# Evaluate this index Integral containing Trig

1. Aug 29, 2012

### bugatti79

1. The problem statement, all variables and given/known data

Folks Evaluate $B_{11}$ given

$\displaystyle B_{ij}=\int_0^1 (1+x) \frac{d \phi_i}{dx} \frac{ d\phi_j}{dx} dx$ where $\phi_i= sin i \pi x$ and $\phi_j=sin j \pi x$

2. Relevant equations
3. The attempt at a solution

I calculate $\displaystyle B_{ij}=\int_0^1 (1+x)[ i \pi \cos(i \pi x))(j \pi \cos(j \pi x)]=ij \pi^2 \int_0^1 (1+x) \left[\frac{1}{2} [\cos(i+j)\pi x+\cos(i-j)\pi x\right ]dx$

$\displaystyle = \frac{ij \pi^{2}}{2} \int_0^1\left [ \cos(i+j) \pi x+\cos(i-j) \pi x +x \cos(i+j) \pi x+x \cos(i-j) \pi x \right ]dx$

Now for the second and last term in the integrand if we substitue $i=j=1$ after integrating we will get a 0 in the denominator ....but the book calculates $B_{ij}=\frac{3 \pi^2}{4}$

What have I dont wrong?

2. Aug 29, 2012

### SammyS

Staff Emeritus
Did you mean to say "numerator" ?

cos(0) = 1

3. Aug 29, 2012

### bugatti79

Well no...looking at the second term $\displaystyle \int_0^1 \cos(i-j) \pi x dx= \left [\frac{sin(i-j) \pi x}{\pi(i-j)} \right]_0^1$ 1 is indeterminate....

I dont know how that answer $\frac{3 \pi^2}{4}$ was evaluated...

4. Aug 29, 2012

### vela

Staff Emeritus
You can't integrate it like that when i=j. Assume i=j, simplify the integrand, and then integrate.

5. Aug 29, 2012

### SammyS

Staff Emeritus
Well, if i = j then you are integrating cos((0)πx), but cos((0)πx) = 1

6. Aug 29, 2012

### bugatti79

Thank you, I thought one has to insert the index values AFTER integration....? Where is the mathematical rule that asserts this?

Cheers

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook