The Iron Age 1915-10-21: Vol 96 Iss 17

FSTABLIS! ciation Secured at oundry, the design of which repre- experience in the requirements of light castings, was completed ithin the past year by the Federal Foundry Com- ny, Indianapolis. The layout of the foundry oper features centrally located cupolas from hich the carrying distance to any pouring floor nes not exceed 100 ft., an inclined cableway from A stove nts a larg’ hanufacturing al By A] ia '; _ hah € Interior we raw materials larging floor. ium of tra olding D ft. The plat 1 t especial ) the large ich the Urner of t he found; storage yard to the cupola an overhead trolley which is a me- ‘ranster for the entire plant and wide floor bays with unobstructed spans of e has an area of 22 acres, having ‘itness for foundry purposes because proportion lying below the level on resent buildings are erected in one property that is available for filling. uilding is 400 ft. long and 180 ft. clined Runway to Elev the Foundry with the Inclined Conveyor to the Overhead Charging Floor, New York, October 21, 1915 VOL. 96: No. 17 oundry Construction for Light Castings Building of Modern Design and Low Depre- Cost—In- ated Charging Floor Moderate wide, subsequent increases in capacity depending upon a lengthening of the building and the installa tion of additional cupolas in the same relative lo- cations as at present. The supporting framework of the building is of light steel construction with but one central row of columns lengthwise of the building, the columns being spaced on 25-ft. cen- ters and dividing the floor space into two parallel the Floor and the Cupola It Serves 90-ft. bays. As can be noted from the accompany- ing cross-section of the building, a monitor-type roof has been adopted of only moderate height above the floor and equally moderate pitch. The building is almost entirely inclosed with glass from a low concrete wall to the The use of a trolley system, and that only above the main aisles of traffic, permits a much lighter overhead con- struction in the building framework than would be possible with cranes of any type. To take up the longitudinal thrusts induced by the trolley-system eaves. 917 ma = ——— Ss Sand Bins icliiieee Cinder} House coienniaeniill oadling > ) > 2 Q and Bins | a | | 7 | aka? S Finisning Buildin General Floor Plan of the Federal Foundry Company’s Plant Showing the Arrangement of the Various + 1915 , vith the Sand Bins on the Left and the Pig Iron ar materials to the scale platforms and the special lattice bracing has been used. ral design is that of W. J. Carter, engi- eveland, Ohio, and represents the very low pproximately 80c. per sq. ft. of floor area. ntic-Marshall Company was the steel con- yround plan illustrates the manner in which iry buildings are arranged around the proper. The sand bins are in two dupli- one for each cupola. On the one side ns the sand is received through doors from ning cars while on the opposite side it e loaded as required into steel boxes which are then carried into the foundry via the trolley which is equipped with a 4000-lb. Pawling & Harnischfeger cage-controlled hoist. Between the and storage bins and the foundry is an areaway which the raw materials, pig iron and coke are ' The standard railroad switch provides for these materials into the plant. The mat- getting the raw materials into the cupolas somewhat unusual by reason of their lo- the center of the foundry. An elevator ed by the side of the cupola, operating be- undry and the charging floors, but the Oven in the THE IRON Center Equipped with Overhanging Hood to Carry Off the Gases, the Racks and the Benches Located on Three Sides of the AGE * : MTT ue ate . a | ve oe Poe / JN a . > d Coke Piles at the Right Industr rs transfer the inclines leading to the charging floors more common plan of trucking the iron and coke across the foundry floor to this elevator has been discarded as an undesirable intrusion upon the normal molding floor operations. Accordingly an inclined runway with a motor-driven cable haul is built on which the small charging cars that are loaded on the industrial track and weighed on the track scales shown in the illustration can be run directly from the yard up to the charging floor. The details of the construction by which the incline and charging floor are incorporated in the build- ing structure is shown in the accompanying cross- section. The charging floor elevators were fur- nished by the Curtis Pneumatic Machinery Com- pany, St. Louis, and have a capacity of 4400 Ib. Artificial heating of the foundry building is pro- vided for in the installation of four Baetz air heating apparatus, furnished by Skinner Brothers, St. Louis, these units being of the indirect type in which air, under fan pressure, is heated in con- tact with steam coils. With a few minor exceptions where hand work on the bench and floor is necessary, the molding in the foundry is entirely by machine. The mechan , but the ical equipment still includes hand squeezcr 920 THE IRON AGE October 1915 greater number ot tne squeezer type, built by and the Arcade Mfg. both machines are of the power the Osborn Mfg. Company Company. The molding floors are laid out on sides of the center aisles ex- tending to the center line of the building and to the walls In area they are approximately 10 x 41 ft. The molding machines are mounted on cas- ters so as to be readily moved in conformity with the not unusual method of working up the sand from the front to the back of the floor laying down the molds behind the machine. This sand is laid down in long piles at right angles to the aisles on which the molding floors front and is mixed mechanically with machines furnished by the Sand Mixing Ma- chine Company, Greenfield, Ohio. These mixers are electrically driven, and an electrical connection is brought down from overhead, at each floor, in a Served by the Plant Monorail Syste conduit with a socket terminal, so that the mixer can be plugged in wherever desired. While water connections are made accessible there is no pro- vision at each floor for wetting down the sand with a hose, for example. The amount of water mixed with each pile of sand is accurately measured for various classes of work and it is accordingly re quired that the water be handled in pails, a cer- tain number to the pile of sand, varying from five to twenty-five, depending upon the weight of the casting. The floor of the foundry is of milled clay la from 10 to 12 in. thick and well tamped. For the designation of each molding floor the serial nun system is employed, the four rows of floors being numbered from 1 to 100, 100 to 200, 200 to 300 and 300 to 400. The one number serves to link ttern Stor ige Vault and the Way » Patterns for Work to be Done Are Laid Out with the 0: 1915 ecate accurately, the man, the work, nd production. As soon as a pattern of the storeroom it is immediately a blackboard to the floor number to ing. A pocket is provided on each for the insertion of a time and pro- n which the record of the workman’s production on that floor is kept from Pouring of the molds is entirely from each molder pouring his own work. were built by the Whiting Foundry Piaait ompany and have 78-in. shells lined They melt enough metal daily for ns of castings but the cupola capacity the tonnage that can be got out on handling of materials, dependence is st entirely upon the overhead trolley. Located Cupolas with the Overhe Inclosed Inclined Runway and a Hot Heats Havine Large Disc rge Pipes and boards are brought in, in trays xes, the castings after being shaken out olds are carried out to the mill room, lt in the cupola is brought to the or elevator, and the cupola drop is to a Sly mill in the yard for the re- metal. The coreroom which ‘eme iseable st an isolated building is rendered con- essible by reason of the trolley service. \ view of the coreroom is shown and with it istration of the type of cores made. The graph presents a better description of the cated core ovens with overhanging hood gases, the racks for temporary storage i the core benches on three sides facing than is possible in the text. For the rk that is being done this arrangement tself both convenient and effective. The the mill room is also illustrated. Most were furnished by the Cleveland Chap- and vary in size as required by the astings. The interesting feature in the drive for these mills was the care- to the speeds at which the mills turn e of the casting, the diameter of the e number of revolutions at which it sociated factors, the proper determi- n results in the casting turning over nanner to make the cleaning action [he application of the trolley serv- rtment is apparent. Coming from the milling room, the trolley con- THE IRON 921 AGE tinues out through the opposite end to the grinding, inspecting and shipping rooms. The pattern storage vault presents an interest ing arrangement of shelves for the storage of pat terns. Most of the patterns for this work are flat and readily stored on shelves with vertical spacing. A flexible numbering system is used for the locating of the patterns which depends on the number of the shelf and the number of the pat- tern. Each section of shelf is numbered in se around the but there is no attempt to return the same pattern to time it is used. A tern number returned the shelf number and date are against the pat tern number which fixes the location } Ciose quence room the same place each record is kept according to pat and when a pattern is entered the pattern at any time. The open central floor space in the vault is used for laying out series of patterns in planning work for the foundry New Protective Relay for Motors For use iz conjunctior with a low-voltage release for automatic, overload and low-voltage protection of alte gt nating-current motors up to 2500 volts and 3000 amp the General Electric Company, Schenectady, N. Y., has developed a new type of circuit-opening oil dashpot relay. The relay is generally used with motors having self-contained compensator control, but sometimes for switchboard service when both low-voltage and time delay overload protection are required. In such a case series relays replace the secondary relays, current trans formers and oil switch tripping coils otherwise required The relay contacts, dashpot and calibrating tube ar inclosed by dust proof stamped steel cove The rela is connected in series with the line with the low-voltage release across one phase in the customary way and the low-voltage coil in series with the relay contact Wher an overload greater than the setting the relay occurs the relay contacts open the circuit through the low-voltage release coil and the moto it out of ircuit If the voltage drops to predetermined per entage of normal the moto also disconnected from the power supply The current calibratior Is Iron ormal iT y Ww twice that figure, and the time adjustment is from The current and time adjustments are accomplished outside of the dashpot with the aid of a screw driver, and are constant, as an adjusting nut is locked in place after each setting is made. The delay recommended for the normal start- ing current is about 15 sec., which affords ample protec- tion to the motor against damage erload or single-phase operatior and the from being opened while the motor is starting 10 sec. to 5 min. on a 25 per cent overload from ov also prevents circuit Ps ; ¥2 # 922 THE IRON AGE Octobe) PUPPE UNIVERSAL MILL German Development for Rolling Beams with Wide Parallel-Face Flanges In the course of tests made in the last ten years to determine the power requirements for rolling blooms and different kinds of shapes Dr. Johan Puppe of the Royal Technical High School of Bres- lau, Germany, conceived the idea of a new type of beam mill. It is of the universal type and con- sists of two stands. One such mill, it is under- stood, has been in operation at the Peine Steel Works, near Hannover, Germany, since June, 1914. The beam blank comes from the blooming mill in the shape of a thick walled H. It enters the lisence 5 7 — . \ , | — =) \ f \ , oh 4 } ~ f ‘ hanhudbut + / ky cabecdaiasall 7 | | ) | | | yong | |R Fig. 1 CT The steel is re ceived by the roughing mill, | Fig 1, in the i seetiieaaiae shape of a thick : walled H and is passed back and forth through the roughing and finishing roughing mill, Fig. 1, where both web and flanges are worked, giving a bloom of a double Y-shape. From the rougher the bloom is going forward and backward through the finishing mill, Figs. 2 and 3. In the first pass of this mill, that is, during the forward movement of the beam, the flanges are straightened by means of the vertical rolls c, Fig. 2, while in the second pass, that is, during the backward movement of the beam through the same rolls, the edges are worked upon at d, Fig 3. The piece is then sent again through the rougher, where the flanges are bent back to the double Y-shape. The beam or girder is rolled down to its final shape by continuously going back and forth through the roughing and finishing rolls until the final section is obtained. Figs. 4 and 5 show the arrangement for a chan- nel beam. Through this method the metal is worked and kneaded more thoroughly, it is con- 1915 tended, than can be done with any ot and it allows for rolling beams or gird most any width of flanges. Special . placed on its making possible the rollin with absolutely parallel faces. The be: signed with a slope of 10 per cent bx and flanges, to insure a smooth co; tween these two parts. This sloped pi in regard to the total width of the flan 30 that there is enough space left for the para! rt f, attaching by means of rivets or bolts r “a connection wanted. Finally, the first co of this type, it is claimed, is small. A general section of a Puppe bean in Fig. 6, which represents a shape co: to a beam about 15 x 15 in. Three stan mills Figs and 3 show tt flange and edge rolling, tively, finishing ! Figs : show _ the equipment channels ee Fig. 5 of shapes, containing 49 variations, ean be rolled in the Puppe mill, it is stated, as it is arranged to day. The smallest size is 160 x 160 mm. (about 6.4 x 6.4 in.), and the largest one 1000 x 380 mm. (about 40 x 15 in.). Since the four rolls of the Puppe mill lie ™ the same vertical plane and move relatively to each other during the adjustment between the different passes, the housings and bearings had to be de- signed so that the ridges of rolls b b, say, in Fi ig} are always exactly in line with the center ° f the web, since otuerwise the flanges of the girder beam would not be rolled with a uniform thick ness above and below the web. That the pressure is distributed uniformly over the whole section © the beam is regarded as proved by the fact that ™® finished beam leaves the last pass straight s0 that a straightening out of the cooled beam is unnece> sary. he Bearing on the working, so called that th ary rolling, the following interest- is been obtained from Fr. Denk, neering Company, House Build- vho represents Dr. Puppe in this of 15 x 15-in. Puppe Beam ference in the work of reduction of luring the rolling process is caused by the difference in speed of the various points of the groove. This difference in speed Se rom Fig. 7, which represents an in- ove in a 35-in. three-high beam mill, rolling 18-in. I-beams. The diameter of the middle roll at point A is 26.25 in., while the diameter for me point on the upper roll is 43.5 in. The ircumferential speed is correspondingly 6.87 ‘and 1.39 ft. per revolution, or, the speed of the upper for point A is 1.66 times as high as the speed f point A on the middle roll. Similarly, at sec- ion B B, the speed of the top roll is 1.29 times that ff the middle roll. Such differences in the circum- ferential speed have a decided influence on the qual- ity of the metal, and the design and layout of such rooves must be made with the utmost care, so that splitting or cracking of the beam, due to in- ernal stresses, may be avoided. A Hand Device for Testing Hardness \ simple device, adapted to determining hardness thas ‘actory degree of accuracy for many manu- ring purposes, is offered by M. F. Turpin in the le Metallurgie for February, 1915. It employs ytruding from the end of a hollow man- by a cylindrical slug of known hardness | seale. A plunger, protruding from the irmounts this slug. When in use the ought into contact with the piece to be tested tree end of the plunger is struck with a ham- omparison is then made of the indentation standardized slug. The device is 30 and 90 mm. long. -~oons ao ak A S er BR VY; 5 YA “IN i/ } 6) T/Ee VY | SES hss; Sy / A ) ees — | *y f 44. i I asts/ idle Rol y VIOTTTIIR A . S 5 ing S ) lustrate Different Peripheral Velocities of Mating Rolls 15 THE IRON AGE 923 Radial Drilling Machine with 2-Ft. Arm The Fosdick Machine Tool Company, Cincinnati, Ohio, has developed a new radial drilling and tapping machine, which will be known as its high-speed manu- facturing radial drilling machine, and for the present will be built in two sizes, having 2 and 2% ft. arms. The machines are designed to meet the demand for a high-speed durable tool, capable of handling a great variety of work, especially in shops where unskilled labor is employed. The unusually wide range furnishes 48 rates of drilling, which are correct for 3/16-in. carbon to 2% in. high-speed drills, in iron or steel, and for boring A Radial Drilling Machine That Is Being Built with 2 and 2%4-Ft. Arms up to 5 in., all with but one speed at the pulley. A metal plate on the arm girdle, for high-speed drilling, corresponds to the indexes on the head and on the speed box. The bushings are of special phosphor bronze, gears under severe duty are of steel forgings, hardened where necessary. Where the speed is high or the duty light, cast-iron or bronze gears of extra wide face are used. The spindle and principal driving shafts are of ham- mered steel, and the column, spindle and arm elevating thrusts are taken on ball bearings. The column and shafting are all finished by grinding. In a recent test a 1-in. high-speed drill was driven through a very hard cast-iron slab 2 in. thick, in 7.8 sec., or at a rate of 15.4 in. per minute, the speed being 550 r.p.m. and the feed 0.028 in. per revolution. A 2%-in. drill was successfully driven through 1%-in. machinery steel at various rates, ranging from 137 r.p.m. with 0.007-in. feed, to 49 r.p.m. with a feed of 0.028 in. per revolution. The 2-ft. machine will drill to the center of a 48-in. circle at the base, which has a working surface of 26 x 31 in. The 2%-ft. machine drills to the center of 60 in., and has a base working surface of 28 x 36 in. Other principal dimensions common to both machines are, base to spindle, 51 in.; spindle traverse, 12 in.; least diameter of spindle, 1 9/16 in.; spindle bore, Morse taper No. 4. The net weights of the machines are 2900 and 3200 lb. respectively. Like the builder’s heavy duty machines the inter- changeable drive has been adopted. The cone-driven machine may be changed to speed-box drive, or vice versa, or a constant or adjustable speed motor may be added at any time, without the necessity of a special base, special speed box, special shafts, or gears in the machine. Tilting, swinging, or round tables of the builder’s standard types can be supplied. 2 ere eee te Phosphorus Limit in Malleable Castings Higher Percentages Than Customary Demon- strated as Beneficial in Some Cases—Important Dynamic Tests—Unsoundness from Shrinkage The failure to include a phosphorus limit in the’ shown in Fig. 1, until finally failur revised standard specifications for malleable cast- fracture. The vise and anvil block ha Y ings recently adopted by the American Society for Structed that the wedge, when struck, r s th Testing Materials is made the nucleus of an investi- force of the blow without upsetting. gation by Enrique Touceda, metallurgical engineer abie 1.—Chemical Analysis of Wedg: of Albany, N. Y., to prove that a slight increase Manufacturer A is not harmful. The author presented the results ct poe b, Set of his work in a paper, “Some Remarks Regarding _ Silico: 0.800 0.810 the Permissible Phosphorus Limit in Malleable ¢)paphorus are sae Iron Castings,” which he read in abstract at the Manganes 0.284 0.290 convention of the American Foundrymen’s Asso- Graphitic hen af 2070 1 Sis 1 ciation at Atlantic City, N. J., Sept. 28, 1915. Mr. Touceda acknowledged his honest and bitter preju- dice against phosphorus which, so far as steel was concerned, was still strong, if not stronger than ever. Still he believes now that “many mysterious a fixed position to deliver an initial blow of 70 { failures of low carbon pounds, but a littk steel, attributed at the [fF tion will prove that « time wholly to a _ phos- Si phorus content slightly above the Bessemer limit, would have been found to have been due to faulty structural conditions had we at that period been possessed of the knowl- edge since gained through the use of the micro- scope.”’ An abstract of Mr. Touceda’s valuable paper follows: this, the vise had to be so designed that the opera: can turn the wedge in such a manner that the we is always falling on the highest part of the end. The tripping pin that releases the hammer j succeeding blow is greater than this amount. This is due to the fact owing to the cur] wedge the weight has ; slightly greater distanee to fall before coming contact with it than wher the initial blow was struck, coupled with th fact that the maximur fect of the blow does not take place on the sectior struck, but on the extreme fibers of a_ sectior distance from the Through the of two proglucers of leable iron castings | furnished with sets of t bars, those from Manu facturer A _ having sisted of 24 wedges. first six of this series ¥ cast from a ladle wh contained 30 pounds air-furnace iron Until several years ago the writer was strongly of the opinion that no good and dependable malleable iron could be made with a phosphorus content in ex- cess of 0.20 per cent- strengthened by the ex- amination of many poor | samples that had come to | hand from various sources. As the true cause of failure a sponding to the mixture was, through st udy and in- Fig. 1 Device for Testing Malleable Iron Specimens, Show- that was being ru! vestigation, thoroughly un ing Wedge Partly Curled foundry at that tim derstood, the fear of phos- next six were cast phorus grew less. It is not my purpose to recommend the same ladle of iron, but after an additior had that the manufacturer of malleable iron castings should certain amount of 20 per cent ferrophosphide had De take undue liberties with this element, especially as it made; the next six were cast after a further addi leads to no saving in cost—since pig iron with a phos-_ the alloy, and the last six after a still further phorus content of 0.18 per cent is as plentiful and as_ tion. In this manner four sets of six wedges ' cheap as that containing 0.30 per cent—but there are obtained, the composition of which differed appro® particular cases where to use a slightly higher percent- mately only as to phosphorus content. At te a age than is customary would result in a better product. when the wedges were made, a %-in. square Dat For the purpose of investigating this problem I exactly square owing to draft), also was © decided to confine my experiments to the testing of spec- Table 2.—Number of 70 Ft.-Pound Blows R imens dynamically, for the reason that if the metal Test Wedges Shown in Fi; proved good under such conditions there could be no an ae he Bigs tS a f Blows question as to what would be shown under tensile test. 7 36 The test bar used was wedge-shaped, 6 in. long by 1 in. + . wide throughout, the dimensions of the base being 1 12 51 x %-in. and of the top, 1 x 1/16-in. +2 4 SPECIAL METHOD OF DYNAMIC TESTING Average 43.33 $6.50 To test a specimen, its base is securely keyed into transverse test, and the analyses in Tal are a slot in a vise free to rotate in the anvil of a drop tained from rillings taken from the bars ind ne hammer. The weight then is raised and the free end the wedges. an Set of the wedge is struck repeated blows; the effect of each Unfortunately, the six wedges belong . » anet blow is to cause the end of the wedge to curl up as_ inadvertently were not sent to me and 1a 924 turn 1915 secure them, but inasmuch as this set straight air furnace iron without the addition, the omission will in no way isions that can be drawn from these s statement can be made safely will be referring to Fig. 2, which shows the of Sets 2, 3 and 4. All of these, with in Set 4, were subjected to 30 blows iter than 70 ft.-lb. each. One of the ke at 28 blows and the other at 27. 9 res shown in Fig. 2 were photographed subjected to this test until failure took ible 2 contains a record of the total num- required to produce fracture. described the method of testing, I believe knowledged that the metal of which Set 4 and which contains 0.388 per cent of of superior quality and is trustworthy The metal of Set 3, which con- per cent of phosphorus, is of still better sufficiently high grade to be suitable for for which malleable iron castings may be poses. is exists in iron as phosphide of iron. It th iron in definite proportion, which com- is dissolved in solid solution in the excess the percentages of phosphorus in these four wedges be calculated to phosphide of iron, we THE IRON 925 AGE set which contained more of the alloy. (Detailed re- sults are given in the original paper.) None of these sets of wedges furnished by Manu- facturer B contains as high a percentage of phosphorus as those furnished by Manufacturer A. The metal from which they were cast was of poor quality, but the practice of this shop has since been corrected. Un fortunately, the wedges were not cast as well as they might have been, which also may account for some of the variations in the results. When a test bar is sub jected to such a severe test as the one used in this in stance, any surface defect on the tension side is bound to influence the bar. It would appear then that when the amount of com bined carbon is very low, the evil effects of phosphorus result to the disadvantage of the test are slow to make themselves felt, and it requires a substantial increase in this element before a corre sponding change in physical characteristics can be noticed. UNSOUNDNESS FROM SHRINKAGE Patterns occasionally are delivered to casting plants from which it is almost impossible to obtain satisfac tory castings free from shrinkage in parts that should be very strong, in spite of the generous use of heavy risers and chills. A shrink results from failure to supply liquid metal to the mold so that as the shell of the casting gradu . [ j a: —— ally builds up with x solid metal, liquid ‘ metal always will t =f be present to fill up af solidly the last = void in the casting ose There ire many 1 circumstances that 150 | operate to make S* difficult, or entire | ly prevent, com plete solidity in the % in. SET 3 average casting. A t bars, | | discussion of these ly referred | is unnecessary to cast at make lear the time and point I desire to the same make, namely, that the four of two furnace vedges, and SET 4 mixtures, one I > trans- aE et much more fluid at they were Fig. 2.—Tested Wedges Representing Sets the same tempera the anvil ture than the ne hammer, on supports 6 in. apart. The fol- other; the former will produce castings from the same cord of the number of 70 ft.-lb. blows re- pattern with less shrinkage than the latter reak them: i have examined castings that have failed and have Rar No: ¢ Rar No Bar No. 4 tested pieces cut from different portions of these cast Sows aire ane ings to prove that these consisted of metal that was as 0 15 tough and strong as malleable iron can be made, and these figures it would appear that as the that failure resulted because of local weaknesses due increased, the evil effects of phosphorus be- “0 unsoundness rom shrinkage re manifest. However. bar No. 3 was subjected Unfortunately, according to present practice, the 70 ft.-lb. blows before rupture took place, — that yield the toughest and strongest mal metal containing 0.325 per cent of phos- leable a none that have : tendency to produce slug bout 1.95 per cent of phosphide of iron, is gish metal. Phosphorus being constant, carbon is that mean test. Bar No. 2, containing 0.252 element in air-furnace iron that contributes more ' phosphorus, or about 1.51 per cent of phos- fluidity than any other, but experience has demon required 20 blows to produce failure, a strated that among other accompanying conditions, the h I consider indicates superior metal ower the carbon approaches that point where the hard carbides in the white iron can still be broken up in INVESTIGATION OF GRAIN SIZE the anneal, the stronger and tougher the finished fort to discover if the grain size had in- prone. . , ; . Oe AN IMPORTANT QUESTION n the increase of phosphorus, a section was : h of these bars, which was polished and If the facts as I have stated them are correct, ther eful microscopical examination and a com-_ I would like to submit this question for your considera ne sample with the other revealed no par-_ tion: Fo. ence in grain size in any of the samples, phorus is presumed to have this effect. making conclusions from one set of tests es were secured from Manufacturer B. is determination of this series of wedges the amount of ferrophosphide used was _to raise the phosphorus as high as I he investigation, and I requested another istings made from patterns poor ig? which he 7 ifacturer 1s not pern tted to alts if omp ated stings made from patterns that have lesig well is skill can suggest, is it better practice to have certain parts of the ustings strong and solid, and other parts weak and insound, or is it better to effect a compromis ind have t castings of fairly uniform strength and solid throughout There surely can be only one answer to this ques- tion. It is wrong to assume that the physical prop- 926 erties of a metal, as shown by tests, on test bars easily cast sound, are representative of the quality of the metal in any part of a casting. The manufacturers who are producing the best product are the ones who are constantly aiming to in- crease the strength of the weakest parts of their cast- ings, the parts in which, due to certain laws, unsound- ness is liable to occur, unless new laws are invoked to make these inoperative. THE IRON AGE Octobe In conclusion, I will acknowledge that ments made are few in number, but the oe presented as one that has treated the su other than an introductory manner. I hop will follow this particular line of investig it can be determined definitely just how to make use of phosphorus and to ascert thickness of section is a factor entering iem. The Cost of Electric Furnace Steel’ Data Based on Operating Experience with the Snyder Furnace—A Special Door to Prevent Heat Losses—The Argument for One Electrode Bs Bs. ae It is convincingly known to only comparatively few makers of steel that electric steel can be substantially below the corresponding furnaces. To render this more widely available, this paper gives the operating costs for two sizes of furnaces in common use and of a type designed by the writer. In each the represent tending over continuous commercial use made at a cost with fuel cost information case costs records ex- for periods of | Cc [ | A, Af | | fT ee Fd fe Ps fed | i 3 oa ey a = | | | | bosy | | = . —_ j The One-Electrode Type of Electric Steel Furnace from 1% to 2 years. As these costs are markedly lower than records of similar character of earlier types of electric steel furnaces, some of the reasons are given why such lower records should be expected with this latter furnace design. DESIGN OF THE FURNACE The style of furnace from which these operating costs were obtained is shown by one of the illustra- tions. It consists of a circular shell lined with re- fractories. The current enters at a at the bottom of the crane and into the furnace by a single electrode e in the center of the furnace. The heat is generated in the are by which the current jumps from the electrode to the slag covering the metal. The electric current leaves the furnace at c through a contact with the metal bath. These operating costs as given do not include su- pervision, as the amount involved is so small, due to the simplicity and reliability of these particular fur- naces, that no additional cost to the plants resulted on passes *From a paper presented at the twenty-eighth general meeting of the American Electrochemical Society in San Francisco. Sept. 16 to 18, 1915 The author is president of the Snyder Electric Furnace Company, Chicago. SNYDER account of such supervision. No overhead, du st and maintenance of buildings, is included. The build. ing overhead cost varies widely from plant to plant and is substantially independent of the sort of furnace used. The metal loss, due to spills and oxidation, not included, as these depend largely on the by the melter. In each of the furnaces from which these costs come, the metal loss has been substantially below 1 per cent of the metallic contents of the charge The cost for scrap used are real prices over long pe- riods of time for special scrap material, and are due to the ability of these electric furnaces to handle low priced material. In comparing these operating costs with other pub- lished records, it should be kept in mind that they do care used Table of Costs of When 10-Ton Output Operation Melting Cold Scrap at *) Output Per Day Per Tor Pe Se 2h Me os veken cee eeewres 4 Tome Gf motel 16 11 BP. sc cksdsces 10 Labor: ) ree $4.00 RED icc oss ck awe ceawe 2.50 meee $6.50 $0 Electricity (per kilowatt hour, 0.7c.) : Furnace $34.70 Substation 4.00 - —— 8 ‘ Supplies eee RR ee $3.00 Electrodes 6.00 ae 9 tf Maintenance: - 7 - et QUE sc sicuntheewess $56.60 35. 3urden : Interest, depreciation and taxes... 18.00 Conversion cost .............-- $74.60 $ Charge BONO bi vnc ceitsxeceneenes $110.00 BOGE sca viasvsssseberes 6.00 ooo — 116.00 ii.t Total cost melted metal........ $190.60 $19 Table of When Melting Cold sera] 6-Ton Output Costs of Operation Output Per Da T Heates im 12 BOUrB..<. si tevsedevcs 4 Tons of metal in 11 hours........ t Labor: PGE ona ck vote nee $3.00 REOMOEG: 6c 6 «eae héWeR CS eam 1.40 3 4 : $4 $ Electricity (per kilowatt hour, 1.4c.) : WUPMROS bu s4scsseeeke . $45.02 ees ae 1.84 ; 6.5 Supplies: s Refractories ... cane Electrodes 3.00 > Maintenance g4 Direct cost ses ia ‘ — $56.8 . Burden: i Interest, depreciation and taxes.. L] Conversion COSt ...cccececees $68 Charge: BOCED ceccvececesses $58.00 ATO = .c inc’ cvkuveanes 2.04 , — hv Total cost melted metal.......- $128 gle heats or runs made under test ier exceptional circumstances. They regular commercial operation dur- industrial service. They include vetween heats, the breakages and the lost time on Sundays and holi- rk on maintenance. ts are taken directly from long con- yperations they are so low compared furnaces of earlier design, do not carry immediate conviction is for have had direct access to similar . earlier furnaces. It is desirable to . etail why such low operating costs i with this type of furnace. ) DISCUSSION OF THE COSTS he cost items in the reverse order from hey appear on the cost statement, the which includes interest, depreciation due to the rapid melting ability of Each of these furnaces has regularly in an 11-hr. shift. ipplies is made up of electrodes and h of these furnaces runs regularly »f electrode per ton of steel poured, an average of 30 lb. for furnaces of large difference is due to the several causes. The most important se of a single electrode in place of Electrode consumption is largely a e burning. Three electrodes of the naturally burn three times as much ectrode. Actually, the ratio is much ee to one for several reasons. (Detailed riginal paper.) Most the refractory wear in an electric furnace The sides are banked up with rela- naterial that keeps the walls from wear- ' wear is largely around the electrodes due to the hot gases that escape between and the roof. With three electrodes this of near the electrode would be expected es as large with a 3-electrode furnace as de furnace. It is further increased that 3-phase current is used with three reduce the leakage current between electrodes of a 3-electrode furnace are apart as practical. The further they the nearer the arcs come to the walls the cost of wall refractories per ton of With these single-electrode furnaces, the arc at the maximum distance from the nsequent low wall wear. safety to operators, the are voltage of a le furnace is substantially double that of ch are of a 3-electrode furnace. The are in a steel furnace is determined of energy it has to radiate per square ally according to Steffens’ law. With furnaces the energy consumed is low, ace is large. Therefore the are tem- This tories are at a temperature between ture and the temperature of the bath, rnace about 50 per cent of the energy by reflection from the lining of the efore, the lower are temperature of ode furnaces means a lower refrac- >S OF HEAT THROUGH DOORS used in an electric steel melting ip of two parts. One part goes into goes with the metal into the ladle. This ‘mount per ton for a definite pouring is the same for all steel melting fur- and electric. The other part repre- st during the melting operation. It he heated air around the furnace or in from the various jackets used. This igh the doors, the roof, the shell and bal 15 THE IRON AGE 927 oo of a plug, as shown by one illustration, so that i furnaces are fitted with a door in the form place of the usual 4%-in. (11.5 em.) of door br sent the resistance of 14 in. (35 cm.) the flow of heat. Thorough tests of the sliding type of door usually used with earlier furnaces show a door loss exceeding 100 kw.-hr. per ton of steel. Much of this is saved by this plug type of We have no experimental results on the actual heat losses with the plug type of door, but our calculations, based on its outside temperature, show probable heat losses of 3.3 k.w. per furnace door. These furnaces have roofs substantially 50 per cent thicker than those of earlier design. Special attention has been given to the design of the shell linings. They are designed for and operate with external surface temperatures at about 60 deg. C. It is practical to hold the hand against the surface of the fur naces. In actual practice the radiation losses are so low that no extra means are used to keep them hot between heats. The furnaces are luted up after the last heat of a shift and when opened for the next day’s operation have lost so little heat as to be ready for im mediate operation. ick, they pre thickness to door outer HEAT LOSS THROUGH ELECTRODES The principal unavoidable heat loss in an electric furnace is through the electrodes. Just as a wire conducts electricity better than air, so a carbon electrode conducts heat much better than would the air in an empty electrode hole. With the 3-electrode furnaces and with the three electrodes of the same size as the single electrode, the heat loss would obviously be three times as great as in the single-electrode fur copper nace. Actually the ratio is much higher, due to the economic necessity of using amorphous carbon ele: trodes in 3-electrode furnaces. As used in the best actual practice, the cross-sectional area to the three carbon electrodes of a 3-electrode furnace is about 15 times the cross-section of the one graphite electrode of the single-electrode furnace. Carbon at tem- peratures has about half the heat conductivity of graphite, so that the heat lost through the three elec- trodes is about 7% times the heat lost through the one electrode. However, with the single-electrode furnace a contact is also required to connect the charge with the circuit. This contact loses some heat, hence the net result is that the heat loss due to the introduc these tion of the current in a well-designed 3-electrode fur- nace is about furnace. 5 times the heat loss in a one-electrode 928 The low labor cost with these furnaces is due pri- marily to the simplicity of the mechanical and elec- trical construction. The smaller of the furnaces is handled entirely by a single man, except when charg- ing, when three helpers are used to shorten the charg- ing time. Each of these furnaces has been left for an hour at a time without attendance, at work at full power. This extra time the single melter uses in mak- ing ready the next roof and in shearing up scrap for easy shoveling. The other material factor in the low labor cost is their rapid melting ability and the consequent rela- tively large tonnage output per man, as shown in the labor items in the tables given, expressed per ton of steel. This output capacity is in turn due to the fact that the electric input very largely goes to melting steel. ‘This effect is cumulative, as the shorter time of heats reduces the radiation time chargeable to each ton, and so releases further energy for melting steel. This in turn reduces the labor cost per ton of output for a definite size of furnace. THERMAL EFFICIENCY In closing, it may be stated that these furnaces show a thermal efficiency not far from double that usual with furnaces of older design. The most reliable figures we can get for regular operation of furnaces of older design operating on a single shift show about 1000 k.w.h. per ton of steel, of which only 33 per cent goes into the ladle with the fluid steel and slag; the single-electrode furnaces under equivalent conditions use only aout 560 k.w.h. per ton of steel, the corre- sponding efficiency being 63 per cent. This is impor- tant commercially. The slow progress of electric steel making in the first 10 years of its commercial existence was due to its high costs. It could not compete with most types of fuel furnaces. Its plea for existence had to be based on quality of product. The advance in this country, where quality was relatively unimportant, was slow. These furnaces of newer design give costs that are . lower than most types of fuel furnaces. The higher quality of electric steel is secured for nothing. The real competition in the past has not been between dif- ferent types of electric furnaces, but between all types of electric furnaces and fuel furnaces. Now that properly designed electric furnaces can compete in cost with fuel melting on steel, their introduction in this country should be accelerated. Corrugated Steel Cleats for Barrels The Co-Z corrugated steel cleat for kegs and barrels is being placed on the market by Hubbard & Co., Pitts- burgh, Pa. It has a corrugation or rib extending through the center of the cleat from to the other. add strength to the cleat and enable it to resist shocks. The cleat is made to suit standard sizes of kegs or barrels, but variations in the width of the head are taken care of in the shape of the ends of the cleats, which adapt themselves to the heads which are not the exact standard size. The cleat can be used on barrels or kegs of any size with either steel or wooden hoops. It is pointed out that the cleats are readily applied and do not interfere with stenciling or marking the barrel for shipment, as this can be done before the cleats are applied. FE EP ST no | one end This is to A Corrugated Steel Cleat Having a Central Raised Reinforcing Rib for the Heads of Barrels and Kegs The multiplicity of cylinders in automobiles is to be discussed at a meeting at Buffalo, N. Y., Nov. 4, of the American Society of Mechanical Engineers by J. G. Vincent, vice-president Packard Motor Car Company. THE IRON AGE Oct <1, 1915 Plain Lathe for Turning } Another contribution to the numh: lathes that have been placed on the n hea that ordinarily do not build this class « hi a has been made by the Earle Gear & Ma Cor CCtiles YO leet eS \ OMpany, A Heavy-Duty Lathe Engine Turning Projectiles Having a Swing Philadelphia, Pa. It consists of a machine that js made for three different swings of 18, 20 and 24 in,, the length of bed being 7 ft. for the first size, which js the one illustrated, and 1 ft. longer for the other two, The lathe when equipped for belt drive has eight different spindle speeds including two secured through the use of back gears. In conjunction with the thre rates of feed, %, 1/16 and 1/32 in. per revolution of the spindle, twenty-four cutting speeds, the highest being 68 ft. per minute, are available. If this rang is not sufficient, the substitution of an adjustable-speed motor for the two-step cone and the countershaft pro vides additional speeds. To enable the machine to be used for performing the nosing operations on shrapnel and other shells a spindle with a 1-in. hole extending entirely through it is pro vided for the headstock, together with a forming at- tachment milled to the contour of the nose. At the faceplate end a recess is bored into the spindle whieh is of sufficient diameter and depth to take any she within the capacity of the lathe and give a chuck gry of at least 4-in. bearing surface. Apparatus for Determining the Critical Point in Iron and Steel A complete apparatus for the accurate determina tion of critical points in iron, steel and non-ferrous alloys has been put on the market recently by @ Scientific Materials Company, of Pittsburgh. Scimatco method, as it is called, is an adaptation sensitive and accurate method for detecting measuring small thermal changes in solid bodies. based upon automatic comparison of the temperatur during heating and cooling, of the metal test specimen with that of another neutral body which has no tran formation points of its own—at least not within te temperature ranges studied. Both are heated and coolet together in one vertical heating chamber under abst- lutely uniform conditions. The critical points are Ts directly on a specially designed duplex instrument, a special chronograph is provided, from the recorés which all kinds of curves may be plotted. The Snyder Electric Furnace Company, \***" states that the electric furnace which it recent’) stalled for S. Fair & Sons, Inc., Saginaw, Mich. ¥* under guarantee that the kilowatt-hour consumptio" r ton would not exceed 880. The first two heats avere®” 620 k.w.h., and the third heat, from a hot furnace, 0" k.w.h. Basing current on 1c. per k.w.h., this means *”" the furnace from the start operated at a cost of * ton less than that guaranteed for it. The Warren Tool & Forge Company, Warren, is working on an order from the British over™ for 80,000 picks to be delivered Nov. 20. Spring Forming Machine ving illustration is a view of an forming and tempering machine nstalled in the plant of the Ford With respect to the forming the straight piece of spring steel is hine is similar in principle and i ne described in THE IRON AGE, 7 d installed in the plant of the Har- eg pany, Racine, Wis. This machine, owever ad of reciprocating consists of six Son wnhed n a central axis, the rotation of the an seat the spring through the oil bath. The ne is , equipped with an automatic ejector ‘) catches the spring when it is released from nea ng clamp after emerging from the oil bath aaa ' Te wring Machine Requiring no Atter Steel on the Feeding Shelf of the f the Finished Product from the Oil Bath it upon the screen shown in the reground to drip. The operation of the entire trolled by a single clutch, the subse- owing automatically. teel is heated in two parallel con- Inserted in the bed of the ourses of brick running the length and mounted on a cam-actuated the travel of which alternately lifts steel and carries them forward as the ses above the level of the furnace them down again as it sinks below the From the placing of the steel upon ble of the furnace until the finished eted, the entire operation of heating, npering is automatic. eea Turnaces. orizontal Two-Head Drilling Machine Mfg. Company, 4824 Lake Street, successor to Fritz A. Schulz, is placing : double-head drilling machine. The ne headstocks on the bed permits of n., While the headstocks are hori- table on gibs to a maximum gap of centers. The spindles have a travel are bored for %4-in. Morse taper. a floor space of 24 x 60 in. the spindle may be set for any c the maximum by a ratchet attach- Pie n the machine is set for a certain ne hand lever control may be easily machine controlled from the foot erator then has both hands free to ed. The driving pulleys are 4 in. THE IRON \ Horizontal Double-Head Drilling Machine Capable of Handling Work up to 30 In. in Length and 10 I Diamet in diameter and 3-in. face and are mounted, as may be noted, between the bearings of each spindle. The machine is especially suitable for light drilling where holes on opposite sides or ends of a piece can be finished in one operation, and it can be used for end milling or counter boring ad vantageously. Using this ma chine is a convenient tool for threading both ends of bar stock. self-opening dies, A Combination Boiler Blow-Off Valve A regrinding and renewable blow-off valve of the angle type has been designed by the Homestead Valve Mfg. Company, Homestead, Pa. It is intended for use with one of the maker’s plug cock ied quarter-turn valves to give a double blow-off arrange ment. By using the combination shown of the two valves a double protection against leakage 1s secured and it is possible to close the valve the boiler, thus enabling the blow-off valve to be repaired without closing the boiler. The blow-off valve nearest dow! are ! atior Ang Blow has a semi-steel body i Quart Tur ‘ with a Monel metal me valve seat and disk and bronze trimmings. The valve is designed to give a large clearance for repacking, and when the valve i open the disk is entirely above the valve port. The disk is reversible and can be reground when necessary Loosening the four bolts