« PreviousContinue »
REPORT OF COMMITTEI IV-ON RAIL
G. J. Ray, Chairman;
J. M. R. FAIRBAIRN, Vice-Chairman;
To the American Railway Engineering Association:
Your Committee on Rail respectfully submits its report to the Twentythird Annual Convention:
(1) Revision of Manual The Committee has been giving consideration to the revision of the specifications for steel rails. It is not ready to report on the subject at this writing but expects to have something to submit on the subject before the time of the convention.
(2) Mill Practice During the past year the Committee has been devoting considerable attention to the details of the rail manufacturing practices of the several rail mills of the country with reference particularly to the influence they have on the properties of the finished rail. Probably the most important item of manufacture that affects the quality of the finished rail is the condition of the steel as it is poured into the molds. A well-made steel thoroughly deoxidized with silicon or other deoxidizer, sets quietly in the molds with a fat top on the ingot. A steel not fully deoxidized, however, effervesces in the mold and sets with a "horney” top on the ingot. The upper third of an ingot cast with effervescing steel is spongy inside with numerous small holes but with only a small central "pipe." Such steel also shows considerable interior segregation of carbon, phosphorus, and sulphur, resulting in a brittleness in the interior of the head of the upper rail of the ingot. The quiet setting steel is free from interior sponginess, the segregation is much lower and the interior steel of the upper third of the ingot is denser and more ductile, but the ingot has a larger central pipe, which shows as a lamination in the web of the rail lower down from the top of the ingot than in rail made from effervescing steel.
The work by Wickhorst in 1912 (see Proceedings American Railway Engineering Association, Vol. 14, 1913, p. 507), showed that about .25 per cent. or more of silicon is required to obtain its full deoxidizing effect and it would seem that specifications should require a minimum of .20 per cent. silicon in rail steel. In this case the upper end of the A rail, with the usual discard of 10 or 12 per cent., may show a small vertical lamination in the web, but the metal in the head will be free from the sponginess and excessive segregation present in rails made with effervescing steel.
(3) Intensity of Pressure The Committee has done little work on this subject recently and indeed is puzzled as to suitable procedure to develop exact information concerning it. Several years ago the Committee submitted the results of careful tests of the distortion of small tapered holes drilled into the side of the rail head at various depths from the top surface. It has been suggested that this work might be repeated, but plugging the holes before applying rolling loads to the rail, the thought being that the flow of the metal may have been influenced by the presence of the open spaces. It is rather uncertain, however, that further conclusions could be drawn from a repetition of the work. It may be that additional experimentation should be carried on by the laboratory method, but the cost of the necessary machinery would be considerable and the Committee has not thought it well to bring up at this time the question of financing the work.
(4) Rail Failures The report on the rail failure statistics for the period ending October 31, 1920, is submitted as Appendix A. The average failures per 100 track miles for all the rails reported on are given below: lear
Years' Service Rolled
224.1 2778 1910
152.7 198.5 1911
77.0 104.4 133.3 176.3 1912
28.9 32.1 49.3 78.9 107.1 1913
25.8 44.8 69.5 91.9 1914
1.2 8.2 19.8 32.9 50.9 74.0 1915
0.7 8.9 19.0 34.2 53.0 82.4 1916
47.7 70.6 1917
5.3 21.6 38.9 66.0 1918
1.6 8.9 27.6 1919
2.0 14.8 1920
3.9 It will be noted that the failures had been showing a gratifying decrease until the World War came on, even though the conditions of service as to wheel loads had been growing more severe. The "war-time" rollings, however, and particularly the rails rolled in 1917, are not showing up so well. Probably the rail performance record of the next few years will show badly, but after that there is ground for hope that the improvement trend will be again taken up.
(5) Methods of Inspection The Committee has been giving attention to the methods of inspection of steel rails as used by the different railroads and in Appendix B presents abstracts of the replies to a questionnaire, a discussion of the methods used for the inspection of steel rails and a schedule of Recommended Practice for Inspection of Steel Rails. The Committee recommends that this latter be approved by the Association for inclusion in the Manual.
In Appendix C is submitted a paper by M. H. Wickhorst, on “Formula for Elongation of Rails in the Drop Test,” in which is given a formula that has been worked out by which can be calculated the elongation in the drop test for any rail section under given conditions of height of drop and carbon content of the steel, or by which may be calculated the height of drop necessary to produce a given elongation on the first blow for any given rail section and carbon content.
(6) Joint Bar Design The Committee has nothing to report on this subject this year but expects to give the matter active attention in the future.
(7) Joint Bar Material The Committee has under consideration tentative specifications for quenched carbon steel track bolts and for quenched alloy steel track bolts. It is expected that these specifications will be put in shape for consideration of the Association at an early date.
(7) Rail Sections The Committee has been giving attention to the designing of a 150 Ib. rail section but is not prepared to report on the matter.
(8) Transverse Fissures In 1916 a bibliography was presented on the subject of internal fissures in rails (Proceedings, Vol. 17, 1916, p. 587), dating from the beginning of the literature on the subject in 1911 to the end of 1915. In Appendix E, the Committee presents an extended bibliography covering the years 1911 to 1920 inclusive.
The Committee expects to gather full reports of fissure failures on all the railroads for two main purposes :
(1st) To have definite information as to the occurrence and distribution of the failures and to be able to know whether progress is being made from year to year in their elimination.
(2nd) To tabulate and study the reports to determine the relationship of the various manufacturing and service conditions to the failures.
For the first purpose a summarized report should be made periodically, say once a month or once a quarter. For the second purpose selected data sufficiently extensive should be used but after the relationships have been fully established, the special tabulations may be discontinued.
The Committee recommends that the schedule of recommended practice for the inspection of steel rails, submitted in Appendix B, be adopted and included in the Manual.
Subjects for Future Work Your Committee recommends that the following subjects be assigned to it for 1922.
1. Study the Manual and submit proposed revisions thereof.
2. Continue the study and report on details of manufacture and mill practice as they affect rail quality.
3. Continue the study and report on the rational relation between intensity of pressure due to wheel loads and resistance of various rail steels to crushing and deformation.
4. Study and report a fair and correct method of reporting rail failures.
5. Continue the study and report on the developments in methods of rail inspection.
6. Recommend designs of rail joints and bolts covering important dimensions affecting interchange of both.
7. Continue the study and report on material for track bolts and joint bars and methods of treatment.
8. Continue the study and recommend sections for rails over 140 1b. per yd.
9. Continue the study of transverse fissures, with special reference to cause and elimination thereof.
10. Report on the most desirable length of rails.
11. Recommend a carefully considered outline of work for the ensuing year.
THE COMMITTEE ON RAIL,
G. J. Ray, Chairman.
RAIL FAILURE STATISTICS FOR 1920
By M. H. WICKHORST
This report deals with the statistics of rail failures collected for the year ending October 31, 1920, furnished by the railroads of the United States and Canada, in response to a circular sent out by the American Railway Association. The information furnished by each railroad showed the number of tons of each year's rolling from each mill, the equivalent number of track miles, and the total number of failures that occurred in each year's. rolling from the date laid until October 31, 1920.
They were reported by the railroads on American Railway Engineering Association form 408 as revised in 1915. (See Manual for 1915, page 104.)
The report covers rollings for 1915 and succeeding years and the ages of the rolling would average in track about the years shown below:
1918–2 years 1916_4 years
1919–1 year 1917-3 years
1920-several months The tonnages and track miles of rail represented by the statistics in this report are as follows: Year Rolled
41,142.60 In previous years the detail results as given by the railroads have been presented, arranged and classified by the mills that made the rails, and for the last two years they were also classified by the railroads that used them. In the present report, however, on account of a desire to save the expense of preparing the detail statements and to reduce the amount of printing, only the general average results are presented.
Lots of less than 1,000 tons (that is, less than 1,000 tons in any one year's rolling for a railroad) were excluded from the tabulations, as they would not materially change the group totals and averages and would unnecessarily extend the tables and increase the work. The method of compiling the statistics was to make blueprints of the reports submitted by the different railroads after checking the calculations and seeing that all the lines were fully filled out, and then cutting the prints along the horizontal lines. The strips constituted the units in the tables, and were sorted and collected into the desired groups for which the totals and averages were calculated.