - 20h, which becomes \frac{du}{dh} = 2h - 20. Multiplying both sides by the differential u, we get \frac{du}{dh} dh = (2h-20)dh. The definition of the differential of u is du \frac{du}{dh} dh. Making this substitution, du = (2h - 20) dh. Let dv = \frac{dh}{(h-10)^2}. Integrating both sides, v = \frac{1}{h-10}.

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[ (h^2-20h) \left(\frac{1}{h-10} \right) - \int \frac{2h-20}{h-10} dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2 \int \frac{h-10}{h-10} dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2 \int dh \right] \implies

100 \int \frac{h^2-20h}{(h-10)^2} dh = 100 \left[\frac{h^2-20h}{h-10} - 2h + c \right]

which is not the same...

Can anyone see what I might have done wrong?

They differ by a constant. Divide ##\frac{h^2-20h}{h-10}## out using polynomial division. Using partial fractions to begin with would have led to an easier integration problem. BTW both solutions look to have a sign error.
 
Last edited:
Try using the substitution $$u = h-10$$
After some simple algebra, you should end up with the integral $$100∫\frac{u^2-100}{u^2} du$$ which should be fairly easy to evaluate, then you can substitute your h-10 back into your solution.
 
Last edited:
try using
$$100 \int \! \frac{h^2-20h}{(h-10)^2} \, \mathrm{d}h=100 \int \! \frac{(h-10)^2-100}{(h-10)^2} \, \mathrm{d}h$$
 
Bashyboy said:
Can anyone see what I might have done wrong?
Your particular mistake was integrating ##dv = \frac{dh}{(h-10)^2}## to get ##v = \frac{1}{h-10}##. That's pretty wrong. I don't know what Wolfram Alpha's problem is if you quoted it correctly.
 
Last edited:
I wouldn't consider that pretty wrong--perhaps just off the mark; after all, I am just missing a mere minus sign.
 
Bashyboy said:
I wouldn't consider that pretty wrong--perhaps just off the mark; after all, I am just missing a mere minus sign.

Why is the Wolfram Alpha answer you quoted also off by the same sign?
 
Why not simply calculating it carefully? It's not allowed to give the full solution in this forum, but here's a hint:

I'd use
\frac{h^2-20h}{(h-10)^2}=\frac{(h-10)^2-100}{(h-10)^2}=1-\frac{100}{(h-10)^2}.
This is trivial to integrate. Mathematica 9.0 under linux gets the correct result. So I wonder, why Wolfram alpha gets it wrong.
 

Similar threads

Replies
10
Views
2K
Replies
15
Views
6K
Replies
105
Views
4K
Replies
8
Views
2K
Replies
3
Views
2K
Replies
22
Views
3K
Back
Top