GIFT OF ENGINEERING LIBRARY OF WILLIAM B. STOREY A GRADUATE OF THE COLLEGE OF MECHANICS CLASS OF 1881 PRESENTED TO THE UNIVERSITY 1922 VANADIUM RAILS A REPORT OF TESTS OF VANADIUM STEEL RAILS WITH COMPARATIVE TESTS WITH SIMPLE CARBON STEEL RAILS OF THE SAME SECTION AND MANUFACTURE 1914 AMERICAN VANADIUM COMPANY VANADIUM BUILDING, PITTSBURGH, PA. VANADIUM RAILS Convinced from results along other lines that the use of vanadium in rail steel will greatly reduce rail failures and at the same time greatly increase the wearing quality of the rails, the American Vanadium Company has had several heats of basic open hearth vanadium steel rolled into rails for test purposes. While no question exists as to the beneficial effects of vanadium in steel, the price until the past year has been too high to make it commercially practicable to use it in RAILS. It is universally recognized that the rail situation is one of the most serious confronting the railroad officials today. Considerations, primarily of safety, and also of economy, urgently demand a decided improvement in the quality of rail steel. The tonnage of rails for replacement purposes has greatly increased. Published estimates from the statistics available place this increase at 50 per cent, over that of a few years ago, or an increase from 1,000,000 tons annually to fully 1,500,000 tons. The railroads and rail makers have tried to meet the situation by increasing the weight of the rail section. At the same time harder, higher carbon steels have been used in the effort to obtain increased strength and wear- ing qualities; until now rails are rolled from practically spring steel, or low grade tool steel. The result, although attended with some degree of success, has failed to lessen rail breakage. On the con- trary, the use of high carbon steel has introduced new causes of failure; namely, internal fissures, commonly known as "silvery oval spots". It is evident that a steel of greater strength, tough- ness and better wearing qualities, together with greater 501801 VANADIUM RAILS homogeneity and freedom from segregation, than simple carbon steel, is required to meet the situation. Through the addition of a small percentage of vanadium to simple carbon standard rail steel, rails of exceptionally great strength, toughness and wearing qualites are obtained. This is shown by the tests given in the following pages in which comparison is also made with corresponding tests from simple carbon steel rails of the same section rolled at the same time by the same mill. Briefly summarized, the salient facts as shown by analysis of the results of the tests are as follows: SUMMARY MANUFACTURE: The manufacture of rails from vanadium steel does not require any change in the usual rail mill practice. The steel rolls clean and shows no tendency to crack or tear. No change is necessary in the gauges ; as vanadium steel takes the same standard gauge as simple carbon steel. Nothing developed in the melt- ing, casting, heating or rolling of these heats of vanadium steel to indicate that the mill output will in any way be reduced. The percentage of yield of rails from the ingots is equal to or even greater than for simple carbon steel rails; and the evidence is that there will be less scrap from cracking or tearing in the rolls, and also that the percentage of second quality rails will be less. DROP TESTS: The vanadium steel rails met all the requirements specified for ductility and deflection and are stiff er or more resilient than the 100-lb. carbon steel rails of the same section, with which they were compared. The vanadium steel rails deflected about the same amount from an 18-ft. blow as the carbon steel rails did from a 15-ft. blow. TENSILE TESTS: The vanadium steel rails show an elastic limit, or useful strength averaging, over 40% higher than for the simple carbon steel rails; though the latter had 50% higher carbon content. The ratio of elastic limit to tensile or breaking strength is 70% or more for the vanadium steel rails as compared with about VANADIUM RAILS 57% for the carbon steel rails. The ductility or tough- ness of the vanadium steel rails also shows higher than for the carbon steel rails. These tests are a direct measurement of the superiority of the vanadium over the carbon steel rails. ALTERNATING IMPACT TESTS: These tests prove that vanadium steel of the high elastic limit shown by these rails has great endurance under repeated stresses and is free from brittleness. BEND TESTS: Although the vanadium steel rails are very much harder than the simple carbon steel rails with which they are compared, the bend tests are equally as good. HARDNESS TESTS: The vanadium steel rails show great uniformity of hardness throughout the entire section; and are from 10% to 35% harder than the carbon steel rails of 50% higher percentage of carbon. This superior hardness, combined with the high elastic limit and the great toughness of the vanadium steel rails indicate a great increase in wear resisting qualities. WEAR TESTS: Comparative tests to determine resistance to wear show a very great increase in wear resisting qualities for the vanadium over the simple carbon steel rails. CHEMICAL UNIFORMITY (SEGREGATION) : The vana- dium steel rails show no segregation, the variations in analysis noted being within the allowable limits of analytical error. The standard high carbon steel rails usually show segregation and in the case of the "A" rails the segregation is frequently very marked. HOMOGENEITY AND DENSITY: Sections from the vanadium steel rails etched with boiling dilute sulphuric acid show a remarkably dense, uniform structure, free from piping, slag and other defects. The contrast in this respect with the etched sections of the carbon steel rails, as shown in the accompanying reproductions of photographs, is very striking. The carbon rails were selected at random by the makers. VANADIUM RAILS i VANADIUM STEEL HEAT 26813, RAIL "A" Shows only a slight amount of piping at the top of the web and in the web. The appearance of the center of the head of this etched section would indicate that the center of the top of the ingot was a little soft or green when bloomed. A few of the ingots from this heat were heavily cropped at the bloom shears on this account. VANADIUM RAILS VANADIUM STEEL HEAT 26813, RAIL "B" Shows a very solid, uniform, dense structure. VANADIUM RAILS VANADIUM STEEL HEAT 27989, RAIL "A" Shows a perfectly uniform, dense structure with no evidence of residual piping. VANADIUM RAILS . VANADIUM STEEL HEAT 27993, RAIL "A" Shows a uniform, dense structure, with only the very slightest indication of residual piping 9 VANADIUM RAILS CARBON STEEL, RAIL "A" Is typical of most "A" rail structures and is really better than many; because more than the usual discard was made from the top of the ingots in rolling this lot of rails 10 VANADIUM RAILS CARBON STEEL, RAIL "B" Shows residual piping, quite marked on the left side of the etching 11 VANADIUM RAILS Three heats of vanadium steel were made to the following chemical specifications : Heat 26813 Heat 27989 Heat 27993 Carbon Manganese Silicon Phosphorus Sulphur Vanadium . 45 to . 65% 1.10 to 1.40% under . 20% under .05% under .05% 4 Ibs. to ton .40 to .52% 1.00 to 1.30% under .20% under .05% under .05% 4 Ibs. to ton .60 to .75% .75 to 1.00% under .20% under .05% under .05% 4 Ibs. to ton In the first two heats, the manganese specified is higher than usual; as previous investigations have shown that with manganese somewhat higher than usual the effect of the vanadium on the physical properties of the steel is still further increased. The third heat, however, conforms to the usual specification for rail steel. The actual chemical compositions of these three heats are: Heat 26813 Heat 27989 Heat 27993 Carbon .... .550% .510% .558% Manganese Silicon ... . . 1.510% .170% 1.110% .120% .780% .158% Phosphorus . Sulphur Vanadium Actual per cent. Vanadium added .015% .019% -148% 168% .010% .029% . 146% 160% .017% .025% .156% .177% Heat 26813 The manganese is higher than called for, due to the percentage of loss in manganese addition not being nearly as great as allowed for in usual practice. Heat 27993 The percentage of carbon is about 5 points below the limit called for, due to various mill delays. PRODUCTION The production percentage, or yield of rails per ton of ingots, is higher than usual; although from the evidence of the tests which follow it seems as though it might be possible to obtain even higher yields. Date 4-22-14 7-27-14 7-27-14 Heat..... Ingots, weight 26813 121,000 Ibs. 27989 104,400 Ibs. 27993 108,000 Ibs. Rails, weight 89,500 Ibs. 80,200 Ibs. 82, 000 Ibs. Rails, number Rails, scrap Per cent, yield 77-lst, 5-2nd none 74.0 73-lst, l-2nd none 76.9 73-lst, 2-2nd none 75.9 12 VANADIUM RAILS All three vanadium steel heats rolled perfectly. The blooms were clean and free from seams and cracks, and no scrap rails were produced. The standard gauges were used; showing that the shrink- age of the steel is the same as for standard carbon steel, and that no change in this respect is necessary. Nothing developed in the heating and rolling to indicate that the mill output would be in any way reduced by the use of vanadium steel. The top portion of some of the ingots from the first heat, 26813, were a little soft or green when bloomed, and were cropped heavily on this account. DROP TESTS When the first heat, 26813, was made, it was arranged to make drop tests on crop ends from both the "A" and "B" rail of three ingots, representing the beginning, middle and end of the pouring. Through misunderstanding, this was not carried out on the second heat, 27989; and only three drop tests were made from this heat, one being from a "C" rail crop; as the top blooms from the ingots were rolled into 3-rail lengths. The arrangement for two drop tests from the beginning, middle and end of the third heat, 27993, was carried out, excepting that the tests were made on "C" rail crops instead of "B" rail; on account of the top blooms all having been rolled into 3-rail lengths. The usual requirements for drop test were followed, excepting that the height of the drop was increased from 15 ft. to 18 ft. for the vanadium rails. Two vanadium rails from the first heat, however, were tested with the height of the drop at 15 ft.; in order to get a direct comparison with the carbon steel rails. One vanadium rail was tested with the flange up. All the others were tested in the usual manner with the head up. Six one-inch spaces were laid off on the bottom of the flange; in order to determine the ductility or stretch after each blow of the drop, the requirements being 5% or 5-100-inch stretch in two adjacent inch spaces. As will be seen, the vanadium steel rails met all the require- ments specified for ductility and deflection. At the same time, they show up stiff er under the drop test than the carbon steel rails. The chemical specifications to which the carbon steel rails were made are: Carbon 62 to . 75% Manganese 60 to . 90% Silicon Under . 20% Phosphorus Under . 04% 13 VANADIUM RAILS DROP TESTS VANADIUM RAILS Heat No. 1 1 :! l| fflQ P J3 5 "o i Deflection in Inches Elongation in Hundredths of an Inch Fracture Per Inch Total lin. 2 in. 3 in. 4 in. 5 in. 6 in. 26813 16 A 15 i 2 0.7 1.4 4 5 3 6 5 8 4 7 4 7 2 5 22 38 Clear 3 3.2 16 19 17 17 14 1 0.9 3 5 4 4 3 3 22 2 1.6 6 9 8 7 4 4 38 26813 16 B 15 3 2.3 8 12 12 10 6 5 53 Clear 4 2.9 13 16 14 10 6 5 64 5 3.7 17 23 20 14 8 6 88 6 4.6 18 19 13 8 6 26813 1 A 18 Flange Up 1 2 1.0 1.7 5 6 6 7 7 9 6 8 5 GO GO 32 Clear 1 1.1 4 6 6 4 3 3 26 2 1.8 8 9 9 6 4 4 40 26813 1 B 18 3 2.7 12 14 11 8 4 4 53 Clear 4 3.6 15 16 14 10 7 5 67 5 3.7 16 14 10 6 4 26813 9 A 18 1 1.0 4 4 5 5 4 2 24 2 1.8 4 4 5 5 3 *Web Piped 1 1.0 3 4 5 5 5 4 26 26813 9 B 18 2 1.8 5 6 8 8 8 5 40 Clear 3 2.6 6 8 10 7 1 1.0 3 4 5 5 4 2 23 27989 1 A 18 2 1.9 4 5 7 10 10 9 45 Clear 3 5 8 11 11 11 9 55 1 1.1 3 4 5 5 5 5 27 27989 6 C 18 2 2.1 5 5 7 10 10 9 46 Clear 3 7 8 12 13 12 10 62 1 1.1 2 3 5 5 5 5 25 27989 14 A 18 2 2.0 5 7 10 10 8 6 46 Clear 3 7 11 16 15 13 10 72 1 1.0 3 3 4 4 4 4 22 27993 1 A 18 2 1.9 4 6 7 8 8 7 40 Clear 3 7 7 9 9 10 8 50 27993 1 C 18 1 2 1.1 3 4 4 5 5 7 5 9 5 8 4 8 26 41 Clear 27993 6 A 18 1 2 1.1 2 3 3 4 4 5 5 7 5 7 5 7 24 33 Clear 14 VANADIUM RAILS DROP TESTS Continued VANADIUM RAILS 6 "o? I C* Elongation in Hundredths of an Inch 1 - J3 . S ti-5 Per Inch $ 1 " P o 6 nS lin. 2 in. 3 in. 4 in. Sin. 6 in. Total Fracture Z 1 1.1 3 4 5 5 4 3 24 27993 6 C 18 2 2.0 6 8 9 9 8 6 46 Clear 3 7 8 11 10 10 8 54 1 1.2 4 5 5 5 5 4 28 27993 14 A 18 2 3 2.2 3.2 6 8 7 9 9 11 8 10 8 10 8 12 46 60 Clear 4 8 9 11 12 13 14 67 1 1.1 3 4 4 5 5 4 25 27993 14 C 18 2 2.1 6 8 10 10 8 6 48 Clear 3 8 11 15 15 11 8 68 CARBON RAILS a ^ S Elongation in Hundredths of an inch fa | 6-2 1| Per Inch Fracture I KQ |3 I s lin. 2 in. 3 in. 4 in. Sin. 6 in. Total 1 1.0 3 4 4 3 3 3 20 i 15 2 1.9 5 6 7 7 7 5 37 Clear 3 Nicked 1 1.0 4 4 4 3 3 2 20 2 15 2 1.8 5 6 7 7 6 5 36 Clear 3 Nicked 3 15 1 1.1 4 4 5 5 4 3 25 Clear 2 Nicked *A very small indication of piping which showed on one side only of the wedge- shaped piece broken out when the rail broke under the drop. In addition to drop tests, one full length "A" rail from each of the three vanadium heats was broken into ten pieces and the fractures carefully examined for evidence of piping. Every fracture was found to be free from any evidence of piping. PHYSICAL TESTS Tensile tests, alternating impact, and bend tests were made from rail crops from each heat, and also wear and hardness tests. The crops from the "A" rails were taken from the top end of the rail. 15 VANADIUM RAILS Tensile tests were also made from the middle section of each of the vanadium steel rails which were broken under the gag press. Corre- sponding tests for comparison were made from an "A" and "B" 100-lb. section carbon steel rail. These tests are shown in the fol- lowing table; and the locations of the tests are indicated on the accompanying illustrations. Due to the more rapid cooling of the rail crops, the tests from these show a little higher in elastic limit than the tests from the mid- section of the three rails broken in the gag press ; as these rails cooled much more slowly on the hot bed. TENSILE TESTS FIG. 1 : Location of Tensile Tests Where 5 Tests Were Made. FIG. 2 : Location of Tensile Tests Where 8 Tests Were Made. Fig. 1 Fig. 2 VANADIUM RAILS Heat No. Rail Test No. Elastic Limit lbs.persq.in. Tensile Strength Ibs. per sq. in. Elongation in 2 in. per cent. Reduction of Area per cent. 1 130000 140000 12.0 22.0 A 2 102500 132500 5.0 5.5 26813 Crop 3 100000 121000 Broke in Fillet 4 107500 130000 Broke in Fillet 5 105000 123000 Broke in Fillet 1 92600 130000 12.5 21.5 A 2 112000 135000 13.0 24.0 Rail 3 90000 125000 14.0 24.5 (Mid 4 98000 130000 14.5 25.5 26813 Section) 5 95000 129000 15.0 30.0 6 105000 137000 13.0 26.5 7 92000 128000 13.0 20.5 8 99000 134000 15.0 28.0 1 92000 134000 12.0 24.0 2 95000 133000 12.0 24.0 B 3 97500 125500 6.0 10.0 26813 Crop 4 100000 126500 Broke in Fillet 5 102500 140000 11.0 20.5 16 VANADIUM RAILS TENSILE TESTS Continued. Heat No. Rail Test No. ElasticLimit Ibs.persq.in. Tensile Strength Ibs. per sq. in. Elongation in 2 in. per cent. Reduction of Area per cent. B 6 105000 145000 10.5 25.5 26813 Crop 7 110000 147500 9.0 18.5 8 110000 147500 10.5 20.5 1 100000 130500 13.0 27.5 2 100000 133000 12.0 25.5 E 3 102500 135000 11.0 20.5 26813 Crop 4 105500 142000 11.5 22.0 5 105000 140000 12.0 25.0 6 110000 140000 12.0 25.0 7 102500 135000 12.0 23.5 8 107000 140000 12.0 25.5 1 92000 127000 13.0 24.5 2 90000 127000 12.0 26.5 A 3 98000 130500 12.0 21.5 27989 Crop 4 97500 129500 13.0 27.5 5 95000 126000 14.0 28.0 6 102000 132500 13.0 20.5 7 88000 122000 12.0 23.5 8 92500 125000 13.0 20.5 1 92000 130000 14.5 26.5 A 2 90000 125000 15.5 30.0 Rail 3 97250 127000 5.0 8.0 JMid 4 90000 138000 9.0 10.0 27989 Section) 5 92500 130000 14.5 25.0 6 90000 126000 15.0 30.5 7 85000 124800 14.0 26.5 8 95000 127500 14.5 29.0 1 90500 123000 11.5 16.0 2 85000 120000 11.5 18.0 3 90500 117000 11.0 16.0 27993 A 4 96000 121000 13.0 20.5 Crop 5 82500 118000 13.0 26.5 6 90000 125000 11.0 17.0 7 85000 122000 10.0 15.5 8 85000 125000 10.0 17.0 1 80000 118000 13.0 22.0 2 90000 126000 12.5 24.5 A 3 87500 125000 11.5 17.0 - Rail 4 80000 123000 14.0 23.0 27993 (Mid 5 85000 125500 12.0 17.5 Section) 6 85000 125000 13.5 24.5 7 80500 122500 13.5 20.5 8 80000 123000 14.0 27.0 17 VANADIUM RAILS TENSILE TESTS Continued. CARBON RAILS Heat No. Rail Test No. ElasticLimit lbs.persq.in. Tensile Strength Ibs. per sq. in. Elongation in 2 in. per cent. Reduction of Area per cent. 1 63000 112500 13.5 20.5 2 66000 114000 13.0 18.5 3 65000 114000 9.0 11.0 4 66000 114500 12.5 20.0 A 5 65000 115000 12.5 17.0 6 65500 117000 11.5 17.0 7 65000 115000 12.5 16.0 8 70000 120000 12.5 18.5 1 60000 123000 10.5 14.5 2 46000 119000 10.0 17.0 3 65000 126000 9.0 11.5 4 72250 125000 9.0 13.5 B 5 75000 125500 10.0 16.0 6 70000 121000 10.5 15.0 7 57000 121000 8.5 13.5 8 80000 124500 10.5 16.0 These tests show a decided increase in elastic limit, or useful strength, in favor of the lower carbon vanadium steel, without sacrifice of ductility. BEND AND ALTERNATING IMPACT TESTS The bend tests were made on rectangular pieces about 8 inches long. The load was applied 6 inches from the fixed end of the test piece. The radius of the jaws holding the bend specimen was not over y% inch, and the edges of the specimens were not rounded. FIG. 3: Location of Bend Tests. FIG. 4: Location of Alter- nating Impact Tests. Fig. 3 Fig. 4 18 VANADIUM RAILS The alternating impact tests are made on bars turned to % inch diameter. The bar is held firmly in a vise, and the upper end moved backwards and forwards by means of a slotted arm, through a total distance of % inch at the rate of 600 movements per minute. The distance from the vise to the slotted arm is 4 inches. Each move- ment is accompanied by a blow on the bar by the slotted arm. BEND TESTS Heat No. Rail Test No. Size Degrees Test No. Alt. Impacts 1 2.125 x .375 50 1 1464 26813 A 3 2.0 x .496 31 2 520 Vanadium Crop 4 2.4 x .455 42 3 1304 5 2.4 x .455 49 5 670 6 1600 1 2.0 x .300 62 1 1444 2 2.4 x .287 21 2 1110 26813 B 3 2.1 x .533 29 3 1136 Vanadium Crop 4 2.3 x .503 35 4 1268 5 2.3 x .503 40 5 1454 6 1540 7 1326 8 1580 1 2.05 x .30 62 1 960 2 2.5 x .35 26 2 790 27989 A 3 2.05 x .55 55 3 964 Vanadium Crop 4 2.2 x .50 51 4 880 5 2.45 x .50 39 5 1210 1 1.8 x .20 61 1 910 2 2.5 x .36 55 2 530 27993 A 3 2.0 x .55 61 3 800 Vanadium Crop 4 2.5 x .50 48 4 784 5 2.2 x .50 52 5 1110 1 1.9 x .30 82 1 1441 2 2.5 x .32 39 2 1250 Carbon A 3 2.05 x .50 57 3 1716 4 2.3 x .50 44 4 1820 5 2.45 x .50 58 5 1300 1 2.0 x .365 47 1 1208 2 2.0 x .275 58 2 980 Carbon B 3 2.1 x .506 47 3 1670 4 2.3 x .502 41 4 1476 5 2.3 x .502 40 4 1230 5 1604 6 1420 8 1440 ALTERNATING IMPACT TESTS 19 VANADIUM RAILS HARDNESS TESTS Hardness tests were made by the Brinell method. This method of determining hardness consists of measuring the impression made by a standard steel ball under a standard load. The tests were made on sections from the same rail crops from which the other tests were made. The locations of the tests are shown in the accompany- ing illustration; and the results are given in the following table. LOCATION OF HARDNESS TESTS 20 VANADIUM RAILS The sections from heat 26813 showed an average hardness of about 340; heat 27989 about 302; heat 27993 about 293; carbon rail "A" about 248, and carbon rail "B" about 269. The vanadium steel rails, although lower in carbon, have greater hardness; and hence can confidently be expected to give correspondingly increased resistance to wear in the track. HARDNESS TESTS VANADIUM RAILS CARBON RAILS Serial No. Rail 26813-A Rail 26813-B Rail 27989-A Rail 27993-A Rail A Rail B 1 302 340 286 286 241 269 2 340 340 302 286 235 269 3 340 340 302 286 248 262 4 340 340 293 286 241 269 5 311 340 286 286 228 255 6 340 340 302 293 248 269 7 340 340 307 302 255 269 8 340 340 307 302 241 269 9 340 340 307 302 241 262 10 340 340 302 293 241 269 11 340 340 302 293 241 269 12 340 340 307 302 262 269 13 364 340 311 302 255 293 14 340 340 307 302 255 269 15 340 340 307 269 217 269 16 387 340 302 293 241 302 17 387 340 293 293 248 302 18 364 340 302 286 248 302 19 364 340 302 293 248 302 20 364 332 293 293 248 302 21 364 332 293 302 248 302 22 340 332 311 311 269 269 23 340 340 311 311 269 269 24 340 340 302 307 269 286 25 340 332 311 286 255 269 26 340 340 311 293 255 269 27 321 340 311 302 262 269 28 321 332 302 311 269 269 29 321 332 311 311 255 269 30 340 332 311 307 248 269 31 340 332 302 307 241 269 32 340 332 311 302 228 269 33 332 340 311 302 241 269 34 332 340 307 302 269 269 35 332 340 293 302 255 269 21 VANADIUM RAILS WEAR TESTS This test is made by rotating a piece 1 in. long by 1 in. diameter between three manganese steel rollers of 3 in. diameter. The two bottom rollers are driven by gears with a different number of teeth; which gives the rollers different speeds and causes the test piece to slip as well as rotate, imitating the action of a car wheel on the rail. The tests were all taken from the head as shown in the illustration. A direct load of 110 Ibs. is applied to the test piece by loading the top roller. In previous tests, a load of 220 Ibs. was used. Owing to the great tendency of soft rails like carbon rail "A" to flow and form a fin or bead which gave trouble, the weight was reduced to 110 Ibs. It was found that the abrasion of the test piece was better with this weight than with the heavier load. The test pieces were weighed before and after test. The loss in weight in millegrams was divided by the original weight of the test piece; in order to Location of Wear Tests obtain comparative figures and allow for variations in weight of test pieces. The tests which follow were all run 50,000 revolutions: Heat Rail Milligrams Loss Divided by Weight of Test Relative Wear 26813 26813 27989 27993 Carbon Carbon A-Crop B-Crop A-Crop A-Crop A B 16.3 13.1 12.8 12.2 28.8 21.1 54 45 44 42 100 73 The relative wear of the vanadium steel rails is practically one- half that of the carbon rails. CHEMICAL ANALYSES OF RAIL SECTIONS FREEDOM FROM SEGREGATION Chemical analyses were made of rails from each of the vanadium heats; to determine whether there was any tendency to segregation, and how the vanadium steel compared in this respect with the carbon steel. For this purpose, drillings were taken from two locations in the head of the rail. One sample was taken from the top corner of the VANADIUM RAILS head, corresponding to the outer portion of the ingot; and the other sample was taken from the junction of the head with the web, corre- sponding to the axial center of the ingot where any segregation present would be most certain to be found. There is no segregation in the case of the vanadium steel rails, the results all being within the limit of analytical error. In the case of the vanadium rails, all the samples were taken from rail crops. CHEMICAL ANALYSES Heat No. Rail Location of Sample Carbon Per Cent. Manganese Per Cent. Phosphorus Per Cent. Sulphur Per Cent. Vanadium Per Cent. 268 13 A Corner of Head Top of Web .55 .56 1.49 .52 .015 .015 .023 .022 .147 .148 26813 B Corner of Head Top of Web .54 .57 .46 .49 .016 .017 .022 .021 .144 .147 27989 A Corner of Head Top of Web .52 .52 .11 .11 .013 .013 .022 .024 .150 .150 27993 A Corner of Head Top of Web .56 .57 .73 .74 .019 .019 .023 .024 .158 .159 Carbon A Corner of Head Top of Web .72 .79 .67 .68 .035 .047 .028 .038 Carbon B Corner of Head Top of Web .74 .81 .68 .69 .011 .012 .035 .041 HOMOGENEITY AND DENSITY (ETCHED SECTIONS) Sections from the rails tested were polished and etched in boiling dilute sulphuric acid and photographed. The time required for etching the vanadium steel rail sections was several times longer than for the carbon rails. The average time required was about 30 minutes, as against 5 minutes for the "A" carbon steel rail and about 15 minutes for the "B" carbon steel rail. The difference in length of time of etching is also a very good indication of the comparative resistance to wear. CONCLUSIONS From the results of the tests made on these three heats, the American Vanadium Company recommend for vanadium steel rails the chemical specification given on the following page. 23 VANADIUM RAILS This specification will give rails with 30% to 50% higher elastic limit, or useful strength, combined with greater toughness and hard- ness than simple carbon steel rails with .62% to .75% carbon content. The vanadium steel rails will show even greater superiority in comparison with lower carbon steel rails of .45% to .60% carbon. Carbon 45 to . 60% Manganese 1 . 00 to 1 . 25% Silicon over . 10% Phosphorus not over . 05% Sulphur not over . 05% Vanadium 4 Ibs. added per gross ton The relatively low percentage of carbon recommended, together with the great freedom from segregation of vanadium steel, should result in the practical elimination of the danger of failure from internal fissures, silvery oval spots. 24 UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. APR 1 2 2003 LD 21-100m-ll,'49(B7146sl6)476 50 ( H UNIVERSITY OF CALIFORNIA LIBRARY