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 The information provided may be out of date.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.

red variables, plot two ghraphs, no dimensions

Please login with a confirmed email address before reporting spam

Hi,
I'm using version 4.1 from today! I'm reproducing one of my models in comsol 3.5.
I've copied every thing in new version but I end up with error of not being able to find jaccobian of one of my variables. I have some questions:
1) why do some of variables appear in red and some in black? Does it mean that error is there or something?
2)how can I plot TWO variables while solving in separate graphs?
3)how can I set dimensions to NONE in comsol 4.1?
Many thanks for your attention.
Elham

3 Replies Last Post 10 gen 2011, 09:50 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 8 gen 2011, 09:57 GMT-5
Hi

if you have the new 4.1, pls check the version w.r.t the patch on the comsol main site (product updates)

And ifyou are all new to 4.1, then really start simple, one physics at the time, go through several of the Model Builder examples, and check the nice videos of v4.

My experience with translation is that these are often rather buggy (even the 4.0 and 4.0a ones! check the release notes too), so I only open the older files t copy the formulas or to learn how COMSOL translates the main items. I always restart my new models from a fresh new 4.1 template !!

Then note that all variable naming convention has changed, and many new names has arrived. For the variables a 3.5 "OldComsolVariableName_smsld" becomes a prfixed "solid.NewCsolVariableName". Then basically eerything of 3.5 exist, but is mainly put at different places, when not renamed. And many new things has appered.

Also the model node tree is to be read from top downwards = programming sequence (but sometimes you need to reloop and some (not that many) changes far down in the tree structure have impliations and changed further up (which often leads to unexpected errors, not always trivial to identify), so in doubt, restart your reading nodeby node and section by section from the top (again and again).

variables in red are naming errors, or formula errors these are fatal errors if you solve
variables in orange are units inconsistency warnings, if you lave the mouse pointer on the field, yuo will be informed briefly, these warnings are not fatal, the model will olve but might lead to errors in the results, all depends if YOU were in control or not, just being lazy not using the correct units.
Functions calls are expected unitless, so add a t[1/s], x[1/m] to void the oange.
Note still sometimes the orange color is "stuck", then jump to another tree node further down and up again, and the issue might get refreshed and resolves y it own.
--
Good luck
Ivar
Hi if you have the new 4.1, pls check the version w.r.t the patch on the comsol main site (product updates) And ifyou are all new to 4.1, then really start simple, one physics at the time, go through several of the Model Builder examples, and check the nice videos of v4. My experience with translation is that these are often rather buggy (even the 4.0 and 4.0a ones! check the release notes too), so I only open the older files t copy the formulas or to learn how COMSOL translates the main items. I always restart my new models from a fresh new 4.1 template !! Then note that all variable naming convention has changed, and many new names has arrived. For the variables a 3.5 "OldComsolVariableName_smsld" becomes a prfixed "solid.NewCsolVariableName". Then basically eerything of 3.5 exist, but is mainly put at different places, when not renamed. And many new things has appered. Also the model node tree is to be read from top downwards = programming sequence (but sometimes you need to reloop and some (not that many) changes far down in the tree structure have impliations and changed further up (which often leads to unexpected errors, not always trivial to identify), so in doubt, restart your reading nodeby node and section by section from the top (again and again). variables in red are naming errors, or formula errors these are fatal errors if you solve variables in orange are units inconsistency warnings, if you lave the mouse pointer on the field, yuo will be informed briefly, these warnings are not fatal, the model will olve but might lead to errors in the results, all depends if YOU were in control or not, just being lazy not using the correct units. Functions calls are expected unitless, so add a t[1/s], x[1/m] to void the oange. Note still sometimes the orange color is "stuck", then jump to another tree node further down and up again, and the issue might get refreshed and resolves y it own. -- Good luck Ivar

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 8 gen 2011, 10:00 GMT-5
Hi again
continuation as the screen gets mad if the thread is too long ;)

plots are one per subnode, but a plot class node (3D, 2D, 1D) migh have several subnode plots ont he same graph.

Basic idea: right-click systematically on all nodes to see what and where you might add something, this adding of sub-nodes is context depedent, so sometimes by changing a physics settings, then a sub-node already visited might gain new sub-sub-node items :)

currentlay there are no units "none", this might appear in a later version, I have heard. Personally I prefer to have the units, as they help to ensure consistency and for far easier model validation/verification. But then I'm doing "engineering simulations" and not pure science or math
--
Good luck
Ivar
Hi again continuation as the screen gets mad if the thread is too long ;) plots are one per subnode, but a plot class node (3D, 2D, 1D) migh have several subnode plots ont he same graph. Basic idea: right-click systematically on all nodes to see what and where you might add something, this adding of sub-nodes is context depedent, so sometimes by changing a physics settings, then a sub-node already visited might gain new sub-sub-node items :) currentlay there are no units "none", this might appear in a later version, I have heard. Personally I prefer to have the units, as they help to ensure consistency and for far easier model validation/verification. But then I'm doing "engineering simulations" and not pure science or math -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 10 gen 2011, 09:50 GMT-5
Thank yo Ivar for your extensive reply.
My model is a math model :2 PDEs and I don't need any dimensions so May be I have to ignore Comsol orange lines.
as you suggested in converting a model from v. 3.5a to 4.1, I started from a new file, but I end up with an error.
I'm posting a new question just now about it.
Cheers,
Elham
Thank yo Ivar for your extensive reply. My model is a math model :2 PDEs and I don't need any dimensions so May be I have to ignore Comsol orange lines. as you suggested in converting a model from v. 3.5a to 4.1, I started from a new file, but I end up with an error. I'm posting a new question just now about it. Cheers, Elham

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.