1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Need help proving an equation for power loss in a dielectric

  1. Apr 9, 2015 #1
    1. The problem statement, all variables and given/known data
    2. Relevant equations

    I am trying to prove the equation
    which can be rewritten as
    \frac{\bar{P}}{V} &= \frac{1}{2}E_0^2\sigma_{AC}\\
    &=\frac{1}{2}E_0^2\ \omega\ \epsilon_0\ \epsilon^{''}_r\\
    &=\frac{1}{2}E_0^2\ \omega\ \epsilon_0\ \epsilon_r^{'}\ \tan(\delta)
    Here $$\bar{P}$$ stands for the time-averaged power loss which satisfies the equation
    \bar{P}=\frac{1}{T}\int_0^T U\ I\ dt,
    where $$T=\frac{2\pi}{\omega}$$ is the time period, $$U=U_0 e^{j\omega t}$$ is the complex sinusoidal voltage, and $$I=j\omega\epsilon^{'}_rC_0U + \omega\epsilon^{''}_rC_0U$$.
    The instructions say to use
    U_0 &= E_0h\\
    C &= \epsilon_r\epsilon_0\frac{A}{h}\\
    V &= A\ h \\
    \tan(\delta) &= \frac{\epsilon^{''}_r}{\epsilon^{'}_r}

    3. The attempt at a solution

    The problem I face is after solving the main integral part, which is like:
    \epsilon^{''}_r*(F(T) - F(0)) + j*\epsilon^{'}_r((F(T) - F(0))
    where $$F(t) = e^{2j\omega t}$$ and I neglected all the constants for simplicity.
    $$F(T)$$ is equal to $$\exp(j*4*\pi)$$ which is 1, making $$(F(T)-F(0))$$ zero and thus the whole equation zero.

    I thought of root mean squaring both U and I to begin with, but this gives out $$\sqrt{\epsilon^{''2}_r+\epsilon^{'2}_r/2}$$ term which doesn't seem to lead to the proof result.

    I have a hunch that I maybe missing some elementary calculation basics. I would really appreciate any help that you can offer.
  2. jcsd
  3. Apr 9, 2015 #2

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Basic problem is with using complex expressions for volts and current to compute power. Can't be done.
    V = V0exp(jwt). Real part is V0 cos(wt).
    I = I0exp(jwt). Real part is I0 cos(wt).
    Multiply: P = VI = V0 I0 exp(2wt). Real part is V0 I0 cos(2wt) which has zero average over T = 2pi/w (no dc term).
    Which we know is wrong since P = V0 sin(wt) x I0 sin(wt) = V0 I0 sin2(wt) = V0 I0 (1/2)[1 - cos(2w)]
    with average P = (V0 I0)/2.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted