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 The information provided may be out of date.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.

What's the meaning of this error information "Undefined value found" ?

Please login with a confirmed email address before reporting spam

What's the meaning of this error information below?In the error, the coordinates are the points of the grid nodes. the model is 1D. the model file has been attached.
----------
Undefined value found.
- Detail: Undefined value found in the stiffness matrix.
There are 15 equations giving NaN/Inf in the matrix rows for the variable mod1.c
at coordinates: (0), (5e-010), (1e-009), (1.5e-009), (2e-009), ...
There are 15 equations giving NaN/Inf in the matrix rows for the variable mod1.phi
at coordinates: (0), (5e-010), (1e-009), (1.5e-009), (2e-009), ...
and similarly for the degrees of freedom, NaN/Inf in the matrix columns
----------


2 Replies Last Post 26 dic 2011, 00:57 GMT-5
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 30 mar 2011, 01:34 GMT-4
Hi

first of all you may replace your an1(x)=1-flc2hs(x-ss,lamda2) by a step function going from 1 to 0, with offset "ss" and smoothing "lamda2/2" this is the new "pre-cooked" are the V4 way ;)

Then to get rid of the unit/orange issue on Initial values, add a phi0(x[1/m]) to make the function call unitless, for the Source term in the general form it's more complex, havent really found out, would neeed quite more time, that I do not have now

Then for me the equations are not restrained enough, you need some Direichlet or Neumann BC conditions more here, but as I'm not familiar enough with your model I'm nut sure what to put

--
Good luck
Ivar
Hi first of all you may replace your an1(x)=1-flc2hs(x-ss,lamda2) by a step function going from 1 to 0, with offset "ss" and smoothing "lamda2/2" this is the new "pre-cooked" are the V4 way ;) Then to get rid of the unit/orange issue on Initial values, add a phi0(x[1/m]) to make the function call unitless, for the Source term in the general form it's more complex, havent really found out, would neeed quite more time, that I do not have now Then for me the equations are not restrained enough, you need some Direichlet or Neumann BC conditions more here, but as I'm not familiar enough with your model I'm nut sure what to put -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 dic 2011, 00:57 GMT-5
Hi! Did this problem finally get solved? I am facing the same problem on a different physical setup. This is the message I get:

"Undefined value found.
- Detail: Undefined value found in the stiffness matrix.
There are 61970 equations giving NaN/Inf in the matrix rows for the variable mod1.R1
at coordinates: (1.53872e-10,1.82966e-10,2.5563e-10), (1.61255e-10,1.78767e-10,2.58712e-10), (1.68638e-10,1.74569e-10,2.61794e-10), (1.54324e-10,1.81999e-10,2.60782e-10), (1.61733e-10,1.77834e-10,2.63944e-10), ...
There are 61970 equations giving NaN/Inf in the matrix rows for the variable mod1.R2
at coordinates: (1.53872e-10,1.82966e-10,2.5563e-10), (1.61255e-10,1.78767e-10,2.58712e-10), (1.68638e-10,1.74569e-10,2.61794e-10), (1.54324e-10,1.81999e-10,2.60782e-10), (1.61733e-10,1.77834e-10,2.63944e-10), ...
and similarly for the degrees of freedom, NaN/Inf in the matrix columns"

I am doing ALE method, setting boundary velocities on "Prescribed Mesh Velocity" as:

vx = R1t * nx,
vy = R1t * ny

and same for second boundary, where I use R2t. R1 and R2 are dependent variables to be solved for, only initial values are provided. Is this the right way to set boundary velocities? (I want radial growth of a spherical shell). If so, what goes wrong with my problem? I don't think I am missing any Boundary Conditions.

I am attaching my model, Ivar, could you help me? Or someone else?
Hi! Did this problem finally get solved? I am facing the same problem on a different physical setup. This is the message I get: "Undefined value found. - Detail: Undefined value found in the stiffness matrix. There are 61970 equations giving NaN/Inf in the matrix rows for the variable mod1.R1 at coordinates: (1.53872e-10,1.82966e-10,2.5563e-10), (1.61255e-10,1.78767e-10,2.58712e-10), (1.68638e-10,1.74569e-10,2.61794e-10), (1.54324e-10,1.81999e-10,2.60782e-10), (1.61733e-10,1.77834e-10,2.63944e-10), ... There are 61970 equations giving NaN/Inf in the matrix rows for the variable mod1.R2 at coordinates: (1.53872e-10,1.82966e-10,2.5563e-10), (1.61255e-10,1.78767e-10,2.58712e-10), (1.68638e-10,1.74569e-10,2.61794e-10), (1.54324e-10,1.81999e-10,2.60782e-10), (1.61733e-10,1.77834e-10,2.63944e-10), ... and similarly for the degrees of freedom, NaN/Inf in the matrix columns" I am doing ALE method, setting boundary velocities on "Prescribed Mesh Velocity" as: vx = R1t * nx, vy = R1t * ny and same for second boundary, where I use R2t. R1 and R2 are dependent variables to be solved for, only initial values are provided. Is this the right way to set boundary velocities? (I want radial growth of a spherical shell). If so, what goes wrong with my problem? I don't think I am missing any Boundary Conditions. I am attaching my model, Ivar, could you help me? Or someone else?

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.