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.
Visualizing total displacement as a function of frequency
Posted 01.04.2010, 07:47 GMT-4 7 Replies
Please login with a confirmed email address before reporting spam
I've build a model in which I apply a pressure to it and run a simulation to see how high is the maximum total displacement as the frequency changes. Now what I'd like to do is to make simple xy-graph with having frequency on the x-axis, and the maximum total displacement on the y-axis, but I haven't found the means to do it yet.
I'm using 3D Solid, Stress-Strain (smsld) module.
Some restrictions:
a) I can't use the "Point evaluation" since the maximum of total displacement is in different place at different frequencies, and also is never at the edges in this model.
b) I know I could use slices, but they're hard to read especially when I'd want to have numerous different frequencies shown in the same picture.
c) I can't use matlab, since my PC has issues with it. (just wouldn't work with my 64 bit vista)
Thanks for any responses in advance.
I'm using 3D Solid, Stress-Strain (smsld) module.
Some restrictions:
a) I can't use the "Point evaluation" since the maximum of total displacement is in different place at different frequencies, and also is never at the edges in this model.
b) I know I could use slices, but they're hard to read especially when I'd want to have numerous different frequencies shown in the same picture.
c) I can't use matlab, since my PC has issues with it. (just wouldn't work with my 64 bit vista)
Thanks for any responses in advance.
7 Replies Last Post 02.04.2010, 03:44 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
there is another discussion here on the min max (try a search) the conclusion was (if I remember right) that currently it's not possible without matlab in V3.5a. It will come in V4 though.
You have the min-max postprocessing feature for plots, but that is not fully what you want.
By the way note that the eigenfrequency displacements are only qualitative (not absolute), dues to their normalisation
Good luck
Ivar
there is another discussion here on the min max (try a search) the conclusion was (if I remember right) that currently it's not possible without matlab in V3.5a. It will come in V4 though.
You have the min-max postprocessing feature for plots, but that is not fully what you want.
By the way note that the eigenfrequency displacements are only qualitative (not absolute), dues to their normalisation
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thanks for the response. I'll just look into the min max, and if that won't work, I'll just do it manually until V4 will be released.
It's fine if the displacement values aren't absolute as long as they can be compared. I can just examine how big the displacement is at a eigenfrequency, and compare it to the other frequences I'm interested in. (i.e. if it says 1m at 500 Hz, and 1mm at 550 Hz it's already great amount of information)
BR,
Marcin Hurkala
It's fine if the displacement values aren't absolute as long as they can be compared. I can just examine how big the displacement is at a eigenfrequency, and compare it to the other frequences I'm interested in. (i.e. if it says 1m at 500 Hz, and 1mm at 550 Hz it's already great amount of information)
BR,
Marcin Hurkala
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
then probably the mass participation factors are of better use, unfortunately the way COMSOL normalises the eigenvectors these are not directly there as with most other FEM pogrammes.
But if your model is simple enough you can calculate them (need matlab though) there is another thread on the forum with all the code, try a search with
"Matlab script for effective modal mass extraction"
Good luck
Ivar
then probably the mass participation factors are of better use, unfortunately the way COMSOL normalises the eigenvectors these are not directly there as with most other FEM pogrammes.
But if your model is simple enough you can calculate them (need matlab though) there is another thread on the forum with all the code, try a search with
"Matlab script for effective modal mass extraction"
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thanks for your answer.
Unfortunately Matlab isn't (currently) an option, but I'll try to keep that in mind in case they'll release new version that is more compatible with Vista 64.
BR,
Marcin Hurkala
Unfortunately Matlab isn't (currently) an option, but I'll try to keep that in mind in case they'll release new version that is more compatible with Vista 64.
BR,
Marcin Hurkala
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
I'm not the IT specialist in 64 it (trying to keep away from it as it catches to much of my time ;) but we use both software on UX and Vista here so it's possible
Good luck
Ivar
I'm not the IT specialist in 64 it (trying to keep away from it as it catches to much of my time ;) but we use both software on UX and Vista here so it's possible
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
The thing is that when we tried to install or use Matlab, we had to turn off all network connections or it would fail. And when we are using floating licenses at the university, it's just not possible to use the program in offline mode. Are you perhaps using 32 bit version of Vista or non-floating licenses? If you're using 64 bit Vista and floating licenses and have no issues with getting them to work, could you tell me which version of Matlab you're using? It might work in our environment as well in that case.
BR,
Marcin Hurkala
BR,
Marcin Hurkala
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
we have a mix of about everything with matlab (because of prising issues , it would be much easier if we could get a reasonnale deal with them, they lack obviously a good competitor), fixed & floating, 32&64 bit, UX differnt blends, MS XP, Vista, even W7 now, on old R14 up to latest 2009, but I have managed to leave off installations for the last 5 years, and I do not want to start again ;).
The only thing I'm not using is in fact MSVista 64 with COMSOL&matlab, because I'm well served so far in UX64.
But next on my list is a dual boot UX/MS-64 workstation for COMSOL+matlab+Solidworks so I can mix and get te best of all (MS for CAD-Comsol, UX for heavy duty COMSOL), so your issue might come up then, we will see.
Good luck
Ivar
we have a mix of about everything with matlab (because of prising issues , it would be much easier if we could get a reasonnale deal with them, they lack obviously a good competitor), fixed & floating, 32&64 bit, UX differnt blends, MS XP, Vista, even W7 now, on old R14 up to latest 2009, but I have managed to leave off installations for the last 5 years, and I do not want to start again ;).
The only thing I'm not using is in fact MSVista 64 with COMSOL&matlab, because I'm well served so far in UX64.
But next on my list is a dual boot UX/MS-64 workstation for COMSOL+matlab+Solidworks so I can mix and get te best of all (MS for CAD-Comsol, UX for heavy duty COMSOL), so your issue might come up then, we will see.
Good luck
Ivar