Effect of cast-iron disc thickness on the reliability of tribological data from pin-on-disc test: Case study of brake friction materials

Abstract

For braking system, rubbing between disc and brake pad generates heat, physical transformation on the contact, and activates many complex solicitations. Therefore, researchers used reduced scale testing to evaluate tribological and thermomechanical performance of friction composite. The pin-on-disc configuration of the tribometer is considered the better solution to reproduce these generated transformation and friction evolution. However, if the pin geometry is chospaden based on representative elementary volume (REV) approach, no research has investigated and justified the choice of the disc geometry and thickness. In this research work, an examination of the critical disc thickness is discussed, highlighting its substantial impact on the friction-wear behavior during brake application. For that, four disc thicknesses were tested, varying on thickness values (10, 15, 20, and 22 mm). Through this experiment, the friction coefficient evolution as function of braking numbers and sliding duration was studied while maintaining pressure and sliding velocity constant. Wear and temperature rise are identified for each disc thickness situation. A 3D thermal model is employed also to simulate the heat conduction and dissipation in brake system for various disc thickness. The temperature predictions are carried out using Ansys software. Experimental investigation presented the effect of the thickness on the friction stability, particularly at high temperature. Numerical results highlight the effect of reducing the thickness of the disc on the kinetic temperature rise through a rapid increase in disc temperature. The thickness modification used for the brake disc enhances the resulted thermal response and tribological brake pad performance, and consequently enhance the reliability of tribometer outcomes.

1. Introduction

In the specific context of the study of brake materials, the pin-on-disc tribometer is widely considered the most effective test configuration. This setup allows for a reliable assessment of the friction coefficient and representative friction-wear mechanisms [1]. The friction coefficient, significantly affected by the temperature, is also affected by various factors such as microstructure heterogeneities and chemical reactions or physical transformations induced by the thermal properties of the material. The complexity and non-reproducibility of this phenomenon are not only due to the activated solicitations (tribological, thermo-mechanical, etc.) at the sliding surfaces but also the significant influence of the appropriate choice of the Representative Elementary Volume (REV) of samples or the tribometer architecture.

Despite the crucial impact of disc dimensions on braking applications, especially concerning thermal phenomena, there is a limited number of critical studies addressing the impact of disc thickness on the tribological behavior of brake materials. Disc geometry, as demonstrated by several authors [2,3], proves to be a critical parameter in braking applications. The cycling of braking introduces inertia, effusivity, and diffusivity of disc materials, which are expressions of disc thickness, controlling behavior of the disc [4] and its performance in temperature diffusion [5].

Throughout service, changes occur in the microstructure of the disc at different depths, involving an oxidative process on the active surface and a plastic strain phenomenon in the volume (hardening from deformation and phase changes) [6]. These thermal, mechanical, and chemical phenomena act synergistically, affecting friction and wear responses, particularly at the temperature of phenolic resin degradation, which should be determined by the tested material's properties [7]. Afzal [8] demonstrated that the thermal properties of the disc significantly impact system's design. A braking test on a 22 mm disc revealed that the thickness affected by braking is around 16 mm, and the thermal evolution at the heart of the disc is not linear, resulting in a thermal gradient [9]. Majcherczak [10] tested a cast iron disc with four different thicknesses: 11, 15, 21, and 31 mm, noting that the disc's thickness has a substantial influence on the maximum temperature reached and the time when the maximum temperature is reached.

In this study, we delve into the potential of utilizing a pin-on-disc test for investigating the tribological performance of friction material while considering the influence of disc thickness. To facilitate this exploration, we manufactured discs with four different thicknesses. Our objective is to conduct friction and wear tests under cyclic loading at medium temperatures, and assess the sensitivity of tribological and thermal behaviors to variations in disc thickness. We proposed a 3D numerical model to complement our experimental efforts. Numerical results were then compared with experimental data obtained from tests conducted on discs with varying thicknesses. Key performance parameters, including disc temperature evolution, were systematically compared between numerical predictions and experimental outcomes. Upon successful validation of the numerical model by replicating experimental results, we employed it to explore different configurations, specifically varying disc thicknesses, aiming to optimize brake disc performance. The integration of experimental results, coupled with numerical disc temperature predictions, provides valuable insights for selecting the most appropriate disc thickness in practical applications.

2. Materials and methods

2.1. Materials

The brake materials under study encompassed cast iron and brake pin friction materials. Our study aims to highlight the impact of disc thickness while creating a representative environment mirroring a real braking system. Additionally, this material has been widely employed by various researchers to study tribological performance of friction materials using the pin-on-disc configuration [11].

The microstructure of the disc is defined by the dispersion of graphite lamellas within the ferrous matrix. without any preferential orientation, as depicted in Fig. 1a. These graphite lamellas possess an overall dimension of 0.5 mm, contributing to a heterogeneous microstructure that induces non-linear mechanical behavior. The graphite is considered a solid lubricant with a high modulus of elasticity and excellent wear resistance compared to other types of cast iron [12]. Moreover, it exhibits high thermal diffusivity, a characteristic determined by the percentage and length of graphite inclusion, as well as the percentage of carbon equivalent. The mechanical and physical properties of this material at temperatures of 25 °C and 250 °C are detailed in Table 1 [13,14].

Fig. 1.

Microstructure of a) gray cast iron and b) brake lining materials.

Table 1.

Material properties of the gray cast iron at different temperatures [14].

Properties	At 20 °C	At 250 °C
Density ρ (g.cm−3)	7.3	7.2
Thermal conductivity λ (W.m−1.K−1)	46.5	44.2
Specific heat capacity Cp (J kg−1 K−1)	438.7	471.8
Thermal Effusivity (J K−1 m−2 s−1/2)	12236	12256
Young modulus (GPa)	3	2.2

The fundamental formulation of the brake lining material, as outlined in Table 2, is formulated using ingredients employed in the manufacturing of brake lining materials. It consists of a mixture of particles and fibers, exhibiting varied sizes ranging from micrometric to millimetric dimensions. The predominant elements are rubber and graphite. The diameters of Barite and Rockwool particles do not exceed 5 μm. Fig. 1b illustrates a heterogeneous microstructure of the components incorporated with a phenolic resin as a binder. This material also exhibits a certain degree of porosity, with pores averaging 20 μm in size. Table 3 provides the mechanical, physical, and thermo-physical properties of the composite material.

Table 2.

Formulation of the friction material in wt % [15].

Classification	Materials	Mass proportion, Wt%
Binder	Phenolic resin	14
Fibres	Glass fibre Mineral fibre (rockwool)	22
	Cellulose
Fillers	Barite Calcium carbonate Chalk	45
Abrasive	Alumina	2
Lubricant	Graphite Coke Black carbon	10
Friction modifiers	Cashew Rubber	7

Table 3.

Mechanical, Physical and thermal properties of the composite (average values) [16].

Properties	At 20 °C
Density ρ (g.cm−3)	2.1 (0.04)
Thermal conductivity λ (W.m−1.K−1)	1.14 (0.012)
Specific heat Cp (J kg−1 K−1)	675 (23.1)
Thermal Effusivity (J K−1 m−2 s−1/2)	1221 (124)
diffusivity D (10−6 m2. s−1)	0.71 (0.008)
Normal Heat swell (%)	0.35 (0.007)
Transverse Heat swell (%)	0.35(0.005)
Normal compression modulus (GPa)	1.8 (0.14)
Transverse compression modulus (GPa)	2.2 (0.15)
Normal maximum deformation (%)	9 (0.2)
Transverse maximum deformation (%)	7(0.09)
Indentation energy (10−6J)	22265

*The value within parentheses indicates the standard deviation.

2.2. Experimental set up

This study aims to evaluate the influence of disc thickness on the braking performance of the friction composite under medium friction conditions as defined by several researches [[14], [15], [16]]. Discs, with a diameter of 80 mm, were utilized, featuring four different thicknesses (10, 15, 20, and 22 mm). The pin sample's shape is a cylinder that was cut from commercial brake pad materials with dimensions of 16 mm thickness and 14 mm diameter (Fig. 2a). The friction-wear test program is structured into two parts, governed by disc thermal evolution (Fig. 2c). The first part aims to elevate the disc temperature (T_d) from 50 °C to 250 °C. The second part involves 30 successive cycles, at a constant sliding velocity and pressure, of "On/Off contact". During the "On contact", The pin rubs against the disc, leading to an increase in temperature of until T_d reaches 250 °C then for the "Off contact", The disc temperature experiences a reduction to Td = 200 °C. (Fig. 2c). The mean friction radius is set at 32 mm The disc and spindle temperatures (T_d, T_s respectively) are measured on the mean radius of contact by K-type thermocouples, as shown in Fig. 2b. A burnish phase precedes the own test to ensure parallel pin/disc contact. During this step, the friction is carried out under moderate conditions so as not to alter the physicochemical properties of the two materials in contact. Burnish is carried out at 0.6 MPa and 3 m s⁻¹ respectively pressure and sliding velocity of and disc temperatures between 50 and 70 °C. Experimental parameters are summarized in Table 4. The choice of this temperatures range is to simulate friction-wear behavior at medium thermal solicitation induced on road braking situation without reaching thermal degradation value of the composite material [14]. To calculate the specific wear rate, the weight of the pin was measured prior to and following each wear test then normalizing wear result by sliding time or distance. In our case, specific wear was determined by equation (1).

$$\mathrm{W}={\Delta }_m/d$$ (Eq 1)

Fig. 2.

Experimental environment and conditions; a) pin-on-disc configuration, b) details of the thermocouples positions, c) friction-wear test protocol [17].

Table 4.

Friction test parameters.

Parameters	Values
Pressure (MPa)	1.2
Sliding velocity (ms−1)	6
Initial temperature (°C)	50
Sliding temperatures interval (°C)	[200, 250]
Sliding cycles number	30

W is the specific wear rate (mg.km⁻¹), Δ_m is the mass loss (mg) and d is the sliding distance for all the 30 cycles (km). Measurements of mass loss were realized after 30 cycles of friction for each sample. Prior to each test, the disc specimens underwent ultrasonic cleaning for 30 min. The average initial arithmetic surface roughness (Ra) of samples is 2 μm.

2.3. Numerical modelling

The current study revolves around the modification of disc geometry to enhance the thermal kinetics of frictional heat while preserving the desired friction level. Specifically, our focus is on adjusting the disc thickness. To achieve this objective, we developed a 3D thermal model utilizing ANSYS software modelling interface, disc and spindle (Fig. 3).

Fig. 3.

3D numerical model.

The interface represents the rubbed surface between friction materials. The key dimensional parameters for each part are outlined in Table 5.

Table 5.

Parameters of disc brake and interface.

Parameters	Values
Disc diameter (mm)	80
Disc thickness (mm)	22
Pin diameter (mm)	14
outer interface diameter (mm)	78
Inner interface diameter (mm)	50

Given that the numerical model relies on a thermal analysis, attention is directed solely towards defining the thermal properties of disc's material. Crucial parameters, such as density, thermal conductivity, heat capacity, etc., need accurate specification in the numerical model. In all computations, materials were assumed homogeneous and isotropic. The properties of the material are detailed in Table 6.

Table 6.

Thermo-physical properties of the disc, spindle [18] and interface [19].

	Disc	Spindle	Interface
λ (W/m1.K1)	12	42	0.007
Cp (J/kg1 K1)	500	480	1000
ρ (kg/m3)	7293	7790	0.1

For the properties of the third body, representing as interface in the model, the values proposed by some researcher [19] are used. He quantified certain properties of interface debris: λ = 0.07 W/m K, ρ = 0.1 kg/m³ and Cp = 1000 J/kg. K with a thickness of third body of 10 μm. The thickness of the third body is consistent with those encountered in tribology [20]. The surface thermal conductance C e = λ (inverse of the resistance) corresponding to the Day values is 7000 W m⁻²K⁻¹. This value is consistent with the orders of magnitude encountered in the bibliography from 1000 to 50000 W m⁻²K⁻¹ [21,22].

In this study, heat generation is modeled as volumetric within the interface. The thermal analysis focuses exclusively on thermal properties. To simplify the investigation, assumptions are taken into consideration.

• -At the level of the interface, the kinetic energy is converted into thermal one due to the frictional forces generated during the contact. No other form of energy is dissipated or lost in the process, [23,24].
• -The geometry and the prescribed conditions around the system possess axisymmetric characteristics.
• -The applied load is considered to exert a constant pressure that is evenly distributed
• -Frictional heat originates at the sliding surfaces between the stationary pin and the rotating disc is dissipated through convection and conduction
• -The pin is characterized as rigid and isotropic, possessing properties that remain unaffected by temperature variations and it is not influenced by conduction.
• -Disc's radiation is not taken into consideration.

2.3.1. Initial conditions

• -Temperature at the Spindle End: The temperature at the end of the spindle is set to the ambient temperature, typically denoted as T₀ = 25 °C.

2.3.2. Boundary conditions

• -Emissivity at the Spindle: The emissivity at the spindle, controlling thermal radiation, is set to a specified value, often represented by the Greek letter epsilon (ε), with a typical value of 0.55.
• -Heat dissipation through convection is applied to the free surfaces of system excluding pin surfaces. Convective heat transfer coefficients specify how heat is dissipated through convection. This convection is defined by convective heat exchange coefficients (Fig. 4a).
• -The heat flux is generated at the interface [18].

Fig. 4.

a) Convective heat exchange coefficients b) Mesh.

2.3.3. Meshing

The mesh (Fig. 4b), created using three-dimensional (3D) tetrahedral elements with 10 nodes, is particularly refined on the friction tracks. Increased mesh refinement near rubbed surface ensures results that are more accurate, improves convergence, and adapts the model resolution to local variations in the studied phenomena.

The thermal study requires the determination of the values of the convective heat exchange coefficients of the free surfaces of the tribometer from the reverse identification method. The convergence of the temperature results at the level of the disc and the spindle is an asset for the validation of the coefficients introduced into the model.

Three continuous wear tests were conducted under a contact pressure of 1.2 MPa and a sliding speed of 6 m/s, lasting for 1500 s. The purpose of these tests was to determinate all convective exchanges on the disc's and spindle's free surfaces by thermal insulation method (Fig. 5).

• -First isolation technique: A refractory concrete added to lime water were employed on the disc's lateral surface (Fig. 5a).
• -Second isolation technique: At the top of the spindle, a thermal sheath is fixed to the spindle by a galvanized steel wire (Fig. 5b).
• -Third isolation technique: Aluminum adhesive tape is used to attach the fiberglass to the spindle in the lower part (Fig. 5c).

Fig. 5.

Experimental setup of the wear test a) test1 b) test 2 c) test 3.

Three configurations were defined to determine the convective heat exchange coefficients (hconv) of the numerical model.

• -Test 1: all surfaces are isolated to determine the convective exchange coefficients hconv1 to hconv 4.
• -Test 2: To determine hconv disc, first isolation is removed
• -Test 3: The hconv spindle can be determined by removing the second isolation

The heat flux (depicted in Fig. 6) was computed using the coefficient of friction obtained at various times during the friction test, utilizing Equation 2. Fig. 6a, b and 6c depicts the heat flux density evolution for respectively test 1, test 2 and test 3. The equation incorporates the mean friction coefficients (μ), pressure (P), and sliding speed (v) for three tests [14].

$$\mathrm{\phi }=\mu *P*v$$ (Eq.2)

Fig. 6.

Heat flux density a) test 1 b) test 2 c) test 3.

The inverse method, as shown in Fig. 7, presents a tool for determining the convective heat exchange coefficients. Initially, a mathematical model is established to characterize the system's behavior. This model typically comprises a set of equations linking the known input parameters to the unknown output responses. Through the acquisition of experimental data from friction tests, the reverse identification process involves determining the parameters or unknown properties of the mathematical model that best align with the observed data. This is achieved by employing optimization techniques to minimize error between model predictions and actual data. An objective function is defined to quantify the difference between model predictions and observed data, and the goal of optimization is to minimize this objective function.

Fig. 7.

Inverse identification approach.

The identification algorithm is utilized to find the set of parameters that minimizes the objective function, providing the best estimate of the unknown properties of the system. The parameters are considered valid as they converge to real values with an error less than the minimal acceptable error. Subsequently, the identified parameters are validated by comparing the model's predictions, using these parameters, with experimental data.

Fig. 8a, b and 8c represents the modeled temperatures and experimental evolution of disc temperature for respectively test 1, test 2 and test 3. The correlation between the temperature levels obtained through simulation and experiments facilitated the determination of convective heat exchange coefficients as outlined in Table 7.

Fig. 8.

Comparing the temperature evolution, the numerical results are represented in blue, while the experimental data is shown in red for three tests: a) Test 1, b) Test 2, and c) Test 3.

Table 7.

Convective heat exchange Coefficients.

	Test 1				Test 2	Test 3
	hconv1	hconv2	hconv3	hconv4	hconv disc	hconv spindle
Convective heat Exchange coefficient (hconv) (W/m2 K1)	140	20	90	110	40	600

The proposed procedure for identifying boundary conditions was validated by successfully determining the heat exchange coefficients. This implies that the finite element method's transient thermal simulation of the braking process exhibits a strong correlation with thermocouple measurements. The numerical approach employed in this study proves to be a highly effective method for investigating the disc's thermal behavior.

3. Results and discussion

3.1. Experimental results

The experimental results obtained from the tribometer are illustrated in Fig. 9a, b,9c and 9d, depicting the evolution of the coefficient of friction in function of the friction distance and cycle number for different disc thickness respectively 10, 15, 20 and 22 mm. It is observed that during the running-in phase, the evolution of the friction coefficient varies considerably among different disc thicknesses. Additionally, it is evident that with increasing thickness, the sliding duration also extends (Fig. 10). Specifically, for the 10 mm thickness disc, the friction time does not exceed 100 s, whereas for the 22 mm disc, it takes approximately 120 s to reach 250 °C. At the conclusion of 30 cycles, the 10 mm thickness disc has a total duration of about 2000 s, while a disc with twice the thickness results in a contact duration of about 3100 s. With a 2 mm increase in thickness, the contact duration exceeds 3800 s. This longer duration provides more opportunity for the friction materials to induce third bodies and the formation of a second plateau. As there are no metallic particles or fibers in the pin composition, a smooth surface is established, contributing to improved friction performance at elevated temperatures [25].

Fig. 9.

Friction coefficient evolution Vs times for 30 cycles; a)10 mm, b) 15 mm, c) 20 mm and c) 22 mm thickness disc.

Fig. 10.

Total pin-on-disc contact duration.

The reduced mass of the thin disc and its lower capacity to store thermal energy contribute to a much higher temperature rise over a short contact period. In contrast, for the 22 mm thickness disc, the generated heat increases with the duration of friction, reaching the temperature threshold of the test for a more extended period. Friction evolution is more stable for the 22 mm disc from the first cycle to the 30th cycle, remaining at 0.38. This stability may be attributed to the material's ability to store high energy, induced by the presence of fewer hot spots on the disc surface [26]. Conversely, for the 10 mm thickness disc, there is a gradual increase in the COF from the first cycle to the 30th cycle, ranging from 0.3 to 0.45 that is attributed of steel oxidized debris coming from the cast iron disc.

The pin temperature calculated on the means friction radius and at the distance of 2 mm far the rubbed surface varied from the disc with 10 mm thickness to 22 mm (Fig. 11). The observed phenomenon can be ascribed to the low thermal properties of the pin, which is empty of metallic elements. In fact, at the first cycles, the temperature of the pin with the small thickness disc do not exceed 230 °C at high temperature (Fig. 12). However, at the end of the cycling, temperature of all pins exceeds 250 °C, especially when 22 mm disc slides with the pin, it rises to 270 °C. It can be explained by the formation of hot bands and deep localized spot more intensive at the mean radius of the contact [27,28]. It is important to note that temperature measurements were conducted from a depth of 2 mm into the contact surface. This temperature does not precisely correspond to that of the friction surface. Additionally, the surface temperature of the pin is consistently higher than the disc temperature. However, the rate of the pin temperature rise is dramatically reduced to the half when we move from 10 to 22 mm of disc thickness (Fig. 13). For the disc thickness of 15 and 20 mm, not so much difference is noticed for pin and disc temperature rise. We conclude the presence of a critical thickness value that well affected the friction and temperature results: beyond 15 mm of thick, the variation of tribological results is more notable; the friction-wear properties of the friction materials are not reliable and not representative of the friction material behavior. The quite stability in the rate of the pin temperature rise for disc thickness less than 15 mm can be explained by the formation of a friction layer of stable oxide in the sliding surface, which modifies the part of the heat entering the pin [29]. It is well known that he decrease of friction coefficient associated to the increase of thermal load knowing as thermal fade behavior [30].

Fig. 11.

Disc temperature evolution for 30 cycles.

Fig. 12.

Pin temperatures evolution for 30 cycles.

Fig. 13.

Rate of the temperature rise of the pin and the disc.

Fig. 14 illustrates the specific wear rate calculated for the different disc geometries at the end of the cycling. The results reveal that the use of 10 mm disc exhibited less wear (6 mg/m). However, the wear rate increased with the increase in disc thickness, reaching 13 mg/m when the 22 mm disc was used. This suggests that the wear rate of the friction material is influenced by the geometrical characteristics of the counterface material and is notably affected by the disc's capacity to store heat at higher temperatures. Additionally, the oxidation rate plays a significant role in wear. The creation of an oxide layer, which is brittle and easily worn off, is accelerated by the increasing pin temperature. This is confirmed by SEM observations of the pin's rubbed surface (Fig. 15). Specifically, for the pin sliding against a disc with a 10 mm thickness, where the temperature rate is the highest, the surface is more covered by second plateaus, acting as a wear-resistant oxide layer (Fig. 15a). The development of these plates is attributed to the role played by rockwool fibers and shots. Carbon particles, appearing as dark gray elements, are more covered, necessary for secondary plate expansion (Fig. 15b).

Fig. 14.

Pin-wear rate Vs disc thickness.

Fig. 15.

Pin rubbed surface against a) 10, b) 15, c) 20 and d) 22 mm disc thickness.

As the thickness of the disc increases, the surface of the pin becomes less uniform and homogeneous. For a disc thickness of 10 mm, more powder plates are observed. Furthermore, for disc thicknesses of 20 mm (Fig. 15c) and 22 mm (Fig. 15d), fewer compacted particles are detected, and third-body plates no longer shield the surface. In fact, the presence powdery third body and reduced size flat plates indicates that the third body tends to disengage from contact during friction. These mechanisms contribute to explaining the highest observed wear rate for the pin sliding against the disc with a thickness of 22 mm.

3.2. Results from the thermal model

A simplified finite element model, comprising a brake disc, a spindle, and the friction track, was utilized to streamline the calculation time and fully leverage the capabilities of the Ansys finite element software. The thermal model was initially calibrated using experimental results, determining the parameters that control the convective cooling of the friction materials. It is noticeable that the temperatures of the calibrated model closely matched the measured temperatures for most of the test duration. With the calibrated model, it became possible to conduct numerical analysis to study the temperature evolution of brake disc.

In this context, with the aim of enhancing the thermal kinetics of the disc, an analysis of the modification of the geometry was carried out. Four models were defined by altering disc's thickness to 22, 20, 15, and 10 mm.

The numerical study (Fig. 16a) demonstrates that the reduced disc thickness of 10 mm induces a rapid temperature rise in the disc brake and because of the increase in disc thickness; the rate at which the temperature rises has decreased (Fig. 16b).

Fig. 16.

a) Numerical disc temperature of 15th cycle. b) Kinetics of temperature rise versus disc thickness.

The relationship between disc thickness and temperature rise kinetics might be influenced by several factors such as thermal capacity, thermal conductivity and its mass. In fact, a thicker brake disc (22 mm) has a higher thermal capacity, which means it can absorb more heat during braking. The additional mass for the disc of thickness of 22 mm can act as a thermal reservoir, absorbing the heat generated during braking. Greater mass can also help dissipate heat more efficiently. This will maintain a more stable temperature during braking and reduces the time required to reach the maximum test temperature (250 °C) from 130s to 67 s. Moreover, the thinner the brake disc, the less it will be able to effectively absorb and dissipate this heat. As a result, a brake disc that is too thin tends to overheat faster and a rapid kinetics of temperature rise (Fig. 16b), which can lead to a decrease in braking efficiency, premature brake p wear, and even permanent disc deformation. Several researchers proves that the rapidly rise in temperature reduces friction and causes thermal cracks on disc [31] on the braking surface and vibrations and noise [32]. A recently study proves that thermal cracks are originated from the rapid increase of disc brake temperature which mainly reduce friction performance [33].

In addition, if there is a rapid temperature increase for the thinner disc, a very smaller temperature increase could cause dynamic stresses, which could lead to thermal deformations [34]. In fact, the deformation of the disc can cause a variation in the coefficient of friction during braking. This factor has the potential to influence the overall functionality of the system: friction and wear responses. This aligns with the experimental findings discussed aboved.

4. Conclusion

In the present study, a critical examination of the choice of disc thickness for the development of the tribological test to investigate dry sliding of brake lining material against a cast iron disc under high conditions has been conducted. The key findings and conclusions are summarized below.

• -In a pin-on-disc configuration, the friction coefficient, sliding contact duration, surface temperature rise, and pin-wear rate are significantly influenced by the disc thickness, especially under high solicitation conditions (250 °C contact temperature, 30 cycles of sliding). Thicknesses of 15 and 20 mm exhibit more similar tribological responses, while 10 and 22 mm thicknesses show noticeable divergence and less similarity in friction-wear behavior.
• -An arbitrary choice of disc thickness cannot yield reliable results and capture the effective behavior attributed to the brake material. Specific modeling was necessary to better understand the influence of thickness on disc thermal kinetics. The thermal model results indicates that thinner discs dissipate more power through friction, compensating for losses by convection more effectively.
• -For a significantly reduced disc thickness (10 mm), heat dissipation is much lower. In fact, thinner disc has a smaller heat dissipation surface area, potentially leading to poorer heat dissipation during prolonged braking and reduced performance due to rapid overheating.
• -Experimental and numerical results demonstrate that a disc thickness of 10 mm results in a loss of efficiency of the friction material due to the high kinetics in temperature rise. Increasing disc thickness can simultaneously improve thermal and frictional properties. For disc thickness equal to 15 mm, the good repeatability of the coefficient of friction, allows for proper consideration of the change in disc thickness.

In conclusion, the brake disc's thickness affects temperature kinetics during braking by influencing thermal capacity and disc mass. It can indirectly affects the coefficient of friction, thereby influencing brake system's performance. Properly sizing the disc is crucial for maintaining stable braking temperatures and preventing the overheating of braking system.

Data availability statement

Upon request, the data will be provided.

CRediT authorship contribution statement

Amira Sellami: Writing – original draft, Methodology, Investigation, Conceptualization. Riadh Elleuch: Supervision, Methodology, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.