1

Braking Performance of Ceramic Coated Discs

Bu-Byoung Kang, Kang-Youn Choe, Hyeong-Jin KimKorea Railroad Research Institute

374-1 Woulam-Dong, Uiwang-city, Kyonggi-Do, KoreaTEL: +82-343-461-8531(Ext328)

FAX: +82-343-461-8561E-mail: bbkang@krri.re.kr

Gérard DEGALLAIXLaboratoire de Mécanique de Lille, Ecole Centrale de Lille, Lille, France

E-mail: Gerard.Degallaix@ec-lille.fr

ABSTRACT For the speed up the train and the reduction of maintenance costs, we need to develop new brake discmaterials with higher frictional performance and longer service life. An interesting alternative is ceramicmaterials. The advantages of ceramic materials are lower wear rate and higher resistance to thermalshocks. However, bulk ceramic materials show low fracture toughness and difficulty in fabricating to therequired shape. On the other hand, the use of ceramic coatings give good solutions to those problems.Plasma spraying technique, which ensures thick ceramic coatings, has been chosen as a good method toproduce coatings on a metallic substrate. In this study, three kinds of brake discs including two coated brake discs and one steel disc were testedunder the same experimental conditions on a reduced scale braking test bench. Braking test bench wasspecially designed to analyse thermo-mechanical and frictional behaviors of two sizes of brake discs instop and hold braking modes. Plasma spray coating technique was used to coat ceramic powder on thediscs. In the test, four commercial sintered brake pads were coupled with discs. Ceramic coated discs haveshown good stability in friction coefficient at high speed and high energy braking conditions. However,ceramic coated discs caused more wear loss of pad mass than the steel disc. It was shown that thermalbarrier effect in ceramic coated discs adjusted the thermal partition between pad and disc. Steel discshowed fluctuating friction coefficient at high speed but less wear loss of pad mass than ceramic coateddiscs.Keywords: Ceramic coated brake disc, Friction coefficient, Mass wear loss, Thermal barrier effect

1. IntroductionFor the speed up of the train and the reduction of maintenance costs, we need to develop new brakematerials for high energy braking conditions with higher frictional performance and longer service life.When high-speed train is stopped from the high speed like 350km/h or more, high energy is dissipatedon the friction surface of pad and disc. And disc and pad are under the high temperature and highbraking pressure condition. So we need to develop new materials with stable frictional performanceand higher wear resistance for the reduction of maintenance costs. Interesting alternatives are ceramicmaterials and carbon materials.

Ceramic materials are chosen for the experimental study . The advantag es of ceramic materials are lowerwear rate and higher resistance to thermal shocks. However, bulk ceramic materials show low fracturetoughness and difficulty in fabricating to the required shape. On the other hand, the use of ceramiccoatings gives good solutions to those problems. Plasma spraying technique, which ensures thick ceramiccoatings, has been chosen to produce coatings on a metallic substrate. Ceramic coatings can endure high surface temperature. Low thermal conductivity of ceramic materialsshows thermal barrier effect. This effect decreases the proportion of heat transferred to the disc substrateand reduces the temperature rise of the disc substrate. This also decreases the amount of plasticdeformation, and metal seizure due to localized high flash temperature during high energy braking. In thisway, the ceramic-coated brake disc leads to stable frictional performance and decreases the wear.However, high proportion of the heat transferred to the pad side and high hardness of disc surface makethe pad be under the severe conditions. Therefore, we should examine the associated phenomena betweencoated disc and pad. In this study , three kinds of brake discs including two coated brake discs and one steel disc were tested

2

under the same experimental conditions on a reduced scale braking test bench. Braking test bench wasspecially designed for stop and hold braking tests.

2. Experiment

2.1 Braking test bench and specimen Reduced scale braking test bench as shown in Fig. 1 was used. The test bench has electric inertiasimulation mechanism and can control spindle speed, braking pressure, braking torque for simulation ofstop/hold/subway braking modes. The cross-section of coated disc and pad are shown in Fig. 2. Thedimensions of discs are shown in Fig. 3. We used two different sizes of discs. The larger disc (Disc1) has345mm outer diameter and 85mm inner diameter, and the small disc (Disc2) has 265mm outer diameterand 85mm inner diameter. We tested two ceramic-coated discs and one steel disc currently used for TGVand 4 commercial sintered pads used also for TGV. The ingredients and specifications of coating powdersare shown in Table 1 and 2

Braking test bench

The cross-section of the disc and the pad

3

2.2 Test methodBraking tests are performed on same experimental conditions and procedures as in Table 3 and 4 fordifferent couples of the discs and the pads. We tested two kinds of test modes, emergency stop brakingand hold braking simulating the conditions of running down the slope. We control the specific contactpressure or initial speed in stop braking test, and control the torque or speed for hold braking test in orderto see the braking performance under different braking conditions. We measured the temperatures of discsand pads with K type thermocouple during the braking period. Temperature measurement positions in thedisc and pad are shown in Fig. 3. We also measured the pad mass wear after each test step of brakingtests. We had bed-in test before the start of braking tests for all couples.

Temperature measurement positions in the disc and the pad

4

Table 1: Materials for tested brake discs and pads

Table 2: Geometrical properties of brake discs

3. Discussion

3.1 Thermal behaviorFig. 4 shows the temperature evolutions of three different discs and pad C under hold braking conditions.Temperature difference between pad and Zirconia coating disc is larger than that between pad and steeldisc. This results from the thermal barrier effects of zirconia coating. Thermal barrier effect of ceramicmaterials results from low conductivity caused by the porosity of materials. In the case of Zirconia disc,much portions of heat is transferred to the pad, so the pad temperature is very high. However in the caseof steel disc, pad temperature is low and even lower than disc temperature. This means that more portionof dissipated energy is transferred to the steel disc compared with the case of the ceramic-coated disc.Fig.5 shows the temperature evolution of discs and pads under stop braking conditions. Zirconia coateddisc shows the biggest thermal barrier effect among three discs.

5

Fig. 4: Temperature evolution of the discs and the pads during hold braking test

Fig. 5: Temperature evolution of the discs and the pads during stop braking test

3.2 Friction coefficients

3.2.1 Hold brakingDissipated energy vs. maximum temperature and mean friction coefficient characteristics of the threecouples with pad B are shown in Fig. 6 and 7. Maximum temperature shown in Fig.6 is the maximumtemperature measured during braking process like one shown in Fig.4, and mean friction coefficient

6

shown in Fig.7 is the average of the friction coefficients during braking process like one shown in Fig.8under different braking conditions.

Fig. 6: Dissipated energy vs. maximum temperature: hold braking test with disc 2

Fig. 7: Dissipated energy vs. mean friction coefficient: hold braking test with disc 2

Temperature of disc and pad increase proportionally with increasing of dissipated energy, and thetemperature difference between disc and pad is not much varying with increasing dissipated energy asshown in Fig.6. The temperature differences between pad and zirconia coated disc that has bigger thermal

7

barrier effect than other discs are the biggest of the three cases. Mean friction coefficients of cermet-padB and zirconia-pad B couples show more stable state than steel disc-pad B couple according to theincrease of dissipated energy as shown in Fig.7 Fig.8 and Fig.9 show friction coef ficient evolution during hold braking test, which was performed for 3to 5 minutes for the simulation of high energy dissipated braking conditions for ceramic-coated disc andsteel disc. Ceramic-coated discs show apparently stable friction coefficient at high energy brakingconditions. Among couples, zirconia-pad B, cermet-pad B, and zirconia-pad C couples show stablefrictional performance in descending order. Steel disc had shown fluctuating friction coefficient. But incouple with pad C the fluctuation is changing with small amplitude about the constant mean value of 0.4,and with pad B fluctuation is very small and gradually increases with time. Steel-pad D couple showsunstable result and the test was stopped at 130sec.

Fig. 8: Friction coefficient evolution during the hold braking test with ceramic coated discs

8

Fig. 9: Friction coefficient evolution during the hold braking test with 28CDV5 discs

3.2.2 Stop brakingFig.10 shows the mean friction coefficient vs. initial contact velocity characteristics of cermet and steeldisc coupled with pad B. Cermet disc show very stable friction coefficient and lower friction coefficientsfor higher contact pressure.

Fig. 10: Mean friction coefficient vs. initial contact speed: stop braking test with disc2

9

Fig. 11: Friction coefficient evolution during the stop braking test with pad C and disc2

Fig.11 shows the friction coefficient evolution in the stop braking test of 30m/s contact linear velocity(2604rpm) and contact normal force of 666N(normal pressure 0.53Mpa) for three different couples withpad C.Cermet disc shows stable friction coefficients from beginning of the test. However, steel disc and zirconiacoated disc shows rooster tailing phenomena of increasing friction coefficient at the end of braking. In thebeginning of the stop braking, three different kinds of disc show similar tendency. But as speed isdecelerated at the end of the braking, zirconia and steel disc show abrupt increase in friction coefficient,while cermet disc shows small increase. The test result can be explained from the reason that cermet disccause faster pad surface wear before material transformation is taking place. But zirconia disc sufferedsurface change and transformation and caused stick-slip at low speed. Cermet disc caused more pad masswear than other discs as shown in Fig.13.

10

Fig. 12: Friction coefficient evolution during the stop braking test with disc1

Fig.12 shows the results of high speed stop braking tests. Cermet disc and pad B couple shows moststable friction coefficients. But pad mass wear is larger than any other couples as shown in Fig.13. In thebeginning, friction coefficient is high as shown in Fig.12 and then decreasing to the stable value duringthe braking and rising again at the end of the test. Evolution of friction coefficient seems to result fromthe reason that the surface layer made of wear debris causes the decreasing of friction coefficient, andtransformation of surface layer and wear debris cause the rising of friction coefficient at the end of thebraking. Rooster tailing phenomena is larger in steel disc than cermet disc.

3.3 Pad mass wearThe pad mass wear per dissipated energy was measured after each braking step shown in Table 3 and 4,and the total pad mass wear per energy is shown in Fig.13 for all tested couples. The pad mass wearcoupled with small disc show the result of the smallest wear in steel disc. The friction couples with largerdisc show similar tendency in pad mass wear, but 1.5 times larger than couples with small disc. This isbecause the result with larger disc includes the test under high normal pressure and high-speed conditions.Cermet disc coupled with pad B show stable friction coefficient but larger pad mass wear. This is becausecermet-coated disc with metallic ingredients, which have high hardness, cause the wear of the pad surfacebefore oxidized friction surface was made on the disc surface and keep the surface under uniformconditions.

11

Table 3: Braking test procedure for disc1

Table 4: Braking test procedure for disc2

12

Disc surface temperature is known as the dominant factor to wear of steel disc. Similarly in this study ,steel disc cause greater pad mass wear loss in high-speed stop braking test and high energy hold brakingtest. But comparing to other disc, steel disc shows smallest pad mass wear per energy as shown in Fig.13.Mass wear of pad A was smaller than that of other pads. But thick hard layer of about 1mm thickness wasformed on pad surface and locally detached from the pad surface. Severe transformation of pad surfacelike surface layer detaching cause great change in friction coefficient and unstable friction coefficient asshown in Fig.8 Pad D show larger pad mass wear than any other pads as shown in Fig.13 and showunstable friction coefficient as shown in Fig.9.

Fig. 13: Pad mass wear loss of the braking pairs after finishing all braking test steps

Pads used in this study are composed of Cu as main component and Fe, graphite, silica, Al2O3 as frictioncontrol elements. All particles are mixed uniformly but the particle sizes are different in each pad. Pad C,B, D have smaller particle size than pad A. Pad A has bigger graphite particles than other pads. Pad C andB composed of small particles show stable friction coefficient and more healthy surface state after testing.

4. ConclusionsThree different kinds of brake discs including two coated brake discs and one steel disc coupled with fourcommercial sintered pads were tested on a reduced scale braking test bench for friction, temperature, wearanalysis. The following results are drawn through the test.

- Ceramic coated discs had shown good stability in friction coefficient at high speed and high energybraking conditions. But Ceramic coated discs caused more pad mass wear loss than the steel disc.- Zirconia disc showed the bigger thermal barrier effect and more stable friction coefficient at high energybraking conditions than cermet disc. Cermet disc showed stable friction coefficient under nearly allconditions but caused larger pad mass wear than zirconia disc.- It was shown that thermal barrier effect in ceramic-coated discs adjusted the thermal partition betweenpad and disc.

13

- Steel disc had shown fluctuating friction coefficient at high speed and high energy braking conditions.but less pad mass wear loss than ceramic coated discs.

In the future, we need to develop new pad materials for ceramic discs to maintain longer service life ofpad. Also we should examine the durability of ceramic coating to assure service life and reliabilitythrough endurance test.

5. BibliographyBu-Byoung Kang, Nam-Uk Baeg, Kyung-Soo Jang, 1998, "Experimental Analysis of Ceramic CoatedBrake Discs for High Speed Train", KSME(KOR), Vol 22, No 5, pp. 821_833.