RD-E: 1200 Jumping Bicycle

After a quasi-static pre-loading using gravity, a dummy cyclist rides along a plane, then jumps down onto a lower plane. Sensors are used to simulate the scenario in terms of time.

The purpose of this example is to illustrate how to use the Radioss description when resolving a demonstration example. The particularities of the example can be summarized using dynamic loading during a four-step scenario where a dummy is first put on a bike, then it rides on a plane to subsequently jump back down onto the ground. The scenario described is created using sensors.

Options and Keywords Used

Shell, brick, beam, truss, general spring, and beam
Sensors on rigid bodies (/SENSOR/RBODY and /RBODY) and monitored volumes (perfect gas) (/MONVOL/GAS)
Quasi-static load treatment (gravity) (/GRAV), kinetic relaxation (/KEREL), restart file
Dummy and hierarchy organization
TYPE7 interface auto-impacting (/INTER/TYPE7) and rigid wall (infinite plane and parallelogram) (/RWALL)
Linear elastic law (/MAT/LAW1 (ELAST)) and Johnson-Cook law (/MAT/LAW2 (PLAS_JOHNS))
Added mass (/ADMAS)
Initial velocity (/INIVEL)

Two types of rigid walls are set up:

A fixed infinite plane (floor)
A fixed parallelogram (springboard)

rad_ex_fig_12-9 — Figure 2. Position of the Rigid Walls

The characteristics of the parallelogram plane are: 2013 mm x 1200 mm. Both rigid walls are tied to allow the wheels to turn.

The infinite plane is defined by the normal vector (

\vec{M M_{1}}

) and the parallelogram by the coordinates of three corners (

M

M_{1}

, and

M_{2}

). For both rigid walls, the secondary nodes are obtained from the tire and rim parts (displayed in green in Figure 3).

rad_ex_fig_12-10 — Figure 3. Secondary Nodes Definition (green) and Profile View of Rigid Walls

Several rigid bodies are created (/RBODY) and activated by sensors for use at the appropriate time and in a chronological manner (sens_ID not equal to 0). Thus, every rigid body is not active at the same time. The activation order is described in the paragraph dedicated to /SENSOR/RBODY. According to their activation time, the rigid bodies are classified in groups which are:

rad_ex_fig_12-11 — Figure 4. Classification of Rigid Bodies (group)

The inertias of rigid bodies are set in local skew frames for groups A, C and D.

Rigid body activation/deactivation
Groups A and B: The rigid bodies are activated during pre-loading up to equilibrium then applied to the initial velocity start. They are activated again just before the impact of the bike on the inferior plane.; During the free fly phase, both the cyclist and the bike undergo a rigid body motion. In order to save the computation time, the motion can be simulated by putting the whole structure into a global rigid body (Group D). The rigid body is deactivated just before landing.
Group C: Three rigid bodies include the dummy, the frame and both wheels (not including the tires). This configuration allows just the wheels to turn, taking into account the active tires action on the plane. This rigid body is activated while the bike is running on the springboard.
Group D: This global rigid body, including all nodes of model is activated as long as the bike is in the free fly phase and is deactivated just before impact on the floor.
Group E: This rigid body is activated before impact ensures the stiffness level of the lower fork.

A 8333 mms^-1(30 km/h) initial velocity (/INIVEL) is applied to all nodes of the model (bicycle and cyclist) in a parallel direction to the high plane at time t = 0.004 s. This initial condition is defined in the Engine file *_0002.rad (start time: 0.004 s) which is run after the quasi-static equilibrium with gravity loading.

Options in Engine file (*_0002.rad):

/INIV/TRA/X/1: initial translational velocities in direction x
8333: of 8333 mm/s
1 338000: on node 1 to 338000

rad_ex_fig_12-12 — Figure 5. Initial Translational Velocities of the Model Bike/Man (30 km/h) at t = 0.004 s

Gravity is applied to all nodes of the model. A constant function defines the gravity acceleration in the Z direction versus time. Gravity is activated by /GRAV.

rad_ex_fig_12-13 — Figure 6. Gravity Function (-9810 mm/s^-2)

The explicit time integration scheme starts with the nodal acceleration computation. It is efficient for the simulation of dynamic loadings. Nevertheless, quasi-static simulations via a dynamic resolution method need to minimize the dynamic effects to converge towards the static equilibrium. Among the methods usually employed, the kinetic relaxation method is quite effective and is activated in the Engine file (*_0001.rad) with /KEREL (Figure 7). All velocities are set to zero each time the kinetic energy reaches a maximum value.

rad_ex_fig_12-14 — Figure 7. Kinetic Relaxation Method

Rigid bodies are activated and deactivated with sensors (/SENSOR/RBODY). A sens_ID flag characterizes the sensors and it is required in the rigid bodies' definition. The five types of sensors used are:

TIME: Activated with time
DIST: Activated with nodal distance
INTER: Activated after impact on rigid wall
SENSOR: Activated with sensor IS1 and deactivated with sensor IS2
NOT: ON as long as sensor IS1 is OFF

rad_ex_fig_12-15 — Figure 8. Events Definition for the Activations and Deactivations of Sensors

At the beginning of the simulation (time=0), the rigid bodies are automatically set to ON, as long as the sensors are not active. Thus, in order to deactivate the rigid bodies at the first cycle, active sensors at time t=0 should be used. Consequently, the rigid bodies are active when the sensors are not active.

Added masses and inertia, as well as the flag for the gravity center, are ignored when a rigid body is managed by sensors. By default, the gravity center is only computed by taking into account the secondary nodes mass (ICoG set at 2). The main node is moved to the computed center of gravity where added mass and inertia are placed. In order to distribute the mass to the dummy over the rigid bodies, option /ADMAS is used.

Table 1. Sensors Used for Simulation
Name	Type	Definition	Rigid Body's Group using Sensor
S1	TIME	Time 0s.	-
S2	DIST	Distance between rear hubs and extremity of springboard equal to 1810 mm.	-
S3	DIST	Distance between rear hubs and extremity of springboard equal to 345 mm.	-
S4	RWALL	When the infinite rigid wall is impacted.	-
SEN(S2,S3)	SEN	Activated with S2 and deactivated with S3	-
SEN(S3,S4)	SEN	Activated with S3 and deactivated with S4	-
SEN(S2,S4)	SEN	Activated with S2 and deactivated with S4	Group A/B
NOT(SEN(S2,S3))	NOT	Deactivated with S2 and activated with S3	Group C
NOT(SEN(S3,S4))	NOT	Deactivated with S3 and activated with S4	Group D

Sensor (S4) is also used for deactivating both the beam type springs modeling links between the feet and pedals (I_shear set to 1). A case could be considered without this sensor to study the risks of automatic pedals.

rad_ex_fig-12-16 — Figure 9. Activation and Deactivation Zones of Sensors and Rigid Bodies

Input Files

Before you begin, copy the file(s) used in this example to your working directory.

RD-E-1200_Bicycle.zip

Model Description

The purpose of this example is to set up a demonstration in which sensors and restart files are used to allow the change of a problem over time.

Subjected to the gravity field, a dummy cyclist rides on a higher plane, then jumps down onto a lower horizontal plane. The problem can be divided into four phases:

positioning the cyclist under the gravity effect
running the bicycle on the high plane
free fly
the impact on the ground

The following system is used: Ton, mm, s, N, MPa

Model Method

rad_ex_fig_12-2 — Figure 12. Meshes of the Models Main Parts

The material of the metallic parts use the Johnson-Cook law (/MAT/LAW2) with the following properties:

Steel Spokes Material Properties: Value
Young's modulus: 210000 $[MPa]$
Poisson's ratio: 0.3
Density: 7.9x10^-9 $[\frac{T o n}{m m^{3}}]$
Yield stress: 185.4 $[MPa]$
Hardening parameter: 540 $[MPa]$
Hardening exponent: 0.32

Aluminum Frame Material Properties: Value
Young's modulus: 60400 $[MPa]$
Poisson's ratio: 0.33
Density: 2.7x10^-9 $[\frac{T o n}{m m^{3}}]$
Yield stress: 90.27 $[MPa]$
Hardening parameter: 223.14 $[MPa]$
Hardening exponent: 0.375

A QEPH formulation (I_shell= 24) is used for tires in order to prevent hourglass deformations. A Belytschko & Tsay element with a TYPE4 hourglass formulation is used for the other shell parts. A global plasticity model is used.

Table 2. Properties and Materials of Main Parts
	Parts	Properties	Materials
Bike	Frame	Shell Q4 - 3 mm	Aluminum - Law 2
	Spokes	Truss - 2 mm²	Steel - Law 2
	Rim	Shell Q4 - 3 mm	Aluminum - Law 2
	Tires	Shell QEPH - 3 mm	Rubber - Law 1
	Hubs	Beam - 900 mm²	Steel - Law 2
	Saddle	Brick	Foam - Law 1
	Pedals	Beam - 900 mm²	Steel - Law 2
	Tube of saddle	Shell Q4 - 3 mm	Aluminum - Law 2
Dummy	Body (limbs)	Shell Q4 - 3 mm	Law 1
Dummy	Joints	Spring (8)	-

Hierarchy organization:

Bike model: 6 subsets comprising 23 parts
Dummy model: 11 subsets comprising 38 parts

Monitored Volumes / Perfect Gas

A perfect gas monitored volume with /MONVOL/GAS is defined to model the pressure in the tires. For further details about monitored volumes, refer to GAS Type in the Radioss Theory Manual.

The main properties are:

External pressure P_ext: 0.1 $[MPa]$
Initial internal pressure P_ini: Front tire: 0.75 $[MPa]$; Rear tire: 0.1 $[MPa]$
Gas constant $γ$: 1.4

All other properties are set to default values.

rad_ex_fig_12-3 — Figure 13. Visualization of a Monitored Volume (yellow part)

Quasi-static loading: gravity effect on initial static equilibrium
The quasi-static solution of gravity loading on structure deformation corresponds to the steady state part of the transient response. It describes the pre-loading case before the dynamic analysis. Therefore, the simulation is divided into two phases: quasi-static response (structure subjected to the gravity) and dynamic behavior (run, jump and landing). The solution is obtained from kinetic relaxation (/KEREL). Gravity is defined by /GRAV.
Contacts modeling
The TYPE7 interface using the penalty method serves to model contacts between the dummy cyclist and the bike. With contact in /RWALL to treat the landing of the bike. Figure 14 illustrates the description of the interface.
Figure 14. Contacts Modeling with TYPE7 Interface (Penalty Method) and /RWALL

A TYPE11 interface models contact between the pedals (beams) and the feet (shells).

Links between man and bicycle

The spring TYPE8 (/PROP/SPR_GENE) general spring property model the links between the feet/pedals and the hands/handlebar.

Stiffness (TX, TY and TZ): 50 $[\frac{N}{m m}]$
Mass: 1e-10 $[Ton]$
Inertia: 1e-5 $[{mm}^{2} Ton]$

A rupture criteria based on displacements is activated by the beams connecting the hands and handlebar in order to simulate the fall of the cyclist after landing.

Left hand: Z = 20 mm
Right hand: Z = 20 mm
Figure 15. Link Right Hand/Handlebar (TYPE8 Springs)
Dummy joints
Figure 16. TYPE8 Springs

The general TYPE8 springs, characterize a spherical hinge with a stiffness given for each DOF. Directions are local and attached to a moving skew frame. Two coinciding nodes define a spring.

Limbs are linked to the springs via the secondary nodes of the rigid bodies, as shown in Figure 17.

rad_ex_fig_12-7 — Figure 17. Connection Rigid Body - Spring TYPE8 - Rigid Body

Wheel rotation

Beam elements are used to attach the wheel to the forks. The rotational DOF is released around the beam axis.

rad_ex_fig_12-8 — Figure 18. Wheel/Forks Junction

Results

The elements included in a rigid body are deactivated. Therefore, the element flags saved in /TH/RBODY provide information on the activation and deactivation of rigid bodies during simulation.

rad_ex_fig-12-17 — Figure 19. Activation and Deactivation of Main Model Parts (elements flag ON/OFF)

rad_ex_fig_12-18 — Figure 20. Distribution of the von Mises Stress on the Frame after Quasi-static Loading

rad_ex_fig_12-19 — Figure 21. Kinetic Relaxation Effect on Kinetic Energy with /KEREL

rad_ex_fig_12-21A — Figure 22. Simulation Phases (impact at t = 4.6 s)

rad_ex_fig_12-21B — Figure 22. Simulation Phases (impact at t = 4.6 s)

rad_ex_fig_12-22 — Figure 23. Dummy Cyclist during Impact on the Ground (shoes not attached)

rad_ex_fig_12-23 — Figure 24. Variation of von Mises Stress for a Shell Element of the Frame