Editorial Type:
Article Category: Research Article
 | 
Online Publication Date: 27 Sept 2021

Self-Excited Full-Vehicle Oscillations Caused by Tire–Road Interaction: Virtual and Real-World Experimental Investigation

Page Range: 18 – 38
DOI: 10.2346/tire.21.20019
Save
Download PDF

ABSTRACT

In this article, self-excited full-vehicle oscillations (power-hops) are introduced. Initially, results of full-vehicle measurements are shown followed by the presentation of a specially build test rig (longitudinal dynamics test rig). Subsequently, these oscillations are investigated by using simulation-based tools within multibody simulation–related full-vehicle modeling. Tire–road interaction is evaluated in this process either by characteristic curves or by a proprietary quasistatic tire model that returns overall tangential forces by evaluating the state of every discretized element within the footprint area.

Keywords: self-excited oscillations; power-hop; tire–road interaction; full-vehicle measurement; test rig development; multibody simulation; tire modeling
FIG. 1
FIG. 1

Test vehicle (left); measurement setup for clearance studies (right) [2].

FIG. 2
FIG. 2

Placement of used measurement equipment.

FIG. 3
FIG. 3

Displacement of M1 in x-direction with and without power-hop [2].

FIG. 4
FIG. 4

Mechanical models to describe self-excited oscillations (right side; [4]).

FIG. 5
FIG. 5

Measured wheel speed signals and TCS intervention [1].

FIG. 6
FIG. 6

Overview of identified types of excitation [2].

FIG. 7
FIG. 7

Estimated slope in μ-slip characteristic.

FIG. 8
FIG. 8

CAD model of the developed test rig (front view) [2].

FIG. 9
FIG. 9

CAD model of the developed test rig (isometric view) [2].

FIG. 10
FIG. 10

Comparison of signals in vehicle testing and test rig: a) throttle signal; b) engine rpm.

FIG. 11
FIG. 11

Comparison of results in vehicle testing and test rig: a) velocities; b) drive shaft torque.

FIG. 12
FIG. 12

Test rig results for different wheel load signals.

FIG. 13
FIG. 13

Road surfaces: mastic asphalt D47 0/11 (left) and concrete C20/25 (right) [2].

FIG. 14
FIG. 14

Derived μ-slip curves on asphalt (left column) and concrete (right column) [2].

FIG. 15
FIG. 15

Influence of road surface texture (concrete: new vs worn) [2].

FIG. 16
FIG. 16

LDP with transparent road surface (left); principal setup (right).

FIG. 17
FIG. 17

Postprocessing results (screenshots): rolling tire (left); Power-Hop maneuver (right).

FIG. 18
FIG. 18

Model interaction in MBS [2].

FIG. 19
FIG. 19

Comparison of wheel speed: measurement, tire characteristic curve, xTire.

FIG. 20
FIG. 20

Block diagram of control algorithm (feed forward) [2].

FIG. 21
FIG. 21

Proof of concept in full-vehicle simulation [2].

FIG. 22
FIG. 22

Section view of measured tire on rim and geometry representation of tire model.

FIG. 23
FIG. 23

Contact patch data: wheel load Fz = 3 kN (left); wheel load Fz = 6 kN (right).

FIG. 24
FIG. 24

xTire result: time signal (first row, Fx; second row, ΔFz) and contact patch data.

Contributor Notes

1  Department of Automotive and Aerospace Engineering, Hamburg University of Applied Sciences, Berliner Tor 9, Hamburg 20099, Germany; and IPE Lübeck, Lübeck 23556, Germany. Email: dirk.engel@haw-hamburg.de
  • Download PDF