Page 1 of 1

Hydraulic Cylinders, Excavator Arms, Hexapod Simulation

Posted: Thu Jul 09, 2015 5:16 pm
by RBD
Tried some Bullet rigid body dynamic simulations entirely and exclusively driven by hydraulic cylinders.
See video description on YouTube for more details...
YouTube video: Homemade Hydraulic Cylinders, Excavator Arms, Hexapod - Bullet Blender
Image

Re: Hydraulic Cylinders, Excavator Arms, Hexapod Simulation

Posted: Sat Jul 11, 2015 3:38 am
by Basroil
Pretty damn nice. Was it direct controlled or were fluid flow and pressure also simulated?

Re: Hydraulic Cylinders, Excavator Arms, Hexapod Simulation

Posted: Sun Jul 12, 2015 1:25 pm
by RBD
Very simplified, no fluid dynamics, just a linear constraint motor (push / pull) connecting rod to cylinder, limited by a max force (≈ max hydraulic pushing/pulling force).

Re: Hydraulic Cylinders, Excavator Arms, Hexapod Simulation

Posted: Mon Jul 13, 2015 4:52 am
by Basroil
Still awesome though. I was just a bit confused by the lack of changes in hose stiffness, is it possible to adjust the rope stiffness in your simulation based on motor feedback? It's certainly more trouble than it's worth, but would be cool to see that extra bit of realistic change.

Re: Hydraulic Cylinders, Excavator Arms, Hexapod Simulation

Posted: Mon Jul 13, 2015 12:26 pm
by RBD
While there is some variability in load pressure, there is always high pressure in hydraulic hoses, it's not like there is a lot of pressure when a piston moves and no pressure when there is no movement.

Having said that, to answer your question:
Each hydraulic hose does have its own stiffness parameter (implemented as constraint angle limits), currently always set to the same value. I could change the stiffness on demand based on the "control valves" regulating the constraints' push/pull motor. But I don't think I could change the stiffness based on external forces applied to the hydraulic cylinders simply because of the current lack of (not implemented) feedback data in Blender (that could be done in a C++ version of course).