Differential Equation, Substitution/Homogeneous

13 replies · 2K views · last reply
Destroxia
Messages
204
Reaction score
7

Homework Statement



Solve the differential equation:

Printed equation numbered 37: dy/dx = (x^2 - 3y^2) / (2xy).


Here is the books answer to the problem:

Equation: |x|^3 |x^2 - 5y^2| = c, with c a constant.


Homework Equations



This is a substitution/homogeneous first order differential equation, which can be converted into a separable differential equation.

The Attempt at a Solution



Handwritten derivation solving a homogeneous first-order differential equation by substitution, showing steps of separating variables and integrating, with equations stacked down the page.


Where am I going wrong? (ignore the paragraph breakers in my work, sorry)
 
The expression you have 4 lines from the bottom ##\int\frac{-1}{5m}dV## is rather meaningless. Rather than making the m substitution, try using a trigonometric substitution.
 
In general, in order for a substitution to be useful, you need to be able to replace all of the old variables with the new variable(s) to get an integral in terms of the new (including the differential)
 
BiGyElLoWhAt said:
The expression you have 4 lines from the bottom ##\int\frac{-1}{5m}dV## is rather meaningless. Rather than making the m substitution, try using a trigonometric substitution.

How would I go about the trigonometric substitution without the radical? Or can you do it without the radical?
 
BiGyElLoWhAt said:
The expression you have 4 lines from the bottom ##\int\frac{-1}{5m}dV## is rather meaningless. Rather than making the m substitution, try using a trigonometric substitution.
The integral in question here, ##\int \frac{2V dV}{1 - 5V^2}## can be done without resorting to a trig substition. As it sits, it's pretty close to ##\int \frac{du}{u}##.
 
Mark44 said:
The integral in question here, ##\int \frac{2V dV}{1 - 5V^2}## can be done without resorting to a trig substition. As it sits, it's pretty close to ##\int \frac{du}{u}##.

Yes, I just used m instead of u, as the derivative of the bottom cancels out all the V terms in the equation, and leaves me just with ms to integrate. I don't understand what I have done wrong.
 
But you haven't gotten rid of all the v's... Ok, yea you did, you just didn't replace the dV with dm. Sorry about that. upon further inspection, you actually have the same answer. At least on the second to last line. Try raising ##e^{\text{Left Hand Side}}=e^{\text{Right Hand Side}}##

It's a matter of simplification, and I actually had to go through and do it to make sure it was the same. Thanks @Mark44 for straightening me out o0)
 
BiGyElLoWhAt said:
But you haven't gotten rid of all the v's... Ok, yea you did, you just didn't replace the dV with dm. Sorry about that. upon further inspection, you actually have the same answer. At least on the second to last line. Try raising ##e^{\text{Left Hand Side}}=e^{\text{Right Hand Side}}##

It's a matter of simplification, and I actually had to go through and do it to make sure it was the same. Thanks @Mark44 for straightening me out o0)

Sorry! I forgot to replace the dV, you are right! I tried raising it to the e, earlier in the day but I suppose I didn't simplify correctly... Thank you for all your help!
 
Mark44 said:
The integral in question here, ##\int \frac{2V dV}{1 - 5V^2}## can be done without resorting to a trig substition. As it sits, it's pretty close to ##\int \frac{du}{u}##.

BiGyElLoWhAt said:
But you haven't gotten rid of all the v's... Ok, yea you did, you just didn't replace the dV with dm. Sorry about that. upon further inspection, you actually have the same answer. At least on the second to last line. Try raising ##e^{\text{Left Hand Side}}=e^{\text{Right Hand Side}}##

It's a matter of simplification, and I actually had to go through and do it to make sure it was the same. Thanks @Mark44 for straightening me out o0)

Sorry to re-awaken this problem, but I have gone through the steps of simplifying it as follows...

Handwritten derivation of a differential equation solution through a series of steps, with integration logarithms, substitution of (x^2 - 5y^2)/x^2, and a final result in the form |x^2 - 5y^2|/|x|^5 = C.


I get this answer instead of the answer to the problem provided up top.
 
No problem.
##e^{2*ln|x|}=e^{ln|x^2|}##
You have a -1/5 exponent that didn't get handled properly.
 
RyanTAsher said:
Sorry to re-awaken this problem, but I have gone through the steps of simplifying it as follows...

Handwritten integration steps with logarithms and a substitution involving (y^2 - 5y^2)/x^2, ending with the constant form |y^2 - 5x^2|/x^5 = C.


I get this answer instead of the answer to the problem provided up top.
Your initial result looked better. You did have some errors in the steps, but the result appears to agree with what the book has.
 
Alternatively, between the 3rd and 4th lines, you put the x^5 in the denominator as if it were negative, but its positive. It should stay in the numerator. I see what you did with the 1/5 now.
 
BiGyElLoWhAt said:
Alternatively, between the 3rd and 4th lines, you put the x^5 in the denominator as if it were negative, but its positive. It should stay in the numerator. I see what you did with the 1/5 now.

Yeah, I just figured that out and submitted my work. I was so intent on seeing it come out the way the book had it, I completely ignored the logarithmic rules. Thank you for your time on this problem!