Keywords

1 Introduction

Tyre grip is crucial for evaluating vehicle handling performance and design effective vehicle dynamics controllers, such as ABS and ESP. Assessing tyre grip in the early stage of its design is therefore of utter importance in the tyre development. Under this point of view, the possibility to assess tyre performance using laboratory tests on compound samples before manufacturing a full tyre appears very attractive. The present paper considers tests carried out on a Laboratory Abrasion Tester (LAT100) on two compound samples (referred as “CompoundA” and “CompoundB”) and Lateral Steady State test performed on MTS Flat-Trac. The CompoundB is a more performing compound than the CompoundA and so it experciences higher value of grip and cornering stiffness.

The LAT100 machine consists of a spinning disc on which a solid rubber test wheel (Grosch wheel) is pressed under a normal load Fz at a slip angle. The tangential velocity of the sample wheel is induced by the disc angular velocity. The traveling speed, slip angle, and load can be varied over a wide range. The average slip velocity in the contact area is created by combining the angular velocity and slip angle [1]. Although the machine was originally designed to assess the abrasion loss of the test wheel according with different experimental test conditions, the objective of this study is to verify if it can also be used to assess its grip by performing tests similar to ones carried out on the MTS Flat-Trac machine on a full tyre.

2 Methodology

As mentioned earlier, the present research investigates the possibility of using the LAT100 machine to measure friction coefficient as a function of load and temperature and assess possible correlations between these measurements and their counterparts on full tyre grip measured on MTS Flat-Trac. On this purpose, an ad hoc procedure was developed for testing compound samples, in the form of a solid wheel, in steady-state cornering conditions. The idea is to replicate on the LAT100 machine tests similar to the ones performed on the MTS Flat-Trac machine on a full tyre, so to compare the outputs. Specifically, two compounds were tested in the LAT100 machine. Tests on the MTS Flat-Trac machine were carried out with two full tyres, having an identical structure, but with the tread made of the two compounds tested on the LAT100.

2.1 MTS Flat-Trac Tests

Only steady-state cornering tests are considered in this work since these conditions are replicable on the LAT100. These tests are performed applying a triangular wave slip angle input at constant vertical load. The test is then repeated by varying the vertical load. Figure 1 shows on the left the cornering (or lateral) force (Fy) vs. the slip angle (\(\alpha \)) curve obtained during tests carried considering three different vertical loads: the blue curve refers to the nominal load, the red curve to the 65% of the nominal load and the yellow curve to the 125% of the nominal load. Data are reported in a non-dimensional form for industrial privacy reasons: the cornering force is made non-dimensional by the vertical load applied during the test, while the slip angle is made non-dimensional by maximum slip angle imposed during the test. From these tests, the following information can easily be obtained:

  • tyre grip: the peak cornering force divided by the vertical load imposed during the test;

  • cornering stiffness: the slope of the lateral force curve near the origin (\(\alpha =\pm 0.5\) deg);

  • the nonlinear dependency of the tyre grip on the vertical load.

2.2 LAT100 Tests

Similarly to the MTS Flat-Trac tests, the procedure involves the application of a triangular repeated slip angle time history profile to the test wheel. In particular, LAT100 tests are performed by placing the test wheel on the actuator (right side) and then repeated with the specimen flipped, i.e. by rotating the specimen \(180^\circ \) with respect to its vertical axis (left side). In fact, since the testing surface is a rotating disc, the velocity profile is not uniform in the contact patch, thus by repeating the test on both sides, the non-uniform speeds distribution, and the consequent conicity effects on the specimen, should be compensated [2]. An example of the results obtained on the LAT100 machine are shown on the right of Fig. 1, which reports the cornering force vs. the slip angle for three different vertical loads: the nominal load (blue), the 75% of the nominal vertical load (red) and the 125% of the nominal load. The same non-dimensional form used for the MTS Flat-Trac data (left of Fig. 1) is used. The influence of the disc speed, the maximum slip angle, the slip angle rate, the powder sprayed on the contact patch and test duration was investigated during the research. Parameters were tuned in order to improve the matching with the MTS Flat-trac data. Results are shown in the next section in terms of grip and cornering stiffness and their dependence on vertical load and test temperature.

Fig. 1.
figure 1

Lateral force vs SSA on MTS Flat-Trac (left) and LAT100 (right)

Full size image

3 Results

In this section, the comparison between LAT100 and MTS Flat-Trac data is reported. From Fig. 2 and 3 is possible to appreciate the similarities and the differences between the grip (Ed. Note: for simplicity of comparison, we will use the not entirely appropriate term “grip” for the LAT100 wheel as well, aware that the more suitable term would be “coefficient of friction”) and the cornering stiffness as a function of the vertical load evaluated on MTS Flat-Trac and LAT100. Data are reported in a normalized form for industrial privacy reasons, in particular the vertical load is normalized on the nominal one and the temperature range on the central value of the test (same for MTS Flat-Trac and LAT100 data). Consequentially, each key performance indicator (KPI) is normalized on the corresponding “nominal value”. LAT100 data were compared with MTS Flat-Trac data to assess the correlation in terms of grip (maximum of the normalized lateral force with respect to the side slip angle) and cornering stiffness (slope of the lateral force trend for null side slip angle). Each graph shows the same scales between MTS Flat-Trac and LAT100 data for grip and temperature representation but different ones for what concern cornering stiffness and loads, since the two machines work at very different loads.

As expected considering the characteristics of the two compounds, grip and cornering stiffness of CompoundB are larger than the ones of CompoundA at each tested vertical load. This ranking is confirmed both by MTS Flat-Trac and LAT100 data. As it can be seen the trend of the grip with the vertical load is well correlated in MTS Flat-Trac and LAT100 tests. Indeed, tyre grip is mainly influenced by tread compound characteristics. Viceversa, the trend of the cornering stiffness on the vertical load differs between MTS Flta-Trac and LAT100 data. For what concerns the cornering stiffness, in fact, the tyre structure (not present in the Grosch Wheel) plays a crucial role. The tyre cornering stiffness is actually given by two contributions: tread and belt/sidewall stiffness. Obviously, the Grosch Wheel’s cornering stiffness is instead determined only by the tread contribution. The influence of temperature was also investigated.

Fig. 2.
figure 2

Grip vs load on MTS Flat-Trac (left) and LAT100 (right)

Full size image
Fig. 3.
figure 3

Cornering Stiffness vs load on MTS Flat-Trac (left) and LAT100 (right)

Full size image

Figures 4 and 5 respectively show the trend of the grip and the cornering stiffness vs. the tread temperature during the tests carried out on the MTS Flat-Trac and the LAT100 for the CompoundB.

Fig. 4.
figure 4

Grip and Cornering Stiffness vs temperature for the CompoundB on MTS Flat-Trac

Full size image
Fig. 5.
figure 5

Grip and Cornering Stiffness vs temperature for the CompoundB on LAT100

Full size image

A good correlation between MTS Flat-Trac and LAT100 data can be seen for the dependency of the grip on the temperature for all the tested loads. For what concerns the cornering stiffness, unless the trend with temperature and load is captured, more remarkable differences can be noticed: the effect of the load is more evident on the MTS Flat-Trac, while the effect of temperature is higher on the LAT100, where it is possible to observe that the temperature range is bigger than the MTS Flat-Trac one, since the first acquisition is at a lower temperature with respect to the MTS Flat-Trac tests: a more severe “Run-In” has to be implemented in order to obtain a preconditioning similar to MTS Flat-Trac tests, matching the entire temperature range.

Figure 6 compares the trend of grip vs. the tread temperature for CompoundA and CompoundB during the tests carried out on the MTS Flat-Trac and the LAT100 at the nominal load. As already noticed, the trend for CompoundB both for grip and cornering stiffness is in good agreement between MTS Flat-Trac and LAT100 data, instead the CompoundA shows a less temperature dependent behavior on LAT100 than on MTS Flat-Trac.

Fig. 6.
figure 6

Grip vs temperature comparison for MTS Flat-Trac (left) and LAT100 (right)

Full size image

4 Conclusions

The objective of this work was to evaluate the effect of different compounds on tyre grip and cornering stiffness through indoor tests at the LAT100 and correlate obtained results with full tyre data collected on a Flat Track machine. The developed procedure offered the opportunity to assess the behavior of two test wheels made of different compounds (CompoundA and CompoundB) as function of the normal load and temperature, performing tests similar to the ones carried out on the real tyre on the MTS Flat-Trac. The results of the LAT100 tests align qualitatively with those obtained from the MTS Flat-Trac tests. While the overall trends of the key performance indicators (KPIs) remain consistent, there is a scaling effect due to the substantial difference between the experimental subjects (Grosch wheel and pneumatic tyre) and the two machines.

In particular, a qualitative correlation between LAT100 and MTS Flat-Trac exists for the grip against the load for both the CompoundA and CompoundB. On the other hand, the trends of the grip against the temperature only finds a correlation for the CompoundB. The behavior of the CompoundA in fact tends to remains almost constant over all the temperature range.

Regarding the cornering stiffness, instead, a qualitative correlation against load and temperature holds for both the compounds.

Although the complexity of the mechanics involved precludes an immediate quantitative correlation, a finite element simulation approach generally allows for the incorporation of structural and geometrical differences between the Grosch wheel and the tyre during validation. This enables the prediction of the tyre grip using the coefficient of friction measured in the laboratory.