The Iron Age 1926-02-18: Vol 117 Iss 7

& THE IRON AGE New York, February 18, 1926 ESTABLISHED 1855 VOL. 117, No. 7 Seamless Tube Mills Analyzed Limitations and Advantages of Five Types and of Nine Combinations of Processes—German Steel Quality Criticized BY PAUL CEBRAT™ TEEL tubes form a substantial part of the pro- nately two-thirds of the production went to two prin S duction of finished steel. The Jan. 7, 1926, An- ipal users. Of all the material used in the former in nual Review Number of THE IRON AGE stated that dustry, 53 per cent or over one-half was pipe and in juring 1925 there was produced in the United States the construction of buildings 12 per cent of the total approximately 3,339,000 gross tons of steel pipe, o1 steel used was in the form of pipe. 10% per cent of the total of 32.5 million tons of steel Corresponding figures for the previous three years rolled. Of this amount 42% per cent was used for oil, show that the tonnage of steel pipe was 11% per cent gas and water and 21 per cent in buildings. Approx f the total steel rolled in 1924, 12 per cent in 1923 and — 11 ne . . 1999 There ic 1 "PASO ’ , ; yr. *With Perin & Marshall, 40 West Fortieth Street, Nev na per nt " ee eT Phere = no reason why this per York entage should decrease and, with the natural growth Bitlet PIERCING | SED OPERAT PROCESS ¢ I C i) 4 a) oot T es : Mannesmann i 10 Fig. 1 Comparison a of Five Different Pro- Ila {| = cesses for Making Ehrhard+ ' s ' ’ i Seamless Steel Tubes, Wellman - Peter: = , ‘ CMG CTETS Two of them with Va- a riations. Four cases aaa —ae a of round billets, two IIb Same Large Diameters of hexagonal billets G1 Ce) es) ee and three of square Il = sh nes he ie Cee eee - ‘ % : billets are shown. Cont, vous Mi} . ‘ ’ Crossed rolls are used a x \\ f for piercing in three \ € . Wa aps OR me. aieenatel woe * 7 . ; cases, Stiefel rolls in 477€. C/ jf ° oceans "4 . one and punching in 2 five. Pilger rolls are Wb » used for finishing im ‘Te OTENG - ai one case, draw or push = 4 hench in two, continu- 3 —_—~ , ue Ve . ‘ —" ous rolls im one, two- 5 thrhord+ ‘ ; ; = oe "aad high rolls in four and 9 - i aie ite inside and outside - IVd e oy - ios re rolls for the last = 3 oo I. Erhardt large Diameters 474 THE ° j — f oo f rolled steel produced, the tonnage of steel “ nerease correspondingly. manufact steel tubes in this lap weld processes are used ure of rolled the Dutt weld and There exists, however, an ever greatest extent. ; tubes which, on account é ail seamie superior qualities, are, for certain purposes, n ex lusively used and demanded. Seamless tubes ir equiré for ince, for oil-well casings, drill pe boiler and superheater tubes, etc. Another large er using seamless tubing in increasing quanti- ties, and replacing with it other forms of steel, is the Building construction also looks umless tubing, it being less expensive rass and unquestionably superior to welded pipe. Seamless Tubes in the United States M 4 IRON Eh rhard+ February 18, 1926 AGE Very little has been published in this country on this subject. One article in THE IRON AGEf by E. R. Kelso described the automatic mill of the Weldless Tube Co., Wooster, Ohio. A series of four articles in THE IRoNn AGE of April 7, 14, 21 and 28, 1921, gave the theory in considerable detail. This was a translation from Stahl und Eisen. An article by E. F. Ross in Jron Trade Reviewt describes a recently-built English push-bench plant and continues with a short but instructive discus- sion of the Pilger process. The following observations, made by the writer dur ing a recent study trip to the Continent, may, there- fore, be of interest. Germany has, since the war and practically for some years before the war, almost entirely eliminated the manufacture of lap-welded tubes, with the exception of the very large sizes (over 20 in. in diameter) which \ few American mills are producing such tubes, are usually water-gas welded. And even boiler shells either Mannesmann or Stiefel type turbine rotor drums and other high-pressure vessels ercing nd ir nnection therewith continuous’ and containers up to 11% ft. in diameter are in Ger Others employ push benches on many now being made seamless. U - TUBE LENGTHS N FEET ; £ : < © © S a = Fig. 3 (at Left)—Diagrammatic Comparison of = 0 a the Practical Limitations on Sizes Made by the Different Processes. Dimensions shown in inché KM represent tube diameters, maximum and m mum; those in millimeters, wall thicknesses. | in\ | maximum and minimum > kK j mm. r t —<_ - > - as e ~ 5 & Smn “a > smi | QO Smm 62 it w > © 4 — +4 } 52" i 3h Fig. 2—Pilger Rolls for Forging Tubes Pierced 3") © ph he o ihe “Ste by the Crossed Rolls of the Mannesmann System. Tn 2mm a 2 7 @ . + ae m vmm In the set shown the finish diameter of the tube > —<- aonaeceres sneleeemenn silicates nie = . ° » ¢ 5 [ II It ll IV Vv is 67 mm. (25% in.) and the roll diameter 315 4 Q > . " 0 mm. (12% im.) OC annesmanr 4 A re 3 a wi =x lwo High M,// Ehrhardt and Wellman-Peters Continuous Mi// ~ bot- , while one small plant has had an antiquated hand- yperated Pilger mill in for a number of which cupped plates are drawn out into tubes or operation years. Seamless steel tube tonnage produced in the United States during the past few years has varied from 118.,- 000 tons in 1921 to 404,000 tons in 1923. Percentages correspondingly have varied from less than 6 to almost 13 per cent of the total pipe and tube production The figures are shown in gross tons in the table below ti _ ir a7 Tut Pers e 6.9 6 67 > 4°9 ++ : 187,44 17,8 y 732,23¢ 257.8 { 1,02 x4 = t ‘ 9 J li 924 x th / nd rt 162 ‘ I pe od pee srnenernaneerneererver seeneennremenpesserisesensanner Production of steel tubes in Germany for January to October, 1925, was 553,139 metric tons, or 6% per cent of the total steel rolled in that period. According to trustworthy information, this production can be di- vided ino 70 to 75 per cent seamless tubes and 30 to 25 per cent small butt-weld and large water-gas welded tubes. Of the total production of seamless tubes not less than 75 per cent are manufactured by the Pilger process and the remainder in Stiefel mills, Ehrhardt mills, push benches and others. In France, now including the Saarland and Alsace- Lorraine, and in Belgium similar relations exist. Eng- land is still making a considerable amount of lap-welded pipe in the medium sizes (6 to 12 in.). The figures for that country are said to be about 50 per cent seam- less and 50 per cent butt and lap-welded tubes. Pilger Process Especially Useful It is particularly the Pilger process which, on the Continent and especially in Germany, has been deve!- acyl Pa ae February 18, 1926 oped to a certain degree of perfection. By this means are produced economically tubes of the desired qualities and of considerably greater length than the common mill length of about 20 ft., which usually command a premium over the price of lap-welded tubes. This has induced some American manufacturers to study this process and its advantages, together with the commer- cial aspects, with the result that two large concerns have purchased Pilger mills in Germany. Other concerns are contemplating similar installa- tions, as it may be expected that not only the normal steady increase in steel tube requirements will be prin- cipally in seamless tubes—except probably in the small sizes commonly used for gas and water piping, con- duits, etc.—but that the market for lap-welded tubes will become continually smaller, seamless tubing being preferred and demanded, where up to now lap-welded material has seemed sufficient. An indication that this assumption may be consid- ered justified lies in the fact that, with an increase in 1925 of 19 per cent in total rolled steel production over 1924, the amount of steel pipe fell from 11% per cent to 10% per cent of the total. This may be due partly to the importation of considerable tonnages of seamless tubing from Germany and Belgium, which was largely used in the oil fields. Various Processes Before going further into details it may be well to review briefly the different methods in use on the Con- tinent for producing seamless tubes. This can best be done by quoting from an article by Ewald Rober of Diisseldorf, published in Stahl und Eisen, Feb. 16, 1922, page 253, although some changes and further develop- ments have taken place since then. This article does not include the automatic mill, which is being widely used in this country, but no mill of this type is in oper- ation in Germany. Mr. Kelso’s article, however, gives a full description of such a mill and it has, therefore, been omitted from this discussion. The following principal methods are in use in the manufacture of seamless tubing, as illustrated diagram- matically in Fig. 1. I—Piercing mill and Pilger mill (Mannesmant! process). Ila—Piercing press and push bench (Ehrhardt and Wellman-Peters process). Ilb—Piercing press and draw bench for large diam eter tubes. Il1I—Piercing mill and continuous mill 1Va—Piercing mill (Stiefel system) and two-high (plug) mill (Swedish process) IVb—Pre-piercing and piercing press and two-high <a e press (Ehrhardt) and two-high mill IVd—Piercing mill and two-high mill V—Piercing press, draw bench and special rolling mill (Ehrhardt system), for hollow bodies of large diameter. The principle of the piercing mill, either Mannes- mann or Stiefel system, is well known and has been described in Mr. Kelso’s article, as well as in the 1921 series. To review briefly, either a cast or rolled round of a somewhat larger diameter than the tube to be rolled, after being thoroughly heated, is delivered to the feed trough of the piercing mills. It is pushed forward until it is gripped between the working rolls of this mill. These conical rolls are inclined 5 to 8 deg. to the horizontal plane, in opposite directions, and are rotating in the same direction. A tapered plug is car- ried and held in position by a rod, the distance between the rolls and the greatest diameter of the plug deter- mining the outside diameter and the wall thickness of the pierced round. The round gripped between the rolls is now forced to rotate between the rolls and, on account of their inclination, is moved forward and pulled over the plug. The rolls, exerting pressure on the outside of the round, while it is rapidly rotating, at the same time moving it forward, cause a stretching of the inner fibers. The rough and irregular hole created is smoothed and lim- ited to a certain size by the pierced round being shoved over the plug. The piercing process is, therefore, the result of three operations: 1.—The rolling between cross rolls, which creates the hole in the center; 2.—The forward movement, which pushes the outer fibers of the round forward between the rolls THE IRON AGE 475 3.—The resistance of the plug, which forces the inner fibers outward along the conical surface of the plug Instead of the piercing rolls, system Mannesmann, conical rolls or conical disks (system Stiefel) are some- times employed, although less frequently, and their operation is based on the same principle. The piercing plug and its supporting rod are held in position against the forward movement of the ingot by a hinged abutment gate. After piercing, this gate is swung open, the hollow round taken off the plug rod and conveyed to the Pilger mill. A new plug is then put on the rod, the gate closed and the mill is in readi- ness for piercing the next bloom. Operation of Eccentric Rolls Contrary to the other tube-rolling and drawing processes, the practice on which the Pilger process is based is carried out by a single pair of rolls, which are fitted with an eccentric groove, as shown on Fig. 2. The thick-walled hollow ingot drawn over a cylindrical steel mandrel or Pilger rod is introduced into the rolls against their direction of revolution. During about one- half of a revolution the two rolls with their eccentric groove force back the hollow bloom a certain distance, at the same time reducing its cross section, whereas, during the next half revolution of the rolls the feeding device attached in front of the housings pushes the block forward again, a little farther than it has been previously thrown back. A compressed air cylinder feeds the hollow ingot into the rolls, an air receiver taking up the recoil when 7 Finished or seized orily io (7 «Reduced in dhamefter mmp “edicedin dameter and wall Hickness , y a € a 7 \; re { , Z F V4 if { rill - co eeir > 22sbta ‘I In D WY Il Ion wW Gas Pipe BOILER TuBES Fig. 4—Steps Necessary to Produce a Finished Seam- less Tube by Each of Four Processes: 1—Mannesmann; lla—Ehrhardt and Wellman-Peters; I]I1—Continuous Mill, and IV—Two-High Mill. To go below 1% in. for gas pipe in process I the tube must be reduced in diam- eter, by hot or cold drawing. The same is true, below 2 in. diameter, in the other processes; below 1% in., these must have wall thickness, also, reduced by draw- ing. For boiler tubes a similar procedure is followed, as shown the blank is pushed back, while a long hydraulic cylin- der moves the whole feeding device forward by steps, corresponding with the advance of the ingot into the rolls at each revolution. The front end of the piston rod of the feeding device bears a long pitch thread. In going through a bearing with a ratchet, this turns the hollow ingot 90 deg. at each forward movement, thereby insuring a uniform working of the rolls on all sides of the tube and consequently uniform wall thick- ness and roundness of the tube. By this alternating forward and backward “Pil- grimage” (hence the name of this process) of the hol- low ingot, it is drawn out to a smooth, thin-walled 176 THE IRON AGE February 18, 1926 rative ttle of it sidering the complaints now raised by seamless tube mill men, no doubt presents increased difficulties. As solution of this problem experiments are being made Piercing Press and Push Bench with “expanding mills,” which are machines built som« ‘ what on the principle of a conical roll piercer Ir Ross : m4 ’ Expanding Large Diameters een four It is claimed that, by pilgering a tube with a wa given in Mr thickness above that finally required, reheating it and ou an 0 putting it through the expanding mill over a specially g, the diameter can be increas t is claimed also that the sam arranged 1 nrocess could be done twice, for instance rolling a ] 1 ¢ayr ; ] ed tapered pl u ime as Ila, but as much as 4 it I n. tube and expanding it to 16 in., and this again t gh one die i 0 in. It is natural to expect that such a process wil ibject the metal to enormous strains and the slightest r uli tl efects in the steel would be enlarged, seams opened uy te Results of experiments with these machines w ) SII i Continuous Mill ( xtreme! interesting. In regard to the stated limit of 1% in. diamete: é thi l hould be modified, as the smallest tubs ame ch wer erved being rolled by this process we rOus mS 1% 11 Tubes made by this process are usually p1 ecreasing luced in lengtt from 39 to 50 ft., but can be mad n¢ evi lesired, tl being one of the adva ‘ j ‘. ' ‘ i LV« Ehrhardt and Wellman-Peters Process Not Economi Above 6 In. Diameter i ( LI; ne mz ‘st tubes are 216 in. eter; the largest, according to the German aut! fer ! \ tated previously, however, it has not I { < pr¢ pey na 6 in. No tubdes ll thickne are made and the usual lengt t 0 ft. Within this range this type of plant ev 1 T competes successfully with the Pilger mi nnace nda economical operation. Process IIb is a devolpment of the former proc Pubs ip t n. outside diameter and 6 in. wall, u verage 18 ft., are now produced. Hig Ire § ntainers and similar articles are usual ( method out 23% in. to 4% itside diameter and by process 1V from about 2% 614% in. outside diameter are produced. Wall thi es of such tubes are not below % in. and not ove 16 in. normally, and &% in. maximum. The rollins ul I] engtl ! rocess III, usually about 23 ft., and Exceptionally Large Diameters Made irge | vy seamless bodies such as boiler shell irbdine j is { fron 12 n. to 11% fit. i Sieac hy Vari ns Pnnniaces l ims, 7 “9 I m 1 2 oa. ” A ‘7s ter, with wall thicknesses from 7/16 in. to 6 ! ngths, not exceeding 11% ft., can be p! roc V. bis i a diagram indicating the steps which have P MI e taken to produce a finished seamless gas pipe or e, after it has gone through the various pr ‘ e! a Vi ‘ pes made by the Pilger proces [) can be considered finished as pilgered dow ¥, ut 1 For smaller tubes a reduction in dian Nnece iry This can be done down to a certall umeter in a reducing mill or to the smallest dian e or eters by hot or cold drawing. Tubes made by processe I] III and IV must be reduced in diameter from about 1ec , , meter down, and on tubes below 114 in. the wa : kr also, has to be reduced by drawing oiler tubes can be finished rolled in the Pilger n , Vn 1 1% ! Vin Réber’s article claims 32 mm ; dl ; (1% in.) as 1 wer limit but, as stated before, ex- aa ; : erience has proved that the pilgering of tubes below 4 so ; 12 not practical nor economical and it is not ; ie ed ’ rolled boiler tubes are sometimes a pm ; ed through sizing rolls for exact diameter. In all hic] a: stidiges rather a ther processes a reduction in diameter, as well as in ll thickness, necessary. I mM] tal I (To he concluded) j What Is “Combustibility of Coke’? Laboratory Conception of Little Value in Actual Blast Furnace Practice—Agreement on Terminology Needed BY RALPH HAYES SWEETSER RACTICAL blast furnace men are now at logger the rate of wing the furnace was deliberately slowed heads with many technical investigators over the jown so that the blast was at only 1 lb. pressure. The meaning of the words “combustibility 1] 1e test were published by Kinney in the June, Because of this misunderstanding, there is danger that 1925, issue of Blast Furnace and Steel Plant under ) - much misleading information will get into technica the title of “Combustior f Coke at the Tuyere Level literature regarding the combustibility of blast furnac¢ f the Blast-furnacs In the article the author said fuels. Already several articles have been } blishe qa i t eX t f penetrat f the con this country and in England which contain statement t , ra ire, it would entirely contrary to the general understanding of bla , nica setaaa ey "_ os : furnace men in this country and on the other side ee a ee The difference in interpretation has been carried int ee . the meetings of the iron and steel section of the Amer can Institute of Mining and Metallurgical Engineer Both imbus furnaces were in blast at the time ind the West of Scotland Iron and Steel Institute. ar the experiment, and the results of the investigation Combustion Zone with Normal Blast Combustion Zone with 1-lb Blast Pressure + wee Analyse of Gases at Tuyere Leve } i. American Rolling Mill Co., Columbu UI! . that . the Amount of Blast Has Little Effect on the t} we of Gases at Tuyere Li cu. ft. per min.; the make, 316 tons of ! : } a Zone vith Ver per day; the consumption of ¢ e (l t ; we s . per ton of pig . : . 1 4 ie fF the » of ft combus has been discussed in the trade journals in this cou 1e heal é he combu and in Great Britain. At the February, 1924, meetings ! né and tl nt at which the combustion of the f the American Institute of Mining and Metallurgica e wa mplete ( med up by Kinney in his Engineers, the discussion of the top becams Var? mmary and co! I as toliov that Arthur G. McKee, blast furnace engineer, Cleve the extent of penet and, said, “It is evident that we all need to get tne se level of the b right definition for the combustibility of coke, b Is¢ ep pon the an we are not all talking about the same thing.” In July 30, 1924, issue o f Coal Review 1s printed T} ! ] right, according to the charts erial No. 2604, Reports of Investigations, Bureau of Nos I ), and e! to accord with other experi Mines, by T. L. Joseph, associate metallurgist he ent f 1 Bureau of Mines and actual practice. The ubject is “Combustibility of Coke and Rat f ( nclusio1 é , is contrary to that of Joseph, who bustion.” In the article the author stat , as quoted a ( “A fast-burning coke will burn Tt istil ‘ f 1! than a slow-burning coke.’ hose ] t 0) e recent articles published on the combusti he combustlor zone The rat ty f ‘ n tne ist furnace, the one that seems f i tt tuyers t fur ¢ re t ( t I 1 ( ne IS1OT tnat of Daniel Sil ; ou , ire entitled “Recent Views of ms 1 aa ' ) I it e Functior before the tl irty-second ses rates determined by the supply of oxy: f West of Scotland Iron and Steel Institute at an the Royal Technica] College, Glasgow, Scotland, Feb Supply of Air and Size of Combustion Zone . i adil tblished in Vol. 32. Part IV S. P. Kinney, Bureau of Mines, has gone far irna i the yu tute On page 56 of this i determine that the size of the mbustior ne Mr. Silla: juoted I N “Combustibility 4 front of the tuyeres of a blast furnace does not vary ne e, a measure of the speed of the coml nation : much under changing conditions in the amount of win gen and carbon to form carbon monoxide 4 1 099 r ay na ‘ ‘ “a nee In May, 12 2 Kinney oe , a ests of Finely Ground Fuel Give Unexpected Results tests at the two blast furnaces of the Am in R f Mill Co., Columbus, Ohio, and determined t! Z4 Another report f1 Great Britain published in the combustion zone in a furnace, working und normal September, 1925, issue of Fuel in Science and Practice pressure and normal volume of blast, and then wher eems to prove that anthracite has greater combusti- ty thar ke At the annual meeting of the Society *Assistant to vice-president \ I R . ~—_ f Chemical Industry, E. C. Evans presented a paper Le EAI» ! 478 THE test of the rez ities -oke de- giving reports on a test of the reactivitie¢ of cok vised by the Federal Research Board and the National Federation of Iron and Steel Manufacturers. A brief lescription tk é state xid i ( Aft “ pot: ( > oe ¥ - cari I 1 wi l } be 4 re ill st 1d) f this test will how exactly Ww hat ” © set nye « between the technical sa t I biast nace men Laboratory Tests Not Conclusive When a blast furnace man talks about the combus it e refers to the rate of burning, at the tuyer he particular coke which he is using, and he is not concerned with the rate of burning of that part ( cC ean ¢ laboratory after it is ground to pass through a 20-mesh sieve. The laboratory experi- ment presupposes preparation of the fuel, so that there will be intimate contact between the finely ground fuel and the gases or the oxygen which will react upon the of the fuel. On the other hand, the blast furnace man has coke f certain chemical composition, size, structure, porosity and hardness, the hardness sometimes having the addi- tional variation of a dense “skin” on the surface. The last furnace man has to burn the coke under certain blast pressures and blast temperatures, and he has to flux the coke according to the amount of ash and sul- yhur in it. He has to handle the volume of blast ac- ‘ording to the rate at which the furnace can be driven, and this rate of driving is sometimes restricted by the carbon ‘ombustibility of coke. If the coke is very combustible, then the maximum limit of the driving depends on whether hi keep the furnace full and whether he an take care of the products and by-products of the lurnace Laboratory and Furnace Conditions Not Comparable The technical Tg y r WU +} f\y lx Alife’ Witfl AiTi©@ Ly research man in his laboratory is ground fuels, and with very small Internal Structure of Steel Knowledge of How to Control These Changes » Sought in Cooperative Work Ir order te develo; a litable combination of trength and toughness in steel for severe service, as airplanes and automobiles, it is heat treated by juenching and tempering processes which alter the in structure of the metal. The more that is knowr ibout the mechanism of this alteration in structure. the more intelligently can heat treatment be yrding to the Bureau of Standards very slowly through the temperaturs it which the structure changes and observing the elec- trical resistance of the steel at different temperatures t is possible to follow greater detail than when the usual method of “thermal analysis” 1s applied. . ternal applied, ac- By heating this change in Such an application of the resistances made by the method late of the has heen research American assoc IRON AGE February 18, 1926 quantities, and he has the added difficulty of having only a certain amount of fuel to burn. On the other hand, the blast furnace man no sooner burns a certain amount of fuel in front of the tuyeres than an equal amount of fuel flows into the combustion zone, is burned, and passes on to the interior of the furnace. It is impossible to compare the two sets of results. Practical blast furnace men appreciate the research work done by technical men, especially those of the United States Bureau of Mines. In some respects, the work recently done constitutes the most progressive in- vestigation that has been made since the test of Sir Lowthian Bell, but further results of laboratory re- search should not be reported in terms that are con- trary to those used in the practice of making pig iron. This article has been written with the aim of bringing practical men and technical investigators in this coun- try and abroad to an agreement on terminology in dis- cussing the combustion of blast furnace coke. Physical Qualities of Coke Affect Combustion H. A. Brassert, Chicago, was probably the first man to investigate the combustibility of cokes. It was first brought to his attention in 1906 (year book American Iron and Steel Institute, 1914, page 27). Brassert says: is the which depends chemical analysis as the physical qualities of the coke It is this rate of progression What concerns the blast furnace rate of progression of the principally combustion, not so much on the that we term combustibility, which is the speed which the carbon molecules in the coke combine with oxygen under given conditions. At the February, 1924, meeting of the American Institute of Mining and Metallurgical Engineers, in discussing the paper of P. H. Royster and T. L. Joseph on “The Effect of Coke Combustibility on Stock Descent in Blast Furnaces,” the author suggested a modifica- tion of Brassert’s definition of combustibility of coke. This is given on page 233, Vol. LXX of the transactions of that society. The writer now wishes to make some slight changes in that definition and would suggest the following wording: The combustibility of a rate blast furnace coke is the gasification of that particular coke in front of the tuyéres of a blast furnace under stand- ard conditions of blast temperature and of blast volume of complete It is not sufficient to compare the combustibility of different cokes in a laboratory when ground so fine that all the particles are easily exposed to the hot air or hot gases. The combustibility of different cokes must be compared under the conditions which exist in those cokes when delivered to the blast furnace. The two charts shown prove conclusively that the size of the combustion zone is not dependent upon the quantity of coke consumed in a given time. Unfortu- nately, there is not at present a means of measuring the exact amount of coke passing through this combus- tion zone, but it could be approximately calculated. Society for Steel Treating, stationed at the Bureau of Standards. This work corroborates the previously known fact that the elements silicon and manganese, both of which are present in varying amounts in steel according to the properties desired, have quite differ- ent effects upon the temperature at which the structural change takes place. It has been known that manganese lowers and silicon raises this temperature. The re- sistance method shows this and also shows that, in- stead of occurring at a single definite temperature, the change occurs over a definite range of temperatures, ind that the range due to manganese can be altered by previous heat treatment tending to make the steel more or less homogeneous in composition, while that due to silicon is not thus altered. Steel sufficiently high in both manganese and silicon shows by the re- sistance method two separate and distinct ranges of temperature at which the structure of the steel changes while the constituent which the metallurgist knows as pearlite is being brought into solution into another constitutent known as austenite, as the steel is heated. SVS wwseerer wwe TS Sse C:C:CrOrmrmCmCmm a ne oe ae, le ele tn, Refractories at Mellon Institute Work of the Fellowship in Research and in Testing —-Consumer and Producer Both Benefited —Notable Publications BY M. C. HE refractories fellowship was established at Mel- [Pin Institute by the Refractories Manufacturers Association in 1917 and has grown constantly in the amount and scope of its work since that date. Ray- mond M. Howe, whose pioneer work in the refractories research field earned for him a wide and enviable repu- tation, was the first incumbent of the fellowship, and he continued in charge of the work until 1923, one year before his death. Until the establishment of this fellowship, there were few attempts by manufacturers to control the quality of their raw materials and products, because of the lack of testing methods and the expense involved in setting up individual laboratories. It soon became evident that the establishment of control tests was an important part of the work to be accomplished and the efforts of the fellowship were largely concentrated in this direction. As confidence in these tests became established and the value of the results became known, the volume of such work increased rapidly until it was necessary to increase the personnel. This expansion has continued until today the testing division alone employs six men and a wide variety of chemical and physical tests are carried out. Research and Testing Work That research work has not suffered unduly by the attention paid to testing is evidenced by the fact that no less than 33 investigations have been completed and the results published. During the early life of the fellowship these investigations were largely concerned with the development of suitable tests and, while there is much that remains to be done on these tests, the information obtained is invaluable. The knowledge gained by the fellowship from tests made on refrac- tory materials and products for individual manufac- turers was a great benefit. It has also frequently been the case that information from tests has indicated the need for research investigations along other lines, and so it has served several purposes. It was to be expected that the problems investigated would be those of the manufacturer rather than the consumer, since the former provided the necessary funds for the work. The manufacturers are to be com- plimented, however, in taking an unselfish attitude toward the work and allowing investigations to be con- ducted that were only of indirect benefit to themselves. A study of the list of published articles, which is ap- *Senior Fellow, American Refractories Institute Fellow- ships, Pittsburgh. BOOZE* pended, will show that a considerable number of the investigations were of this nature. With the growth of the fellowship and because of the confidence placed in it, the scope of its work nat- urally became broader. Technical problems arose which were referred by the manufacturer or consumer to the fellowship, complaints of peculiar nature were investi- gated, new products were developed and data requiring peculiar tests were requested on specific brands of brick. Until recently there was no sharp separation of re- search and testing work, with the result that the re- search investigations were not pushed rapidly to completion and suffered whenever the volume of testing work became unusually heavy. However, when the fellowship was transferred from the Refractories Manu- facturers Association to the newly formed American Refractories Institute, it was reorganized to include research and testing divisions, each to be operated separately from the other. The complete organization of the fellowship is shown on the preceding page. Work of the Testing Division The testing division is self-supporting through charges that are made for individual tests, these charges being adjusted as nearly as possible to just meet the expense involved. In the event that there is any surplus, the balance will be used to help defray the expense of the research division. The tests that comprise most of the work done by this division are complete and partial chemical analyses, fusion tests, load tests, spalling tests and reheating tests. In addition to these, tests are frequently made for cold strength, porosity, absorption, specific gravity, coefficient of expansion, slag penetration, screen an- alysis, rate of shrinkage, softening point of slag-brick mixtures and physical analyses of refractory mixtures. Requests have been received for data on relative ther- mal conductivity, diffusivity, hot transverse strength, chemical solubility, disintegration by carbon monoxide and a number of other properties requiring the devis- ing of new testing methods and special apparatus. In several cases the testing division has been called upon by individual manufacturers to develop products for special purposes with a fair degree of success. Another phase of the work, included under testing, is the investigation of complaints by consumers and visits to plants of manufacturers. In a number of such instances the fellowship has been able to be of material benefit through definitely locating the cause of the complaint or the manufacturing difficulty. In other cases the fellowship has been able to determine the wi PT MOU nl Organization of A. R. T. Fellowship DIRECTORS, RESEARCH COMMITTEE MELLON INSTITUTE Am.Ref. Inst MC.Booze, Senior Fellow HEMICAL DEPARTMENT PHyYsicar DEPARTMENT Andrews, Student Asst. C.G.Denney,JuniorFellow %.Mck.Swain, Junior Fellow W.R.Kerr, Junior Fellow K.W.S. Smith, Student Asst. C.S. ADVISORY COMMITTEE Am. Ref. Inst. RESEARCH DIVISION S.M.Phelps, Junior Fellow H.F. RobertsonJunior Fellow A.C. Hughes, Student Asst. . =~ % . shy , LR THE IRON AGE February 18, 1926 You Should Know About the Technical the Refractory Manufacturers Associa- > ries Kur e Refractories R. M. Howe, July reve at e f Hand-Made Fire Brick August I re I f Fire ¢ y Refra } ; . i t Interpretation of Results R. M r Electric Furnaces.’’—R. M. Howe Rotary Cement Kilns.”"—R. M i he tesearcl Work YT) } on y ‘ Data « th Disintegration of Blast ey I Refr tories.”’ R. M. Howe, December, 1920 es , . Ss i Spalling.’—R. M. Howe and R. F \ few r 1921 I | ( J. Spotts McDowell and + + ’ I J . 192] y wh L sé Plastic Clay Grog in Preventing , Sears M. Hows nd S. M. Phelps, February 1 I I fc S 1 Brick February ive I B Mad from Ganister, Flint é nin¢ ( Mixtur with Special Refer é na there R. M. How i Mark Sheppard : vel n Tables May, 1921 : a ie rmation Regarding Fus ( M »9] e rt f Diaspor, Bauxite M. How i | F. Fer n, N 7 . \\ 1} t Strength of Refractor n I M. How S. M. Phely nd I I Fergu r I | } , l I n the Characteristics ? nore fs } | M. H ind W R. Kerr, Feb earch d ’ f I Bri larket n Latin America.”’—1922 VI Institute Heat ! n, with Special Reference to the e for the fellow St Fired R. M. Howe and S. M. Phelp ¢ i t uid , [Tr + + I ctories R. F. Fer ore eT ne ‘ ; g r ] . . . ; , A Stu the Slag Test R. M. Howe, Febru e! ! l ! nrterence Wwitn an ary 1992 ; sen Fresh Lump, Air-Slaked and Hydrated ; ; ' Lime the M if ture Silica Brick.’’—R. M. Howe S. M ps, Februa 192 New Tests and sSpechications ; B oO Fir B k.’ ‘ I tne ourning I lay “ire rich , ¢ . Pe th. 1 ‘ ‘ ¢ MeOH larch. 192 ymmittee that tl} fellows} ip Importance of Determining the Burning Character udy of the tests for refractories. t f Ref: tory Clays M. C. Booze, July, 1923 ‘ I make re I r devise new experiment Made to Determine the Value of the relial nf t lata ¢ Pat ed Rebuffat Process for the Manufacture of 1 ri€ I r! i I ind data tnat : s 5 : Si I I M. C. Booze, July, 1923 ‘ f i I if VO! nas already : : : ‘ } . Pr a Use of Pyrometers in Burning Fire Brick M. ‘ | ‘ put t ] iearly ent tr 1 Ve I T } Pa Cally ¢ I Zz December, 192 = _ ; les. t , Checker Brick for Resisting Alkaline Slags WV ne tl test nave reacned tne ptal tage ui. I Z J i 1924 ’ : au rmation has beer ptained t the Deformation under Load of Fire Clay Refrac re I a avallable product n the tests, it will ries M. C. Booze, July, 1924 é Write specificatior r those classes of “Report of Tests on ‘Faulkner’s Bond’.’"—M. C é é ynd ns al irate | Vv} It I J 1924 y more important ser e could be rendered Effect f Grind and Burn upon Shrinkage and ndustri¢ interested in retractori than carry Weight of Fire Clay Mixtures.”"—M. C. Booze, July kind of work ar ( eratior f al] o% nest] icited Fire Clay Brick for the Open-Hearth.’ M. ¢ ' Booze, September, 1924 the Ret é Fellows} “Discussion on the Disintegration of Clay Refra es in Iron Blast Furnaces.’’—S. M. Phelps, Septem > si I LJoF “Some Aspects of the Industria] Application Refractories.’ M. C. Booze, 1924 hs reaaiegs Report on Some Refractories Plants and Labora me . Visited in England, Scotland and Germany.” M. C. Booze, December, 1924 Common Sense and Common Refractories.”—M. C Eff: f Red Hearts in Fire Clay Brick.”—M. C. “Clay Products M. C. Booze, February, 1925. \ Study of the Factors Involved in the Spalling f Fire Clay Refractories with Some Notes on the ad and Reheating Tests and the Effect of Grind on I paration and e of Shrinkage M. C. Booze and S. M. Phelps, February, a a ee ae February 18, 1926 THE IRON AGE 481 “The Relation of Structure and Composition t “Research Possibilities in Refractories."—M. C. Thermal Efficiency of Refractories When Used in Re Booze, 1925 generators.’’—S. M. Phelps, April, 1925 Note—Limited quantities of reprints of some of these “The Chemical and Physical Properties of Fire ticles are available and may be had by addressing the Clays from Various Producing Districts.’ M G American Refractories Institute, 2202 Oliver Building, Pitts- 3oo0ze, October, 1925 irgh Scrap in Blast Furnace Burdens Lower Coke Consumption and Higher Output, Says a German Investigator—Effect on the Pig Iron N exhaustive discussion of the effect of scrap in the blast furnace charge is contained in two re- cent issues of Stahl und Eisen (Dec. 10, Dec. 17, 1925). It is by E. Bormann at Hoerde, Germany, and s taken from his doctor of engineering dissertation, presented at the University of Berlin. 3efore the war the use of scrap in blast furnace ‘harges was infrequent. During the war it was found necessary because of the lack of foreign ores, and in some plants it was carried so far that only enough ors + + _ J | | ant j / a a a ae 50 Ot QO ¢ v ' cv - a A sami wen Claiwend ervenrT i ry NAaATG Fig. 1—Coke Consumption and Daily Output with Varying Charges and flux was used to give slag volume sufficient for yperation. This complete change in the character of the charge had a great influence on operating conditions and results. Above everything could be noticed a saving in fuel and a considerable increase in output. When Making Bessemer Iron In Table I are shown results obtained on a furnace under the same operating conditions, while making Bessemer iron. Table I—Effect of Scrap While Making Bessemer Iron Iron Iron Daily in Ore inScrap Coke, Lime- Output ———————__—_—., Consump-__ stone Slag, it In the Charge tion, Used, Weight, Metric fest PerCent PerCent PerCent Per Cent PerCent Tons 1 100 00.0 120.0 50.0 n.d 137 2 64.5 35.5 117.0 56.5 48.0 152 ; 43.5 56.5 91.5 44.7 48.0 180 40.8 59.2 89.0 41.8 47.0 189 39.6 60.4 85.0 39.7 46.9 198 6 30.0 70.0 80.0 37.3 42.0 202 7 25.0 75.0 83.5 46.0 42.0 222 S 21.7 78.3 93.9 44.8 41.0 196 ) 19.0 81.0 95.5 44.2 42.0 212 Test No. 1 shows the results with an ore charge used as a basis of comparison. Then follow charges containing varying amounts of scrap. The scrap was taken at 95 per cent iron in all cases. The coke ana stone consumption and the slag weight are figured against the ton of pig iron. The results are strikingly shown in Fig. 1. The lowest coke consumption is shown when 70 per cent of the iron in the charge came from scrap. With more scrap than this it rises. The largest daily output was given with 75 per cent of the iron charge from scrap, an increase of 62 per cent com- pared with an all-ore charge. The figures in Table I, regarding slag volume, are also very interesting. The next part of the paper gives carefully worked out heat-balances on five blast furnace runs, the iron in scrap in the charge rising from 0 with No. 1 to 59.8 per cent in No. 5. Unfortunately it was not possible to have perfectly comparable conditions. The furnaces were different and operating under varying conditions. Test No. 1 was on basic Bessemer iron, the others on Bessemer iron. In the consideration of results the variables were harmonized as much as possible. Table II shows some of the results, taken from the complete heat balances. The weight of charge per ton of pig iron steadily Fig. 2—Influence of Increasing Scrap ( harges on the Carbon Monoxide and Carbon Dioxide in the Gases dropped as the scrap increased, the total decrease of No. 5 compared to No. 1 being 33.5 per cent. The author discusses slag volume and reaches the conclu- sion that in no case should it be less than 40 per cent of the weight of the pig iron; that is, for each ton of pig iron the weight of the slag should not be less than 0.40 ton. He next discusses the gases, shown diagrammati- cally in Fig. 2, where it is seen that the carbon-dioxide Table Il--Results from Heat-Balances on Five Furnaces Test Ni l 2 3 4 5 Lengtl f test in days 7 6 6 6 6 Fur capacity, cu. ft. 22,530 16,000 13,243 16,000 16,000 Hearth cross section, sq. ft ere 1,636 1,033 872 1,033 1,033 Pig iron produced: Carbon, per cent.. 3.09 3.46 3.34 3.65 3.51 Manganese, per cent 1.86 1.05 1.05 0.87 0.81 Silicon, per cent 0.23 2.15 2.02 2.29 2.95 Phosphorus, per cent.. 2.67 0.112 0.112 0.106 06.104 Sulphur, per cent 0.10 0.03 0.03 0.02 0.026 Iron in ore in charge, per cent és . 100 65.4 64.6 42.3 40.2 Iron in scrap in charge, per cent .. ; 0 34.6 35.4 57.7 59.8 Iron in the scrap, per _ cent ; : 94.0 92.6 96.0 96.0 Slag per ton pig, tons 0.84 0.64 0.565 0.41 0.39 Gas analysis, vol. per cent: Carbon-dioxide . » Bee 7.16 6.70 5.09 4.95 Carbon-monoxide 28.7 30.51 31.90 2.36 32.92 Hydrogen 2.8 5.41 5.31 3.96 4.0 Methane . 0.20 0.18 0.43 0.33 Nitrogen 56.7 56.74 55.91 58.16 57.80 steadily decreases and the carbon-monoxide increases. With 60 per cent iron from scrap in the charge, the CO, in the gases has dropped 58.5 per cent from the highest value. The increase in CO naturally brings about an increase in heating value. The author next discusses exhaustively the various THE IRON s of the | ilances and then passes to a care- nsider gas samples taken from different he ! ew arr it Test No. 3 of I] The results are rather theoretical but of ere Advantages and Disadvantages i? t it tne 1IS¢ f£ cx nsiderable 8 ng n the furnace shaf ind throat tem- ! n and higher wa he use of scrap wered about itput reaches 60 per great importance February 18, 1926 AGE in regard to cost of production. Further, the greater fuel value of the waste gases and the larger excess of energy thereby available is profitable. The disadvantages are chiefly: The difficulty of loading and handling the scrap, which brings about higher labor costs, particularly with light scrap such as turnings Greater loss of heat and use of cooling water, due the higher shaft and throat temperatures, Changes in the composition of the pig iron, due to uniform scrap Lower total carbon in the pig iron, which is, how- ever, not particularly noticeable. These disadvantages make the blast furnace opera- tion more difficult, but seem to be outweighed by the advantages. G. B. W. Phosphides in Manganese Steel Effect of High-Phosphorus Ferromanganese—Long Continued Annealing Recommended as a Cure I f Stak ind ) P} sda it : . E. Piwowarsky } rr ,y f I 2 ] len i iimits er cent, manganese i os ceé t Ktched ' I ) Phot 1 ne 2 (R gh t)j— 1 in Fig 1 Etched wit} 0.06, and silicon 0.20 to 0.40 per cent. There is usua no difficulty about meeting the sulphur requirement lx cause MnS, which is produced, slags off, but sometimes hard to keep the phosphorus down for ferromanganese available frequently runs 0.30 per « The examined a_ high-phosph steel which did not Its ar the author and ve ana ove! give satisfactory results. } gra ph Sod ate ~-¢ > v « re ; . Fig. 4—About ; F the Same Lo- . 2 more cation as Fig _ r 2 3% tA - ’ we BK : 3 1 and Etched . wages Cee with Picri¢« a . Acid a ~ ’ of > ¢ . snpeneane se . + 22d ae mens 3 Py February 18, 1926 ysis was carbon 1.39, manganese 14.2 and phosphorus 0.13 per cent. The structure of the steel, as cast, is shown in Fig. 1. In addition to the austenite (white) and the martensite needles, there may be seen a eutectic structure between the crystals. A similar appearance is sometimes found in ordinary manganese steel, but in this case the im- pression of a ternary or quaternary eutectic is strong and it may be presumed that free phosphide as well as free carbide is present. In order to test for this, selec- tive etching was carried out by Matweieff’s method for phosphoric pig irons, using both alkaline and neutral sodium picrate. The results of etching part of the same area shown in Fig. 1 are given in Figs. 2 and 3, and in the original paper are also shown at higher magnifications. The presence of free phosphide as well as carbide is clearly THE IRON AGE 483 shown. It was assumed by this time that the free phos- phide was the cause of the trouble. After annealing a small sample for 15 hr. in a vacuum at 1050 deg. C. (1920 deg. Fahr.) and quenching in cold water, the surprising result was reached that neither undissolved carbide or phosphide was left. The structure was homo- geneous austenite as shown in Fig. 4. This work was repeated with another sample of the same steel. Free phosphide was found again. After annealing this sample for 5 hr. at 1100 deg. C. (2010 deg. Fahr.) and quenching in water, phosphide and car- bide had practically disappeared but fine inclusions were present at the grain boundaries. The author’s conclusions were that free phosphide might be present in high-phosphorus manganese steel but that, with long continued annealing, it would go in solution. G. B. W. Big Construction Year Forecast Building Shortage Not Yet Overcome, Together with Normal Growth, Assures Out- lay of Five Billions, Says Research Body A LTHOUGH the building shortage in the United States has been largely overcome, there is no rea- son for a marked recession in building, according to the Copper and Brass Research Association, 25 Broadway, New York. For 1926 another year of great activity can be expected. Probably $5,500,000,000 will be spent for new construction, repairs and replacements. Indica- tions are that the next two years will see some $10,- 000,000,000 spent on buildings. During 1925 $5,792,000,000 was spent for construc- tion. For the third successive year, building in the United States exceeded $5,000,000,000. Estimates pre- pared by the association show a total construction ex- penditure of over $17,000,000,000 for the three years ended Dec. 31, 1925, as follows: 1923, $5,923,000,000; TN REND of Building in the United States, 1910 to 1925. Normal yearly growth represents the building requirements which accumulate an- nually because of increased popula- tion, and repairs and replacements. The normal growth line, being in terms of money, is necessarily affected by irregularities in the buying powe? of the dollar. The shortage of build- ing at the beginning of each year, as shown in the curve, represents the difference between the total building requirements and the total building expenditures for the preceding year. It will be noted that the shortage built up during the war years has not been nm wrt wo BILLIONS OF DOLLARS 0 entirely wiped out 1910 1924, $5,341,000,000; 1925, $5,792,000,000; total, $17,- 056,000,000. The year 1925 almost equaled the record of 1923 when practically $6,000,000,000 was spent to catch up with the shortage which the war years brought. Normal Amount Placed at Over Four Billions Careful study of all available data and estimates by others set the normal volume of the country’s build- ing at $4,400,000,000. Of this amount $2,500,000,000 are for some 375,000 new housings to meet increasing population demand. The remainder is for construction that follows as a logical step in the growth of cities and communities, such as schools, churches, stores, hospitals, theaters, etc. The curves on the accompanying chart show the trend of construction for the past 15 years and give some index of future. As the building shortage is not entirely overcome and the normal growth shows no indication of receding from the present high figures it is apparent that our annual building program for some years to come will approximate $5,000,000,000. Present Yearly Requirements Five Billions The annual requirements for the country, at its present state of prosperity, can be figured approxi- mately as follows: 375,000 housings $2,500,000,000 ( } net? ¢ 1,900,000,000 Depre itior repair replacement 800,000,000 $5,200,000,000 Once the building shortage is overcome the annual needs will be almost identical with the normal growth. I9I5 1920 1925 The volume of requirements, as measured in money, will be affected by fluctuations in the value of the dol- lar, but at present values, an outlay of some five bil- lions yearly seems assured. Complaint Against Midland Steel Products Co. Dismissed WASHINGTON, Feb. 16.—The Federal Trade Com- mission has announced dismissal of its complaint against the Midland Steel Products Co., Cleveland. The complaint had charged the co