DISCUSSION ON TIES

(For Report, see pp. 233-262.)

Mr. W. A. Clark (Duluth & Iron Range):-As stated by the Standardization Committee, this Committee has represented the Association in a conference on the unification of tie specifications under the auspices of the American Engineering Standards Committee.

Subject 2 was combined with Subject 5, as they were closely connected, and I will ask Mr. Burton, Chairman of the Sub-Committee handling those subjects, to present the report.

Mr. W. J. Burton (Missouri Pacific) :-These subjects were referred to the Committee last year, and the Committee, after study of what statistics were available in printed form, decided it would be necessary or at least desirable to solicit data, supplemented by opinions, from as many roads as possible. Such material was collected last year and was published. However, it was received so late in the year as to make it impossible to analyze it or draw any conclusions from it, and the report this year is an attempt to make use of the information which was received and presented to the convention last year. The answer to both questions is entirely dependent on one particular factor-the life of the tie, that is, before the proper answer can be given, we must know the life of the tie of the particular kind, based on the most reliable data which the Committee can secure. The answer for any particular location or particular condition depends upon knowledge of results secured elsewhere.

The Committee has found, as has everyone else who has gone into the subject, that the data available is not satisfactory for application in specific cases. Our tie statistics, generally, are more or less complete as to the conditions of the treatment of the tie, and the details of just what the tie was to start with, but they are not at all satisfactory as to the conditions of use. The Committee feels that the life of a tie is fully as dependent on where and how it is used, as on what it was to start with. The greatest difficulty lies in comparing traffic on different railroads in different parts of the country, and in comparing climates on different roads in different parts of the country. We have in the past attempted to obtain these comparisons strictly by the number of years of life reported. But what is reported as heavy traffic on one road may be very light traffic on another road, and as for climate, even though a road may be in a territory in which the general climatic conditions are favorable, the local conditions may be such, as regards producing decay, that is not favorable to long life in the ties. In some places the local conditions are different from those of the surrounding country.

With these considerations in mind, the Committee is recommending the use of a standard test tie as a means of avoiding the difficulties of comparison between different parts of the country, the idea being that the standard test tie used in test installations will serve as a sort of

thermometer to measure the climate, and also to measure the traffic. We may have a climate in which the standard tie will resist decay for six years; the six years will define the climate, and then there will be another climate where the standard tie will resist decay for eight years, etc. We intend to use the standard tie as a yardstick, not only to measure the climate, but also the traffic, and to use the results obtained with the test tie as a standard for comparing results from other kinds of ties. The Committee is recommending that for the standard test tie, a certain particular tie shall be designated.

The results of the data secured by the Committee are tabulated, beginning with page 239. The table on page 239 is made up from the returns obtained in last year's report, and we have eliminated from consideration all reports of ties which failed largely through mechanical wear; in other words, we are attempting to get at the relative resistance to decay of untreated timber. The results are averaged up in the fifth column, taking white oak as 100. The average life reported throughout the country for the white oak ties, untreated, is slightly over eight years. The table on page 239 may be said to represent the relative resistance to decay of untreated timber, as determined from the data in the hands of the Committee.

As a means of comparing the resistance to mechanical wear of different ties, the Committee has given considerable consideration to various formulae in which the physical properties of timber are taken into account, weighed, etc., but has come to the conclusion that the data we have, considering the relative crudeness of the whole subject, does not warrant such refinement. The Committee believe that satisfactory results can be obtained by considering the specific gravity of the timber as the measure of relative strength.

The tie is quite different from most parts of the railroad as regards design. The shape of the tie, the size of the tie, and the amount of timber contained in it remain substantially constant throughout its life, but the quality of the material varies from new to decayed. With most other parts of a railroad, the physical characteristics of the material remain nearly constant (this statement does not apply unqualifiedly with all timber), but ties are subject to much more severe decay conditions than any other part of the structure of the road.

The table on page 240 was prepared from the tabulation in Bulletin 556 of the Department of Agriculture (Forest Service), which gives the results of physical tests of various woods, and Table II is intended to justify the use of the specific gravity as the measure of mechanical strength. The mechanical life of the tie may be considered to be dependent upon how much reserve strength it has to start with. When the tie is decayed it has no strength left. It fails because its strength is then insufficient, and the Committee feels justified in using specific gravity as the measure of mechanical strength, based on this table. The data is also shown by Diagrams A and B, particularly Diagram B.

This diagram indicates that the modulus of rupture, which, for any particular cross-section of the tie, may be considered as the measure c the beam strength of the tie, varies according to the specific gravity.

The Committee considers the beam strength as fundamental in tie design, because if the beam strength is insufficient, there is no remedy. If it is not strong enough as a beam we cannot use it. If it is deficient in crushing strength across the grain we can use larger tie plates. If it is defective in spike holding power, we can modify this by the design of the fastenings. The beam strength is essential, and we consider the modulus of rupture the most important physical characteristic of tie timber.

Table III on page 242 is a combination of the decay charactertistics of various timbers as shown in Table I, and the mechanical characteristics as shown in Table II. Table III is perhaps not easily understood at the first glance, and the Committee itself has had considerable discussion over it.

The headings of the columns under the general headings of "Decay Failures" which read "Climatic Conditions Such that Under Light Traffic White Oak Ties Fail by Decay" might be headed "Measure of the Climate." We have used six years, eight years, and ten years' life of white oak as the basis for this table, these ties being used under very light traffic.

For instance, in column 2, we start out with the white oak tie at 100, corresponding to eight years. The other woods in column 2 are listed in proportion to the results given in column 5 of Table I, except that the values have been slightly modified.

The values for the several kinds of timber in columns 4 to 9 are based on the specific gravities of the various timbers, but starting with the white oak at 50, corresponding to a life of four years, which life the Committee considered about the shortest average life where mechanical wear is the determining factor.

Columns 10, 11 and 12 are headed in a manner similar to columns 1, 2 and 3, or in other words, column 12 means that the ties are subjected to a climatic condition such. that a white oak tie under very light traffic will fail in ten years. It does not mean that the white oak ties under the particular traffic fail in that time, but it is such a climatic condition as would cause the white oak ties to decay in ten years if allowed to decay and that the actual failures are partly decay and partly mechanical. The table is prepared as though the failures are fifty-fifty, decay and mechanical. We do not know of any way of ascertaining that the tie fails three-quarters by mechanical wear, and one-quarter owing to climatic conditions, or any other certain proportion. The best that can be said is that it fails both by mechanical wear and by decay.

The Committee submits some examples as to how the data in Table III, together with the diagrams given on pages 245 and 246, may be applied. These two diagrams were published by the Committee a year

or so ago in connection with a report to the A. R. A. on substitute ties, and are merely diagramatic representations of the compound interest formula, which formula was recommended for computing the annual cost of ties. The diagrams on pages 245 and 246 are for ties not having a salvage value, and they would not apply to substitute or other ties having a salvage value.

The recommendations to the Committee are given on page 238. (Mr. Burton read the recommendations referred to.)

Chairman Clark:-I move that the recommendations be adopted.

Mr. V. K. Hendricks:-It seems to me that by considering the relation of the average life of ties to the percentage of renewals, the Committee could get much valuable information as to tie-life from a study of the actual annual renewals on the different railroads, going into more detail than has ordinarily been the case. It would be very desirable to make such an investigation, although the subject is a large one to go into. I have been studying a paper by Mabel E. Thorne, Statistician of the Forest Products Laboratory, which was presented to the American Wood Preservers' Association, in January, 1918, the results of which could be applied to good advantage in many cases.

Mr. W. H. Courtenay (Louisville & Nashville) :-This matter of getting reliable information about ties is one of great difficulty. The Committee's own table on page 242, appears to me to indicate that. I understand column 1, in situations where a white oak tie will last six years, that cypress tie lasted 5.4 years. That is certainly not true of certain parts of our country. A white oak tie produced in our territory near the Gulf Coast will not last there more than six years. ordinarily.

I doubt if the average life will be as much as six years. I have actually seen cypress ties that were laid over 30 years ago; there could be no doubt about the age of the ties, as they showed the change of gauge made in 1886, and that was conclusive proof. Under ordinary traffic a so-called black, red or yellow cypress tie will last more than 25 years. I have seen cypress ties which have been in service much more than thirty years. They will not stand under heavy traffic. The life of cypress ties under heavy traffic is wholly determined by mechanical wear. Even though ample sized tie plates are put in, the traffic will drive them down. I mention that to show the extreme difficulty of getting correct information about ties.

Years ago the Louisville & Nashville used on some divisions one species of the exclusively. We obtained reliable information then, and were in the habit of computing every year the average renewals per mile in periods of four, eight and sixteen years immediately preceding the date of making the computation, and determining whether the superintendents requested too many or too few ties, and in forming our judgment in that regard we relied on these averages, but since the United States Railroad Administration began to use almost any and all kinds of

ties, the changes have been so great that the matter has been greatly complicated.

As to the propriety of using a 6-inch tie for making a test, I doubt it. Mr. Burton-Mr. Courtenay has spoken of the difficulty of getting satisfactory data; the Committee fully appreciates this fact. We would be very glad to receive data from the L. & N. or any other railroad which has not furnished it. Last year we tried to obtain information about the cypress ties on the L. & N., but did not receive reply containing it. We realize that this whole subject is one which, with our present knowledge, cannot be treated as though mathematically exact.

As regards the comparison between white oak and cypress in column 1, the white oak is shown as 75, which corresponds to six years. The cypress is shown as 90, which corresponds to 7.2 years. The interval may not be large enough between the white oak and cypress, but the 90 is arrived at from the returns which the Committee received last year, which included, in the case of cypress, returns from fifteen or twenty roads.

As regards the choice between the six-inch and the seven-inch test ties, the Committee has had considerable discussion. At first it considered the possible use of both the 6 and the 7-inch ties, but this seemed to defeat the purpose of the standard, and we finally decided on the 6-inch 8-foot tie, because of its shorter life as compared with the 7-inch tie, and because therefore, we will obtain results more quickly. A further consideration was that the 6-inch tie is more largely used, as shown by the returns to the Committee, and it is, perhaps, the tie which is most largely used, taking the country as a whole.

The standard size, particularly in the West and Middle West, has been 6-inch by 8-inch by 8 feet. The Committee felt, while it was not the best tie, perhaps it might be the best we could select for a standard test tie.

Mr. Courtenay-I can furnish reliable data on ties laid years ago. Our lines extend from the Ohio and the Mississippi River to New Orleans. On the Gulf Coast, between Mobile and New Orleans, it is hardly worth while to put a white oak tie in. It is hardly worth while putting in a long leaf yellow pine tie unless it is treated, but the cypress tic will give long life in that section.

It is a difficult thing to do, and it will cost a tremendous amount of money to give precise figures as to the life of ties. Since we have been treating ties with creosote oil and chloride of zinc, and distributing them all over the system, mostly due to the great difficulty of getting ties in the last few years, we had a number of them taken out, but we have no record what ties were taken out, and I think it would be a matter of very great value to the railroad if we could get accurate statistics about the life of ties on our own road, and about the only way to do it is to keep a life record of each individual tie, which is a rather large job, and particularly as you will have to rely upon the foreman to do that. Fore