Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

The penetration of ac electric field through the dielectric film

Please login with a confirmed email address before reporting spam

Hello~
I am studying a subject about the the penetration of the ac electric field through the dielectric film (PDMS) by adjusting the frequency to find a cuf-off frequency. I used the AC/DC model_Quasi-Static, Electric and set the ac pontential as v*sin(2*pi*f*t). However, the distribution of the electric field did not change as the frequency is varying. Have anyone done the similar subject before, could you please share your experinces?

Thanks.

5 Replies Last Post 25.07.2012, 00:57 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.05.2012, 01:02 GMT-4
Hi

you should decide between the time series with amplitudes explicitely written as v*sin(2*pi*f*t) and frequency domain, harmonic developments in omega=2*pi*freq. where you define the frequency (or sweep range) the amplitude and phase, and get amplitude or rms values back.

Check the physics and the equations, the frequency sweep is much computational lighter than a true time series over several periods

--
Good luck
Ivar
Hi you should decide between the time series with amplitudes explicitely written as v*sin(2*pi*f*t) and frequency domain, harmonic developments in omega=2*pi*freq. where you define the frequency (or sweep range) the amplitude and phase, and get amplitude or rms values back. Check the physics and the equations, the frequency sweep is much computational lighter than a true time series over several periods -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.05.2012, 12:27 GMT-4
Hi Ivar,

Thanks for your reply. I hvae setted the amplitude, frequency and time in the table of constant, when I gave the boundary condition as v*sin(2*pi*f*t). However, I am still confused with developing the harmonic domain. Could you please explain more detailedly? Thanks.
Hi Ivar, Thanks for your reply. I hvae setted the amplitude, frequency and time in the table of constant, when I gave the boundary condition as v*sin(2*pi*f*t). However, I am still confused with developing the harmonic domain. Could you please explain more detailedly? Thanks.

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.05.2012, 14:50 GMT-4
Hi

if you select a time dependent solver (and only if that) then COMSOL defines "t" that it steps along by the solver, then you can define BC's depending on time "t" such as V0*sin(2*pi*freq*t) (freq must also be defined typiclly via a Parameter or via a Parametric Sweep stepper.

If you choose the harmonic solver => Frequency Dependent Study then you define ony V0 for the mplitude, and freq is set in the solver "range" field (or a frequency sweep i.e: range(10,2,100)*1[Hz]). but "t" is NOT defined for Frequency Dependent solvers so it should NOT appear, or you must fix it via a Parameter call, but it will not change during the study

--
Good luck
Ivar
Hi if you select a time dependent solver (and only if that) then COMSOL defines "t" that it steps along by the solver, then you can define BC's depending on time "t" such as V0*sin(2*pi*freq*t) (freq must also be defined typiclly via a Parameter or via a Parametric Sweep stepper. If you choose the harmonic solver => Frequency Dependent Study then you define ony V0 for the mplitude, and freq is set in the solver "range" field (or a frequency sweep i.e: range(10,2,100)*1[Hz]). but "t" is NOT defined for Frequency Dependent solvers so it should NOT appear, or you must fix it via a Parameter call, but it will not change during the study -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24.07.2012, 11:39 GMT-4
hi ivar

i have tried time domain to produce ac signal . i used 5*sin(2*pi*f*t) as voltage input and i goes as u said. f = 1 MHz but but my final voltage came in negtaive. and final results was distorted.
please tell me stepwise how to do this in frequency domain where i should put omega and t value.
please help me.

hi ivar i have tried time domain to produce ac signal . i used 5*sin(2*pi*f*t) as voltage input and i goes as u said. f = 1 MHz but but my final voltage came in negtaive. and final results was distorted. please tell me stepwise how to do this in frequency domain where i should put omega and t value. please help me.

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.07.2012, 00:57 GMT-4
Hi

in frequency domain you have no "t" you define the frequency and you use equations deveopped for the frequency omega, in fact 2*pi*f, it's like a Lapalce development of your equations. Therefore you drop also the sin() from your BC and you give ONLY the amplitudes (possily the phase(s) if required), and you get amplitudes as replies (sometimes you have/(had?) RMS values check the doc

--
Good luck
Ivar
Hi in frequency domain you have no "t" you define the frequency and you use equations deveopped for the frequency omega, in fact 2*pi*f, it's like a Lapalce development of your equations. Therefore you drop also the sin() from your BC and you give ONLY the amplitudes (possily the phase(s) if required), and you get amplitudes as replies (sometimes you have/(had?) RMS values check the doc -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.