Mutual Inductance - Wavy Loop

Loop 1 1 is a circle:

x 1 = cos θ 1 y 1 = sin θ 1 z 1 = 0 0 θ 1 2 π x_1 = \cos \theta_1 \\ y_1 = \sin \theta_1 \\ z_1 = 0 \\ 0 \leq \theta_1 \leq 2 \pi

Loop 2 2 is a "wavy circle":

x 2 = cos θ 2 y 2 = sin θ 2 z 2 = 3 + 2 sin θ 2 0 θ 2 2 π x_2 = \cos \theta_2 \\ y_2 = \sin \theta_2 \\ z_2 = 3 + 2 \sin \theta_2 \\ 0 \leq \theta_2 \leq 2 \pi

If Loop 1 1 carries 1 1 unit of electric current, what is the magnetic flux through Loop 2 2 ?

Details and Assumptions:
1) Magnetic permeability μ 0 = 1 \mu_0 = 1
2) This link may be helpful (hyperphysics website)
3) Give your answer as a positive number


The answer is 0.1106.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Karan Chatrath
May 10, 2020

Consider a point on loop 1:

r 1 = cos θ 1 i ^ + sin θ 1 j ^ + 0 k ^ \vec{r}_1 = \cos{\theta_1} \ \hat{i} + \sin{\theta_1} \ \hat{j} + 0 \ \hat{k}

And that on loop 2:

r 2 = cos θ 2 i ^ + sin θ 2 j ^ + ( 3 + 2 sin θ 2 ) k ^ \vec{r}_2 = \cos{\theta_2} \ \hat{i} + \sin{\theta_2} \ \hat{j} + (3+2\sin{\theta_2}) \ \hat{k}

The arc length elements on each loop can be computed by evaluating the total differentials:

d r 1 = ( sin θ 1 i ^ + cos θ 1 j ^ + 0 k ^ ) d θ 1 d\vec{r}_1 = \left(-\sin{\theta_1} \ \hat{i} + \cos{\theta_1} \ \hat{j} + 0 \ \hat{k} \right) \ d\theta_1 d r 2 = ( sin θ 2 i ^ + cos θ 2 j ^ + 2 cos θ 2 k ^ ) d θ 2 d\vec{r}_2 = \left(-\sin{\theta_2} \ \hat{i} + \cos{\theta_2} \ \hat{j} + 2\cos{\theta_2} \ \hat{k} \right) \ d\theta_2

Finally, we can compute the elementary magnetic vector potential at any general point due to loop 1:

d A = μ o I 4 π ( d r 1 r r 1 ) d\vec{A} = \frac{\mu_o I}{4 \pi}\left(\frac{d\vec{r}_1}{\lvert \vec{r} - \vec{r}_1\rvert}\right)

The total magnetic potential vector due to loop at any point can be evaluated as such:

A = μ o I 4 π L 1 d r 1 r r 1 \vec{A} = \frac{\mu_o I}{4 \pi} \oint_{L1} \frac{d\vec{r}_1}{\lvert \vec{r} - \vec{r}_1\rvert}

Now, we know that the magnetic field at that point can be computed as such:

B = × A \vec{B} = \nabla \times \vec{A}

The elementary surface area through the loop is say: d S d\vec{S} . This means that elementary flux is:

d Φ = B d S d\Phi = \vec{B} \cdot d\vec{S}

The total flux can be computed by integrating over the surface area of loop 2 as such:

Φ = S 2 B d S \Phi = \oint_{S2}\vec{B} \cdot d\vec{S} Φ = S 2 ( × A ) d S \implies \Phi = \oint_{S2} \left(\nabla \times \vec{A}\right) \cdot d\vec{S}

Applying Stoke's theorem leads to:

Φ = S 2 ( × A ) d S = L 2 A d r 2 \Phi = \oint_{S2} \left(\nabla \times \vec{A}\right) \cdot d\vec{S} = \oint_{L2} \vec{A} \cdot d\vec{r}_2 Φ = L 2 L 1 μ o I 4 π d r 1 r 2 r 1 d r 2 \Phi = \oint_{L2} \oint_{L1} \frac{\mu_o I}{4 \pi} \frac{d\vec{r}_1}{\lvert \vec{r}_2 - \vec{r}_1\rvert}\cdot d\vec{r}_2

Substituting expressions and simplifying the integrand gives a function f ( θ 1 , θ 2 ) f(\theta_1,\theta_2) and the integral looks as such:

Φ = 0 2 π 0 2 π f ( θ 1 , θ 2 ) d θ 1 d θ 2 \Phi = \int_{0}^{2 \pi} \int_{0}^{2 \pi} f(\theta_1,\theta_2) d\theta_1 \ d\theta_2

I performed this entire calculation process using a script of code and have only provided an outline. The answer evaluates to:

Φ 0.1106 \boxed{\Phi \approx 0.1106}

The use of magnetic vector potential is a really neat trick to use especially when it is difficult to parameterise the surface on which the loop of interest lies. Thanks for posting these problems. I learned something new in the process.

Karan Chatrath - 1 year, 1 month ago

Log in to reply

Exactly. This was something of a revelation to me. For years, I have wanted to calculate inductances for crazy loops and coils, but couldn't because the interior surface was difficult or impossible to parameterize. As it turns out, the interior surface is irrelevant, thanks to the magnetic vector potential. I posted this pair of problems in order to illustrate that.

Steven Chase - 1 year, 1 month ago

@Karan Chatrath very nice and explanatory solution. Can you see, in my method I have use some tricks?

A Former Brilliant Member - 1 year, 1 month ago

Log in to reply

Yes, I did. I want to understand what you did in a better way, so I have requested for more details.

Karan Chatrath - 1 year, 1 month ago

Log in to reply

@Karan Chatrath I have encountered a little error in my solution. Therefore I think you have asked me to elaborate Right??? Mere solution me kuch galat hai kya ap pehle hi bata do??

A Former Brilliant Member - 1 year, 1 month ago

Log in to reply

@A Former Brilliant Member I asked for the solution as I wanted to understand it better. I could not solve the question by parameterising the surface on which the second loop lies, that's why.

Karan Chatrath - 1 year, 1 month ago

Log in to reply

@Karan Chatrath @Karan Chatrath I have not parametrise anything special. I just place the loop in every z coordinate. And take it's average from 1 to 5.nothing else.

A Former Brilliant Member - 1 year, 1 month ago

Log in to reply

@A Former Brilliant Member Do you get the correct answer to the second problem by applying this method?

Karan Chatrath - 1 year, 1 month ago

Log in to reply

@Karan Chatrath @Karan Chatrath even I don't tried the second problem. Because I know that this loops are not placed symmetric in part 2.Therefore if i try, it will lead to 4 integrals which wolfram can't solve. Therefore I leave it. Kyo bhaiya aap hi dekho jo second problem hai usme unsympathetic position hai. And one more thing. Mujhe maaf kardo kal jo maine galti ki.

A Former Brilliant Member - 1 year, 1 month ago

F u n P r o b l e m \textcolor{#BA33D6}{FunProblem} As always, it is a pleasure to solve your problems. Thank you for another good one.

The wavy circle is changing its position according to the curve s i n θ 2 sin \theta_{2} . Its range is from z = 1 z=1 to z = 5 z=5
Take a z = γ z=\gamma a coordinate . From the centre of loop 2 2 move β \beta rightwards and take a elementary strip of width d β d\beta Find field at this point ( β , 0 , γ ) (\beta, 0,\gamma) . Multiply it with d A = 2 π β d β dA=2π \beta d\beta Now by doing this , average it over γ \gamma , γ = 1 \gamma=1 to γ = 5 \gamma=5
After that you will reach this expression ϕ = 8 1 1 5 0 1 0 2 π β ( 1 + β c o s α ) ( γ 2 + s i n 2 α + ( c o s α β ) 2 ) 1.5 d α d β d γ 0.11 \large \phi = 8^{-1}\int_{1}^{5} \int_{0}^{1} \int_{0}^{2\pi} \frac{\beta(1+\beta cos \alpha)}{(\gamma^{2}+sin^{2} \alpha+(cos \alpha-\beta)^{2})^{1.5}} d \alpha d\beta d\gamma \approx 0.11
I will elaborate my solution, if asked by anyone.


I would like to see an elaborate solution. Please share if possible. Thanks in advance.

Karan Chatrath - 1 year, 1 month ago

Log in to reply

@Karan Chatrath Yes sir why not. I will .

A Former Brilliant Member - 1 year, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...