Видео

@@xzarrandiko Yeah, I also did it in Abaqus. Sometimes double-check is helpful.

Whats the difference of setting the mesh thickness in the section menu instead of the Thicken Mesh on the Meshing Parameters step?

Not necessarily. You can mesh right away with some defaults (usually way too coarse). You can also add only the Meshing Parameters. Then it will be meshed with Netgen. To use Gmsh, you have to choose its meshing objects.

@@EngineerAAJ The Thicken Shell Mesh feature creates a solid mesh from a surface. Shell section definition only assigns the thickness as a property for shell elements.

I used this parameter in the previous tutorial (40). It's just for the visualization. Internally, it offsets the nodes of GAPUNI (spring-like) elements but it doesn't affect their properties.

Reference points work only when assigned to rigid body constraints. But if you just want to measure something, you can request history output for an existing node (split the model into segments if you don't have any vertices and thus nodes in the desired locations).

Yes, you can import and work on meshes. Individual nodes can be selected too. Just change the selection mode in that "Set selection" window (click "More" first to show its mesh-based selection options).

@FEAnalyst tnks and thank you for the istantaneous reply

PrePoMax doesn't support beam elements but the FEM workbench in FreeCAD does (3D ones at least).

Sounds like a missing/incorrect contact definition or just a deformation scale factor larger than 1.

Sounds like missing or incorrect contact definition.

@@FEAnalyst Thanks for your reply. THat's what I think, but I am using everything identical to your tutorial. Only difference I have seen is that when I search for contact pair, it doesn't automatically find it unless I increase the search distance to unreasonably high value.

@@pathik036 Then maybe the units (dimensions) are incorrect. Try to query the geometry. It's hard to say what else can be wrong without seeing the model.

@@FEAnalyst Is there a way I can send you the model?

@@pathik036 Try posting a link to some hosting website with this file uploaded. Let's see if RUclips allows this.

Currently, PrePoMax doesn't support ortho- and anisotropic materials but they are available in CalculiX and thus can be added via Keyword Editor. The trickiest part is defining orientations. CalculiX also has composite shell elements for modeling of laminates.

Probably not exactly this (because I stick to my rule of always using hand calculations for comparison) but I may cover something similar in the future. Currently, the closest video to that is the one about Hertz contact but it's purely elastic.

Do you mean the CalculiX User's Manual with the description of the *CREEP keyword syntax and creep models in CalculiX? It's available here for the newest ccx version: www.dhondt.de/ccx_2.22.pdf

@@harishnanjundaswamy3600 It’s Matej’s website but yes - it’s temporarily down because the website of the University hosting the PrePoMax website was hacked. If you want to download a new version of PrePoMax, it’s available in the News section of the PrePoMax forum. No need to log in: prepomax.discourse.group/t/prepomax-website/2108/2

@@harishnanjundaswamy3600 It’s Matej’s website but yes - it’s temporarily down because the website of the University hosting the PrePoMax website was hacked. If you want to download a new version of PrePoMax, it’s available in the News section of the forum. No need to log in: prepomax.discourse.group/t/prepomax-website/2108/2

@@nazimyildizpau Thanks. Answering your questions: 1. Thermal conductivity is part of the material properties needed for FEM calculation based on the Fourier’s law. For steady-state analysis, this is sufficient. For transient analysis you also need specific heat and density. 2. The cases I show in my tutorials aren’t actual designs, only simple examples made up by me while trying to keep possibly reasonable values.

@@ellipse7 No, it’s made only for Windows and you need a virtual machine or an emulator like Wine.

Dziękuję. W tym przykładzie po prostu wybrałem sensowną skalę po serii testów. Ale zwykle amplitudę imperfekcji przyjmuje się m.in. na podstawie rekomendacji z norm. Jest to generalnie ułamek grubości powłoki. Tu akurat ok. 20%. Ale powłoki są zazwyczaj bardzo wrażliwe na imperfekcje, więc warto sprawdzić wpływ innych wartości na wyniki.

Only by approximating it as a proper film coefficient (there are different values for natural and forced convection in various media). To simulate it more accurately, you would need connection with CFD solver like OpenFOAM which can't be done with PrePoMax at the moment but there's a preCICE library that you can try using to couple CalculiX with OpenFOAM.

@@FEAnalyst thank you :)!

No need to download it separately, it's included with PrePoMax.

Dziękuję, faktycznie lubię wyszukiwać takie niestandardowe przypadki w literaturze.

"a" in the formulas is the radius of curvature of the spherical shell.

Można to uprościć i policzyć bez kontaktu zadając tylko odpowiednie wymuszenie, ale chodziło tu o pokazanie pełnego podejścia z kontaktem, który sprawiał w tym przypadku sporo problemów, więc była okazja omówić jego niestandardowe ustawienia i różne przydatne sztuczki.

ruclips.net/video/sE5HKcAgXnw/видео.htmlsi=5kEJCSIoUx152We4

This case was problematic for contact formulation in CalculiX and required non-default stiffness specification which is recommended in many cases anyway.

@@FEAnalyst linear soft contact hasn't been necessary in that case... hard contact with friction should give good results because of no penetration

I'd love to see few more tutorials... cylindrical shell in hydraulic press, 3-point bending beam, CNC cutting, bullet penetration of steel plate and sheet metal stamping... 😎

@@jakubsobka2621 The problem here was with convergence, sometimes reducing contact stiffness helps.

Thanks. Please check my reply to the recent comment by rishabhbhatt3840, I explained it there.

There are no special features for modeling of rigid pavements in PrePoMax/CalculiX but it should be doable. The trickiest aspects would be the choice of material models and interactions between the layers plus correct meshing.

What kind of plasticity definition are you using ? Is it just perfect plasticity or with hardening (bilinear or multilinear) ? What's the level of equivalent plastic strain in those locations ? Do the values look reasonable ?

A gdyby miał Pan opisać charakterystykę naprezeniowo -odkształcenia za pomocą 3 punktów charakterystycznych... Jakie były by to wartości?

To jest uproszczony biliniowy model bez wzmocnienia jaki czasami stosujemy w Abaqus. Tylko dla umożliwienia obliczeń analitycznych, bo normalnie dajemy wzmocnienie i trzecim punktem jest wytrzy. na rozciąganie.

@@FEAnalyst Czy w Ansys ;> ??

Nie, wysokość jest 2*h (na rysunku).

@@FEAnalyst Przepraszam... "Drobny druczek"

@@jakubsobka2621 A co konkretnie się nie zgadza mimo zgodności z obliczeniami analitycznymi ?

@@FEAnalyst Wzór na naprężenia w przekroju rozciąganego ok. Na temat drugiego nie potrafię się wypowiedzieć. Generalnie chodzi o wartość przyłożonego obciążenia i zestawienie wyników analitycznych i tych wykonanych w programie.

@@wgmg7746 Dzięki, pewnie jeszcze wrócę do tego podejścia, bo jest najlepszym sposobem na uzyskanie regularnych siatek hexa z Gmsh.

@@FEAnalyst A może raz obróbka w Gmesh i import posiatkowanego do PrePoMax?

@@wgmg7746 To pokazałem w tutorialu nr 20. Od tamtego czasu PrePoMax zyskał możliwość tworzenia siatek hexa dzięki wbudowanym algorytmom Gmsh, więc do samego Gmsh już tu nie wracałem i pewnie raczej nie będę jeśli da się to samo zrobić w PrePoMax. Chyba że trafi się coś, czego tu jeszcze nie ma a Gmsh potrafi. Ale to raczej rzadkie przypadki.

Thanks. In this case, I just selected a reasonable value after a bunch of tests. But normally, imperfection amplitude is selected following the recommendations from engineering codes, among the others. It's a fraction of the shell thickness. Here, it was around 20%. But shells are usually very sensitive to imperfections so it's good to check different values.

@@FEAnalyst thank you for the information

Yes, I covered that in tutorial 20. But since then, PrePoMax gained the hex meshing capability.

@@FEAnalyst i got U man:-p Regały good tutrial but... U don't tell US whole true ;-p Take a look on the lecture about vessel model that one called Hertz contact;-P R U checkin' US? ;>

@@jakubsobka2621 I'm not sure what you mean about those two other tutorials (thin-walled pressure vessel and Hertz contact), especially since they are quite different from each other. The stress concentrations and mesh sensitivity there ?

@@FEAnalyst I understand the way of your thinking but.. U R changing the topić the way is comfortable 4 U... We're not talking about the way of meshing elements.... I mean somethong more...

@@jakubsobka2621 Then just tell me directly what you mean about those other two tutorials and I'll refer to that.