Advanced Electricity and Magnetism Advanced Electricity and Magnetism Physics Help Forum 
Apr 27th 2011, 10:05 PM

#1  Member
Join Date: Nov 2010
Posts: 40
 Please help in vector potential involve line integral with log. This is a new question, I am just using the old thread because the tittle applys. I want to find the vector potential A at origin due to a current segment I(t)=kt flowing along x axis from b<x<a where b>a. This mean from b to a on the left of the origin. I did not write the second part because that is not part of the question. But the book said it is: Please help me on this, thanks Alan 
 
Apr 28th 2011, 04:42 AM

#2  Forum Admin
Join Date: Apr 2008 Location: On the dance floor, baby!
Posts: 2,468

(I had to edit your quote because the system is a bit feral when it comes to how many images a post can have.)
Your integral is slightly off. <For negative x.
(Plus the arbitrary constant, of course.) This is the source of the extra negative sign outside the ln() function.
Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.
See the forum rules here.

 
Apr 28th 2011, 08:41 AM

#3  Member
Join Date: Nov 2010
Posts: 40

Originally Posted by topsquark (I had to edit your quote because the system is a bit feral when it comes to how many images a post can have.)
Your integral is slightly off. <For negative x.
(Plus the arbitrary constant, of course.) This is the source of the extra negative sign outside the ln() function.
Dan 
Is sgn(x) the sign of x?
Is this because if x>0 then x=x and dx/dx=1. But if x<0, then x=x and dx/dx=1. So if we do substitution of U=x, dU=sgn(x)dx?
So .
Thanks
Alan
Last edited by yungman; Apr 28th 2011 at 08:45 AM.

 
Apr 28th 2011, 09:42 AM

#4  Forum Admin
Join Date: Apr 2008 Location: On the dance floor, baby!
Posts: 2,468

Originally Posted by yungman Is sgn(x) the sign of x?
Is this because if x>0 then x=x and dx/dx=1. But if x<0, then x=x and dx/dx=1. So if we do substitution of U=x, dU=sgn(x)dx?
So .
Thanks
Alan 
Exactly. Note that (integral)dx/x will involve lnx. This might have been another problem with your result, but since both a and b are negative and it is (b)/(a) that enters the logarithm this does not matter.
Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.
See the forum rules here.

 
Apr 28th 2011, 06:00 PM

#5  Member
Join Date: Nov 2010
Posts: 40

Originally Posted by topsquark Exactly. Note that (integral)dx/x will involve lnx. This might have been another problem with your result, but since both a and b are negative and it is (b)/(a) that enters the logarithm this does not matter.
Dan 
Can you give an example of this? I don't quite get what you saying here.
Thanks

 
Apr 28th 2011, 06:49 PM

#6  Forum Admin
Join Date: Apr 2008 Location: On the dance floor, baby!
Posts: 2,468

Originally Posted by yungman Can you give an example of this? I don't quite get what you saying here.
Thanks 
Sure. Let a and b be positive. Then
In this case we need to beware of that negative sign on the a. The absolute value gets rid of it so we have
Similar to your problem, if we have
the negative signs cancel out anyway, so there is nothing to worry about.
Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.
See the forum rules here.

 
Apr 28th 2011, 07:04 PM

#7  Member
Join Date: Nov 2010
Posts: 40

I am confused again, my original question is like your last example:
How can it be your original answer that I thought I understand that
Thanks
Alan
Last edited by yungman; Apr 28th 2011 at 07:07 PM.

 
Apr 28th 2011, 07:07 PM

#8  Forum Admin
Join Date: Apr 2008 Location: On the dance floor, baby!
Posts: 2,468

Originally Posted by yungman How do you post the equation again? The method you showed don't work anymore!!!
I am confused again, my original question is like the last example 
Right. I don't mean to make too much out of this. All I'm saying is that the two negative signs cancel so you don't need to worry about the absolute value. Nothing more than that.
To post an equation using LaTeX copy the following line:
[img]http://latex.codecogs.com/png.latex?*LaTeX commands go here*[/img]
Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.
See the forum rules here.

 
Apr 28th 2011, 07:09 PM

#9  Member
Join Date: Nov 2010
Posts: 40

Originally Posted by topsquark Right. I don't mean to make too much out of this. All I'm saying is that the two negative signs cancel so you don't need to worry about the absolute value. Nothing more than that.
To post an equation using LaTeX copy the following line:
[img]http://latex.codecogs.com/png.latex?*LaTeX commands go here*[/img]
Dan 
I modified the last post, please read it again. Somehow the Latex work now!!!

 
Apr 29th 2011, 08:33 AM

#10  Forum Admin
Join Date: Apr 2008 Location: On the dance floor, baby!
Posts: 2,468

Originally Posted by yungman Note: I have removed your quote. Apparently there is a limit to the number of tex statements we can put in the posts and the ones in the quote count. 
Oh, I'm sorry. First we have to change this back to the original integral:
This last step follows from both limits on the integral are negative, x is negative so sgn(x) = 1. And the negative signs in the ln cancel. Also, we can remove the absolute value bars since both a and b are positive.
Now recall that
So your integral becomes:
Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.
See the forum rules here.

  Thread Tools   Display Modes  Linear Mode  