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 The information provided may be out of date.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.

Is it possible to have a varying geometry?

Please login with a confirmed email address before reporting spam

Hi all,

I am looking at building a model which describe the distribution of potential (band bending) within a semiconductor.
The potential distribution is given in the following function

V = V0 * (1-(x/W))^ 2 at 0 < x < W
V = 0 at x >W

Where
W = Ld * ( 2* V0 * F / (R * T) ) ^ (1/2)

and Ld, F, R and T are all constants

V0 will be a variable which I would like to make a step change later in the model.

The problem is that when I try to define this in a piecewise function, it all work well if my V0 and W is defined as a parameter having a fix value. When I start to make V0 and W as a variable, my piecewise function does not work.

Any suggestion on how to solve this problem?
Thank you

Best regards,
Chin Kin

4 Replies Last Post Apr 8, 2011, 7:03 a.m. EDT
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 Apr 7, 2011, 11:17 a.m. EDT
Hi

what about trying an analytical function of the type

y=x/W=x/sqrt(Ld*(2*V0*F/(R*T))

V(y) = V0*(1-y)^2*(y<1)

and restrain to y>=0
--
Good luck
Ivar
Hi what about trying an analytical function of the type y=x/W=x/sqrt(Ld*(2*V0*F/(R*T)) V(y) = V0*(1-y)^2*(y=0 -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 7, 2011, 1:34 p.m. EDT
Thank you Ivar. It works, but not when I would like to implement a function to V0 under time dependant study.
The way I introduce the time dependant study is I let the function

V(y) = V0*step1(t)*(1-y)^2*(y<1)

where step1 is the step function which I am interested to look at in a time dependant model.

The error I obtained is undefined variable, t.
Is it not possible for comsol to recognise that t is part of the time dependant function?

Please help.
Thanks

Best regards,
Chin Kin
Thank you Ivar. It works, but not when I would like to implement a function to V0 under time dependant study. The way I introduce the time dependant study is I let the function V(y) = V0*step1(t)*(1-y)^2*(y

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 Apr 8, 2011, 12:57 a.m. EDT
Hi

normally yes, but I have noticed that if you do not have any transient study defined, or even loaded then "t" is often not recognised.

Workaround: define a "Parameter" as t = 0[s] then it normally works. As a Parameter is read once at execution of the solver it can be overwritten later by COMSOL without any errors

--
Good luck
Ivar
Hi normally yes, but I have noticed that if you do not have any transient study defined, or even loaded then "t" is often not recognised. Workaround: define a "Parameter" as t = 0[s] then it normally works. As a Parameter is read once at execution of the solver it can be overwritten later by COMSOL without any errors -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 8, 2011, 7:03 a.m. EDT
Thanks. Problem solved.

Best regards,
Chin Kin
Thanks. Problem solved. Best regards, Chin Kin

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.