# Another Straight Forward DE

1. Jul 9, 2012

### Nano-Passion

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

Find a 1 parameter family of solutions for the differential equation yx^2 dy - y^3 dx = 2x^2 dy

3. The attempt at a solution

yx^2 dy - y^3 dx = 2x^2 dy
x^2 dy - y^2 dx = 2x^2/y dy
dy - y^2/x^2 dx =
- y^2/x^2 dx = 2/y dy - dy
Divide all by y^2
dx/x^2 = (2dy/y - dy ) * -1/y^2
dx/x^2 = -2dy/y^3 + dy/y^2
Integrate
-1/x = y^2 - y + C

Where y ≠ 0 & x ≠ 0

While the book gives an answer of
(cx+1)y^2 = (y -1)x
with the same domain

2. Jul 9, 2012

### BruceW

dx/x^2 = -2dy/y^3 + dy/y^2
This line was correct, but then integrating this does not lead to your next line:
-1/x = y^2 - y + C
I think it was maybe a mistake? I've done similar mistakes loads of times.

3. Jul 9, 2012

### SammyS

Staff Emeritus
Check the integration for y.

What is $\displaystyle \int (-2y^{-3}+y^{-2})\,dy\ ?$

4. Jul 9, 2012

### Nano-Passion

Thank you Bruce & Sammy. It seems mental integration isn't always reliable.

So my new answer is -1/x = y^2 - y^-1 + C

But I still don't have the answer in the book (cx+1)y^2 = (y -1)x

5. Jul 9, 2012

### SammyS

Staff Emeritus
Hopefully, you mean -1/x = y-2 - y-1 + C.

6. Jul 9, 2012

### Nano-Passion

Hmm, forgot to fix that too!

7. Jul 10, 2012

### SammyS

Staff Emeritus
Does that mean you now have the book's answer?

8. Jul 10, 2012

### Nano-Passion

Yes! By distributing the variables within parentheses and then dividing by x*y^2.

Thank you. I am not sure why the book chose to put it in that form though.

9. Jul 10, 2012

### BruceW

Nice work! The book might have chosen that form to test rearranging skills. Also, there are no reciprocals which is always nice.

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