Calculating Capacitance of a Parallel Plate Capacitor Using Integration

  • Thread starter Thread starter sankalpmittal
  • Start date Start date
  • Tags Tags
    Capacitor
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
8 replies · 3K views
sankalpmittal
Messages
785
Reaction score
27

Homework Statement


See the image:
http://postimg.org/image/q550yqqd1/

Left click to enlarge...

Homework Equations



For a parallel plate capacitor, Capacitance C=KεoA/d

A is area of plate and d is distance between the two plates..

The Attempt at a Solution



I do not know how to approach. I divided the capacitor along the diagonal into two halves. For the first half,

dC1 = K1εodA/dl

I took a strip of edge dx and length "l" perpendicular to the capacitor. Area of that strip= (x+dx)2-x2=2xdx..

dA=2xdx

Hence,
dC1 = 2K1εoxdx/dl

Now taking an angle θ, tanθ=d/a= (d-l)/(a-x)...

Am I going in the right direction ?

Please help !

Thanks in advance...
:smile:
 
Last edited by a moderator:
Physics news on Phys.org
I can't clearly picture how you're dividing up the device. In my mind's eye I can see taking vertical slices of width dx across the image. At some horizontal position x, a slice will effectively consist of two capacitors in series, one with dielectric K1, the other with dielectric K2. "Plate" area is given by a*dx, and "plate" separation for each capacitor can be determined with a bit of geometry.
 
gneill said:
I can't clearly picture how you're dividing up the device. In my mind's eye I can see taking vertical slices of width dx across the image. At some horizontal position x, a slice will effectively consist of two capacitors in series, one with dielectric K1, the other with dielectric K2. "Plate" area is given by a*dx, and "plate" separation for each capacitor can be determined with a bit of geometry.

Ok, So I have a differential form,

dC2=K2εoadx/distance, as per the hint you gave. I also understand that the two capacitors are in series. So what should I replace K with ?

And about geometry... I took angle θ between diagonal and an edge. Then,

tanθ = d/a = y/x = (d-y)/(a-x)

where, the capacitor with dielectric constant K2 is positioned at distance x and its distance from the one edge(horizontal) to the boundary of second capacitor.

Am I on the right track ? Thanks..
 
sankalpmittal said:
Ok, So I have a differential form,

dC2=K2εoadx/distance, as per the hint you gave. I also understand that the two capacitors are in series. So what should I replace K with ?
You don't replace K, rather you need to find expressions for the two capacitors in terms of their individual K's and geometry, then combine them in series to yield a net capacitance for that "slice". The slices are all in parallel...
attachment.php?attachmentid=60497&stc=1&d=1374767115.gif


And about geometry... I took angle θ between diagonal and an edge. Then,

tanθ = d/a = y/x = (d-y)/(a-x)

where, the capacitor with dielectric constant K2 is positioned at distance x and its distance from the one edge(horizontal) to the boundary of second capacitor.

Am I on the right track ? Thanks..

I believe so... can you write expressions for the two capacitances of a slice?
 

Attachments

  • Fig1.gif
    Fig1.gif
    2.6 KB · Views: 1,073
gneill said:
I believe so... can you write expressions for the two capacitances of a slice?

gneill,

As per the image you gave,

dC1 = K1εoadx/d1

dC2= K2εoadx/d2

Correct ?

Now dC=dC1*dC2/(dC2+dC1) ?
 
sankalpmittal said:
gneill,

As per the image you gave,

dC1 = K1εoadx/d1

dC2= K2εoadx/d2

Correct ?

Now dC=dC1*dC2/(dC2+dC1) ?

Right. You'll want to express d1 and d2 in terms of some common variable over which you can integrate.
 
gneill said:
Right. You'll want to express d1 and d2 in terms of some common variable over which you can integrate.

Ok, so replacing d2=d-d1,

dC1 = K1εoadx/d1

dC2= K2εoadx/(d-d1)

Now, dC=dC1*dC2/(dC2+dC1)
dC=K1K2εoadx/{K1(d-d1)+K2d1}

Now to what limit shall I integrate this expression over ? Looks like I will have to eliminate d1.

By geometry, I got,

d(a-x)/a = d1

Correct ?

P.S. d is distance between plates as given. dC is small capacitance. dx is small distance between slices as per your image. (Thanks.)
 
Last edited:
sankalpmittal said:
Now to what limit shall I integrate this expression over ? Looks like I will have to eliminate d1.

By geometry, I got,

d(a-x)/a = d1

Correct ?
Sure. So you're integrating w.r.t. x. You should be able to see from the diagram the limits for x.
 
gneill said:
Sure. So you're integrating w.r.t. x. You should be able to see from the diagram the limits for x.

Thanks a lot gneill ! :smile:

I integrated the expression in limits of x from zero to "a" and got the correct answer ! Thanks once again.