lazza79 Posted February 29, 2016 Share Posted February 29, 2016 Hi guys, I was trying to achieve the "evolution" of my waterfall/river scene by making the water to destroy a wall but I have a problem with the constraints: the wall starts to collapse way before the fluid reaches it and the cause is that the constraints are broken. I made this small hip file to test a similar scene in a smaller scale but I still have the same problem with the constraints: right now, as you can see from the file attached, at frame 2, all the glue constraints of the wall are broken and the wall starts to collapse. The way I would like to achieve this it would be to have the wall as an RBD object (or fractured object, still don't know the difference very well) and the river bed as a static object (to which the wall should be constrained to). I watched the bullet masterclass but with no help on this matter. Can you guys help me? Is this the correct way to do this? Why am I getting those huge forces on frame 2? Is it due to the gravity? I am sure that the solution is easy, I am convinced that I just lack some knowledge... Thank you for your help! Cheers, Andrea constraint_test.hip Quote Link to comment Share on other sites More sharing options...
blaykenadeau Posted February 29, 2016 Share Posted February 29, 2016 Hey,I downloaded your sample file to test. It looks like the glue constraint collapses under its own weight. I gradually increased the glue strength (Under Glue Constraint Relationship), and it stopped collapsing on itself at a value around 27,000 (previously 10,000). Quote Link to comment Share on other sites More sharing options...
lazza79 Posted February 29, 2016 Author Share Posted February 29, 2016 Interesting! I tried to reduce the size of the wall and actually, without touching the constraint, it holds. Thank you very much!!!! FINALLY I got it! Cheers! Andrea Quote Link to comment Share on other sites More sharing options...
lazza79 Posted March 1, 2016 Author Share Posted March 1, 2016 Thank you very much for your helps guys, I appreciate it! Proceeding with my scene, I have another issue that I cannot still figure out: I am trying to simulate a fluid destroying a wall. The problem is that as soon as one particle hits the wall, it explodes! I tried to dial down the velocity scale but with no success. Any idea? I am attaching my scene... Cheers!Andrea week03_AL_waterfall_tests.hip Quote Link to comment Share on other sites More sharing options...
blaykenadeau Posted March 1, 2016 Share Posted March 1, 2016 (edited) Hey again!EDIT: I forgot to mention that I think your wall is exploding because it is inserted into the terrain via the hole you cut out of the terrain and is having collision issues. I noticed when I disconnect the static object (Terrain) that the wall/water no longer explodes on impact. Also take note that the water wont push the wall blocks unless Feedback Scale is above 0 in the FlipSolver>MotionVolume>Solver tab (recommend setting it to 1).To solve keeping the wall in place without intersecting the geometry, you might try cutting the bottom of the wall off using the terrain (The opposite of what you have now), and grabbing the blocks closest to the terrain and gluing them in place with a second glue constraint. Let me know if you want an example.- Blayke Edited March 1, 2016 by blaykenadeau Quote Link to comment Share on other sites More sharing options...
lazza79 Posted March 1, 2016 Author Share Posted March 1, 2016 Thanks Blayke! unfortunately I tried to follow your advice but still doesn't work. I tried to cookie out the wall - the terrain and I even moved the wall up by a tiny bit (0.1) to be sure it won't intercept but is still exploding. Ah, I also set the feedback to 1. This is the updated file, obviously you can edit it or put sticky notes in it... whatever you prefer! Cheers, Andrea week03_AL_waterfall_tests2.hip Quote Link to comment Share on other sites More sharing options...
blaykenadeau Posted March 1, 2016 Share Posted March 1, 2016 (edited) I've set up a new file with an example of a method that works for me. I commented using sticky notes, but let me know if I've left anything vague. Wall_Flip_Interaction_Test.hip Edited March 1, 2016 by blaykenadeau Quote Link to comment Share on other sites More sharing options...
lazza79 Posted March 1, 2016 Author Share Posted March 1, 2016 GREAT!!! Using your method it's working! Now I have to better understand why my method didn't... Anyway this is the result (for now) looks nice (attached). Thank you so much!!!! If you happen to be in London I'll buy you a beer!!! Andrea test_breaking.mov Quote Link to comment Share on other sites More sharing options...
lazza79 Posted March 2, 2016 Author Share Posted March 2, 2016 Hi Blayke, proceeding with the implementation of my test scene, it seems that the particles are too strong and they provoke a lot of "jittering". I tried assigning to the wall a higher density (even 4000) and lower (700) but with no significant improvements. Maybe I should try assigning a mass to every piece (primitive)? Ideas? Cheers!Andrea Quote Link to comment Share on other sites More sharing options...
blaykenadeau Posted March 2, 2016 Share Posted March 2, 2016 Hey Andrea,If you mean you see jittering in the wall chunks as they are pushed by the water, I've had similar problems in my scenes. If I'm remembering right, it got better with more substeps/lower particle separation. I can take a look at the file if you want. Quote Link to comment Share on other sites More sharing options...
lazza79 Posted March 2, 2016 Author Share Posted March 2, 2016 if you want, you are more than welcome (hip attached). Today I was trying to change a setting in the FLIP object, setting "use volume velocity for collisions" instead of point but the result is the same I wonder if we should use something that can transfer the velocity from the particles to the wall (advection?) but I don't know how to do it... I can see that the particle pass through the wall but they don't really push it. Actually it looks like they are pushing it vertically... Thanks! Andrea week03_AL_waterfall_e.hip Quote Link to comment Share on other sites More sharing options...
blaykenadeau Posted March 2, 2016 Share Posted March 2, 2016 (edited) I'm assuming is was the dopnet3 node that is the most recent iteration. The Feedback Scale was set to 0, and the glue strength for the base of your wall wasn't set high enough and was popping up from the terrain. Adding a few zeroes onto the glue strength should fix it. This got rid of most of the jittering - but the wall chunks were still being thrown around by the water regardless of their density. If you switch the Merge1 node to "Left Input Affect Right Inputs", it reacts better (Still need to mess around with the density, but it doesnt fling out of the simulation area.)I'm not sure why the simulation works better with the merge node set to that instead of mutual. Edited March 2, 2016 by blaykenadeau Quote Link to comment Share on other sites More sharing options...
.jpg.018a3ccc7a43f4bbdbeb8201c39f4f99.thumb.jpg.aa578da6fe2b7f5c8b26d654da58dd3f.jpg)
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.