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:
2 decades ago
4 août 2009, 05:22 UTC−4
Hi Jon
Indeed the grid ratio between X and Y seems to be limited to 3 orders of magnitude (in 2D without "EQUAL" axis spacings i.e. doble click on the lower border word "EQUAL" at the side to the "GRID" then rescale the scren).
Well it has some logic, you cannot really draw on the screen with a resolution higher than the pixel dots.
As the drawing feature does not acept units I usually draw my parts in the default unit (SI [m] for me) and then use the "scale" function to convert to [mm] or [um], then I do mostly not bother about axis and grid scale settings, furthermore I draw most of my items by entering the numerical values, not with the mouse this does not need any particular grid scale factor.
If not in COMSOL I use my CAD software (SolidWorks for me).
I beleive you can set the axis and grids locally do your mouse drawings at small scale and then reset to a larger axi/grid settings and finalise the "macro items". You can also use the scale factor only in on direction (non homogenious scaling). There are several ways around.
If you are working in 3D, you should also check your repair / knit surface absolute tolerance, if you use these features, as it might make some of the small details dissapear
Do not forget that you can easily get numerical roundoff errors with 6 order of magnitude scale factors between your items, so check your results carefully.
Good luck
Ivar
Hi Jon
Indeed the grid ratio between X and Y seems to be limited to 3 orders of magnitude (in 2D without "EQUAL" axis spacings i.e. doble click on the lower border word "EQUAL" at the side to the "GRID" then rescale the scren).
Well it has some logic, you cannot really draw on the screen with a resolution higher than the pixel dots.
As the drawing feature does not acept units I usually draw my parts in the default unit (SI [m] for me) and then use the "scale" function to convert to [mm] or [um], then I do mostly not bother about axis and grid scale settings, furthermore I draw most of my items by entering the numerical values, not with the mouse this does not need any particular grid scale factor.
If not in COMSOL I use my CAD software (SolidWorks for me).
I beleive you can set the axis and grids locally do your mouse drawings at small scale and then reset to a larger axi/grid settings and finalise the "macro items". You can also use the scale factor only in on direction (non homogenious scaling). There are several ways around.
If you are working in 3D, you should also check your repair / knit surface absolute tolerance, if you use these features, as it might make some of the small details dissapear
Do not forget that you can easily get numerical roundoff errors with 6 order of magnitude scale factors between your items, so check your results carefully.
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
16 avr. 2010, 11:10 UTC−4
Hi Ivar,
In view of your post below, your constants, do you leave them in SI units too?
Thanks
Hi Ivar,
In view of your post below, your constants, do you leave them in SI units too?
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
17 avr. 2010, 03:30 UTC−4
Hi
Yes I use as far as possible only SI, I have seen in prerelease of V4 at the last conference, that you might change basic units and start in mm or um, which is easier for your graphics, but then for your physics it's easy to get a scale factor out of range.
=> always double check by two means your results ;)
Have fun Comsolng
Ivar
Hi
Yes I use as far as possible only SI, I have seen in prerelease of V4 at the last conference, that you might change basic units and start in mm or um, which is easier for your graphics, but then for your physics it's easy to get a scale factor out of range.
=> always double check by two means your results ;)
Have fun Comsolng
Ivar