Made two simple test models, lets say cantilever plate, one meshed using plates and another using solids. Run them without fluid region and with fluid regions with same setting (height, density). Compare results and you will see if both meshes produce close frequencies both in "dry" and "wet"...
This make sense to add RBE2 element ability to expand so it do not produce artificial stress when used for example in bolt modeling or if RBE2 element used for beam to plate transition.
Thermal expansion is just change of linear dimensions, in don`t require cross section or thickness only...
RBE2 elements have thermal expansion coefficient for thermal stress problems.
Also this remark may be interest.
By default CASE CONTROL command RIGID=LINEAR so maybe this is cause of problem.
You have couple of options to account added water mass.
Simpler one is to use MFLUID region as shown in this video.
More resource consuming method is to use specific elements that represent water.
In both cases check Nastran documentation file adv_dyn.pdf located...
It is hard to understand your setup only from text description. Can you post some pictures or maybe share input .dat file for some simple test model.
If you have one side fully constrained than you definitely have some stress in your model.
UPD. Maybe try 3-2-1 constraint method instead of...
Make a hole in the center of roller, create RBE2 spider and apply enforced rotation load to a spider node. As bonus with hole in center you can make nice structural mesh of roller.
UPD. In general use enforced translation and rotation to set up movement of modeled tools.
Ugh...
In Femap you can use tool Model->Output->Process to multiply all results by "-1" but i don`t know if NX has similar tool. Again in Femap it is possible to expand dynamic results from Amplitude/Phase to results at discrete phase angles so you get results from positive to negative...
You can also look into Openradioss capabilities of importing Nastran deck into it. Some comments in the Web mention that Openratioss is good at reading Nastran input files (.dat, .bdf .nas etc.).
Better not trust users that do not understand how Frequency Response Analysis works.
In Frequency Analysis you define oscillating load that change from positive to negative value according to sine function. In result of Frequency analysis you obtain Amplitude and Phase of some value of interest...
Hello brut3. Can you show us pictures of your mesh or describe what type of mesh you want to get? Maybe you try to create solid mesh on this thin walled part and get errors.
GregLocock
It moves, here is zoomed view.
I made nonlinear analysis (SOL106, Nastran) with geometric nonlinearity.
Yep, that`s what I get when use animated view with setting Linear-Full in Femap. Here is better animation with intermediate nonlinear results.
I am not a Strand7 but I assume that you can and know how to fix any of 6 DOFs in any particular node in your mesh. Than you should use 3-2-1 method to apply additional support to model.
https://www.digitalengineering247.com/article/free-floating-fea-models/
Here is an example, plate located in...
I am not familiar with this exact type of Nastran solver - SOL400 but in general Normal Modes is linear solution only. You can find modes either for closed or open gap. You can use prestressed state from nonlinear static analysis but it still represent only one contact state - either open o...
For initial guess you can decompose complex loading into basic directions and think what modes they excite and what not. Than use higher of possible frequencies. Like you have bending and axial load and axial frequency if bigger than use it.
Than to prove that your timestep is small enough you...
...mode period.
I have made error in my description. What i mean is you have oscillation period T=7.752E-05. You calculate timestep dt as (1/20)*T=T/20=7.752E-05/20=3.88E-06 sec.
Now you can estimate whole simulation time as Load Pulse duration + 10 oscillation periods (again rule of thumb, 10...
...same order as one period of oscillation (T=1/lambda) of rod axial mode. To capture transient behavior you should use timestep not greater than 1/12*T or 1/20*T. Think of it as you have one sine wave with period T and you need some number of points to capture it. So 6 points per half sine wave...
In Nastran axis for axisymmetric elements is hardcodded so you do not need to define it. Your axisymmetric mesh (element types CQUADX4, CQUADX8, CTRAX3, and CTRAX6) should obey those rules (taken from Nastran Quick Refrence Guide, QRG.pdf).
Maybe if O-ring is made from some hard material than it if possible to use formulas for wedge, for example from here.
https://engineeringstatics.org/Chapter_09-block-and-wedge-friction.html
From your pictures P is much lower than Fy but for wedge you should get much bigger P than initial load...
