The Iron Age 1933-03-23: Vol 131 Iss 12

Teadle, tion is farting m the 1g tool it play mater- jioning mn this to the turing at re- n de- e dies le and press- intage 2rmits oper- eater, from ocally then ‘azine lanks ed or ; con- iction thout lock | the neta] ilver com- l are that 1 the nery pints nter- y re- out ec are rac- even y of rica- side izes end- cial for ani- all 385 1SS., rear ‘ing ical ince in has sive ily, om- rer. and ..THE IRON AGE.. ESTABLISHED 1855 MARCH 23, 1933 Vol. 131, No. 12 Rustless steel, aluminum and other resistant metals will compete with wood for the estimated domestic market requirements of 12,000,000 kegs. These are rustless steel drums used in Germany. ETURN of beer to legitimate marts of trade promises more than first aid to unbalanced Federal budgets and thirsty citizens. Many industries expect to benefit by the revival of an old industry, chief among them the producers and fab- ricators of steel, non-ferrous metals and alloys. Metal has long been an important feature of the braumeister’s art; pos- sibly it should be amplified to “metal and cleanliness.” A pamphlet en- titled “Treatise On the Brewing of Beer,” published in London, 1878, de- votes many pages to advice on keep- ing utensils clean. “The copper,” it is pointed out, “should be cleaned after each brewing as it will keep it bright, when it is used but seldom.” The “copper” refers to the large kettle in which the wort is boiled, and which then, as now, in most brew- eries is of sheet copper. Even today, reference to these utensils is to “coppers,” for that metal and a few iron tubs used in mashing and sparg- ing the malt were formerly about the only metals used in brewing. In the past decade, or more, how- ever, the ancient art of the metal- lurgist has been contributing new materials for the equipment needed by the equally ancient art of the braumeister. Rustless steel, nickel and its alloys and aluminum have joined copper and ordinary steel in European and other foreign brew- eries, Equipment and modernization of breweries in the United States are ex- ~ What Beer Means to Our Metals Business By GEORGE S. HERRICK pected to require many millions of dollars, but much of this expenditure, according to the brewers, themselves, will be from earnings and in conse- quence, spread over a period of years. However, absolute essentials of im- mediate increase in output promise to furnish a _ substantial market for equipment, and, in general, equip- ment in the brewery means steel, non-ferrous metals and alloys. Large capacity tanks and vats com- prise the major part of the equip- ment, and while considerable quan- tities of wood were used in the past for such purposes, the trend in brewing practice has been definitely a mm ANY new metals and alloys have been developed and have gained industrial acceptance since the enactment of prohibi- tion. With the resumption of our brewing industry, following the legalization of beer, there will be extensive opportunities for plant modernization which will take ad- vantage of these interim develop- ments. Some of these are indi- cated by current European prac- tice in those countries where brewing has not been interrupted. This article outlines some of the possibilities and probabilities of metals consumption in America with the return of beer. vvyv 46l toward use of metals, such as stain- less steel, nickel and its alloys, alumi- num, gilass-lined steel, copper and metal-lined concrete tanks. Copper is most widely used of all metals for the cooking kettles, because of its desir- able heat conductivity. The metal requirements of the brewing field are apparent from a few statistics on its pre-prohibition and post-prohibition condition. In the decade preceding prohibition, there were 1700 to 1800 breweries operat- ing in the United States, the average being 1307, with the peak reported at 1847 in 1915, and the low for these ten years at 669 in 1919. In 1914, according to Government statistics, the brewing industry rep- resented a capital investment of $858,861,000, and the average annual production in that decade, 1910-1920, was 57,308,476 barrels of 31 gallons each. Today it is estimated by brewers and engineers in contact with the field that there are not more than 200 breweries available or operating under license, to which must be added an unrecorded number of illegally operated plants, most of which, how- ever, are inadequately equipped and producing so-called “cold-water beer.” Based on these records of past per- formance, R. H. Huber, vice-presi- dent of Anheuser-Busch, Inc., St. Louis, and _ vice-president of the United States Brewers’ Association, makes a conservative estimate of 40,- 000,000 barrels annually when beer is again legal. This is predicated on Fifty thousand lbs. of copper were required for each one of these four brew kettles with their steam and other ducts. This installation is in the Jacob Ruppert brewery in New York 66,105,455 barrels produced in the peak year, 1914, a 24 per cent great- er demand through increased popula- tion (Bureau of the Census), or 81,- 970,764 barrels, halved by an esti- mate that the industry will revive less than 50 per cent in the first two years. So it would appear that the brew- ing industry may to a great extent fulfill many of the expectations of metal producers, fabricators and builders of equipment for its use. 12,000,000 Kegs Needed Metal instead of wooden kegs are being used by certain breweries in Germany and other foreign countries, a development of the past few years, which promises substantial tonnage business in this country when beer is legalized. Even though the familiar wooden kegs of pre-prohibition days are gen- erally readopted, the cooperage indus- tries claim a survey indicates that 17,000,000 to 19,000,000 such kegs were in circulation prior to 1919, and that there are less than 200,000 beer kegs in the United States at present. Based on the fact that about 75 462—The Iron Age, March 23, 1933 per cent of all beer in 1919 was dis- tributed in kegs of 31, 15%, 7% or 3% gallons capacity, it is estimated that 40,000,000 gallons of beer annu- ally would call for about 12,000,000 kegs. With four steel hoops on each, there would be about 108,000 tons of strip steel used, not to mention steel dowel pins and rivets. Stainless steel lined tanks in a Now, if these kegs were to be of rustless steel, aluminum, nickel or nickel-clad steel, it becomes obvious that substantial tonnages of metal would be required, not to mention considerable activity on the part of metal fabricating shops. Brewery Tanks and Vats Large Within the brewery itself, however, there is much heavy metal equipment in addition to piping, cooling and heating coils, filters, skimmers and small yeast tanks. It is the installa- tion of this heavy equipment requir- ing the fabrication of sheets or plates, to which the attention of the metal industries is particularly di- rected as a prospective new field with tonnage requirements. Primarily, the heavy, large capacity units are the mashing and sparging tubs, fermen- tation vats, brew kettles and the stor- age tanks. In addition to the heavy equipment, there are, of course, the filling and bottling machinery, washing equip- ment for kegs, labeling machines and small utensils, such as pails and dip- pers and the table tops in labora- tories. It is in such small equip- ment, especially dippers, pails and similar objects, that the chromium and nickel plating fields may enter in supplying some material. Even the existence and continued use of old wood tanks are expected by metal producers to result in some business from lining the wooden bar- rels and tanks with sheet metal. Then, too, certain breweries have used concrete storage tanks, which are lined with corrosion-resistant sheet metal, when the tank is poured. In general, however, storage tanks con- sidered by brewers are likely to be of ordinary steel plate construction with glass lining, already in wide use; stainless steel, which has been found satisfactory in Germany; nickel or a nickel alloy, or nickel-clad plates, or aluminum. Both solid nickel and aluminum storage tanks have been used in England and on the Conti- nent. Nickel and nickel alloy, especially tiled brewery room in Dortmund s k t ~ —— SO, ae ve of 1 or vious netal ntion 't of ge over, ment and and alla- juir- or the di- with the the nen- itor- ent, and uip- and dip- ora- 1ip- and ium ‘in ued me ar- ee Rustless steel beer storage containers in a Danish brewery in the “clad” form, where 10 to 20 per cent of the total thickness of a steel plate will be pure nickel, rolled on, appears likely to be a consider- able factor in brewery equipment of the new era. Then, too, there is Monel metal, which has been used in certain brewery equipment abroad and been found desirable for piping and similar purposes. A possible new form of nickel for use in breweries is the alloy contain- ing about 81 per cent nickel, 12 to 14 per cent chromium and 7 to 8 per cent iron. It was developed primar- ily for use in the dairy field, but its corrosion-resistant qualities and strength are such as to suggest that it will prove adaptable to the brew- ers’ requirements. Copper Widely Used in Brewing Copper, along with iron, and later steel, was the original metal used in brewing for centuries, not only for the large kettles generally known as “coppers,” but in most other equip- ment where the liquid is in motion. Piping of various diameters to trans- port the liquid from point to point in the processes, the heating and cooling coils and many other units of equip- ment are of copper and continued wide use appears likely with the re- sumption of brewing in this country. It is noteworthy that in Germany, where other metals and alloys have found acceptance, the breweries are said by the Metal Research Institute to be consumers of 1500 to 1600 tons of copper tubing annually, merely for ordinary purposes of operation. That copper represents a substan- tial tonnage of metal in the average brewery is evident from the fact that a 700-gallon brew kettle with its steam and other ducts requires some 50,000 Ib. of the metal in sheet form. Depending upon the size of a brew- ery, there will be from two to as many as six such kettles installed, ranging in capacities from 200 to 700 gallons each. It is estimated that in four kettles and for the piping, cool- ers, filters and other purposes in the Jacob Ruppert brewery in New York there is close to 500,000 pounds of copper. After all, copper is certain to find wide continued application in brew- eries if for no other reason than that it has been used for generations and been found good and sufficient by brewers of the past. Aluminum Also Used in Breweries Still another metal has been found highly satisfactory for use in brew- eries—aluminum. Tests and long use in European breweries indicate that it has no effect on the taste, appear- ance, or other properties of beer, and its comparatively low cost is an addi- tional factor in its use. It requires no lining or other coating to prevent direct contact with the beer, either when it is in motion or in storage tanks. Aluminum equipment has been used by European and British breweries for long periods, according to the sellers, without appreciable deteriora- tion, and in certain instances it has been in service for more than 20 years. One of the principal uses for which aluminum is being recommended is in metal drums, the possible substitutes for the old-fashioned beer keg. As in the case of nickel and stainless steel, aluminum has been used by European brewers and has proved satisfactory in fermentation vats, storage, filling and water tanks, set- tling, pasteurizing, sugar cooking and dissolving apparatus, filters, ladles and gages. In the field of beer transportation, aluminum has served for metal drums and has been fabri- cated into railroad and truck tank cars and used for bottle crates and carriers. A fourth metal is coming into wider use today—rustless steel, the 18 per cent chromium, 8 per cent nickel alloy of steel. It has found its widest acceptance in Germany, but other foreign breweries have in- stallations of this material, and there have already been a few small uses of rustless steel in American brew- eries. The earliest tests of the high chrome nickel alloy steel were made by German brewers about 1921. Sat- isfactory results brought wider ap- plication and about 1925 the rustless alloy began to spread, not only in Germany but in German-made brew- ery equipment exported to overseas countries. It has been used for tubing, large clearing, mixing, filling and storage tanks, as lining in fermenting tanks, and in fact rather generally throughout the brewery. The produc- ers of the alloy are inclined to recom- mend it for any or all equipment. The storage tanks, which represent the really large tonnage use of metal in breweries, range in capacities in Germany from 2000 to as much as 40,000 gallons, with the average ca- pacity from 6000 to 8000 gallons. These capacities are comparable to the sizes of storage tanks in existent American breweries. A typical installation in a modern German plant of medium size in- (Concluded on Page 482) Outlets for metals will exist in the distribu- tion as well as the manufacture of beer. Monel metal drainboards, sinks and other sur- faces are seen in this bar in the Hotel Inver- may, Vancouver, B.C. The Iron Age, March 23, 1933—463 ROM time immemorial there has been no basic change of principle in design of sheet and tin mill housings. The same breakage troubles occur year after year, and still no one has come to the front with a well-thought-out cure. The contours of roll necks have gone through an evolution, but the designs of housings remain the same. No particular location on these hous- ings has been immune from break- age, and we have gone merrily on and just changed housings, replacing them with another one of the same faulty design. This procedure is accounted for more or less by no housing break- ing consistently in any one given loca- tion, and, while this fact in itself is very mysterious, it is highly ac- centuated by the fact that the break generally shows clean metal. As these breakages occurred, designers, from time to time, have added additional sections to the outside contour of the housing, but this has in no way proven a cure-all. Some years ago the author cor- rected roll breakage by taking out metal, and it will be shown later in this article that such a procedure will lead to a solution of this mysterious breakage difficulty in housings. It is to be noted at this point that the > ~= @ _ Fig. 1.—Sketch of roll housing showing irregular inner contour. 464—The Iron Age, March 23, 1933 By WILLIAM H. WARREN Vice-President, Lukens Steel Co., Coatesville, Pa. design of the housing as a whole has been influenced by these mysterious failures, and while it is entirely pos- sible that the housing as a whole is too strong, yet the housing locally is too weak. By increasing the local strength we are in position to de- crease the average strength, and thereby effect economical changes. Celluloid Models Used to Arrive at Rational Design Use of celluloid models, in an ef- fort to arrive at a rational design, cannot be too strongly recommended. A transparent celluloid model under stress will give an accurate picture, through the use of polarized light, of the stress distribution encountered in any given physical shape. The same celluloid model, when subjected to a certain technique, will very complete- ly show the behavior of the structure as a whole, and, what is most im- portant, will take into account what- ever rigid joints may exist. This is quite important, as it is well known 7-4 _ that the subject of rigid frames js extremely difficult to handle analyt- ically, Fig. 1 depicts the general manner of construction. The very irregular inner contour is the conventional method by which the brasses holding the rolls have been mounted. Fig. 2 shows a model of the same structure illuminated by polarized light when under stress. It can be seen that there are 14 possible places at which a fatigue crack may start. Measure- ments of these high local stresses show that the maximum stress may be as much as six times the average stress, and we are immediately con- fronted with the question that once having eliminated these high local stresses, how much can we reduce the weight of the structure as a whole. Irregular Inner Contour Eliminated Let us consider for the moment a frame or housing proper with this irregular inner contour eliminated. Fig. 3 is a photoelastic picture of this same housing with its interior profile smoothed out by eliminating the chuck mountings. It will be shown later that these chuck mountings can be produced, by means of welding, in such a manner as to make the inner Fig. 2.—Celluloid model of structure shown in Fig. 1, illumi- nated by polarized light when under stress. Welding in the Steel Industry\—F col sm pr cer in on av try —Rational Design as Applied to Mill Housings 5 1S lyt- ner ular nal ling pure hen lere 1a ire- ses nay age ‘on- nce real the contour of the housing effectively smooth. We are now confronted with the problem of analyzing this housing as a rigid frame. It is generally con- ceded that the ideal structure is one in which the average stresses at any one point are no higher than the average stresses at any other point. When we look through the various sections of an average mill housing, we find that the shoulder holding the housing nut is only one-fifth as strong as the gross section through the side of the housing, which is far from be- ing ideal. The glaring inconsistency in this design can be brought out at this point by noting that the shoulder holding the nut never fails, while the failures that have been experienced have all been through the gross sec- tional parts of the housing. It is, therefore, reasonable to assume that having eliminated the causes of the failures through the sides and bottom, as shown in Fig. 2, it will be possible to reduce the total weight of the housing considerably by proceeding with a free mind to design this hous- ing as a structure known to be free from local weaknesses. Due to the fact that the loads on a “jump” mill are very indeterminate by the very nature of the impact a tr tr HOTOELASTIC studies of stress distribution in mill hous- ings reveal local stresses that may be as much as six times the average stress. Adding metal to the outside contour has not cured the mysterious breakage. By eliminating local weaknesses, however, the service life of the housing may be greatly extended and the weight considerably re- duced. With a service life of eight to 10 years, the welded steel housing developed from these studies is a source of def- inite economies. This article, the first of five on the subject of “Welding in the Steel Industry,” by Mr. Warren and another authority, is one of the papers submitted for the sec- ond are welding prize contest sponsored by the Lincoln Electric Co., Cleveland vveyv loading and temperature effects, it will be impossible to easily predict what maximum load this structure will have to stand. It will be neces- sary, therefore, to be guided by past gram Fig. 3.—Photoelastic picture of housing with interior profile smoothed out by eliminating chuck mountings. Fig. 5.—Bending moment dia- as é a - wit é wry Fi) IC’ \ fil Y LIASTIC d, experience. Since we have eliminated localized weaknesses, we are in a posi- tion to compare the average strengths of various component parts with a view to finding the weakest average section which has stood up in service. If we now design a frame in which the stress at any point does not ex- ceed the rupturing stress of this weakest serviceable section, we have a rational method along which to proceed. Simple calculation shows that the shoulder on the nut is the weakest section upon which we can depend for average stress. We are thus confronted with the problem of producing a frame which will self- contain this minimum serviceable load. If we take into account, in the design of this frame, the fact that the joints in the final structure will transmit bending moment, we find ourselves solving a problem which is statically indeterminate. The solution of such a frame can be laboriously worked out by the use of Clapeyron’s Theorem of Three Moments, but it is much easier to make use of celluloid models, to which, through a micro- scopical study, can be applied Max- well’s Reciprocal Theorem. This theo- rem, and its application to celluloid (Continued on Advertising Page 10) L Gage /nserted maul At Reaction Foint in Question - Gage Produces A, M: croscope atload ore tr yp mo / a foint Reads d, Pid, (Maxwells Pe d, Reciproc a a. /heorem Fig. 4.—Application of Max- well’s Reciprocal Theorem to celluloid models (of structure). shown by celluloid model. The Iron Age, March 23, 1933—465 Mass Prod Large, specially built 12-stage milling machines complete the machining in one cycle. HE changes in design of freight ears and other rolling stock during recent years have caused many corresponding changes in the foundry industry. In general, railroad castings have moved up into the higher quality grades, with alloy steels and malleable predominating. Some of the parts have been rede- signed to use rolled steel shapes, and no single design today is permanently established. Journal boxes of many different types are in use, but pro- duction of the type to be described in this article runs into hundreds of thousands, making it possible to place its manufacture on a mass basis in line with good practice in most any other field. The plant here referred to was laid out with the definite prospect of pro- ducing established products and it was therefore possible to keep the path of travel of parts in process to a minimum. Storage bins served by a railroad siding are located close to the melting furnaces. Roller con- 466—The Iron Age, March 23, 1933 veyors, cranes, and tractors pick up the materials and carry them through the plant with but little manual han- dling. In the production operations specially designed labor-saving equip- ment and short-cut methods have been generously used. An interesting case illustrating this is found in the core room. Under full operation this department has pro- duced some 40,000 cores a day with a total force of core makers of 37. A special steel-top circular table is provided for each core maker and the location of these tables in respect to the ovens has been worked out to provide maximum convenience. Care- fully prepared core sand is delivered to the circular tables from overhead spouts. Each core maker receives his sand at his right, fills his cores in front of him, and then places his completed cores to his left, but in- stead of moving the cores away from him as he completes others, he him- self moves around the table to the right, using up the sand ahead of him uction Methods for Mallabl a» & & HE inevitable jolts in freight- car operation make it advis- able to use malleable and steel castings for couplers, journal boxes and many other parts. Hundreds of thousands of mal- leable journal boxes are required to fill the needs of American rail- roads. This high demand has made it possible to put opera- tions on a highly efficient quan- tity production basis. Some of the features at one large mal- leable foundry are described in the accompanying article. The description is of actual opera- tions, but the name of the foun- dry is withheld by request. vvryv and leaving behind a string of com- pleted cores. Before he makes the full circuit these completed cores are collected by a special handling gang using the rack car train shown in one of the illustrations. This train operates on narrow-gage tracks set into the concrete floor and serving all of the core makers’ stations. The tracks extend into the various core ovens so that, having once placed the cores on the trays of the trailer cars, it is unnecessary to disturb them until they are baked and delivered to the foundry floor. Long trains of six or eight cars of cores are shifted by means of small storage battery trac- tors, one of which is shown in the illustration. Reserves Held to a Minimum In the core department as in all other departments of the plant the stocks of material in process are held to the absolute minimum. A some what unusual policy has been devel- oped to make this possible. Each day, foremen on the foundry floor turn in requisitions for cores and other materials required on the fol- lowing day. These requisitions are accurate and are based on an intimate knowledge of the orders in process and the established rate of produc- tion. They go direct to the planning department where they form the basis of production schedules and of oF ders sent to the other departments. If a department, for instance, changes its product from a small to a large casting, the changed require ments for different materials are esti- mated and orders given to supply the new demand. It may be that an extra man may be required in the shakeout gang. More fuel, more ant om oo Ch Ba. Aalpble Castings reight.- advis- 1 steel ournal parts, f mal. quired n rail- d has Opera- quan- ne of mal. ed in The )pera- foun- st. f com- es the "es are y gang wn in ; train ks_ set erving The s core ed the r cars, 1 until to the six or ed by trac- n the n all t the » held some- level- Each floor and fol- | are mate ocess »due- ning asis o's ents. ince, ll to lire- esti- the an the nore By HERBERT R. SIMONDS scrap, or more pig iron may be needed. Perhaps the changed work will call for a larger supply of mold- ing sand, and the man in charge of the preparation and delivery of such sand receives his new schedule and adjusts his production rate accord- ingly. When the foreman of the core de- partment receives his output sched- ule he immediately adjusts his pro- duction to fit it, making allowance for any under- or over-production on the previous day. Variations in labor requirements are taken care of by shifting men from one department to another, and because of the large scope of operations there is sufficient flexibility to make the daily schedul- ing practicable. The core makers operate on a combination standard and bonus system of pay, the actual rate of bonus changing each month for the entire-department in accord- ance with variations in the percen- tage of rejections. On extremely small cores, girls have been found more efficient than men, and when the force stood at 30 core makers, 10 of these were girls. Material Handling Simplified Molding sand is handled and con- ditioned by a night gang. A machine cutter is used for the sand along the uninterrupted aisles of the main floor. At other places cutting is done by hand, and the actual conditioning Flasks are of uniform design with lugs for handling and rims to take clamps. The copes are reversible. continues throughout the day with new sand available to each molder if desired. The floor sand is gathered up from time to time and tempered and mixed with new sand and bond- ing materials in the usual way. Nor- mally the consumption of new sand ere S Steel rails imbedded in the concrete floor make a smooth track for core handling. averages 350 lb. per ton of good cast- ings. Sand is received in drop-bottom cars on an elevated spur from which it is dumped into 750-Ib. trailer cars. These are taken by industrial trac- tor to final destination in the plant. When the plant was operating at full capacity, two men with this equip- ment were able to handle all the sand used by the molders and two other men handled the sand used by the core department. The sand mixing and treating is done on a level below the main foun- dry floor. From that point small bucket elevators carry the treated sand to a distributing system from which it passes into hoppers located at each molding station. Miscellaneous materials, including fuel, are handled by means of small steel open-box trailers equipped with hard rubber tires and hauled by elec- tric tractors. These material trains travel on well marked aisles through- out the plant and congestion is avoided by having the aisles all ar- ranged for one-way traffic. The scrap and pig iron storage (Concluded on Advertising Page 10) The Iron Age, March 23, 1933—467 URS EI ren ee Caterpillar Diesel pulling a 5S-ton drag to level the new air field (Bolling Field, Va.) jus outside Washington, D. C. The drag was made by Mr. Farrow of the Arundel Corpn. of 16-in. channel iron and is 35 ft. long. Selection and Heat Treatment of Steel for Caterpillar Tractors OST manufacturers are aware M that the interruption of the steady flow of manufactured product is one of the most serious causes of losses, due to lowered shop efficiency, relative increase of fixed charges per unit of product, and the loss of sales through non-delivery or slow delivery. Not every manufac- turer has gotten down to the funda- mental causes of interrupted produc- tion of which he has control but has too often assumed that things would “iron” themselves out. As a matter of fact, even most diligent study and effort in one’s own plant is many times insufficient to bring a remedy for production delays and stoppages unless the cause is traced back to the origin of the material. On _ the other hand, we do not wish to under- emphasize the importance of a man- ufacturer’s first “setting his own house in order” before looking else- where for trouble. But, let us assume that the produc- tion rate for a certain group of forg- ings is continually falling short of requirements. In such a case the actual rates have been set for the worst conditions being encountered. Thus, in machining a nickel steel gear blank, it was found that the auto- matic lathe was only doing one-half of ‘what was required by the re- mainder of the machine line. Investigation disclosed that the microstructure was a great variable in the annealed structure of the blanks. Hence, the machining pace was set for the worst. Beginning at the steel mill, testing for inherent chemical and physical characteristics before rolling, changing the forging 468—The Iron Age, March 23, 1933 By G. C. RIEGEL Chief Metallurgist Caterpillar Tractor Co. Peoria, Ill. process from batch heating to con- tinuous heating, limiting the size of the stock for the upsetting opera- tion, designing furnaces for con- trolled heating and cooling rates, we have been able to produce a definite and consistent microstructure with a corresponding narrow hardness range, thereby enabling the shop to more than double its previous production rate on the slowest machining opera- tion. This, however, was only an indirect benefit, or by-product, of the primary purpose to which we have set our- selves of producing a stronger and better gear. On this point, we shall elaborate. Failure of Gears Investigated About seven or eight years ago, sporadic trouble was encountered with failure of a final drive gear on one of the largest tractors. Knowing that the majority of these gears served satisfactorily, our metallurgical de- partment insisted that the forging source keep the product of mill heats separate and that they be correspond- ingly marked. While forging quality alloy steel had been used conforming to the limits of S.A.E. chemistry, yet some 40 pieces out of one mill heat gave trouble in service over a period of two or three years. From these unsatisfactory gears, and from gears out of a mill heat which had given no trouble, samples were taken for examination. The ex- amination consisted in fracture tests, hardness uniformity, etch tests, .mi- cro-cleanliness, impact tests, and ten- sile tests. The impact tests were cut from the gears, without annealing, by the use of rubber-bonded saws and by finishing the specimens by grind- ing. The tensile specimens were cut from the annealed gears and sub- sequently treated to the same hard- ness as that in the gears. A number of questionable facts were found in the gears from the mill heat from which the failures had occurred. What the tests disclosed is shown in the table below. RESULTS OF TESTS GOOD GEAR Velvety Total range, 3 points Rock. “CC” on section Flow lines symmetrical Etch Fairly clean Mostly above 5 ft.-lb Impacts R. of A. above 27 percent Tens. tests Fracture Hardness Micro-cleanliness POOR GEAR Dry, finely crystalline Total range, 5 points Rock. “C”’ on section Non-symmetrical Considerable non-metallics All below 5 ft.-Ib. Little reduction and mostly flat breaks wi ai ae eos oa we wreskw es FF ‘a.) just orpn. of FS served al de- orging heats spond- juality yrming ry, yet 1 heat period gears, | heat im ples he ex- tests, 3, .mi- d ten- re cut -aling, rs and grind- re cut sub- hard- umber nd in from What n the yreaks ao a & XPERIENCE of a builder of large tractors in selecting methods and specifications for the control of the quality of steel purchased and for its heat treat- ment is related in this article. The company’s method of controlling heat treatment and the equip- ment used are also discussed. The article is based on an ad- dress by the author before a meeting of the Tri-City chapter of the American Society for Steel Treating at Moline, IIl. oe You will notice that I have men- tioned positive as well as negative evidence from actual service tests. One may easily be misled by ignor- ing positive evidence of service. Parts which “stand-up” in service may be no better than some which fail un- less we know by tests. Likewise, we may use entirely too expensive ma- terial or methods, unless we know these limitations. Controlling Steel Quality Again, sweeping conclusions must not be adduced from too little experi- mental evidence. As has been said, “One swallow does not make spring,” or “a drink,” as the modificationist quotes it. It was not, therefore, im- mediately possible to put into prac- tice the restrictions which seemed ad- ie Quenching final drive sprockets for tractors. visable to us in controlling steel qual- ity at the sources. A number of mill heats from several mills were tested by these standards. Fortified with this accumulated evidence, it became possible to go to the mills and ask for the kind of steel which gave the satisfactory performance. Continuous mechanism for chain belt link hardening furnace, showing discharge into draw furnace elevator. With due regard to the more help- ful and cooperative spirit.of the steel manufacturers today, we were con- sidered at that time to be asking a most impracticable procedure. Our executives asserted that we wished to buy steel by mill heat lots, that we should receive samples from the bot- tom of the first ingot after the usual discard, and in a like manner from the top of the last ingot, thus repre- senting, as we hoped, the minimum of quality to be applied on our orders. Minimum Requirements Set Up After upsetting the flow lines in these steel samples (billets), we pro- ceeded to take specimens from the “centers,” which gave us flow lines transverse to the length of the speci- mens. From these values, we at- tempted to set up our minimum re- quirements. Anything above these minima should, of course, be the con- stant aim of the steel source to fur- nish. We are glad to state that, on more than one occasion during the several years in which such steel specifications have been in effect at the mills before the steel is rolled on our orders, the steel manufacturers have agreed to further restrictions and the raising of the minimum values previously called for. Typical of such specifications are those reported by the speaker in THE IRON AGE of Dec. 3, 1931, for S.A.E. No. 1045 steel. The majority of our purchases of carbon steel are based on such restrictions and all of the alloy steel for gears, pinions, etc., are so restricted. To illustrate how this applies to S.A.E. No. 2345 and its (Continued on Advertising Page 12) The Iron Age, March 23, 1933—469 rf | at the | “ J | an | the _ tio’ | a ter | acning . me { 1 | | thi | me | as vone ti | 19 bu \ to Us | me! — fer tior By WALTER S. GIELE : tak int P Un EXTER S. KIMBALL, Dean on r of Engineering at Cornell Bef University, has pointed out sis that the transfer of skill from the to man to the machine rather than me! the application of power has been the pla significant thing about the mechan- dis ization of industry. The application ind of power to manufacturing processes ( is, nevertheless, a quantitative meas- tot ure of the extent to which that trans- sie fer of skill to the machine has been ing found profitable and is also a quan- the titative measure of the growth of the ne Chart 20—The Growth of Mechanization as Measured by the Application of Power investment in manufacturing equip- chi Th The Growth of Mechanization as Measured by the Application of Power in the Manufacturing Industries of the United States sta 1849 1859 1869 1879 1889 1899 1904 1909 1914 1919 1921 1923 1925 1927 1929 1931 gr Aggregate hp. capacity ho: —millions ......... 2.346 3.411 5.939 10.098 13.488 18.675 22.291 29.328 31.211 33.094 35.733 38.826 43.079 no Hp. capacity of motors is driven by current gen- pr erated within plant ‘b DE” eae wase ne 1.151 3.068 4.939 6.969 8.822 10.255 11.220 12.376 ( Hp. capacity of motors ne driven by purchased pa current—millions 442 1.749 3.885 9.283 13.366 15.869 19.132 22.776 th Hp. capacity of electri- 7 fied equipment—mil- ta lions ...... . 1.593 4.817 8.824 16.252 22.188 26.124 30.352 35.152 of Hp. capacity of prime movers in factories Ee eee 18.406 20.041 19.729 19.904 19.693 20.155 me Adjustment for excess " of motor over genera- ae tor capacity—millions -454 1.360 2.060 4.520 5.560 6.510 7.620 8.890 an Adjusted aggregate hp. co capacity—millions 13.034 17.315 20.231 24.808 26.169 27.534 29.223 31.206 34.189 th Wage earners—millions. .957 1.311 2.054 2.733 4.252 4.713 5.468 6.615 6.896 9.000 6.947 8.778 8.384 8.350 8.836 6.512 Hp. per wage earner ac unadjusted 1.142 1.248 1.397 2.145 2.465 2.825 3.235 3.4533 4.495 3.770 4.265 4.648 4.875 th Hp. per wage earner— ta adjusted i 2.383 2.620 2.935 2.756 3.765 3.135 3.490 3.740 3.853 a S re “< Motors driven by current generated within plant 6 - 12 16 17 19 21 21 21 ay Sg Motors driven by current purchased from outside sources 3 9 17 38 41 44 49 53 QI oe Total electrified ac ruker ee ee 9 21 33 55 60 65 70 74 ec BO Non-electrified 91 79 67 45 40 35 30 26 T 6 Total horse power ied 100 100 100 © 8§=©100 100 100 100 = =§=100 a wo Pe ry ' — Bureau of the Census—see Statistical Abstract, 1932, p. 348, 730, 732; Commerce Year Book, 1932, p. 218; 1921 Interpo- al ated. ° si eae te aici ert tale eer ee Sn nonncnunEnpennenrenernneeee ee eee ee ere ere eee eee ea ee ee ee eee ee ee ee ee ee ee 470—The Iron Age, March 23, 1933 sv Dean ornell | out n the than m the chan- ation “esses meas- rans- quan- f the quip- 1931 erpo- ! am & — is the sixth installment of Mr. Giele’s series embracing a factual study of mechanization and employment. In this chapter the author deals with mechaniza- tion and investment over the long term period and their relation to collective and individual employ- ment. One significant fact revealed in this article is that the period of most rapid advance in mechaniza- tion was not the past decade, 1920-1929, as generally supposed, but the decade of the 1890’s. wv S ment that has come with the trans- fer from hand to machine produc- tion. Chart 20 illustrates the course taken by the application of power in the manufacturing industries of the United States, both as a whole and on a unit or per wage-earner basis. Before proceeding to the detail analy- sis of this chart it will be necessary to give consideration to the funda- mental changes which have taken place in the method of generating, distributing and applying power to industrial processes. Chart 21 shows, first, the gross total horsepower capacity of prime movers installed in the manufactur- ing plants of the United States plus the gross total horsepower capacity of motors operated on current pur- chased from outside sources. The Excess of Motor Capacity Over Generator Capacity Where individual motors are in- stalled to drive single machines or groups of machines the aggregate horsepower capacity of such motors is greater than the capacity of the prime movers which drive them as, obviously, not all the motors con- nected will operate at full load ca- pacity at any one instant. In fact, this is one of the incidental advan- tages of the electrical transmission of energy. In order, therefore, to arrive at a more nearly accurate measure of the extent of the application of power and to obtain figures which will be comparable throughout the range of this study it is necessary to compute adjusted figures to compensate for the excess of motor (connected load) Capacity over generator (prime mov- er) capacity. The United States Bu- teau of the Census has computed an average figure of 72 per cent of ag- gregate motor capacity in estimating corresponding generator capacity. This adjustment has been applied to that portion of the power capacity which js represented by motors oper- ated on current purchased from out- side sources. — 8 3 2 B8 BS 8B x Saggrks © © © 2 © Bh A Di BGOAGS . ft ee 7 | | Horse Power Capacity HT A 40.0 | Adjusted for Excess of Motor NY fl | Capacity over Generator Capacity 4 Is 30.0 I +4 _ + - i. ir —— ae F Ff 20.0 a m + — | = a fe" f| Xx . | | | |e? Y | | 3 ¢ LY & i all | “er S | | Cy) |x © — >) | | | € S S — | o § 4 & 539 re o a0 +++ 4 | | | ¢f fp %70 +— — { f- | _ +0. { fly || | bi z @ IR [| [8 ; Sis DB 5.0b + ; & c.. ++ S : oS ah PSHE * 4.0--— ——— +++. ff —F a» : SHS : sf] fg oe 3.0 —— = + —_+—. x 2 & % a > 2.0} t | —+# x | | Ay 7s | | | r ws | | e | | | | fig | | >< LO} f—_—__+ ___}___}| _j ids 09 | + am aah iT TTT 08} — ey im —+ Resnicnitie e TTT TT 07} SestsaiinsMiagadhelmeceesanctcl | Es | tee 5 a ee si sataieianba | | q e | TTL | 0.5 ete = ee, tf ba bo LT Chart 21—The Growth of Installed Horse Power Capacity by Classes Electrification has made rapid progress since the turn of the cen- tury. The relative slope of the lines representing the total power and elec- trified equipment respectively illus- trates this fact graphically. Even more rapid was the develop- ment of the relative amount of cur- rent supplied to manufacturing in- dustries by the public utilities as shown by the line representing motor capacity operated on current pur- chased from outside sources. The development of alternating current apparatus with its advan- tages in long-distance transmission and facility of transforming from higher voltages to lower has been a factor in this growth. The rate of expansion in the use of purchased current was highest from 1904 to 1909. A distinct re- tardation is evident for the succeed- ing period 1909 to 1919, and an even The Iron Age, March 23, 1933—471 more marked and further retardation of rate for the period 1919 to 1929. It seems probable that some of the larger users of power found it eco- nomical to generate their own, espe- cially where waste steam could be utilized or where exhaust steam could be applied to heating or to manufac- turing processes. This is a neat il- lustration of an application of the law of diminishing returns. Chart 22 shows the changes in the relative proportions of power gen- erated by prime movers and mechan- ically transmitted within the plants that use it, power obtained from cur- rent purchased from outside sources and power generated and electrically applied within the manufacturing plants. The arrangement of items in this chart has been varied from the ar- rangement in that shown in Fig. 21 in order to show more clearly the | | alll Chart 22—The Changing Distribu- tion of Power as to Application and Source Chart 23 (at right)—Comparison of Rate of Mechanization in Manu- facturing in Last Decade with That of the 1890’s (H.P., adjusted, per wage earner) The Growth of Wealth in the United States MILLIONS OF DOLLARS 1900 1904 1912 1922 Grand Total All Forms of Wealth 88,517 107,104 186,300 320,804 j 2,541 3,298 6,091 15,783 Source—Bureau of the Census—See Statistical Abstract, 1932, p. 271. relative stability of the proportion of electrically applied power generated within the users’ plants. Having analyzed the component items we may return to a consider- ation of Chart 20. We note first, that, after adjusting the gross fig- ures to allow for the effect of the change toward electrical distribution, the power capacity increased in fair- ly uniform geometrical progression (i.e. was compounded annually) 1869 to 1909 and again in fairly uniform geometrical progression 1909 to 1929. The rate of increase was, however, distinctly slower in the latter period than in the former period. The Power Capacity Per Wage Earner The more significant measure, how- ever, of the progress of mechaniza- tion with respect to the application of power lies in the unit figure on a per wage-earner basis. The suggested trend of mechaniza- tion as measured by the application of power per wage-earner is shown 472—The Iron Age, March 23, 1933 in Chart 23, which is derived from Chart 20, and based, therefore, on the adjusted aggregate horsepower. The general characteristic is that of the well-known “S” curve. It is probable that this is in fact a com- posite of a series of successive and overlapping “S” curves reflecting, for groups of enterprises, the successive rise and relative decline of major developments. Such major develop- ments would be, for instance, the in- troduction of the electrical transmis- sion of power or the development of scientific management, interchange- able manufacture and so forth. The deviation from indicated trend in 1919 does not represent a relative reduction in power capacity. It rep- resents an abnormal increase in work- ing forces which came with the two and three-shift operation common in so many industries at that period. The quotient horsepower per wage- earner was reduced by an abnormal and temporary increase in divisor rather than by any significant change in the dividend. _— ee oe : oe a So, .too, the deviation in 1921 re- flects a drastic reduction in working forces rather than an abnormal in- crease in power application. Wage- earners were laid off but equipment already installed remained. Maximum rate of power applica- tion occurred in the depression dee- ade of the 1890’s, not in the period 1927-1929, the culminating years of our biggest boom. In fact, should we draw a line through the abnormally low point in 1919 and 1929 we would still find such a distorted rate for that decade lower than the indicated rate for the 1890’s The portion of the curve represent- ing the development since the turn of the century, while nearly fiat, is convex upward, as indicated by the tangent drawn through the points at 1927 and 1929. This characteristic suggests a decreasing rate in the progress of the application of power on a per wage-earner basis. The Increase in the Provision of In- vested Capital Per Wage-earner Another aspect of the growth in the provision of power to supplement the physical effort of the wage-earner is in its implication with respect to the invested capital provided pe? wage-earner. Power is provided for the sole purpose of driving the ma- chines of production. The growth in power generating capacity is, there- a owe me KB Fe CO FO ee DAD AH) 21 re- orking al in- W age- pment yplica- 1 dee- period rs of ld we mally would e for icated asent- turn at, is y the its at ristic 1 the power of In- er th in ment arner ct to per 1 for ma- th in here- | | || | Boe De OS oe 8 Chart 24—Trend of Investment in Manufacturing Equipment, Past Decade Compared with 1890's. (Indicated by H.P. per wage earner) fore, roughly a measure of the growth in the production machines of every sort that are to be driven by that power; a measure not only of their numbers but also in a gen- eral way of their productive capac- ity and of the capital investment they represent. The dollar value of the capital in- vestment in any enterprise or group of enterprises, whether appraised on a basis of first cost less depreciation and obsolescence or on a basis of re- placement, must vary with the price level. The decision whether or not to install additional or improved equip- The Growth of Relative Total Fixed Capital in Manufacturing in Terms of 1880 Prices Index Numbers 1899 = 100 : Index Index Year Number Year Number 1899 100 1911 216 1900 107 1912 226 1901 114 1913 236 1902 122 1914 244 1903 131 1915 266 1904 138 1916 298 1905 149 1917 335 1906 163 1918 366 1907 176 1919 387 1908 185 1920 407 1909 198 1921 417 1910 208 1922 431 Source--Douglas & Cobb, “A Theory of Production,” Supplement to American Economic Review, p. 145. a a 300 lOO Chart 26—Growth of Total Relative Fixed Capital in Manufactur- ing. (In terms of 1880 prices. Index numbers, 1899 = 100) 350,000 —] 7 — 300,000+— i 200,000} ation 2,509 ++ —— Chart 25—Growth of Wealth in the U. S. Upper curve represents total wealth, lower curve the portion represented by manufacturing equip- ment and tools ment at any particular time will be influenced by consideration of the pre- vailing levels of business activity, of commodity prices and labor prices and the relative difficulty or ease of obtaining capital funds for the pur- pose. It seems probable, therefore, that power capacity is a more reliable basis for evaluating the relative changes in capital invested in manu- facturing equipment than would be the dollar valuations. Power Capacity as a Measure of Investment Here, however, the capital repre- sented by manufacturing equipment is more accurately depicted by the aggregate total capacity of prime movers plus total connected capacity of motors installed than by the ad- justed figure which gives effect to the excess of motor capacity over gen- erator capacity. The tendency with the more flexible electrica