So, I made a bunch of spheres (they don't bounce yet) that are supposed to fall and hit the plane beneath them, and stop. All of them, big and small, do exactly that... except one. This one keeps falling right through the plane. First I changed all the physics in the physics tab to make it work, and when nothing changed, I deleted that one, copied directly off of another ball, and placed it where it was supposed to go. And it fell through. Its an exact copy of the other one. Does Shift-D only copy the physical properties of something, and I was just unaware of this?

  • $\begingroup$ A blendfile will probably help find the issue. Upload to: blend-exchange.giantcowfilms.com $\endgroup$ – TLousky Oct 21 '15 at 12:49
  • $\begingroup$ As in send you the blender file that has the issue? $\endgroup$ – BookWyrm17 Oct 21 '15 at 13:39
  • $\begingroup$ You can host it online through the link specified above, then attach it to a comment here, or edit your question to include this link. That way anyone who wants to help will be able to download and examine the file. $\endgroup$ – TLousky Oct 21 '15 at 13:41
  • $\begingroup$ The blend-exchange keeps giving a "fatal error". I have no clue what to do to send the file. :/ $\endgroup$ – BookWyrm17 Oct 21 '15 at 14:19
  • $\begingroup$ Have you tried applying the scale on the faulty ball? $\endgroup$ – PGmath Jan 19 '16 at 16:42

I've been having the same problem with objects falling through a cube surface even with a box collider. This post - Why is this passive rigid body allowing objects to pass through it? - has a really helpful comment from @satishgoda. If you go into the Scene tab in the Properties window, and increase the Steps per Second (I needed 300 steps) that should help prevent some of your smaller objects from falling through.

  • $\begingroup$ increasing steps worked great, thanks $\endgroup$ – Neil Nov 22 '19 at 19:23

There are a few things that could go wrong. Here are some common problems:

  • Plane isn't "solid" try using a cube or the Solidify modifier
  • Bounding volume isn't "box" for the floor
  • Bounding volume isn't "sphere" for the spheres

Here is a short YouTube video I made where it worked in 2.76. I hope it helps.


  • $\begingroup$ Well all the spheres hit the plane except one. And that one just falls through. $\endgroup$ – BookWyrm17 Oct 21 '15 at 14:14
  • 1
    $\begingroup$ As I demonstrate in the video make the Plane solid. $\endgroup$ – Kristopher Ives Oct 21 '15 at 14:26
  • $\begingroup$ Yeah but my plane interacts with all the rest fine! I even made the new one a copy of one of the ones that hits and it still falls though! I want to understand it, not just make a quick fix. $\endgroup$ – BookWyrm17 Oct 21 '15 at 15:39
  • 1
    $\begingroup$ It's because the "Box" modifier around a plane doesn't properly describe how that object can pass through it. If you crank up the iterations or modify the time-stepping it may stop falling through, but the reason it's falling through is because the simulation can't tell if it was falling from above or below the plane if there is too much error in the simulation. Error increases as more objects have to push each other off one another. By making the Plane a box or setting "Margin" on the plane for the box bound you reduce this error by how thick the floor box is. $\endgroup$ – Kristopher Ives Oct 21 '15 at 16:10

I've had issues with physics objects falling through a plane before when the Scale hadn't been applied to it, so make sure you have done that — CTRL+A > Scale


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.