Stuck? 
support@mecway.com
support@mecway.com
Mecway
FEA
Victor
About
- Username
- Victor
- Joined
- Visits
- 4,722
- Last Active
- Roles
- Administrator
Comments
-
Thanks for investigating @kuhl. That's still very confusing, hope you don't mind more questions: Are the models set to output any files when solving like matrices or table exported to .csv? Is step 1 necessary to reproduce it or it also fails if y…
-
That's a strange one. I hope you can identify what led to it if it happens again. There should be no communication between different instances of Mecway and they should be protected from each other by Windows even in the case of bugs.
-
@kuhl, I can probably do that. Could you give me more details? For instance, is it really the components in global coordinates or transformed to the orientation of the SCL line?
-
Is this the internal solver or CCX? In either case, it's not supposed to start solving again if it's already solving, but there could be a bug. Can you reproduce the problem?
-
@Sergio cm^4 and cm^2 done for v20.
-
That's certainly new to me. Thanks for discovering and sharing it! I've found some improvements: You can make the contact using the Elastic option in bonded contact instead of writing the cards by hand. The only special change needed is to add the…
-
Yes, CCX shells are only 1 solid element in the thickness direction. It shouldn't do much to solving time though because a hex20 solid has 20*3=60 DOFs while a quad8 traditional shell has 8*6=48 DOFs. Yes, hex20 can perform well with a high aspect …
-
@prop_design, solids can have element orientation for orthotropic and anisotropic materials, but yea, they don't have composites unless you build up the layers explicitly in the mesh, which isn't really practical for too many layers.
-
Since this comes up a lot, I thought I'd make a comparison between solids and shells to get a feeling for the error. Here's the well-known Scordelis-Lo roof benchmark but made more extreme with 1/10th the usual thickness, giving it a ratio of radius…
-
You can make it using a formula. Here's an example. I haven't really validated it. Unfortunately, it just show the angles as a contour plot, not visually.
-
Good find @disla. I wondered how it decides linear and found that the wigglier the displacement function is, the more increments it does, with different PEEQ for each. I also tried linear functions with different numbers of points in *AMPLITUDE and …
-
@Sebastianmaklary that does look pretty shocking having necking not occur at the notch. I have the same thought that if it's just small numerical error deciding the location, the notch should overwhelm it. I also tried it with the stress-strain curv…
-
Sorry, I misunderstood the problem. Yea that's odd that the two results end up with different plastic strain after necking. I get 2.214 vs 2.223 max. PEEQ which seems like a smaller difference than @Sebastianmaklary got? I managed it without refi…
-
I don't see any asymmetry in the model so I don't think you can identify any particular location as the right or wrong place for necking to occur. As @Sebastianmaklary pointed out, the constraints are also supposed to be symmetry BCs so if they're r…
-
Only with nonlinear analysis type (quasi-static or not) or a nonlinear material being used (v19). v18- would also do it if a nonlinear material just existed but wasn't on any elements.
-
Yes, the "time step" input has two meanings - intitial time step size for automatic time stepping and the only (constant) time step size for DIRECT. That's an interesting observation about necking initiating anywhere. I wonder if this sensitivity a…
-
That increment size of zero seems suspicious. I wonder if you've run into a bug with the time point accidentally coinciding with what the automatic increment put it at. When CCX exits abruptly, it's usually crashed, which you can imagine might happ…
-
Automatic time stepping does seem to have a problem of ignoring the loads so it can skip over big changes and then struggle to recover or ignore brief loads. What you've done by specifying time points to force it to evaluate at those sounds like a g…
-
Since it works at low speed, I'd ramp the speed - eg. enter it as "123 * t" instead of "123". Then you should see the solution up to the point of failure, if any. Turning off quasi static causes loads to be ramped automatically, so the sudden appli…
-
That's not normal. Though maybe if your model has a small mesh and large number of time steps, it could be slower due to possible overhead of calling Pardiso. It should be clearly faster with a big linear static model. Maybe it doesn't include the …
-
That's fantastic @harryvanlangen. I notice it seems to preserve the equal sizes of the elements, so it it kind of "wrapping" the mesh over the surface without stretching it more than it has to?
-
I don't have a clear idea of why it's like that. Though 20 and -20 are z at the ends of the path. 8 could be velocity - 20m/(8m/s)=2.5s to cover 20 m?. Here's an example of unittriangle. Now the constant in front means force instead of pressure and…
-
The solution isn't static, but the dynamic response is at too high a frequency to see much. Reduce time step size. A quicker way to make a moving load is using a function of both position and time. I've done that in the attached file using the heav…
-
No need. It's exactly the same archive.
-
Instructions for installing the compilers and other tools and building CCX are in here, reproduced below %programfiles%/Mecway/Mecway19/ccx/ccx_win64_mkl_pardiso_source_2.19.zip Instructions for building the following on Windows: 64 bit CCX wit…
-
@hooshsim have you tried compiling the CCX included with Mecway? That's the best option for big models if you don't have enormous amounts (~64GB) of RAM.
-
@disla what a clever workaround bonding a slice of the opposite mesh to make them match! Never heard of that before.
-
The elastic support is just linear springs so acts symmetrically in tension and compression. Version 19 (beta here https://mecway.com/download/mecway190_Beta5.msi ) has elastic compression-only supports for CCX which sounds like what you're looking…
-
I agree it's probably faster in typical cases to do something like that. I wonder about the worst cases though. What if the first 10 modes are all in the highest ND? Is that possible? Would it be much slower than finding all 10*ND modes? Maybe such …
-
Yea, I wondered about a heuristic. But what data can you build it from to be general enough? Maybe somebody comes along with a completely unexpected structure that looks like a Polish word used in English and it ends up performing worse than simply …
