Search Documents

By William R. Hart

I was this morning an interested listener to the remarks of Mr. Ashbel Welch in regard to his designing a new section for steel rails, in 1866; and for the salze of the truth of history, and in order t:, give the credit to an American (and where it rightfully belongs) of higning the first section for steel rails, which was intelligently adapted to that material; I beg to state the following facts: On the 14th of August, 1866, Mr. Ashbel Welch gave me an order for 200 tons of steel rails, to be made bbr the Camden and Amboy Railroad, from a section designed by him. A copy of this section 1 append herewith.

Jan 1, 1881

By R. W. Hunt

The old plan was to increase every part of a rail much in the same proportion. But each part should be in proportion to what it ha.; to do. The head should be deep in proportion to tile amount. of traffic and the lowness of the rate of interest on its cost. The body need only he strong enough to carry the head after it is well worn down, and that deperlds on the weight of the machinery, and in the case of steel has little to do with the volume of traffic, except SO far as that affects weight of machinery. Ad on most railroad systems, the same niacliinery is used on main lines with heavy traffic, and bra11ct:es with light, I suggested in 1874 that each systern adopt the same body of rail for both, and make the head deeper on the main lines, shallower on the branches. This plan was adopted on the Pennsylvania Railroad. Bandberg seems to recognize this in his patterns of 1878. E. W. HUNT, TRoy, N. Y.: Again Dr. Dudley preaents to our cousideration a series of most carefully conductetl experiments, and ~llile I fully appreciate the labor and thought this work has cost him, I must still hesitate to accept his deductiono. This paper dilliirs from the previous one in that it denls esclrieively with the wearing qualitiei. of steel rails. Dr. Dudley gives his reason for this change in the following statement: " With the improvement in maintenance of way which has characterized the Pennsylvania Railroad during the last five or six yearn, the remova1 of rails from the track from the first two of these causes (i. e., broken and crushel), has, if I am right, quite notably diminished. This certainly is true with regard to broken rails. And if, as time advances, the number of crushed rails shall diminish, both because of the continued improvement in maintenance of way before referred to, and because, owing to improved and better methods at the steel works, there are fewer crushed rails caused by physical defects in the steel, the question of the wearing capacity of steel rails obviously becomes the allimportant one." Certainly the condition of the roadhd has much to do with mils brealring, crushing, and wearing. If the Pennsylvania and other railroads have done and are doing their part, Dr. Dudley oourteously admits that the steel works have performed part of theirs. But as far as I am informed, the formulas which they have used have been and are still quite wide of the one he continues to recommend. The rails which are not breaking or crushing to so great an extent as formerly, contain higher percentages of both carbon and manganese.

Jan 1, 1881

By O. Chanute

all of' level atid grade curve excepting Nos. 897, 898, 899, and 900, because of the impossibility of grouping them in the same nlonner, no two Iraving the chemical composition and physical prnprtiw as tho only v:iriubles. Phosphorus ~tnits have been taken to show the relatia h:wdnes. Of these 16 groups, 10 decidedly indicate slower wear firr tile harder rail. Of the remaining 6 groups, three (groups 2, 9, and 15) come well within the limits of error (.25 pound per yard) inherent in the palculation of the data as pointed out by Dr. Dndley; they catrnot, therefore, be considered tls exceptions. Group 2 and 15 arc but little outside of the limita of error. Rail 923 of group 9 is obviously abnornlal. This is probahly due to its being overheated, which the chemical analysis shows might easily be the case, and which the physical tc~ts and its low specific gravity tend to c-nfirm. We inry say, then, thnt 13 of these 16 groups fairly indicate, if they do not definitely point out, th:kt, under the same condition% tbe harder rail gives the slower wear. Of course, these comparisons are too few to enable us to arrive at positive ~?o-nclusions; but indications thus obtained are, I believe, fir more vsluitblc and tresttvc~rthy than those fhzt averages rvottld give us, mnde up from any ~lumhr of rails under intloy difi~rcal, conditions. Finally, it, seems to me, that the conclusions arrived at earlier in my remarks, together with, and confirmed by the last, show strong evidence that, under the same conditions, the harder rail will givc the slower wear. 0. C~~anute, New York CITY: We are very much shlipl, 1 ant sure, to Mr. Sandberg for his paper upn " Rail Ryaifications and Rail Inspertion in Europe." We have in thp United States bithhrto Been inspecting mils sonlewl~at haphazard, and we ore glad to gtbt tfie results of Mr. Sandberg's long experience. We recognize that he was an)ong the first, if not the very firwt, to iipply mare rittional and scientific rules to the designing of irou rails., niora recently, to adapt tlitve rules to the clesigtting of steel I':I~]R, to conform to the oapaljilities and rqniretnento of this new n1:ttcrial. Although we have generally adopted, in this country, sectinns for hteel rails which marly of our railway men think even bottar than those of Mr. Snlldlmrg, he is neverthelees the Ieat3er in foot.iito11~ we have followed ever since it has been est~blished

Jan 1, 1881

C. P. Sandberg, London, Eng. 1 think we should all be grateful to the Pennsylvania Railroad Company, and to their chemist, Dr. Dudley, for spending so much time and money in order to solve an important question; nor are me less indebted to the public spirit which leads them to impart tlie experience thus gained. In Europe nothing similar has hitherto been done, either by a company or by any private individual. In England the great railway companies employ their engineers in other ways, and do not keep a chemist or specialist to study rails; bat even if they did they would probably not publish the results. The chief professional engineers are so occupied with their private practice of railroad construction generally, that they cannot be expected to devote their time and energies to so special a question. The German Railway Union has, in my opinion, missed the real object of its elaborate researches by falling into the error of specifying costly, impracticable tests for rails—tests at the same time which do not infallibly expose the impurities that may be in the rails under examination. There has, therefore, been nothing done on this side of the Atlantic in the way of exhaustive study to determine the best composition for steel rails. America should consequently have full credit for the enlightened example which she has set. I wish, however, to make a few remarks on Dr. Dudley's second paper,—& paper which follows very much upan the same lines of argument as the first red three years ago, upon which I have previously touched in my pabr on the same subject read at the meeting in August last. If Dr. Dudley had discovered a new metal he could hardly, I think, have taken wore pains to prove that he was correct in his first formula for the best chemical colnposition of steel rails. But I hare seen so much done in the way of proof by experiment that I am inclined to think that almost anything can be proved by experiments and by satnples. If Dr. Dudley would only give up his formula, which would not suit any country—not even America—

Jan 1, 1881

By R. W. Raymond

to same extent, the assumptions tinderlying Dr. Ilu?ley's con alusions. These amrrmptions are: that 'the loss of metal per million tons of trafffc, depends, first upon the circrtnistances under which the rail is worn; secondly, upon the proportions of its various constituents, ancl that the anlorlnt of loss is afEect~d by eacb constitnent in a cer tain ratio, so that if the circunlstances were the same for all tlie mils, the wear per nliilion tons might he upproxi91ately expre~sed by an equation of the form I + Cc + Pp + SR + Mm = R, io wtlioh I is an unknown valne, dependi~.rg on the quantity of iron (and other ingre dients not determined) in the rail, and assumed to be constant for all the rails ; c, p, s, and m are the experimentally determined quan tities of carhon, phosphorus, sulphur, and manganese, as tabulater1 in YIates VT and VII of Dr. Dudley's paper; and R is the experirnen tdlly dcterlnincd wear per million tons; while C, P, S, and M are onknown wfficients. In other words, it is assunled that the differ ence in wear is proportional to a difference in c, p, s, or m, and that the effects of a change in the amonnt of either are independent of the amount of the rest-—within the experinlental limit5 of the problem. It is also assr~med that sulphar, copper, and other ingredients, not determined, are qually present in all the rails, or at least {lo not affect the wear, since I is taken as constant. Dr. Dudley tacitly assumes the coefficient6 C, P, S, and M to be positive; and assigns certain valries to them, by means of which he redom t.he equation to the form I + Pa: = R, x being equal to cc + Es +Mm i Pp, or tjje uanjbe,. of so-called phosphorus units in the rail. But we do not have R directly given under constant circumstances. The varying weights (per engine, car, or train) and the varying epds, we must neglect for lack of data, and assume that in these rwpects the rails examined have been treated srlbstantially alike. But there remain the varying conditions as to track, which mnst he elirninsted. Dr. Dudley gives six clrlsses, four of which contain 8 rails each, and two 16 rails each. Beginning with the 6~st (tangent grade), take the mean of the figures expressing wear for that clam; and continaing tliis process through the other five, we obtain six expressions in quantity of loss 'by wear, which we assome tc, represent, in their differences, the varying conditions of the trarli, with respect to grade and direction. We now take the mean of these moans (giving double weight to the gmiip containing 16 ohwrvafieras). This general rnean is the 10 of metal per million tons of trafio for a rail of avorage constitution under mean ciroumstances.

Jan 1, 1881

By C. B. Dudley

In a work published in 1838, De Morgan, the author of the article on " Probabilities," in the Encyclopedia Metropolltimn, says: " The method of least squares is not yet introduced iutb the affairs of common life,.....that many thi~gs which are denlonstrable only by the higher branches of mathematics are looked upon as useless." And he mentions an instance in which "a method was suggested for estimating the relative importance of boroughs, by considering their population and ~wntributions to the revenue con)-

Jan 1, 1881

By C. P. Sandberg

C. P. Sandberg, London, Eng. 1 think we should all be grateful to the Pennsylvania Railroad Company, and to their chemist, Dr. Dudley, for spending so much time and money in order to solve an important question; nor are me less indebted to the public spirit which leads them to impart tlie experience thus gained. In Europe nothing similar has hitherto been done, either by a company or by any private individual. In England the great railway companies employ their engineers in other ways, and do not keep a chemist or specialist to study rails; bat even if they did they would probably not publish the results. The chief professional engineers are so occupied with their private practice of railroad construction generally, that they cannot be expected to devote their time and energies to so special a question. The German Railway Union has, in my opinion, missed the real object of its elaborate researches by falling into the error of specifying costly, impracticable tests for rails—tests at the same time which do not infallibly expose the impurities that may be in the rails under examination. There has, therefore, been nothing done on this side of the Atlantic in the way of exhaustive study to determine the best composition for steel rails. America should consequently have full credit for the enlightened example which she has set. I wish, however, to make a few remarks on Dr. Dudley's second paper,—& paper which follows very much upan the same lines of argument as the first red three years ago, upon which I have previously touched in my pabr on the same subject read at the meeting in August last. If Dr. Dudley had discovered a new metal he could hardly, I think, have taken wore pains to prove that he was correct in his first formula for the best chemical colnposition of steel rails. But I hare seen so much done in the way of proof by experiment that I am inclined to think that almost anything can be proved by experiments and by satnples. If Dr. Dudley would only give up his formula, which would not suit any country—not even America—

Jan 1, 1881

By Rich Akerman

as 1866, I wish to assure both these gentlemen that I had not seen the section when I designed mine, :md even if I had, I should not have then dared to put it forth as a standard for English rail makers; tind the soundness of my opinion has a simple but conclusive proof in the statement made concerning the extra price which was at that time required for their production. I need not say that I highly approve of the original Welch section, for it is in principle similar to the design which I made ufteen yeam subsequently; and from a professional point of view I would fully concecle to Mr. Welch the credit of first designing the modern steel-rail section. From com- mercial, matter of tict, and practical point of view, he was, however, before his time by many years; or to put it in another way, rail making was ten years behind its time when the new metal steel was first introduced for rails. If his design has been incorrectly identified with my name (as the sections are similar in principle, though they differ in thickness of flange and web), I beg to say (1st) that I much regret the misun- Understanding which has arisen through no fault of niine, aud (2d) that no confusion has been intentionally created by me. It is a pity that the section designed so long ago has not been made public at an earlier date, as it would make a really goal stand- and pattern, anil even now be quite equal to present requirements. In conclusion, I wish to say that I have never desired to claim any invention or novelty by the publication of my sections; they Were put forth as being such as English railmakers could and would produce at the ordinary market. price, and as being the best and most suitable to be so obtained. Had Mr. Welch or Mr. Holley put forward theirs ten or fifteen years ago, I should have been the first to recommend theru. In fact I should have been glad to see their narnes associated with rail patterns in their own country, in the same way as mine has been. PROFESSOR RICH. AKERMAN, STOCKHOLM, SWEDEN: i AGREE with your (Mr. Sandberg's) views that it is too early yet to stipulate only one definite chemical composition in the rails. Such a stipulation, I think, might he justified, if it were not for the manganese; but this metal changes the properties of the iron aud steel in tho same direc- tion ae carbon, silicon. and phosphorus, by increasing its tensile strength, stiffness, hardness, and brittleaess. If the iron and steel does not only contain carbon and manganese but also phosphorus,

Jan 1, 1881

By Ellsworth Daggett

The ore smelted in the Winnamuck furnace during the year 1872 consisted, for the most part, of oxidized ores from the Winnamuck mine, only sixty tons of outside ore (from the Spanish mine) having been smelted. The latter, like the principal Winnamuck ore, was oxidized or so-called carbonate ore. There was mixed with these oxidized ores three hundred to four hundred tons, or 7 to 10 per cent. of galena, some of which was mined with the oxidized ore, while a part was mined separately from the lower portion of the mine, and afterwards mixed with the ore, with a view of preventing the formation of deposits of metallic iron in the furnaces. The average assay in silver of all the ore handled was 51.46 oz. per ton, most of it existing as chloride of silver. The lead contents were 34.98 per cent,, all, or nearly all, in the form of carbonate of lead. The relative amount of silver is not at all constant, the best silver ore often being poorest in lead. The predominant gangue was silica, several determinations of which have been made on representative samples, yielding in three such samples 26, 38, and 58 per cent. silica, the latter test being ore containing but little lead. The average contents in silica are about 35 per cent., with 6 to 7 per cent. sesquioxide of iron, and small quantities of alumina and lime. Mechanically, the ore was very fine, and so thoroughly disintegrated that it presented few distinguishing characteristics, rendering sorting or separating of ore from waste difficult, and often impracticable. Experiments on a small scale have been tried, with a view to separate the silica by washing; but these were unsuccessful, as the finest slime, requiring a long time to settle in still water, contained a large amount of silver; and on a careful sizing and washing of the sands and coarser parts, the silver contents were found to be less dependent upon specific gravity

By Edmund C. Pechin

The Lucy furnace, owned by Messrs. Carnegie, Kloman & Co., and located on the Alleghany River, on the outskirts of Pittsburgh, is a splendid modern furnace, 75 feet high, and 20 feet bosh. She had been working well on low grade ores of about 50 per cent., producing daily 68 to 75 tons. There was on stock 500 tons of Republic ore, one of the purest and best of the Lake Superior ores, averaging over 68 per cent. of iron, which had been procured for the purpose of making a trial for Bessemer iron. This was charged by itself, and Mr. Skelding, the founder, reports that he did not succeed in getting a single cast when it came down, before the furnace chilled from the hearth to the top of the boshes, some 25 feet. Every effort was made to save her, without avail, and the disagree-

By C. P. Karr

For several years, the Bureau of Standards has been working in cooperation with an Advisory Committee on non-ferrous metals on various phases of the production and testing of zinc bronzes. The work has dealt especially with the type of metal known as "Admiralty bronze" in England and as "Government bronze" in this country and consisting of 88 copper, 10 tin, and 2 zinc. Several reports have been made on this subject and these should be considered in connection with this paper.' First Series of Five Foundry Tests A very brief summary of the first series of five foundry tests has been made12 but for the sake of clearness the results are given here in different form. The five foundries cooperating in this investigation were: Ohio Brass Co., Mansfield, Ohio; Titanium Alloy Mfg. Co., Niagara Falls, N. Y.; Packard Motor Car Co., Detroit, Mich.; National Cash Register Co., Dayton, Ohio; Bureau of Standards, Washington, D. C. The object of the work was to determine whether uniform results could be obtained with this alloy in different foundries working under definite instructions. One of the participating foundries prepared, from virgin metal, a stock of ingots of the composition 88 copper, 10 tin, 2 zinc, and at the same time poured nine test bars from this melt of virgin metal. The results of the physical tests of these bars are shown in Table 1. Examination of this table shows that the highest and most nearly uniform results were obtained when the virgin metal was cast flat in chill molds and the lowest and- most variable results were encountered when it was cast, vertical in green-sand molds. In this, as in all other tables, the mean deviation was obtained by the usual method of taking the average of the differences of the individual values from the average values.

Jan 1, 1921

By F. H. Daniels

IN this paper it is my intention to call your attention to a few of the many producers using blast, now in common use in Sweden, and also those constructed by the Washburn & Moèn Manufacturing Company at their Quinsigamond Works. During a prolonged stay in Sweden the writer had an opportunity

Jan 1, 1881

By F. G. Smith

A few years ago, the writer encountered a problem that, at first, seemed to be due to peculiar conditions affecting grain growth. Large cups made from heavy metal failed in the first drawing operation in the process of making seamless tubes. The metal, as cast, appeared to be sound and strong; it rolled successfully and withstood the first cupping operation, but on the first draw it failed, with a crystalline fracture. The failure appeared in the slightly worked part of the bottom of the cup, and microscopic examination revealed exceptionally large grains. Knowing the general conditions governing recrystallization of brass, which were later published by Mathewson and Phillipsll an investigation was outlined with the idea of including various conditions under which new grains might form, and which would include the effect that a previous anneal might have upon the anneal following. These conditions were produced by annealing specimens of brass at five different temperatures, indenting them with a 10-mm. ball under four different pressures, and then reannealing at each of the five temperatures five specimens each of which had been annealed at one of the five temperatures. Preparation of Specimens The material used was cartridge brass 0.400 in. thick, having the following composition: Copper 66.60 per. cent., zinc 33.35 per cent., lead 0.02 per cent., iron 0.03 per cent. This was rolled with 35 per cent. reduction after a 700" C. anneal. Fig. 1 shows the structure of this original bar on which all subsequent anneals were made. Pieces 1 in. (2.5 cm.) wide and 3 in. long were prepared and annealed for 1/2 hr. at 450°, 550°, 650°, '750°, and 850" C. Each of these pieces are indented under the Brinell test with a 10-mm. ball at 500, 1000, 2000, and 3000 kg. pressure. Then, sets of five pieces from each of the original anneals were reannealed at 450°, 550°, 650°, 750°, and 850" C.

Jan 1, 1921

By P. D. Merica, H. Scott, R. G. Waltenberg

The remarkable phenomena exhibited by the aluminum alloy known as duralumin were discovered during the years 1903—1911 by A. Wilm1,2 and have been described by him and by others. 3, 4,5.6 The unusual feature of this alloy is the fact, as was shown by Wilm, that it can be hardened quite appreciably by quenching from tempera-

Jan 1, 1921

By W. A. Gibson, Zay Jeffries

It has been known for a number of years that certain aluminum alloys could be hardened by quenching from a temperature of about 500' C. Immediately after quenching the total increase in hardness is not marked, but the hardness increases gradually so that after about 4 days, it has reached a maximum. The actual increase in hardness probably continues for weeks or even months, but in order to detect the differences after 4 days, it is necessary to make tests at rather long intervals. So far as is known to the authors, Alfred Wilm was the first to discover these properties in aluminum alloys.' The aluminum-copper alloys are amenable to this treatment and the addition of magnesium makes the heat-treatment effect more pronounced. Until recently heat treatment of aluminum alloys has been confined to the worked alloys in the form of sheet or extruded products. The alloy that has been most used in this connection is known commercially as duralumin and has about the following composition: copper not exceeding, 6 per cent.; magnesium not exceeding, 2 per cent.; manganese not exceeding, 1 per cent.; commercial aluminum,2 balance. A typically good alloy in this range consists of: copper, 4 per cent.; magnesium, 0.5 per cent.; manganese, 1 per cent.; commercial aluminum, balance. This alloy can be rolled into sheets or extruded; and in this condition it can be heated in the ordinary furnace atmosphere to a temperature near 500° C. and quenched in water. After about 4 days it will have the following physical properties: tensile strength 55,000 to 60,000 1b. per sq. in. (3866 to 4218 kg. per sq. cm.); elongation in 2 in., 15 to 20 per cent. No harmful effects of the furnace gases or even the air, if the heating is effected in an electric furnace in an air atmospherc.

Jan 1, 1921

By C. F. Smart

The heat treatment of the bronze composed of 88 per cent. copper, 10 per cent. tin, and 2 per cent. zinc, has been investigated by H. S. and J. G. S. Primrose,' and also by the U. S. Bureau of Standards.2 The results obtained by H. S. and J. G. S. Primrose on dry-sand castings subjected to reheating followed by air cooling, or by quenching in water, are shown in Figs. 1 and 2. The values for tensile strength have been converted from tons per square inch to pounds per square inch. Figs. 3 and 4, have been plotted from values obtained by the Bureau of Standards on material of the same composition and treatment. Both investigations show that the ductility of the cast bronze is improved by heating to 700' C. and cooling in air, the strength not being impaired by this treatment. They are not in agreement, howevcr, as to the effect of quenching, Messrs. Primrose reporting a decrease, and the Bureau of Standards an increase, in both strength and ductility when the cast bronze is heated to 600' or 700' C. and quenched. The present investigation was undertaken to study the effect of annealing followed by different rates of cooling, and, in particular, the

Jan 1, 1921

By Rossiter W. Raymond

The manufacture of ferromanganese in the blast furnace having been the subject of considerable attention in the Institute, I beg to put on record a contribution to the discussion from a quarter hitherto not directly represented, namely, the St. Louis furnaces, near Marseilles in France, belonging to a company of which M. Jordan, a member of the Institute, is Managing Director. These furnaces commenced in 1862 to manufacture specular pig (fontes facettes) containing from three to six per cent. of manganese. In 1864 the regular production of spiegeleisen (fontes miroitantes), containing seven to ten per cent. of manganese, was inaugurated. At that time the Rhine was the only locality of this manufacture, and the Bessemer works of England and France obtained from Gernlany all their spiegel. From 1864 to the present time, the St. Louis furnaces have been advancing steadily the percentage of manganese in their product. From ferromanganese of thirty per cent. they had arrived, about two years ago, at a product containing seventy to seventy-five per cent., and for some time past they have been turning out what, as M. Jordan remarks, may fairly be Termed fonte de manganese, or manganese pig, instead of iron, since the percentage of iron it contains is not more than 8.5, while the percentage of manganese is 85, or ten times as great. This very high grade material has a peculiar appearance, entirely different from that of ordinary cast iron or spiegel. It is granular, and steely gray in color. The following analyses, communicated to the Sociètè des Ingenieurs Civils, by M. Jordan, show the composition of various products of the St. Louis furnaces:

Jan 1, 1879

By F. E. Bash

Constantan is an alloy of copper and nickel that is extensively used, under a number of trade names, as a resistance wire with a low temperature coefficient of resistance, and one of the elements of base-metal thermocouples, iron usually being the other thermoelement, although for low-temperature work copper is also used. Before the war, constantan was imported, in common with many other technical necessities. Since it required considerable time to receive the constantan after it was ordered, fairly large stocks were kept on hand; but after the beginning of the war no more was imported and the stocks of foreign material were exhausted about the same time that the steel and munition manufacturers began to need great quantities of it for thermocouples for the heat treatment of steel. To meet this demand, the International Nickel Co. began manufacturing constantan that met all the requirements of the users of resistance wire but did not give the same electromotive force against iron as the German material and consequently could not be used with instruments calibrated for a definite e.m.f.-temperature relation. It so happened that all the constantan made had an e.m.f. lower than that necessary to meet the specifications of the Leeds & Northrup Co. for whom this investigation was carried out. If some had been higher and some lower, good wire could have been obtained by a process of selection, but as it was no wire at all could be procured. To remedy this situation, a thorough investigation was started with the close cooperation of the International Nickel Co. and the Electrical Alloys Co. The specifications to which the constantan had to conform were that it should give an e.m.f. of 47.40 millivolts against pure iron at 1500" F. (816" C.) with the cold-junction temperature at 0" F., or a proportional e.m.f. at temperatures near 1500" F. The method of checking a wire was as follows: It was first welded to a piece of iron wire; Armco iron was found to be satisfactory but iron with greater than 0.1 per cent. carbon would give a lower e.m.f. against the same piece of constantan than pure iron. The couple was then insulated and the fire end was tied to the

Jan 1, 1921

By F. E. Bash

Previous to the war, this country depended on Europe for its supply of a number of alloys of great importance in the manufacture of electrical apparatus and equipment. When this source was cut off shortly after the beginning of hostilities and the small stocks in the United States were exhausted, either inferior substitutes had to be used or the manufacture of the apparatus suspended. Since electrical equipment was of primary importance in the prosecution of the war, the substitutes had to be used until the composition and processes of manufacture of the alloys could be determined. Included among these alloys was one generally known as manganin, which is composed chiefly of copper and manganese and which, beside having a low-temperature coefficient of resistance, has a low thermal electromotive force against copper. These qualities make it valuable for the manufacture of electrical instruments and machines in which it is used in the shape of wire, rod, and strip. As a result of these circumstances, an investigation was started to determine how manganin could be made on a commercial basis so that a product of uniform quality and the necessary electrical characteristics could be produced. The object was to procure an alloy with temperature coefficient of resistance not greater than 1 X 1X per " C. and a thermal e.m.f. against copper no greater than that of the imported material. It is characteristic of manganin that the temperature-resistance curve is not a straight line but has a peak, which varies in position with different samples; it is most desirable, however, to have it at a temperature of 25' C. if possible. The coefficients of manganin from all records available ranged from a large negative to a large positive value. The composition as reported by different authorities is given in Table I. Prof. Matthew A. Hunter, of Rensselaer Polytechnic Institute, made up a number of small melts of manganin in a graphite crucible and the one that he reported as having the best cocfficient gave the following analysis: Copper, 83.6 per cent.; nickel 2.5 per cent.; manganese, 13.6 per cent.; iron, 0.3 per cent. An analysis of a sample of German manganin gave the following: Copper, 82.6 per cent.; nickel, 3.8 pcr cent.; manganese, 12.8 per cent.; iron 0.7 per cent.

Jan 1, 1921

By P. D. Merica, A. N. Finn, R. G. Waltenberg

CeRtain compositions of the light, i.e., aluminum-rich, alloys of aluminum with magnesium and copper have become quite well known within the past ten years under the name of duralumin. These alloys are used for rolling and forging and can be so treated as to develop quite remarkable mechanical properties. Thus a properly heat-treated alloy containing about 4 per cent. of copper and about 0.5 per cent. of magnesium, rolled into sheet or rod, will have a tensile strength of approximately 55,000 1b. per sq. in. (3866 kg. per sq. cm.) with an elongation in 2 in. (50.8 mm.) of about 15 per cent. This alloy, duralumin, was developed by Wilm,1 and its properties more fully described by Cohn.2, The authors considered it worth while to investigate the mechanical possibilities of the light alloys of two somewhat analogous ternary series; namely, of aluminum-magnesium-nickel and aluminum-magnesium-manganese, to study the effect of variation of composition upon mechanical properties within the alumimum-rich group of the aluminum-magne-

Jan 1, 1921