Thank you for your kind reply! It really helps.
I just read the user manual of BCB and got some questions.
How can I import the solid mesh model in blender and apply constraints on indiviual elements? I tried to import tetrahedron mesh made by Abaqus, but the only surface mesh was imported in the blender. Or, Is there any method to applying rigid body for each tetrahedron elements?
I want to simulate collision-fracture situation like the reference video below.
In the user manual, it said constraints will locate in the bounary box. How can I locate constraints in faces? Because I need to use non-rectangular type of mesh.
Reference: Bullet Constraints Builder User Manual Version 2.32 for BCB v3.27 11. November 2018
The BCB does its own discretization, so you need to use the source mesh without tetrahedralization. It is possible to use prediscretized elements, though, but tetrahedrons are not recommended as the BCB is designed to work best with box-like shapes.
Constraint placement and breaking thresholds are computed automatically based on each element pair’s properties and contact area. However, manually placement of constraints is possible too, the BCB will consider already existing constraints. For instance in the video motor and hinge constraints have been used for the wheels of the car and were placed by hand.
Hi, this is not about the modifier itself, but it’s related, since I think many people here are interested about simulations as well. I hope you can take a look at this part of the simulations system being improved.
Hello again
I’m trying to simulate fracture using FM after porcelain was dropped.
But It seems the breaking threshold doesn’t work. I applied generic constraint.
It always disconnecting all individual elements.
Could you check my .blend file?
@bloox64 Yes, the FM team members and some associates are working to some degree on many aspects of blender 2.8 physics. There is a lot there that needs to be done that directly relates to a better FM in 2.8 master.
For example, Scorp, Jens, Juan, Bzzt and I are all working on OpenVDB and remesher integration for example.
Scorp and I have been keeping an eye on Jacques’ nodes work and code. I’ve been working still on FEM particles plus divergence free SPH, which has been a long term project of mine.
I’m also working on the FM pie menu update for 2.79 that will also be a 2.8 pie menu for FM and for physics to speed up the workflow.
So ya, we are addressing as many things as we can in order of importance to FM but also that eventually will help out 2.8 master as much as we can. You may have to stick with branches for a while…
Keep in mind though, this is a long term thing…no short and quick answers here. Check out Jaun’s Bone-Studio YT channel for the most up to date work the FM team is doing for 2.8 physics in general. It will be a while for FM itself to be in 2.8 master as mentioned in prior posts.
Since I’ve posted that, I’ve discovered that the engine blender uses is called “bullet”, which is actually a standard used by many softwares such as C4D, Maya or Houdini, and the other most used alternative is PhysX from nVidia (Max uses a derivated from it) so not bad at all!!
The weird thing is that it doesn’t have static friction, or it… can’t be acessible!?
I’ve also tested Maya and C4D, the first one has the option for it, while C4D doesn’t either, maybe there are different versions of bullet as well or maybe not everything has been implemented in blender?
I’ve been experimenting with the fracture modifier helper tools recently, particularly the fake cloth and fake fluid. I’m trying to create a softbody tearing/fracturing effect, like this:
The fake cloth, with some constraint changes works quite well, but it’s not ideal as once the pieces fracture, it’s just hollow. The fake fluid almost gets there, with use constraints option, but the UV is getting lost.
Does anyone have other ideas or suggestions on replicating this effect in Blender?
You might still need to tweak the breaking and deformation settings. Just in case the mesh shows some holes after loading, just start the simulation. After the first frame change they will be gone.
Basically I combined BMesh voronoi+boolean fracturing with the fake cloth setup here (instead of using Voronoi + Bisect, which is the default for fake cloth (and is hollow))
The mesh isnt perfectly smooth tho, but looks like it isnt in Houdini either.
That’s a pretty cool result! All things considered, it works pretty well.
Thanks for the scene file, its been useful to see what you did with the constraint settings.
The molecular script produces a decent result also and I was wondering if I could find a way to convert the particles to vertices, hypothetically, could the result be used with the bind option in the fracture animated mesh settings, to drive the fractured donut?
I couldn’t find much info on the animated mesh options, so I might be way off with my expectations!
I’m experiencing a super weird bug where my object is getting duplicated. basically i have a cube inside a cube where the inner cube is animated to expand breaking the shell cube. http://pasteall.org/blend/index.php?id=52385
Hello. it’s me again sorry. I’m just wondering if there’s a better way to fracture this
i was only able to reach 600 pieces before it stops letting me go any higher by crushing/not responding for hours. Also how can i make the inner faces have a different color? my workflow is fracture then export alembic to blender 2.80. here’s the blend file http://pasteall.org/blend/index.php?id=52465
If it’s a scan you’re fracturing, with a high poly-count, try remeshing it or decimating it, I found heavy meshes ‘upset’ FM when fracturing a photoscanned angel-gravestone I made.
Try the decimate modifier- the mesh looks like a dogs dinner in wireframe but seems to keep the UVs OK.
Just tried it with one of my scans- 1/4 million faces down to 84k odd, still showing it’s texture map ok.
Undid it and knocked it down to 20k odd- still showing image fine (loads of detail lost in the mesh but!)
Then again you are smashing it to bits so loss of fidelity might not be such an issue…
In fact: there you go-
1000 shards & intact UVs, 250k poly mesh (already remeshed from the original scan) decimated down to about 40k then fractured.
And the Fracture modifier should create a new material slot called whatever_inner for the internal faces, throw a material on that, it should have it’s own UVs - the above clip’s object seemed to.
Check ‘face sets’ in the Alembic export and you should get the surfaces (material-slots, sorry- ex Lightwave user!) in 2.8x, I find leaving ‘normals’ checked screws up things surface wise myself- so un-check it, but see how things go.
I’m looking to simulate a wooden barrier with metal legs getting hit, and so obviously I want the legs to deform, and the wood to fracture - are there any tutorials out there for this?
Flipping the normals (they were inverted for some reason) and setting the origin to geometry seemed to help here. You might need to remove the FM before and re-add it after doing this. Splinters use some scaling tricks internally, to stretch the cells. Probably i missed to test for inverted normals and origins outside the geometry or so.
