Journal of Mechanical Science and Technology 26 (1) (2012) 235240 /content/1738-494x DOI 10.1007/s12206-011-0921-y Investigation of thermo-structural behaviors of different ventilation applications on brake discsMesut Duzgun*Automotive Engineering Department, Faculty of Technology, Gazi University, 06500, Ankara, Turkey (Manuscript Received November 12, 2010; Revised July 20, 2011; Accepted September 18, 2011) - Abstract One of the most common problems related to brake discs is overheating, which negatively affects braking performance especially un-der the continuous braking conditions of vehicles. Ventilation applications on brake discs can significantly improve the brake system performance by reducing the heating of the discs. In this study, the thermal behaviors of ventilated brake discs using three different con-figurations were investigated at continuous brake conditions in terms of heat generation and thermal stresses with finite element analysis. The results were compared with a solid disc. Heat generation on solid brake discs reduced to a maximum of 24% with ventilation appli-cations. The experimental study indicated finite element temperature analysis results in the range between 1.13% and 10.87%. However, thermal stress formations were higher with ventilated brake discs in comparison to those with solid discs. Keywords: Brake discs; Heat generation; Thermo-structural behaviors; Ventilation applications -1. Introduction Ventilated brake discs or rotors are known as high-performance brakes, and are produced by making hollows or slots (or both) of different shapes on disc surfaces and side edges. Ventilated brake discs were originally tested on racing cars in the 1960s, and they have been employed widely in the automotive and railway industry using different designs 1, 2. During braking, kinetic energy is converted to heat. Around 90% of this energy is absorbed by the brake disc and then transferred to ambient air. Solid brake discs dissipate heat slowly. Therefore, ventilated brake discs are used to improve cooling by facilitating air circulation 3, 4. They generally exhibit convective heat transfer coefficients approximately twice as large as those associated with solid discs 5. There are numerous studies related to ventilation applica-tions on brake discs. Zuber and Heidenreich 6 manufactured three different ventilated brake disc constructions from carbon fiber-reinforced ceramic matrix composites (CMC) and com-pared their strengths. Antanaitis and Rifici 7 proved that the 90-hole cross-drilled pattern improved heat rejection capabil-ity of the disc between 8.8% and 20.1% depending on the vehicle speed. Aleksendric et al. 8 showed the ability of a ventilated brake disc rotor in dissipating thermal flow by finite element analysis (FEA). Venkitachalam and Maharudrappa 9 conducted flow and heat transfer analysis of six different types of disc configurations by computational fluid dynamics (CFD) and recommended ventilated brake discs for high-speed vehicles. Park et al. 10 designed a helical surface in-side vanes for a ventilated brake disc. They optimized Rey-nolds (Re), Prandtl (Pr), and Nusselt (Nu) numbers for their design and obtained improvements to a maximum of 44% in heat transfer. Improvement in brake fade resistance and higher braking performance in wet conditions are some other useful aspects of ventilated brake discs. However, they also have some disadvantages. Cracking, is one of them and this a phe-nomenon that has been correlated to stresses during braking 11. Kim et al. 12 showed the maximum von- Mises stress generation of actual fatigue cracks located on ventilated brake discs of railway vehicles by thermal stress analysis. Similarly, Bagnoli et al. 13 performed FEA to determine the tempera-ture profile and to estimate the von- Mises stress distribution that arises during braking for fire-fighting vehicles. Hwang and Wu 14 investigated temperature and thermal stress in a ventilated brake disc based on a thermo-mechanical coupling model. Decreasing the brake temperatures and/or re-designing the hub-rotor unit were some considerable conclusions of Mackin et al. 15 to eliminate cracking in brake rotors. Heat generation also affects thermo-mechanical instability of brake discs 16. Previous literature focused on heat and stress formations on ventilated discs. In this study, FEA was used to investigate the This paper was recommended for publication in revised form by Associate Editor Dae Hee Lee *Corresponding author. Tel.: +90 312 2028650, Fax.: +90 312 2120059 E-mail address: .tr KSME & Springer 2012 236 M. Duzgun / Journal of Mechanical Science and Technology 26 (1) (2012) 235240 thermal behavior of three different ventilated brake designs: cross-drilled (CD), cross-slotted (CS), and cross-slotted with side groove (CS-SG) discs. The results were then compared to a solid (SL) disc. An experimental study was also performed to verify the FEA results. 2. Thermo-structural FEA For FEA, three-dimensional (3D) constructions of brake discs, brake pads, and their assembly designs were modeled with 1/1 scale in a software program, and then imported into another software program for the interactive thermo-structural analyses. Brake discs and brake pads were modeled by Quad-ratic Hexahedron mesh types. Quadratic Hexahedron mesh generations are known for their accuracy and computational efficiency 17. A frictional contact pair having the element type of Quadratic Quadrilat-eral Contact was defined between disc-pad interfaces. Fig. 1 shows the mesh models of the disc-pad systems. Grey cast iron, a commonly used disc material, was used for the brake discs. The mechanical and thermal properties of the brake discs and pads are given in Table 1. Ventilated brake discs were designed according to the pro-peller-shaped methodology for the hole and slot locations. For the CD disc, five holes with 5.2 mm diameter were arranged at equal intervals on an arc with a length of 60.54 mm. These holes were duplicated in groups of 20 on the disc surface. Thus, a total of 100 holes were drilled on the disc surface of the CD disc. For the CS disc, 20 channels (6.9 mm wide and 67.3 mm long) were placed on a solid disc surface. Finally, the CS-SG disc was designed by making a groove (4 mm wide and 15 mm deep) on another CS disc edge. Hence, a path opening to the outer disc side edge was obtained to provide better air circulation. 2.1 Thermal analysis For the thermal analysis, the ambient temperature was as-sumed at 22C and the disc surface temperature was 100C prior to braking, which was related with cold braking per-formance 8. The current study assumed that heat dissipation from the brake disc to the atmosphere occurs via convection, also known as Newtons law of cooling. Convection is gov-erned by Eq. (1), where Q is the rate of heat transfer (W), h is the convection heat transfer coefficient, A is the surface area of the rotor (m2), Tsis the surface temperature of the brake rotor (C), and Tis the ambient air temperature (C). The convection heat transfer coefficient is applied to the body of the brake discs as the boundary condition. Thus, to increase heat transfer from the brake discs and to reduce the disc sur-face temperature on the total surface area of the brake discs, the heat transfer coefficients were gradually increased by em-ploying ventilation applications. .( )QhAT Ts= (1) The heat transfer coefficient associated with laminar flow for solid or non-ventilated brake discs was derived by Eq. (2) (for Re2.4x105) 5, where D is the outer diameter of the discs (mm), Re is the Reynolds number, and kais the thermal conductivity of air (W/mC). 0.550.70( / )RehkDaR= (2)Meanwhile, the heat transfer coefficient associated with laminar flow for ventilated brake discs was approximated by Eq. (3) (for laminar flow condition, Re 104) 5, where Pr is the Prandtl number, dhis the hydraulic diameter (mm), and l is the length of the cooling vane (mm). The hydraulic diameter (dh) is defined as the ratio of four times the cross-sectional flow area (wetted area) of the hole and slots in ventilated brake discs divided by their wetted perimeters as illustrated in Fig. 2. 1/3 0.331.86( ) ( / ) ( / )hRePrdlkdaR hh= (3) In this condition, Re number is associated with the velocity of the air flow present in the hole and slot-shaped vanes as determined by Eq. (4), where ais the density of air (kg/mm3), mais the mass flow rate of air (m3/sec), and Vaverageis the aver-age velocity (m/sec). Table 1. Mechanical and thermal properties of brake discs and pads. Mechanical and thermal properties Disc Pad Youngs modulus (N/mm2) 110000 1500 Poissons ratio 0.28 0.25 Density (kg/m3) 7200 2595 Thermal expansion (1/C) 1.1e005 6.6e005 Tensile ultimate strength (N/mm2) 240 - Compressive ultimate strength (N/mm2) 820 - Coefficient of friction 0.35 0.35 Thermal conductivity (W/mC) 52 1.212 Specific heat (J/kgC) 447 1465 (a) SL disc (b) CD disc (c) CS disc (d) CS-SG disc Fig. 1. Finite element mesh models. M. Duzgun / Journal of Mechanical Science and Technology 26 (1) (2012) 235240 237 Re ( / )dmVa a averageh= (4)The average speed can be calculated by Eq. (5), where nTis the revolutions per minute (1/min, rpm), D is the outer diame-ter of the disc (mm), d is the inner diameter of the disc (mm), Aoutis the outlet area of the hole or slot-shaped vanes (mm2), and Ainis the inlet area of the hole or slot-shaped vanes (mm2). 220.0158 ( )VnDdAAaverage out outinT=+(5) Moreover, the air flow rate mais determined by Eq. (6): 22 30.00147 ( ) ( sec) .mnDdAma inT= (6)2.1.1 Experimental study The experimental study was performed to examine tem-perature changes on disc surfaces. For this purpose, the venti-lated brake discs were manufactured from the Alfred Teves (ATE) solid discs to obtain their design characteristics. The CD and CS discs were manufactured on a three-axis CNC Vertical Machining Centre. The side groove for the CS-SG disc was made on a CNC Turning Centre. The disc tempera-ture outputs were measured on a brake test system. This sys-tem consists of a caliper mechanism as seen in Fig. 3, a piezo crystal force measurement system for pedal and brake force variations, a driving engine, a gear box, and indicators for brake/pedal forces and temperature. For the temperature measurement, a thermocouple mounted on the caliper system was used. The power of the motor was 4 kW. The rotation of the discs was on a clockwise direction. ATE 501 FF brake pads were used in the experiments. Ex-periments were conducted under continuous braking condi-tions at a constant pedal force of 250 N, and eight periodic measurements of braking temperature were executed at 30, 60, 90, 120, 150, 180, 210 and 240 s. 2.2Structural analysis Disc temperatures obtained by thermal analyses were im-ported into the structural analytical models as boundary condi-tions. In the structural analyses, 9689 N equivalent to 250 N pedal force was applied on each top surface of the brake pads at a revolution speed of 60 km/h (or angular velocity of 52.56 rad/s) as seen in Fig. 4. The rate of the disc was assumed to be constant and the time required for a complete stop was 240 s. The value of the pedal force was obtained by Eq. (7) 18, where Ffis the force on each front cylinder piston, F is the force on the foot pedal, Afis the cross-sectional area of the front pistons, Amis the cross-sectional area of the master cyl-inder, pnis the number of pistons, 2.3 is the pedal leverage ratio, and 2.75 is the effect of servo unit. . . .2,3.2,75fFA pnfFAm= (7) This computation was done according to the test equipment used in the experimental study. The pedal force leverage ratio and the servo unit effect are the values of the test equipment. Analyses were simulated according to the equivalent (von-Mises) stress distributions. Fig. 2. Wetted areas (hydraulic diameter, dh), length of hole and slot-shaped cooling vanes, and inlet and outlet areas for air flow of venti-lated brake discs used in this study. (a) SL disc (b) CD disc (c) CS disc (d) CS-SG disc Fig. 3. Brake caliper and brake discs. Fig. 4. Structural model for the brake discs. 238 M. Duzgun / Journal of Mechanical Science and Technology 26 (1) (2012) 235240 3. Results and discussion 3.1 Heat generation Experimental and FEA results are given in Table 2 for the generation of heat on the disc surfaces. Disc surface tempera-tures increase with increasing braking time for all disc con-figurations. However, heat generation is remarkably reduced by ventilation applications. The friction coefficient between brake pad and disc surfaces decreases depending on the tem-perature rise 19, 20. Hence, maintaining friction properties of the pads at high temperatures is possible by self-ventilation on the discs under continuous braking conditions. Fig. 5 shows the temperature distributions on the disc sur-faces by FEA at the end of 240 s. The maximum temperature generation occurred in the middle regions of all disc configu-rations, similar with related studies 12, 14. However, the temperature region moves from the middle region of the disc to the inner region due to the additional side-cooling in the CS-SG disc configuration. The maximum heat generation on solid disc surfaces is reduced around 4% in the CD disc design. On the other hand, while the maximum heat generation on solid disc surfaces is reduced around 19% in the CS design, it is reduced around 24% in the CS-SG design. Thus, it is bene-ficial to investigate the thermal stress behavior of these de-signs. 3.2 Thermal stresses Fig. 6 shows thermal stress distributions on the disc surfaces by FEA. Ventilation applications increase the thermal stresses on the brake discs. Maximum thermal stress is localized on the corner of the inner and outer edges of the solid disc surfaces. In the case of the CD disc configuration, the maximum stress is located in the inner surfaces of the holes. For the CS and CS-SG disc configurations, maximum stress formations are mainly localized around the slot surfaces close to the inner and outer points of the disc. Hence, the specific regions, where maximum stresses are seen, should be strengthened to prevent potential crack and fatigue problems. One of the possible solu-tions is the design of different heat dissipation surfaces for ventilated brake discs. Therefore, more homogenous heat convection from the rotor can be provided for the disc surfaces. Table 2. Experimental and FEA results for disc temperatures. SL disc CD disc Time (sec.) Exp. temp. (C) FEA temp. (C) Exp. temp. (C) FEA temp. (C) 30 124 137.49 156 169.99 60 189 174.97 219 200.02 90 238 212.44 254 230.06 120 276 249.92 291 260.09 150 318 287.40 321 290.13 180 358 324.87 345 320.16 210 378 362.35 355 350.20 240 395 399.82 371 380.23 CS disc CS-SG disc Time (sec.) Exp. temp. (C) FEA temp. (C) Exp. temp. (C) FEA temp. (C) 30 128 139.66 125 130.89 60 184 165.68 168 155.11 90 213 191.70 201 179.32 120 240 217.72 227 203.52 150 273 243.73 253 227.70 180 289 269.74 269 251.87 210 314 295.75 285 276.03 240 325 321.76 286 300.18 (a) SL disc (b) CD disc (c) CS disc (d) CS-SG disc Fig. 5. Heat generation on different kind of brake discs. (a) SL disc (b) CD disc (c) CS disc (d) CS-SG disc Fig. 6. Thermo-structural behaviors of different kinds of ventilated and solid brake discs. M. Duzgun / Journal of Mechanical Science and Technology 26 (1) (2012) 235240 239 This is proven by the CS-SG design where the maximum stress generation on the ventilated brake discs is reduced. In this study, additional thermal stress dissipation on disc surfaces was also investigated. Fig. 7 shows the dissipation characteristics. Heat dissipation surfaces between hole and slot arcs of CD, CS and CS-SG discs were determined to examine stresses where dense friction occurs between the pad and the disc. The maximum stress locations for ventilated brake discs are seen in Fig. 6. The thermal stresses are reduced from the center point between the disc and the pad surfaces toward the inner and outer regions of the disc surfaces for the SL configu-ration. However, stress distributions on the CD disc surfaces are more homogenous in comparison with the solid disc. For the CS disc configuration, while the stresses are reduced on the frictional surfaces, they gradually increase toward the in-ner and outer surfaces of the discs, similar with the CS-SG disc configuration. These results will cause a stable wear rate on the pad and disc surfaces of the SL and the CD discs. On the other hand, wear rates will be unstable with the CS and the CS-SG discs. 4. Conclusions In this study, three different ventilated discs were modeled and their thermo-structural behaviors investigated. The ex-perimental study verified the FEA results for heat generation on the disc surfaces. The following conclusions were drawn: Heat generation on the solid brake disc surfaces are reduced to a maximum of 24% by ventilation applications. The ex-perimental study verifies the finite element temperature analy-sis results in the range between 1.13% and 10.87%. This result will positively affect the braking performa