Development and Testing of High Performance Wheel Steels

Transcription

1 Development and Testing of High Performance Wheel Steels ABSTRACT Scott Cummings, Semih Kalay Transportation Technology Center, Inc., Pueblo, Colorado, USA Eight types of high performance wheels are being evaluated by Transportation Technology Center, Inc. (TTCI) as part of the Association of American Railroads (AAR) Strategic Research Initiatives (SRI) Program to prevent wheel failures. The goal of this project is to develop and demonstrate the benefits of high performance wheel steels, specifically focusing on improvements in resistance to wear and fatigue. Wheel manufacturers from around the world including Griffin, Lucchini, OneSteel, Standard Steel, Sumitomo, and Valdunes have donated high performance wheels for this project. TTCI s high performance wheel, known as the SRI wheel, is also being tested. Replacement of wheelsets is the major freight car maintenance expense item in North America, accounting for 51 percent of all freight car maintenance dollars in Wheel tread damage (shelling or spalling) and wheel wear are the two primary causes for wheelset replacement. Potential benefits of several of the high performance steels in comparison to the current standard AAR Class C steel were indicated during a laboratory evaluation. Improved yield strength, microcleanliness, and hardness are three criteria in which the high performance wheel steels exceed those of the AAR Class C wheel steels. No safety concerns have been noted with any of the high performance wheels. In August 2009, a revenue service test was initiated by installing a large sample of each type of wheel under a unit coal hopper train. Axle load is the standard for North American freight: 32.5 tonnes. The intended test duration is approximately three years to allow sufficient time for a statistically significant number of wheelset removals. Data from visual wheel inspections and from wayside detectors is used to monitor performance of the wheels and cars in the test train. Inspections of the wheels in the revenue service test train were conducted after approximately 23,000 miles (37,000 km) of service and again at 88,000 miles (142,000 km). At this relatively early stage in the test, the high performance wheels are performing similarly to AAR Class C wheels in terms of rolling contact fatigue (RCF) and shelling. The AAR Class C wheels with one type of tread conditioning brake shoe exhibit more RCF and shelling than other combinations of wheel and brake shoe. The average wheel wear rate is extremely low for all types of wheels. In addition to the revenue service test, a limited number of wheels are undergoing a durability test at TTCI s Facility for Accelerated Service Testing (FAST). Axle loads of 32 tonnes and 36 tonnes are being tested. Wheel/rail friction is carefully controlled and the wheels are evaluated for wear and tread damage at least twice annually. Minor shells and RCF cracks have appeared, and wear rate differences are noticeable between different wheel types. The wear rate is much higher for all wheel types in the durability test at FAST compared to the revenue service test. This is believed to be due to the cars at FAST always running loaded in a loop made up of 70 percent sharp curves. INTRODUCTION AND BACKGROUND As part of the AAR s SRI program to prevent wheel failures, eight types of high performance wheels are being evaluated through laboratory testing and service testing. The goal of this project is to develop and demonstrate the benefits of high performance wheel steels, specifically focusing on improvements in resistance to wear and fatigue. In North American railroad operations, replacement of wheelsets is the major freight car maintenance expense item. Half (51 percent) of all freight car 1

2 maintenance dollars are spent on wheelset replacement. Wheel tread damage (shelling or spalling) and wheel wear are the two primary causes for wheelset replacement. 1 Griffin, Lucchini, OneSteel, Standard Steel, Sumitomo and Valdunes donated high performance wheels for this project. OneSteel is participating with two steel compositions. TTCI s high performance wheel, known as the SRI wheel, is also being tested. With the exception of the SRI wheel, a generic naming convention will be used to identify each manufacturer s wheels. The evaluation of the high performance wheels consists of three phases which overlap to some degree. First, laboratory testing was conducted on each wheel steel including measurements of mechanical properties, microcleanliness, and residual stresses. Next, the wheels were installed in loaded cars at TTC and subjected to a drag braking test and are currently involved in a durability test at FAST. The durability test is being conducted on a small sample size of each type of wheel. The operating conditions of the durability test are intended to accelerate wear and fatigue damage on the wheels. The third phase of testing for the high performance wheels is the revenue service test, which began in August The revenue service test is being conducted on a larger sample size to quantify the benefits of each type of high performance wheel in comparison to the current standard AAR Class C wheels. None of the testing completed to date has indicated any safety concerns related to the high performance wheels. SRI Wheel The SRI wheel steel was developed by TTCI under the SRI program to meet specific criteria. These criteria are as follows: Room temperature yield strength greater than 130 ksi (900 MPa) Room temperature fracture toughness greater than AAR Class C As-manufactured tread surface Brinell hardness between 380 and 420 Limits for standard AAR wheel microcleanliness test procedure 0.1 average volume percent, voids plus oxides 0.2 maximum volume percent, voids plus oxides 0.2 maximum volume percent, sulphides Laboratory testing showed that the SRI wheel met all of the design criteria. A detailed description of the development process has been reported previously. 2 TTCI was awarded a United States patent for the SRI wheel steel in 2009, and it is available royalty free for use in North America. 3 Laboratory Testing Tensile tests were conducted at multiple temperatures on samples extracted from the rim of each wheel type. North American freight wheels are used as the brake drum in the braking system, and therefore can be exposed to temperatures greater than 900 F (480 C) while in service. 4 Such temperatures promote wheel shelling by not only degrading the steel s strength, but also by relieving and sometimes reversing the beneficial compressive hoop stresses imparted during manufacturing. 5 Tensile tests conducted on each wheel steel at temperatures between -40 F and 1000 F (-40 C and 540 C) show Type 6 steel has a yield strength 40 ksi to 50 ksi (275 MPa to 350 MPa) greater than AAR Class C steel at all temperatures tested. Both Type 5 and the SRI wheel steel had yield strength approximately 30 ksi (200 MPa) greater than AAR Class C at most temperatures. The other wheel steels had similar yield strength to AAR Class C. Yield strength is an important characteristic in resistance to fatigue damage associated with wheel shelling. Shakedown theory predicts that stronger steel can sustain higher traction loads without accumulating fatigue damage. 6 Based on the yield strength of the wheel steels, Types 5, 6, and SRI would seem to have an advantage in resisting shelling compared to the other wheel types. 2

3 Preliminary results from the durability and service tests of the wheels do not show this advantage yet. Figure 1 shows the relationship between temperature and yield strength for each type of wheel steel. 7 Yield Strength (ksi) Temperature ( C) SRI Class C Figure 1. Summary of the Tensile Properties for High Performance and AAR Class C Wheels (Values for AAR Class C at 800 F and 1000 F (427 C and 538 C)) 7 A microstructure evaluation determined that seven of the eight high performance wheel types were comprised of a pearlitic microstructure (similar to AAR Class C), but Type 6 was comprised of a bainitic microstructure. Previous test results of rail wear rates at FAST have shown that steel with a bainitic microstructure wears more quickly than steel with a pearlitic microstructure in the wheel/rail contact environment. 8 Where applicable, appropriate American Society of Testing Materials (ASTM) procedures were used for the laboratory tests conducted on the high performance wheel steels. 9,10,11,12,13 Table 1 shows the hardness values of the high performance wheel steels and AAR Class C wheel steel. The internal hardness results are from an average of 20 measurement locations along the cross section of the wheel rim. The surface hardness results are an average of three measurements taken along the wheel tread. Standard AAR microcleanliness testing was conducted on six samples from one wheel of each wheel type. Despite differences in microcleanliness between wheel types, all of the wheels tested easily met current requirements for AAR Class C material. At room temperature, all high performance wheels had fracture toughness values above AAR Class C wheel steel. Note the significantly higher fracture toughness of Type 6 (bainitic steel). Saw cuts were made on one wheel of each type to determine the state of residual hoop stress. All wheels were manufactured with compressive residual hoop stress in the rim. Table 1. Hardness, Microcleanliness, and Fracture Toughness Results Wheel Type Average Internal Hardness (Brinell) Average Surface Hardness (Brinell) Percent Sulfides (maximum) Percent Oxides + Voids (maximum) Percent Oxides + Voids (average) Fracture Toughness - 40 F (ksi inch) Fracture Toughness 70 F (ksi inch)

4 SRI AAR Class C AAR Class C Criteria Durability Test N/A >= 320, <= 366 <= 0.75 <= 0.75 <= 0.1 N/A N/A A durability test of the high performance wheels began in 2008, involving six cars in the test train at FAST to monitor the performance of the wheels in a quasi-normal service environment and to identify any potential safety issues. Approximately 48,000 miles (77,000 km) were accumulated on the wheels in the durability test through the end of The FAST train is turned regularly and the direction of travel around the loop (clockwise/counterclockwise) is also varied so that every wheel accumulates mileage in the leading and trailing positions of a truck and on the inner and outer rails. Although the 2.7-mile (4.3 km) long High Tonnage Loop where the test train operates is largely comprised of 5- and 6-degree curves (350 m and 290 m radius), few wheels are removed from the train for tread damage causes. This is most likely due to the careful control of rail friction combined with minimal use of train brakes. All of the wheels in the durability test are nominal 36-inch (914 mm) diameter, the standard size for operations at 32-tonne axle load. Three of the cars involved in the durability test are loaded to 286,000 pounds (32-tonne axle load). Prior to the initiation of the durability test, a drag brake test was conducted at TTC using these three cars with one wheelset of each high performance wheel type and four control wheelsets with AAR Class C wheels. It is important to note that the AAR Class C wheels used in the drag brake and durability test were not new at the start of the test. The cars were pulled on another loop track at 25 mph (40 km/hr) with the brakes applied for a total of more than 250 miles (400 km) traveled. During the drag brake test, wheel temperatures were measured once per lap (every 9 miles (14 km)) using a wayside wheel temperature detector. Previous studies show that when wheels are heated with tread brakes, the wheel tread temperature can be estimated at 150 F (65 C) hotter than the value reported by the wayside detector. 14 Estimated maximum wheel tread temperatures ranged from 500 F to 825 F (260 C to 440 C). Freight car brake systems do not distribute the heat load evenly to all wheels in the car. Thus, each wheel involved in the brake test was exposed to a different temperature history. While not ideal, this is an unavoidable outcome of a drag brake test. Some of the wheels were heated sufficiently during the drag brake test to relieve beneficial residual stresses. Other than heat discoloration, no damage was visually observed on any of the wheelsets involved in the drag brake test. The other three cars used in the durability test are loaded to 315,000 pounds (36-tonne axle load). These cars are equipped with 36-inch (914 mm) diameter high performance wheels and placed in the test train for durability testing. Heavier axle loads affect the fatigue life of a wheel by increasing the stress at the wheel/rail contact patch. Cars at this heavy load are intended to be used with 38-inch (965 mm) diameter wheels to provide a larger wheel/rail contact patch and reduce the contact stress. Thus, placing the 36-inch (914 mm) diameter high performance wheels under cars loaded to 315,000 pounds (36-tonne axle load) provides a higher stress environment for the wheels compared to normal service conditions. Special wheel mounting procedures were used for the wheelsets in the 315,000- pound (36-tonne axle load) cars to fit the high performance wheels on the wheel seat of the AAR Class G or AAR Class M axles necessary to carry these heavy loads. Inspections of the wheels involved in the durability test are typically conducted every 10,000 miles (16,000 km). As part of each inspection, a wheel profile of each wheel in the durability test is hand 4

5 measured at the same circumferential location. This allows the wear rates of the wheels to be calculated. Cars are occasionally cut out of the test train for servicing, and so the mileage of the wheels in the durability test is similar, but not identical. Delays in the manufacturing of the SRI wheels necessitated a later durability test start date for these wheels, and therefore they have accumulated 8,000 fewer miles (13,000 km) than the other wheels. A second drag brake test using the same procedure was conducted for the SRI wheels immediately prior to their entry into the durability test. During each inspection, TTCI personnel visually inspect the entire circumference of all 48 wheels in the durability test. No spalling or evidence of wheel slides has been found. Table 2 is a categorized summary of the wheel tread conditions of all of the wheels in the durability test after 48,000 miles (77,000 km). No conditions were observed on any of the wheels in the durability test that would qualify as condemnable under the AAR Interchange Rules. 15 All wheels categorized as either Shells or RCF Cracks or No Visible Damage. RCF cracks are a potential precursor to wheel shelling. These cracks may either wear off or grow, connect, and turn into shells. Any wheel with at least one shell with a minor axis (smallest dimension) larger than 1/8 inch (3 mm) was placed in the Shells category. All other wheels were placed in the RCF Cracks or No Visible Damage category. Figure 2 shows an example of the wheel tread condition of a wheel in each of these categories. Flange Flange RCF Cracks Figure 2. Example of Wheel Tread Condition Considered RCF Cracks or Nothing Visible (Left) and Shells (Right) Wheel Type Table 2. Durability Test Inspection Results after 48,000 Miles (77,000 km) Wheel Count RCF Cracks or Nothing Visible Wheel Count (Percent) 5 Shells Wheel Count (Percent) (83%) 1 (17%)

6 2 6 6 (100%) 0 (0%) (100%) 0 (0%) (100%) 0 (0%) (75%) 1 (25%) (100%) 0 (0%) (100%) 0 (0%) SRI 4 4 (75%) 0 (0%) AAR Class C 8 4 (50%) 4 (50%) Both of the shelled high performance wheels are installed under 315,000-pound (36-tonne axle load) cars. All of the Class C wheels are installed under 286,000-pound (32-tonne axle load) cars. Only two of the 24 wheels under 315,000-pound (36-tonne axle load) cars are completely free of RCF cracks and shells. Nine of the 24 wheels under 286,000-pound (32-tonne axle load) cars are completely free of RCF cracks and shells. The presence or absence of RCF cracks and shells on the wheels in the 286,000-pound cars is not well correlated with the wheel temperatures from the drag brake test. Figure 3 shows the wear rates of the wheels in the durability test based on hand measured wheel profiles. The wear rates are substantially higher for the wheels in the durability test compared to the wheels in the revenue service test. The AAR Class C wheels were not new at the start of the test, and therefore it is not unexpected that they have more shelling and a reduced wear rate (wheels tend to wear most rapidly at beginning of their life) compared to the other wheel types. Type 6 wheels (bainitic) show a higher tread wear rate than the other pearlitic wheel steels being tested, but show a comparable flange wear rate. The SRI wheels in the durability test have a higher flange wear rate compared to the other wheel types, but a comparable tread wear rate. Results from the durability test should not be considered as relevant as the results from the revenue service test due to the small sample size and unique operating characteristics. 0.5 Average Wear Rate (Inches/100,000 Service Miles) Tread (Flange Height) Flange Width Wheel Type 7 SRI Class C Revenue Service Test Figure 3. Average Wear Rate of Wheels in Durability Test after 48,000 miles (77,000 km) 6

7 From mid-2009 through 2010, a revenue service test of the high performance wheels had accumulated approximately 100,000 miles (160,000 km). The revenue service train used to test the high performance wheels is composed of steel hopper cars owned by the Union Pacific Railroad (UP). The light weight of these cars is in the range of 61,000 pounds to 65,000 pounds (271 kn to 289 kn), and the cars are rated for a gross rail load of 286,000 pounds (32-tonne axle load). Build dates on the cars range from 1979 through Immediately prior to the test, the cars went through a rebuild program consisting of a truck upgrade to AAR M-976 qualified Barber S2-HD-9C split wedge trucks with primary suspension shear pads and D5 spring nest. The cars were equipped with polymer center bowl liners and long travel constant contact side bearings with nominal a 6,000-pound (27 kn) preload. The brake arrangement is a body-mounted rod-through-bolster design with the dead levers connected to the bolsters. Slope sheet empty/load devices provide a 50% reduction in brake cylinder pressure when the cars are empty. High friction composition brake shoes (abbreviated as Cmp ) were installed on all cars equipped with the high performance wheels and 16 control cars with AAR Class C wheels. An additional 18 cars are equipped with AAR Class C wheels and one of two types of tread conditioning shoes (called TC-A and TC-B in this report) intended to reduce the number of wheels removed from service for shelling. This was done to be able to compare the life of high performance wheels not only to that of AAR Class C wheels with composition brake shoes, but also to the wheel life of AAR Class C wheels with tread conditioning brake shoes. Tread conditioning brake shoes were not paired with any of the high performance wheels for two reasons: (1) to maximize the sample size of the high performance wheels paired with composition brake shoes, and (2) to minimize the test duration required to experience a significant number of wheelset removals. Stencils on each test car indicate which shoe type to apply when the shoes are in need of replacement. The UP is in the process of replacing all of the TC-B brake shoes with TC-A brake shoes and updating the stencils on the cars to indicate this change. The remaining cars in the train were rebuilt prior to 2009 and do not have new wheels, and therefore are not considered part of the test. Cars with wheels involved in the test have been stenciled to indicate that the wheels are part of a test and instructing the shop personnel to call a UP helpdesk phone number before removing the wheelsets. The UP helpdesk will provide special wheelset handling instructions in order to save the test wheelsets for inspection and also to avoid sending the high performance wheels to a wheel shop for truing and reapplication under a non-test car. Table 3 contains a count of the wheelsets and brake shoes in the revenue service test. Because not all types of wheelsets were available in multiples of four, some cars were equipped with more than one type of wheel. Wheel types were never mixed within a wheelset. Prior to test initialization, one AAR Class C wheelset was found to have visible, but noncondemnable tread damage. This wheelset is not considered part of the test. Thus, the revenue service test had 82 cars with 328 test wheelsets and one car with three test wheelsets and one wheelset not part of the test. Table 3. Wheelsets and Brake Shoes in the Revenue Service Test Wheel Type Brake Shoe Type Wheelset Count AAR Class C Composition 54 AAR Class C TC-A 40 AAR Class C TC-B 32 Wheel 1 Composition 29 Wheel 2 Composition 28 Wheel 3 Composition 29 Wheel 4 Composition 30 Wheel 5 Composition 25 Wheel 6 Composition 26 Wheel 7 Composition 25 7

8 SRI Wheel Composition 13 Total 331 Visual inspections of the wheels in the revenue service test were conducted after 23,000 miles (37,000 km) and 88,000 miles (142,000 km) of accumulated revenue service. While the cars in the test train have largely stayed together, not all of the cars with test wheels were present at either inspection. Accordingly, not every wheel in the test has been inspected. During these inspections, the majority of the wheel tread surface was viewed, excluding where the rail or the brake shoe blocked access to the tread. The inspectors were specifically looking for RCF cracks, shells/spalls, and any indications of wheel sliding. When the visual inspectors identified shells or spalls on a particular wheel, a nondestructive testing technician documented the condition by applying etchant to look for martensite and by performing a series of surface hardness tests near the affected area. Post inspection review of the notes, photographs, etching results, and hardness values were used to determine whether the damage on a wheel tread was the result of a sliding event (spalling) or fatigue (shelling). A wheel slide event that results in a spall is not reflective of the performance of the wheel. Thus, wheels with spalling damage were excluded from further analysis. Wheels without spalls were placed into one of two possible categories: Shells or RCF Cracks or No Visible Damage. The same criterion for categorizing the wheels was used in both the durability test and the revenue service test. Any wheel with at least one shell with a minor axis (smallest dimension) larger than 1/8 inch (3 mm) was placed in the Shells category. All other wheels without spalls were placed in the RCF Cracks or No Visible Damage category. The reader is again referenced to Figure 2 showing examples of wheel tread condition categories. Table 4 is a categorized summary of the wheel tread conditions in the revenue service test after 88,000 service miles (142,000 km). Figure 4 shows the categorized percentages of unspalled wheels per wheel type. None of the wheels inspected had shells large enough to be deemed condemnable under AAR rules. 15 Wheel Type Table 4. Revenue Service Test Inspection Results after 88,000 Miles (142,000 km) Inspected Wheel Count Spalled Wheel Count Unspalled Wheel Count 8 RCF Cracks or Nothing Visible Wheel Count SRI C, Cmp C, TC-A C, TC-B Shells Wheel Count Data from a wayside wheel profile detector was compared to wheel profile data measured prior to the initiation of the revenue service test in August In this manner, average wear values of the flange width and tread (flange height) were established for each wheel type. Figure 5 shows the average wear rates categorized by wheel type and brake shoe combination. The flange width measurement is reported at a radial location relative to the worn tread surface, and thus, is somewhat dependent on the shape of the wheel profile.

9 The average wear rates of the wheels in the revenue service test are extremely low. Wear rates tend to be highest initially, and then lower as the wheels develop profiles that produce conformal contact with the rail. 16 Though the wear rates for many of the high performance wheels are higher than AAR Class C wheels at this early stage of the test, the wheel with the highest wear rate (Type 1) would still be projected to have a wear life longer than 500,000 miles (800,000 km). One Type 6 wheelset was removed for thin flange at 40,000 miles (64,000 km); this data is included in Figure 5. Percent of Wheels Inspected 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Shells RCF Cracks or Nothing Visible Wheel Type Figure 4. Tread Condition of Wheels in Revenue Service Test after 88,000 Miles (142,000 km) 7 SRI C, Cmp C, TC A C, TC B Average Wear Rate (Inches/100,000 Service Miles) Tread (Flange Height) Flange Width SRI Wheel Type C, Cmp C, TC A C, TC B Figure 5. Average Wear Rate of Wheels in Revenue Service Test after 91,000 Miles (146,000 km) Wayside wheel impact load detector (WILD) data from the test cars shows minimal problems with impact loads. Eight wheels from the revenue service test with known spalls have WILD readings in excess of 70,000 pounds (310 kn). Wheels with readings greater than 90,000 pounds (400 kn) are condemnable.[aar, 2011] If these or other wheelsets from the test train are removed for cause, they 9

10 will be set aside for inspection and to ensure that non-aar Class C wheels do not become mixed with the general wheel population. CONCLUSIONS An evaluation of eight types of high performance wheels is underway. Laboratory testing showed that three of the wheel types (Type 5, Type 6, and SRI) hold an advantage in yield strength over AAR Class C wheels and the other high performance wheels in the test. Higher yield strength should result in improved resistance to wheel shelling. AAR Class C wheels and all of the high performance wheels have a pearlitic microstructure with the exception of wheel Type 6, which has a bainitic microstructure. A durability test using a small sample size of wheels at TTC provides information about the wheels performance in an environment with accelerated wear and fatigue due to nearly continuous curve negotiation and high axle loads. Though few of the high performance wheels in this test have developed large shells after 48,000 miles (77,000 km), the majority show some degree of RCF cracks on the tread surface. The most important aspect of the high performance wheel evaluation is the revenue service test involving a large sample size of each wheel type. An inspection of the majority of the wheels involved in this test after 88,000 miles (142,000 km) revealed that nearly all the wheels are in good condition without large shells. At this early stage in the test, the high performance wheels show similar performance compared to AAR Class C wheels in terms of RCF and shelling. The average wheel wear rate is extremely low for all types of wheels in the revenue service test. AAR Class C wheels with type TC-B tread conditioning brake shoes are showing more RCF cracks and shelling compared to all other wheel and brake shoe combinations. ACKNOWLEDGEMENTS TTCI thanks the UP for supplying and operating the revenue service test train and the wheel manufacturers for donating the wheels involved in the test. TTCI also thanks MWL Brasil for forging the SRI wheels used in the testing. REFERENCES 1. Railinc Car Repair Billing Database, Cary, North Carolina. 2. Robles Hernández, F.C. et al. September Properties and Microstructure of High Performance Wheels. 8 th International Conference on Contact Mechanics and Wear of Wheel/Rail Systems. Firenze, Italy. 3. Robles Hernández, F.C. and D.H. Stone Railroad Wheel Steels Having Improved Resistance to Rolling Contact Fatigue. US Patent 7,599,999, July 14, United States Patent Office, Washington, D.C., USA. 4. Cummings, S Service Wheel Temperatures and Car Condition in Relation to Thermal Mechanical Shelling. RTDF , Proceedings of 2008 Fall Conference of the ASME Rail Transportation Division, Chicago. 5. Stone, D. and S. Cummings Effect of Residual Stress, Temperature and Adhesion on Wheel Surface Fatigue Cracking. RTDF , Proceedings of 2008 Fall Conference of the ASME Rail Transportation Division, Chicago. 6. Bower, A.F. and K.L. Johnson Plastic Flow and Shakedown of the Rail Surface in Repeated Wheel-rail Contact, Wear, Vol. 144, pp Lonsdale, C. and S. Dedmon. November Fatigue Testing of Microalloyed AAR Class C Wheel Steel. IMECE , Proceedings 2006 ASME International Mechanical Engineering Congress and Exposition, Chicago. 10

11 8. Sawley, K. and R. Jimenez Track Wear Tests of Bainitic and Pearlitic Rails: Final Results. Technology Digest TD Association of American Railroads, Transportation Technology Center, Inc., Pueblo, CO. 9. ASTM Standard E8 / E8M 09, Standard Test Methods for Tension Testing of Metallic Materials, ASTM International, West Conshohocken, PA. 10. ASTM Standard E21-05, Standard Test Methods for Elevated Temperature Tension Tests of Metallic Materials, ASTM International, West Conshohocken, PA. 11. ASTM Standard E45-05e3, Standard Test Methods for Determining the Inclusion Content of Steel, ASTM International, West Conshohocken, PA. 12. ASTM Standard E399 08, Standard Test Method for Linear-Elastic Plane-Strain Fracture Toughness KIc of Metallic Materials, ASTM International, West Conshohocken, PA. 13. ASTM Standard E (2008), Standard Practice for Determining the Inclusion or Second- Phase Constituent Content of Metals by Automatic Image Analysis, ASTM International, West Conshohocken, PA. 14. Cummings, S., H. Tournay, and K. Gonzales. March Wayside Wheel Temperature Detector Test, Technology Digest TD , Association of American Railroads, Transportation Technology Center, Inc., Pueblo, CO. 15. Association of American Railroads Field Manual of the A.A.R. Interchange Rules, Rule 41, Washington, D.C. 16. Wu, H., B. Madrill, and S. Kalay New Wheel Profile Design and Preliminary Service Tests Results. Technology Digest TD Association of American Railroads, Transportation Technology Center, Inc., Pueblo, CO. 11