Mathias - Here’s another source of information that you might find useful: https://academic.oup.com/mnras/article/498/1/1062/5893804
Dan --------------------------------------------- Bernard A. and Frances M. Weideman Professor NVIDIA CUDA Fellow Department of Mechanical Engineering Department of Computer Science University of Wisconsin - Madison 4150ME, 1513 University Avenue Madison, WI 53706-1572 608 772 0914 http://sbel.wisc.edu/ http://projectchrono.org/ --------------------------------------------- From: [email protected] <[email protected]> On Behalf Of mathias dalby larsen Sent: Wednesday, April 16, 2025 1:25 AM To: ProjectChrono <[email protected]> Subject: [chrono] Re: Help with Rolling Friction Behavior in PyChrono SMC Model Thank you this was exactly what i needed. onsdag den 26. marts 2025 kl. 20.09.35 UTC+1 skrev Radu Serban: Hi Mathias, The most complete SMC model is implemented in the Chrono::Multicore module. I never got around to bring up the corresponding SMC model in the core Chrono module to the same level (although that is on one of the many TODO lists I keep). Compared to the Chrono core implementation (see the implementation in ChContactSMC.h<https://urldefense.com/v3/__https:/github.com/projectchrono/chrono/blob/ca60083b2de78732f398adb895f1540ccfadd8b2/src/chrono/physics/ChContactSMC.h*L45__;Iw!!Mak6IKo!JsKD6Y-yJ9a3xuK2mDAsOhZgZVUoI3KeQ1hJJHWfZDy2pcwRjW0KHtP1S0qzEkX6jLZs-SKuRz6lZ3iuL6eOZUOwiA$>), the SMC contact model in Chrono::Multicore (see ChIterativeSolverMulticoreSMC.cpp<https://urldefense.com/v3/__https:/github.com/projectchrono/chrono/blob/ca60083b2de78732f398adb895f1540ccfadd8b2/src/chrono_multicore/solver/ChIterativeSolverMulticoreSMC.cpp*L56__;Iw!!Mak6IKo!JsKD6Y-yJ9a3xuK2mDAsOhZgZVUoI3KeQ1hJJHWfZDy2pcwRjW0KHtP1S0qzEkX6jLZs-SKuRz6lZ3iuL6eHvVfbMQ$>) has the following two main additional features: * A multi-step tangential displacement model (see this paper<https://urldefense.com/v3/__https:/www.researchgate.net/publication/281538010_On_the_Importance_of_Displacement_History_in_Soft-Body_Contact_Models__;!!Mak6IKo!JsKD6Y-yJ9a3xuK2mDAsOhZgZVUoI3KeQ1hJJHWfZDy2pcwRjW0KHtP1S0qzEkX6jLZs-SKuRz6lZ3iuL6dvqSpzlA$> for details on why that is important in SMC); and * Rolling and spinning friction With the above, the kind of simulations you are interested in will give the expect results. In fact, such behaviors are implemented and tested in a series of tests that are included in the Chrono distribution: see the unit tests for multicore SMC contact<https://urldefense.com/v3/__https:/github.com/projectchrono/chrono/tree/main/src/tests/unit_tests/smc_contact/multicore__;!!Mak6IKo!JsKD6Y-yJ9a3xuK2mDAsOhZgZVUoI3KeQ1hJJHWfZDy2pcwRjW0KHtP1S0qzEkX6jLZs-SKuRz6lZ3iuL6eFD4D46Q$>. Having said all of that, please note that the Chrono::Multicore module is currently *not* SWIG wrapped and as such not available through PyChrono. Best, Radu From: [email protected] <[email protected]> On Behalf Of mathias dalby larsen Sent: Monday, March 24, 2025 10:12 AM To: [email protected] Subject: [chrono] Help with Rolling Friction Behavior in PyChrono SMC Model Hello, I'm currently working on a PyChrono model using the SMC (Smooth Contact) method to demonstrate how rolling friction behaves in simulations. The setup I'm aiming to replicate is illustrated in the attached diagram, where the goal is to observe different states of motion: stationary, pure rolling, pure sliding, and combined rolling and sliding (as shown in the second image). However, in my implementation, I am only able to achieve a combination of rolling and sliding, when chaning the friction values. Additionally, even with the wall set at a 0° incline and a friction coefficient of 0.25, the sphere continues to roll indefinitely instead of coming to rest. Any guidance or insights into what might be wrong with my setup would be greatly appreciated. PS.From my countless tests, it seems like the material data is only applied during the first contact, however, I'm not quite sure. Best regards, Mathias Dalby Larsen Code used import pychrono as chrono import pychrono.irrlicht as chronoirr import numpy as np # Simulation Parameters sphere_radius = 0.2 # meters sphere_mass = 5.0 # kg initial_velocity = 0.0 # m/s (parallel to incline, pointing up) incline_angle = 5 # degrees (fix the floor at this angle) static_friction = 0.25 # μs kinetic_friction = 0.25 # μk Rolling_friction = 0.50 # Rolling friction coefficient measure_time = 4.5 # Time after which velocities will be measured sim_time = 10 wall_depth = 15.0 # From wall dimensions # Convert incline angle to radians alpha_rad = np.radians(incline_angle) # Create system sys = chrono.ChSystemSMC() # Check the current normal contact force model print("Normal contact force before setting:", sys.GetContactForceModel()) # Should return default value # Try setting it to Hooke #sys.SetContactForceModel(chrono.ChSystemSMC.Hooke) # Set Hooke model explicitly # Try setting it to Hertz sys.SetContactForceModel(chrono.ChSystemSMC.Hertz) # Set Hertz model explicitly # Try setting it to PlainCoulomb #sys.SetContactForceModel(chrono.ChSystemSMC.PlainCoulomb) # Set PlainCoulomb model explicitly # Try setting it to Flores #sys.SetContactForceModel(chrono.ChSystemSMC.Flores) # Set Flores model explicitly # Check if it changed print("Normal contact force after setting:", sys.GetContactForceModel()) # Should print 1 (Hertz) # Check the current tangential contact force model print("Tangential contact force before setting:", sys.GetTangentialDisplacementModel()) # Should return default value # Try setting it to None #sys.SetTangentialDisplacementModel(chrono.ChSystemSMC._None) # Set None model explicitly # Try setting it to OneStep #sys.SetTangentialDisplacementModel(chrono.ChSystemSMC.OneStep) # Set OneStep model explicitly # Try setting it to MultiStep sys.SetTangentialDisplacementModel(chrono.ChSystemSMC.MultiStep) # Set MultiStep model explicitly # Check if it changed print("Tangential contact force after setting:", sys.GetTangentialDisplacementModel()) # Should print 1 (Hertz) # Create Collision detection system and solver sys.SetCollisionSystemType(chrono.ChCollisionSystem.Type_BULLET) sys.SetSolverType(chrono.ChSolver.Type_BARZILAIBORWEIN) sys.GetSolver().AsIterative().SetMaxIterations(200) sys.UseMaterialProperties(True) # Set gravitational acceleration sys.SetGravitationalAcceleration(chrono.ChVector3d(0, -9.81, 0)) # Create a material for the inclined wall wall_mat = chrono.ChContactMaterialSMC() wall_mat.SetStaticFriction(static_friction) wall_mat.SetSlidingFriction(kinetic_friction) wall_mat.SetRollingFriction(Rolling_friction/2) wall_mat.SetRestitution(0.50) wall_mat.SetYoungModulus(1e9) wall_mat.SetPoissonRatio(0.3) wall_mat.SetSpinningFriction(Rolling_friction * 20.0) wall_mat.SetKn(1e9) wall_mat.SetKt(1e9) # Create the inclined wall (static floor) wall = chrono.ChBodyAuxRef() wall.SetFixed(True) wall.SetInertiaXX(chrono.ChVector3d(0.1, 0.1, 0.1)) wall.EnableCollision(True) # Apply correct rotation using `QuatFromAngleAxis` rotation_axis = chrono.ChVector3d(1, 0, 0) # Rotate around X-axis wall_rotation = chrono.QuatFromAngleAxis(-alpha_rad, rotation_axis) wall.SetRot(wall_rotation) # Position the wall correctly wall.SetPos(chrono.ChVector3d(0, 0, 0)) # Add collision shape wall_shape = chrono.ChCollisionShapeBox(wall_mat, 2, 0.1, wall_depth) wall.AddCollisionShape(wall_shape) # Add visual representation wall_vis = chrono.ChVisualShapeBox(2, 0.1, wall_depth) wall_vis.SetColor(chrono.ChColor(1, 0, 0)) # Red for visibility wall.AddVisualShape(wall_vis) # Add to system sys.Add(wall) # Create the rolling sphere sphere = chrono.ChBody() sphere.SetMass(sphere_mass) sphere.SetInertiaXX(chrono.ChVector3d(2/5 * sphere_mass * sphere_radius**2, 2/5 * sphere_mass * sphere_radius**2, 2/5 * sphere_mass * sphere_radius**2)) # Function to calculate the circle's center position def get_circle_position(radius, theta): y_c = -(radius+0.05) * np.sin(theta) z_c = (radius+0.05) * np.cos(theta) return y_c, z_c # Get the position of the circle center circle_pos = get_circle_position(sphere_radius, alpha_rad) sphere.SetPos(chrono.ChVector3d(0, circle_pos[1]+0.01, circle_pos[0]+0.01)) # Set initial velocity along the incline sphere.SetPosDt(chrono.ChVector3d(0,initial_velocity * np.sin(alpha_rad), initial_velocity * np.cos(alpha_rad))) # Add sphere collision and visualization sphere_mat = chrono.ChContactMaterialSMC() sphere_mat.SetStaticFriction(static_friction) sphere_mat.SetSlidingFriction(kinetic_friction) sphere_mat.SetRollingFriction(Rolling_friction/2) sphere_mat.SetRestitution(0.50) sphere_mat.SetYoungModulus(1e9) sphere_mat.SetPoissonRatio(0.3) sphere_mat.SetSpinningFriction(Rolling_friction * 20.0) sphere_mat.SetKn(1e9) sphere_mat.SetKt(1e9) # Assign material directly to the body sphere_shape = chrono.ChCollisionShapeSphere(sphere_mat, sphere_radius) sphere.AddCollisionShape(sphere_shape) sphere.EnableCollision(True) sphere_vis = chrono.ChVisualShapeSphere(sphere_radius) sphere_vis.SetColor(chrono.ChColor(1, 1, 0)) # Yellow sphere sphere.AddVisualShape(sphere_vis) # Add sphere to system sys.Add(sphere) # Set up the Irrlicht visualization vis = chronoirr.ChVisualSystemIrrlicht() vis.AttachSystem(sys) vis.SetWindowSize(1024, 768) vis.SetWindowTitle('Rolling Ball on Inclined Wall') vis.Initialize() vis.AddSkyBox() vis.AddCamera(chrono.ChVector3d(-2, 0, -0.5), chrono.ChVector3d(0, 0, 0)) vis.AddTypicalLights() # Simulation loop time = 0.0 dt = 10e-6 # Keep accurate physics time step render_interval = 1000 # Render every 10th step for 10x faster visual effect step_count = 0 realtime_timer = chrono.ChRealtimeStepTimer() lin_vel = None while vis.Run(): if step_count % render_interval == 0: # Render only every 10th step vis.BeginScene() vis.Render() vis.EndScene() # Advance simulation sys.DoStepDynamics(dt) time += dt step_count += 1 # Adjust real-time sync for 10x faster visualization if step_count % render_interval == 0: realtime_timer.Spin(dt * render_interval) # Measure velocities at exactly measure_time if time >= measure_time and lin_vel is None: # Only measure once lin_vel = sphere.GetLinVel().Length() ang_acc = sphere.GetAngVelLocal().Length() # Angular acceleration magnitude Rollslide = abs((sphere_radius * ang_acc)) print(f"Measured at {measure_time}s -> Lin Vel: {lin_vel}, Rot Change: {ang_acc}, Rot rollslide: {Rollslide}") force = sphere.GetContactForce() torque = sphere.GetContactTorque() print(f"Time: {time:.3f} - Contact Force: {force.Length()}, Contact Torque: {torque.Length()}") [https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image001.png?part=0.1&view=1][https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image002.png?part=0.2&view=1] -- You received this message because you are subscribed to the Google Groups "ProjectChrono" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/projectchrono/SA1P220MB1275A64A7BE884B285864FFDAEA42%40SA1P220MB1275.NAMP220.PROD.OUTLOOK.COM<https://urldefense.com/v3/__https:/groups.google.com/d/msgid/projectchrono/SA1P220MB1275A64A7BE884B285864FFDAEA42*40SA1P220MB1275.NAMP220.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer__;JQ!!Mak6IKo!IBQgS7fUmMwUVLU_nkkK6qZGuHR5easzklZgtkSA9NDWNMtz9r7RJ4VlvVWz7UrhEy-iiRS-1wtwgxUv7-hPT1GkFA$>. -- You received this message because you are subscribed to the Google Groups "ProjectChrono" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]<mailto:[email protected]>. To view this discussion visit https://groups.google.com/d/msgid/projectchrono/64c9dcfe-60fc-4959-82c6-16cb6e9d92a3n%40googlegroups.com<https://urldefense.com/v3/__https:/groups.google.com/d/msgid/projectchrono/64c9dcfe-60fc-4959-82c6-16cb6e9d92a3n*40googlegroups.com?utm_medium=email&utm_source=footer__;JQ!!Mak6IKo!JsKD6Y-yJ9a3xuK2mDAsOhZgZVUoI3KeQ1hJJHWfZDy2pcwRjW0KHtP1S0qzEkX6jLZs-SKuRz6lZ3iuL6eOPWgK9w$>. -- You received this message because you are subscribed to the Google Groups "ProjectChrono" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/projectchrono/DM8PR06MB770395BFBE2DED6FE3E1F9B3B1BD2%40DM8PR06MB7703.namprd06.prod.outlook.com.
![https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image001.png?part=0.1&view=1][https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image002.png?part=0.2&view=1](https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image001.png?part=0.1&view=1][https://groups.google.com/group/projectchrono/attach/5e91f84ca3335/image002.png?part=0.2&view=1){kind=link}