How Do You Integrate dx/sqrt(x^2 + a^2)?

  • Thread starter Thread starter sokratesla
  • Start date Start date
  • Tags Tags
    Form Integral
Click For Summary

Homework Help Overview

The discussion revolves around solving the integral \(\int \frac{dx}{\sqrt{x^2 + a^2}}\), which arises in the context of evaluating electric potential in electrodynamics. Participants explore various methods and substitutions to approach this integral.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss different substitution methods, including trigonometric and hyperbolic substitutions. There are mentions of using identities and transformations to simplify the integral. Some express confusion about the relationship between the integral and its application in electrodynamics.

Discussion Status

The conversation includes multiple attempts at solving the integral, with some participants providing hints and alternative methods. There is recognition of differing answers from various computational tools, prompting further exploration of the integral's properties. No explicit consensus has been reached, but several productive directions have been suggested.

Contextual Notes

Participants note the context of the problem as part of an electrodynamics exam, with some expressing concern about their ability to solve integrals despite being advanced students. There is a discussion about the use of resources and formulas during exams.

sokratesla
Messages
21
Reaction score
0
integral in the form dx/sqrt(x^2 + a^2)

Homework Statement


# Hi, this is my first post!
# I have to solve an integral in my electrodynamics study for evaluation of an electric potential. They are in the form [tex]V(\mathbf{r})=\frac{1}{4\pi\epsilon_0}\int \rho(\mathbf{r'})/\abs(\mathbf{r}-\mathbf{r'}) d\mathbf{r'}[/tex] In order to find the potential on the top of a conical surface I have to solve an integral in the following form:

Homework Equations


[tex]\int \frac{dx}{\sqrt{x^2+a^2}}[/tex]

The Attempt at a Solution


I solved another problem of the kind
[tex]\int \frac{x dx}{\sqrt{x^2+a^2}}[/tex]
but stucked with this one.
Can you give a hint, or help in order to solve this integral.

http://integrals.wolfram.com/index.jsp
Mathematica gives [tex]\log(x+\sqrt{x^2+a^2})[/tex]
 
Last edited:
Physics news on Phys.org
Okay, the attempt at a solution?

Do you own a handbook on mathematical identites, have you search the internet and/or your books on one-dimensional calculus?
 
Have you tried any trigonometric or hyperbolic function substutions? Might work out to be arcosh or arsinh of some not so pretty thing, my guess.

Let's see, int(1/sqrt(x^2+a^2))dx = arsinh (x/a) , a>0
 
Last edited:
yes that is correct; [tex]\ln(x+\sqrt{x^2+a^2})[/tex]
it is.
 
malawi_glenn said:
Okay, the attempt at a solution?

Do you own a handbook on mathematical identites, have you search the internet and/or your books on one-dimensional calculus?

# Thanks for the reply!
# Here is my attempt. I found this substitution in my Calculus Book and thought it may be useful.
[tex]x\equiv a\tan(\theta)[/tex]
so the denominator becomes:
[tex]a\sqrt{\tan^2(\theta)+1}[/tex] and integral becomes [tex]\frac{\cos{\theta}dx}{a}[/tex]
[tex]dx=a\sec^2(\theta)[/tex]
and at last the integral after substitution is:
[tex]\int \frac{d\theta}{\cos{\theta}}[/tex]
 
Last edited:
Okay, now I am confused what you want to calculate.

[tex]V(\mathbf{r})=\frac{1}{4\pi\epsilon_0}\int ..[/tex]

or

[tex]\int \frac{dx}{\sqrt{x^2+a^2}}[/tex]

??

What is the form of the charge distribution?
 
bel said:
Have you tried any trigonometric or hyperbolic function substutions? Might work out to be arcosh or arsinh of some not so pretty thing, my guess.

Let's see, int(1/sqrt(x^2+a^2))dx = arsinh (x/a) , a>0

# Interesting, Maxima gives this as an answer. Mathematica and Maxima seems to give different answers but, then, may be this expressions are equal.
 
malawi_glenn said:
Okay, now I am confused what you want to calculate.

[tex]V(\mathbf{r})=\frac{1}{4\pi\epsilon_0}\int ..[/tex]

or

[tex]\int \frac{dx}{\sqrt{x^2+a^2}}[/tex]

??

What is the form of the charge distribution?

# Second one. First one is the general formula for vector potentials. But I reduced my problem to the second integral.
 
Okay, then:

[tex]\int \frac{dx}{\sqrt{x^2+a^2}} = \ln(x+\sqrt{x^2+a^2})[/tex]
 
  • #10
sokratesla said:
# Thanks for the reply!
# Here is my attempt. I found this substitution in my Calculus Book and thought it may be useful.
[tex]x\equiv a\tan(\theta)[/tex]
so the denominator becomes:
[tex]a\sqrt{\tan^2(\theta)+1}[/tex] and integral becomes [tex]\frac{\cos{\theta}dx}{a}[/tex]
[tex]dx=a\sec^2(\theta)[/tex]
and at last the integral after substitution is:
[tex]\int \frac{d\theta}{\cos{\theta}}[/tex]

Yes, whenever you see something like x2 + a2, the first thing you'd like to try would be the substitution: x = atan(t).

After arriving at:
[tex]\int \frac{d (\theta)}{\cos \theta}[/tex], we notice that the power of the cosine function is 1, ie, odd, so, we'll use the u-substitution: [tex]u = \sin \theta[/tex], like this:
[tex]\int \frac{d (\theta)}{\cos \theta} = \int \frac{\cos \theta d (\theta)}{\cos ^ 2 \theta} = \int \frac{\cos \theta d (\theta)}{1 - \sin ^ 2 \theta} = \int \frac{du}{1 - u ^ 2}[/tex].
Hopefully, you can go from here, right?

----------------------------

Or, a different way to tackle this integral is to use Euler Substitution.
Let [tex]t = \sqrt{x ^ 2 + a ^ 2} + x[/tex]
Take the differential of both sides yields:
[tex]dt = \frac{x}{\sqrt{x ^ 2 + a ^ 2}} dx + dx = \left(\frac{x}{\sqrt{x ^ 2 + a ^ 2}} + 1 \right) dx = \frac{x + \sqrt{x ^ 2 + a ^ 2}}{\sqrt{x ^ 2 + a ^ 2}} = \frac{t}{\sqrt{x ^ 2 + a ^ 2}} dx[/tex]

Isolate x's, and t's to one side, we have:
[tex]\Rightarrow \frac{dt}{t} = \frac{dx}{\sqrt{x ^ 2 + a ^ 2}}[/tex]
Now, integrate both sides gives:
[tex]\int \frac{dx}{\sqrt{x ^ 2 + a ^ 2}} = \int \frac{dt}{t} = \ln|t| + C = \ln|x + \sqrt{x ^ 2 + a ^ 2}| + C[/tex]

----------------------------

Or, to make it more convenient, you can try to learn this integral by heart, it's pretty common.

[tex]\int \frac{dx}{\sqrt{x^2+a^2}} = \ln(x+\sqrt{x^2+a^2}) + C[/tex]
 
Last edited:
  • #11
sokratesla said:
# Interesting, Maxima gives this as an answer. Mathematica and Maxima seems to give different answers but, then, may be this expressions are equal.

y = arcsinh(x/a)

sinhy = x/a

(e^y - e^-y)/2 = x/a

Multiply by e^y and rearrange.

e^(2y) - 2x/a e^y - 1 = 0

Use the formula for the roots of a quadratic and simplify.

e^y = (x +/- sqrt(x^2 + a^2))/a

y = ln(x + sqrt(x^2 + a^2) - lna

So, up to a constant of integration, the two answers are the same.
 
  • #12
malawi_glenn said:
Okay, then:

[tex]\int \frac{dx}{\sqrt{x^2+a^2}} = \ln(x+\sqrt{x^2+a^2})[/tex]

# Yes, it is the answers. But everywhere what I found is the answer. But I want to learn the intermediate steps too. o:) Otherway I'll feel like I'm cheating in the exam if I memorize the table without learning its proof.
http://www.sosmath.com/tables/integral/integ11/integ11.html
# Here there are two answers.

# I tried the substitution x=a*tan(u) and changed the integral to du / cos(u) but stopped there.
 
  • #13
sokratesla said:
# Yes, it is the answers. But everywhere what I found is the answer. But I want to learn the intermediate steps too. o:) Otherway I'll feel like I'm cheating in the exam if I memorize the table without learning its proof.
http://www.sosmath.com/tables/integral/integ11/integ11.html
# Here there are two answers.

# I tried the substitution x=a*tan(u) and changed the integral to du / cos(u) but stopped there.


hehe yeah, but is it an exam on integral calculus or electrodynamics?..
Are you not allowed to use collection of formulas and so on?
 
  • #14
let x = a sinh (psi)
dx = a cosh (psi) d(psi)

then int(1/sqrt(x^2+a^2))dx = int (acosh(psi)/ sqrt((a^2)(sinh(psi))^2 + a^2)) d(psi)

= int (a cosh (psi)/ a cosh (psi) ) d(psi)
= int (1) d(psi)
= psi+C

but, x = a sinh (psi)
sihn (psi)= (x/a)
(psi) = arsinh (x/a)

so, int(1/sqrt(x^2+a^2))dx = psi+C = arsinh (x/a) +C

my guess is that ln(x+ sqrt (x^2+a^2)) and this differs by a constant.

sorry I don't know how to type maths symbols, or this would have looked nicer.
 
  • #16
thanks, malawi_glenn. =)

but before I learn all that, the ln(u) thing probably came from the fact that sinh(u)+cosh(u)=e^(u).
 
  • #17
VietDao29 said:
After arriving at:
[tex]\int \frac{d (\theta)}{\cos \theta}[/tex], we notice that the power of the cosine function is 1, ie, odd, so, we'll use the u-substitution: [tex]u = \sin \theta[/tex], like this:
[tex]\int \frac{d (\theta)}{\cos \theta} = \int \frac{\cos \theta d (\theta)}{\cos ^ 2 \theta} = \int \frac{\cos \theta d (\theta)}{1 - \sin ^ 2 \theta} = \int \frac{du}{1 - u ^ 2}[/tex].
Hopefully, you can go from here, right?

# Thanks for these important advises. du/(1-u^2) seems much simpler. I think I can solve it looking my Calculus book.

# And many thanks to everybody! Have a nice day!
 
  • #18
malawi_glenn said:
hehe yeah, but is it an exam on integral calculus or electrodynamics?..
Are you not allowed to use collection of formulas and so on?

# It's and electrodynamics exam. In general, the exams are openbook or instructors give some important formulas. But they are all physics books and formulas.
# And, I think, it's a shame for last year physics students, like me, not able to solve basic integrals.:shy:
# Thanks for your help.
 
  • #19
malawi_glenn said:
Okay, then:

[tex]\int \frac{dx}{\sqrt{x^2+a^2}} = \ln(x+\sqrt{x^2+a^2})[/tex]

Actually this is true for
[tex]\int \frac{dx}{\sqrt{x^2 \pm a^2}}=\ln(x+\sqrt{x^2 \pm a^2})[/tex]
 

Similar threads

  • · Replies 105 ·
4
Replies
105
Views
11K
Replies
4
Views
3K
Replies
7
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 22 ·
Replies
22
Views
3K
Replies
3
Views
2K
Replies
9
Views
3K
  • · Replies 9 ·
Replies
9
Views
3K
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K