Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
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.
Microwave heating module - Microwave Oven, time varying frequency source
Posted 2013年6月6日 GMT-4 10:53 RF & Microwave Engineering, Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 4.3a, Version 4.3b 2 Replies
Please login with a confirmed email address before reporting spam
I'm almost new in comsol and I'm not very practice with this software.
I'm training with the Microwave Heating module, in particular I'm evaluating the behavior of the microwave oven considering the model available with the sw pack (rf - microwave oven).
As most of you probably know better than me, the model emulate the beahvior of a mw oven excited by a magnetron, simulated through a rectangular port placed behind a properly designed waveguide. The default analysis is a 5 seconds frequency transient @ 2.45 GHz (time step 1 s).
My question is the following: is it possible to perform a study defining a time varying source? in other words, my wish is to analyze the behavior of the oven (in term of potato's temperature distribution) considering a frequency transient analysis in 5 seconds but with several random frequencies excitation (in the band 2.4-2.5 GHz), varying, for example, one frequency each second (or each time step).
The idea is to generate a time varying field distribution and evaluate the heating of the potato at the end of the whole simulation (i.e. 5 seconds).
If any of you did manage this trouble or if you have any suggestion please write me.
Bye
Giovanni
Please login with a confirmed email address before reporting spam
Actually is possible to defined a function of frequency_vs_time in comsol.
Afterwards you use this function in the frequency box, for example frequency_vs_time(t) -> t is the global variable for time in comsol.
It is also possible to the same with the input power of the waveguide. I have done this for a case I am working at the moment.
Mine only problem is to do this in function of the average temperature of the potato, I have always a error window. I have posted concerning this issue, but still without answer.
- I think you jsut have to create this frequency_vs_time function before the computation with a random variation of the frequency.
Kind Regards,
Duarte Albuquerque
Please login with a confirmed email address before reporting spam
thanks for your quick and really bright reply.
Following your suggestion, I tryed to run a simulation varying the frequency as follows:
supposing to simulate a 5 seconds (step 1 s) frequency transient, I built several rect function (in global definitions), each one defined as rect_i(t-i), with i=0:1:5; in the frequency box of the study settings I wrote down something like freq1*rect1(t)+freq2*rect2(t)+.... The aim of this intricate mechanism is to manage the frequency in every time step of the simulation (not random).
Do you think it could be considered a good trick or did you envisage a smarter solution to generate a deterministic f(t) function?
Probably (I'm still thinking about it) this approach will provide you a starting point in solving your problem with temperature....by the way, is too early for me to give suggestion about Comsol problem!!!
Bye
Giovanni