The Iron Age 1933-02-23: Vol 131 Iss 8

bons, itain- dif. tical ature n of and com- pro- nge- le to has be- this all This the the of ade 1 of y in and the de- ent rds, for for ese ved im- an- er se ad, ny ll . THE IRON AGE.. ESTABLISHED 1855 FEBRUARY 23, 1933 Vol. 131, No. 8 Reducing Polishing Costs UMBLING, or barrel rolling, T for the purpose of polishing and “coloring” metal parts, is one of the oldest mechanical processes, yet in its aspects as a mechanical art it is one of the newest. Some of the old-line New England novelty and no- tion manufacturers have been employ- ing tumbling for such articles as safety pins and eyelets for half a cen- tury, but it is only within the last eight or ten years that tumbling has been taken up generally as a finish- ing method, and only since the be- ginning of the present business de- pression that managements have rec- ognized the art involved and _ its adaptation to an infinite variety of parts for all kinds of products. Examining into the vast literature of metal finishing I find very little published data on tumbling, and prac- tically no specific information on the classification of work to suit the dif- ferent types of barrels, abrasives, and other variables. In no place do I find the process put down in a system of rules of skill, or as an “art.” The quickened interest in tumbling during the past three years has arisen from the desire to eliminate or les- sen the comparatively expensive hand operation of polishing, which on cast- Ings is frequently very costly. This drive to save hand labor during a period when many operators are un- employed has been a strange paradox of the present depression, but it is a recognized fact in the progress of the machine age, and will eventually be met by measures which are outside the field of discussion of this article. A few years ago there were quite definite lines drawn in regard to by Tumbling By GEORGE S. BRADY SE of the tumbling, or barrel rolling, process for polishing and “coloring” metal parts has increased, especially during the past three years, because of sav- ings it makes possible in eliminat- ing or reducing hand-polishing operations. Successful applica- tion involves consideration of a number of variables, many of which are discussed in this article. The information given covers both dry tumbling or coloring and wet rolling. what class of parts or pieces were adaptable to labor economy by tum- bling. But at the present time there is no dividing line, and every day sees new parts finished by this method. As in the practice of any other art, the fundamentals consist in the recog- nition of the variables, and the ap- plication of proper skill to these vari- ables to produce the desired results. In commercial production tumbling there are the following variable factors: 1—Size and shape of piece 2—Kind of metal 3—Nature of finish desired 1—Type of barrel to use 5—Kind of abrasive to use 6—Speed of the barrel 7—Quantity of work in the barrel 8—Time required in barrel. 303 It is no more possible to give a set of rules to cover the first three items than it would be to try to give def- inite rules for the blending of colors to produce an effect on a painting. Every individual part to be finished by tumbling requires separate analy- sis, and having once established the operations for a part they should be entered on the operation or routing cards in the same manner as the de- tailed heats and quenching media for the heat-treatment of steels. In plants where hundreds of different parts are passing through the polish- ing department it is not sufficient to designate the operation merely as “tumble.” Sheet Metal and Screw Machine Parts Prepared for Plating Small sheet-metal parts made of cold-rolled steel or of rolled brass, or screw-machine products turned from drawn metals, almost always have surfaces free from heavy scratches and are satisfactory for plating if an even shine or “color” can be produced uniformly over them. Except for the cheapest class of work, “coloring” by an acid pickle is not satisfactory be- cause the acid accentuates the roll lines or tool marks. This class of work, therefore, constitutes the first group of application to tumbling. To obtain rapid results and an even “color” in tumbling on sheet-metal and screw-machine work it is neces- sary to eliminate bad tool marks, which is easily accomplished if brought back to the source. Since this class of coloring is usually done only with dry sawdust or leather shavings as the abrasive, it is not possible to polish out the tears caused by tight forming tools, the burrs from dull blanking dies, or the deep scratches caused by dull cutting tools. But by referring this bad work back to the manufacturing departments actual economies have been obtained as well as better work by the resulting im- provement in the tools. The increased use of tumbling has thus many times resulted in improving workmanship in the whole plant. Dirty Sawdust Will Leave Blemishes If the above class of work is oily it is given a short 10 or 15 min. tum- ble in old sawdust to remove the oil, since dirty sawdust in the finishing operation will leave blemishes that will later show in the plate. Only new, clean sawdust is used for finish tumbling. This material is a special hardwood sawdust, usually maple, sold to the metal-finishing trade. The use of scrap sawdust from the wood- working department is not advisable as a general practice. The amount of sawdust employed in the barrel de- pends upon the shape and size of the parts, and should be only sufficient in quantity to cushion the pieces to pre- vent scratching. Clean, dry, fine sand may also be employed for “color” tumbling. For ordinary dry tumbling or col- oring 4 to 6 hr. tumbling in a stand- ard tilted barrel at about 30 r.p.m. is sufficient, but for high coloring a much longer period of time may be required. Experimentation is neces- sary to determine the proper amount of sawdust, as the proper proportion to suit the shape of the piece will help to shorten the time of tumbling. Screws and screw-machine parts that have been nicely machined can be brought to a fine finish in less than 2 hr. of tumbling if enough sawdust is used to prevent the pieces from tumbling on themselves. The speed of the barrel can be brought up to a point just below that at which cen- trifugal force would act to whirl the mass together. For parts having fine threads or small holes a fine grade of sawdust is required to prevent clog- ging in threads, slots or holes. Links and formed wire parts are very adaptable to dry sawdust tumbling, it being only necessary to govern the quantity in the barrel and the amount of sawdust to prevent hooking to- gether and bending. It must be re- membered that dry sawdust tumbling is essentially a “coloring” operation, and although applicable to a vast vari- ety of work, is not a process that will remove burrs or heavy scratches. Factors Governing Wet Rolling of Small Parts An important consideration in sheet-metal and small-parts tumbling is that of wet rolling. Two factors govern the requirements of this proc- ess as apart from the cheaper method of coloring by dry tumbling; i.e., shape of the piece and degree of pol- ish. Sheet-metal pieces having con- siderable flat area require a grinding 304—The Iron Age, February 23, 1933 or burnishing that cannot be produced dry. Small parts whose outside posi- tion on the product calls for a smooth high polish must also be burnished in the tumbling process. But wet rolling will kill the original finish of rolled brass or steel parts, and a dry saw- dust tumble may be necessary as a second operation. For wet rolling the liquid burnish- ing medium consists of a solution of potash, soap, sodium cyanide, cream of tartar, or one of the proprietary cleaning compounds. Most screw-ma- chine parts and formed-wire parts are of a nature that permit rolling on themselves in the solution without the use of any solid abrasive. But larger pieces, especially sheet-metal parts with flat surfaces, require the use of abrasive powder, sand, or other burn- ishing material. Frequently, the ad- dition of pumice or emery to the solu- tion is sufficient to remove burrs and tool scratches, but the larger pieces require balls to act as a rolling cush- ion. Considerable skill and judgment are needed in the determination of the best rolling media. Punchings, as Well as Round Balls, Employed Commercial balls for rolling are marketed in a number of sizes from 1/32 to % in. in diameter, but gen- erally two sizes are sufficient for the work of the average plant: the 3/32 in. size for small parts and the 5/16- in. size for larger parts. In plants where there are large quantities of small punchings produced as scrap material these punchings are used in- stead of balls and have advantages over the round balls for many types of parts. It is the custom to use steel punchings for brass parts and brass punchings for light steel parts. Irregular-shaped punchings can be employed, and in the tumbling of heavier steel pieces, such as forged gun parts, iron “jacks” and pebbles may be used. The proportion of balls or punchings to be used depends upon the size and shape of the parts being tumbled, and can only be determined by experimentation. While some parts will burnish by rolling on them- selves with few or no balls, others require sufficient balls to keep them from hitting and scratching and to eliminate “nesting.” The amount of liquid in the barrel for wet rolling is only sufficient to cover the work. The proper speed of the barrel for wet rolling depends upon the bulk and weight of the part, and should be be- low the point of centrifugal effect. Barrel speeds vary from 20 to 50 r.p.m., and must be determined in- dividually for each part. Some plants do a great amount of rolling for pe- riods of upwards of 24 hr., but it will be found possible to reduce these times greatly by devoting time to ex- perimentation with quantities, abra- sives, liquid and speeds. Fine abra- sive, or even sawdust added to the solution, is sometimes effective in re- ducing the time necessary to obtain a high polish. In other cases the total time may be reduced by changing progressively from one tumbling to another of different solution and abra- sive. Oblique Barrel Popular for Wet Rolling The “oblique” barrel, that is, a bar- rel operated at approximately a 25 or 30-deg. tilting position, has become the most popular for the wet rolling process, as it is stated that the hori- zontal barrel is too “abusive” for many parts. While the oblique bar- rel gives a two-way rolling action, the horizontal barrel gives a one-way throwing motion. The oblique barrels are usually of wood, and are octagon in shape to aid the tumbling action. However, it is the opinion of some experienced op- erators that the older type of hori- zontal barrel has been too hastily abandoned before the art of correct combination was understood. One of the largest New England users, whose polishing operations embrace several thousand different parts of widely varying nature, has had remarkable success with the horizontal steel and cast-iron barrel run at fairly high speeds. This plant does not attempt to complete the tumbling operation at one load, but breaks it down into two or more separate operations with dif- ferent abrasive solutions. Although this method requires more handling, a finer polish is obtained in less time. “Cut-Down” Rolling of Castings “Cut-down” rolling as a first tum- bling operation on castings is a proc- ess that has served to eliminate a large amount of hand polishing. The horizontal barrel is generally used for this work because the parts can ordi- narily withstand severe tumbling. Balls, punchings, pebbles, iron “shot,” or “jacks” are used to assist the cut- ting down of the scale of the castings. Parts of different sizes and shapes can be mixed in the same barrel, and by their action on one another will perform the same operation as the ball or slug. The second operation in the tum- bling of castings is the rolling in sand. The abrasive material used is the ordinary coarse builders’ sand with enough water added to form a paste or heavy mud. It will be found that bulky parts are likely to ride around in the barrel without much tumbling, and some experience is needed to de- termine the most efficient speed of barrel for each type of part. The ordinary speed for oblique barrels 15 (Concluded on Advertising Page 12) phas\ it mz of th Th comp not busi gath recul bein; and lar i in h reco: norn and cove lows of t gatl higt cens busi defi be 1 is i bas! ries cha tre! cha bac bee nor 71 for bee pre Wi nts vill ri- or ir- he ay of ‘id P- ri- ly ct of se al le id ot at 10 f. EFORE proceeding to the de- tailed discussion of the specific data applying to the various phases of the manufacturing industry, it may be well to take a general view of the character of the data. The period of the collection and compilation of the census data does not coincide with the period of the business cycle. The census data are gathered at regular intervals, but the recurrent phases of the business cycle, being the net resultant of innumerable and divergent factors, occur at irregu- lar intervals. Col. Leonard P. Ayres, in his Chart of Business Activities, records no less than 19 periods of sub- normal activity of varying dufation and extent within the time interval covered by the present study. It fol- lows, therefore, that the data of some of the census years must have been gathered in a period of abnormally high business activity and of other census years in a period of subnormal business activity. If, then, a clearly defined rate and direction of movement be revealed in the charted data there is indicated what is most probably a basic trend which underlies and car- ries through the impact of the cyclical change. The evidence of fundamental trend will be especially strong if the charted lines exhibit a tendency to fall back into the trend path after having been diverted by the effects of ab- normal conditions. The fitting of trend lines and of formulas to time series of data has been developed as a mathematical process in the statistical method. Where the series is known to follow a fixed law such mathematical meth- ods are exact. The accumulated amount of principal plus interest for a sum of money at a fixed rate com- pounded at fixed intervals furnishes an example of a phenomenon that fol- lows a fixed law. It can be plotted and a curve or a formula can be fitted precisely by exact mathematical meth- ods. Such a curve can be extended (projected, extrapolated) in either di- rection to any degree. The data with which we have to deal, however, follow no such precise mathematical pattern. They repre- W hat the Machine Has Done to Us By WALTER S. GIELE sent, in fact, the net results of count- less variable influences, which follow no fixed laws with respect to them- selves or to each other. Under these conditions no precise mathematical method is warranted or here at- tempted. Where, however, the magni- tudes in a series show an unmistakable tendency, a trend line may be sug- gested by inspection alone. Within the time limits of the data, general direction and either acceleration or retardation in the rate will be pretty clearly indicated. Even though exact time and exact extent may not be cal- culated, reasonably certain inferences may be drawn as to the approach to upper or lower limits and possible re- versals of direction. The discussion of the fitting of trend lines is fully cov- ered in Mills’ “Statistical Methods,” page 288. Changes in Growth of Industrial Employment During the earlier development of this country’s manufacturing indus- tries the number of wage earners em- ployed in those industries increased at a rapid pace, following very nearly a geometrical progression. That is to say, each year’s increase in numbers HIS is the third article in the series by Mr. Giele in which he presents a factual study of the long-term records and relation- ships of mechanization, wages and employment. Preceding articles appeared in THE IRON AGE of Feb. 2 and 16. In this chapter the author de- velops the fundamental trends in employment and wages in the manufacturing industries and de- duces from them that we were ap- proaching, even prior to the de- pression, a stationary rate of in- crease in employment but not of wages. was a nearly constant percentage of the total number employed at the end of the preceding year, from 1849 to 1889. The increase was rather faster than the average rate under the stimulus of the Civil War in the decade from 1859 to 1869, and rather slower in the next decade, which in- cluded the long depression of the 1870’s. During this total period of 40 years, the growth of manufacturing industry absorbed an ever-increasing proportion of men from the forests and farms, the gain in industrial em- ployment being 345 per cent as against a gain of but 165 per cent in popu- lation. The depression of the 1890’s brought an abrupt change of trend in employ- ment even though there were more wage earners employed in manufac- turing industries in 1899 than there had been in 1889. The first decade of the new century, however, saw a re- sumption of industrial expansion ac- companied by a world-wide rise in commodity prices, and the increase in factory employment resumed its for- mer rate. The five-year period from 1909 to 1914 was marked by many events dis- turbing to business sentiment, cul- minating in the panic following the outbreak of the World War in the lat- ter year. Industrial employment dur- ing the five-year period from 1909 to 1914 fell back to the rate of growth characteristic of the 1890’s, being con- siderably influenced by the many strikes that marked this period. The period from 1914 to 1919, including as it does the “war era” and the post-war boom, is marked by a rate of increase in industrial employment only very slightly faster than the rate of in- crease during the first decade of the century. The sharp post-war depression is reflected in the sharp drop in em- ployment in 1921. By 1923 employ- ment had very nearly regained its 1919 peak in number of wage earners, a level which was approximated in 1929, though 1931 showed a level of employment lower than that of 1921. The departure from the uniform per- centage (straight line) rate of in- The Iron Age, February 23, 1933—305 I EE Ere oa 849 1859 1869 1879 1889 __1899 eee + > 80] | | | + - a | gees oe wack = ; “"| ANNUA “) cq } wa pt, -* a (Hundreds 01 +-—+1909 t—4 1914 9 92 9 9 9 Q 93! +— 41904 Chart 11—Growth of total wages and number of wage earners in our manufac- turing industries and record of individual annual wages crease which became evident after 1889 is now revealed as a permanent change in characteristic and seems very clearly to point to an approach to an upper limit. Such an approach to or attainment of the upper limit in numbers employed in manufactur- ing industries does not, however, of necessity forecast an alarming con- dition of unemployment. It may in- dicate merely that, with a total popu- lation rapidly approaching a station ary rate of growth, the great indus- try groups will approach a state of equilibrium in which there will be a nearly constant distribution of occupa- tions. Agriculture, the oldest indus- try, has probably passed that state of equilibrium. It seems logical to ex- pect that the manufacturing indus- tries would be the next to reach it. The aggregate wages paid to the wage earners in manufacturing indus- tries, like the number of wage earners, exhibits an almost uniform rate of growth in geometrical progression from 1849 to 1889, but the rate of growth of wages is faster than the rate of growth of number of wage 306—The Iron Age, February 23, 1933 earners. It is here to be noted that this aggregate includes all wages paid to all wage earners in manufacturing industries, and thus evens out the dis- equilibrium between industries and oc- cupations. The depression of the 1890’s brought a change in the rate of increase in wage payments more abrupt than the change in the rate of increase in wage earners, but the entire 15-year period from 1899 to 1914 shows substantially the same rate of increase as did the 10-year period from 1849 to 1889. The first decade of the new century was one of bumper crops and high crop prices so that, there being little incentive to leave the farm, the factory labor supply was not so urgently augmented from the ranks of farm laborers, and an approximate equilib- rium was reestablished between de- mand and supply. The sudden halt in immigration from the European countries involved in the war, followed by the removal of mil lions from the ranks of industrial em- ployees for military and naval service with our own entry into the conflict combined with enormous demand tor munitions and supplies, completely up- set that equilibrium. The aggregate of wage payments fairly shot upwards from 1914 to 1919. The intersection of the lines of wage earners and wages on the chart has no significance other than that the average annual wages per wage earner passed the one thou- sand dollar mark at that point. The depression of 1921 brought a reduction in wage disbursements as payrolls were reduced with the reduc- tion of working forces, but 1923 saw a total higher than that of 1919 to be exceeded in turn by the total of wage disbursements in 1929. No Permanent Change in Wage Trend Indicated The abrupt drop shown for 1931 in both wage earners employed and wages paid will, in all probability, be extended through 1932 and possibly not materially raised in 1933. There is, however, nothing in the data so far available to indicate that this is differ- ent in kind from the sharp drop shown in 1921, though greater in extent and duration. There is surely nothing in the ac- cumulated data of 80 years’ experience to indicate that the lines will not re- turn to the level of the suggested trend. Wages per wage earner as here com- puted and charted are the average an- nual money wages per wage earner employed in manufacturing industries. As such the figures eliminate the effects of discontinuity of employ- ment, and are independent of varia- tions in hours worked and hourly rates of wages. They represent an average for all of the manufacturing indus- tries. They do not reflect the experi- ence of the individual as it may be influenced by locality, industry or oc- cupation and apply only to those em- ployed. They do reflect the amount of money the average wage earner re- ceived in wages for the year. As the census figures for 1859 were collected two years before the out- break of the Civil War and those for 1869 four years after its close they do not reveal a probable temporary rise during the war period. There is, therefore, indicated a steady rise from 1849 to 1889. The lower demand for workers in proportion to the potential suppl 1890": earne after atas In howe norm ply read: aver most serie annu in t 1914 65-y ann close geol 1.36 W just pres ann turl to ] enc to nu rel th: Wi Wi tor 1p- of rds ion res er res u- ly re ar .Y- wn nd LC ce id, n- he a- re supply in the depression decade of the 1890's is reflected in a wage per wage earner lower in 1899 than in 1889; after which the increase was resumed at a somewhat faster pace than before. In the period from 1914 to 1919, however, with the sudden and ab- normal disturbance between the sup- ply of and the demand for labor al- ready noted, there occurred in the average Wage per wage earner the most violent shift noted in any of the series considered in this study. The annual wage was practically doubled in this period, rising from $590 in 1914 to $1,168 in 1919. During the 65-year period from 1849 to 1914 the annual rate of increase adheres quite closely to a trend line representing a geometrical progression at the rate of 1.36 of one per cent per year. While there were some wage read- justments during the post-war de- pression of 1921, the average rate of annual increase was but little dis- turbed and for the decade from 1919 to 1929 shows an almost exact adher- ence to a straight trend line parallel to the trend preceding 1914. There seems to have occurred not so much a change of trend as an offset in the trend line. The single offset, however, is equivalent in effect to the entire ac- cumula’ed increase from 1859 to 1914. Current Wage Decline a Temporary Dislocation? The abrupt decline in average an- nual wages per wage earner in 1931 reflects part-time employment rather than reduction in wage rates. More will be written of this in connection with real wages and the cost of living. John H. Van Deventer, editor of THE IRON AGE, suggests that the over- expansion of manufacturing equip- ment and the over-expansion of mar- kets were not the choice of industrial management. It was, perhaps, the result of an effort to meet the condi- tion of a war-time wage scale in a world otherwise on a peace-time level by increasing the rate of turnover. The 1921 depression had gone far in the readjustment of the commodity price levels, but had no permanent ef- fect on the price of labor. The effort Was to meet the depressed selling price by a reduction of unit labor cost with- out a reduction of wage rates. 1849 1859 1869 + DOU D— ~~ eaNINNINIANS D RDHAHAMNMA 1899 904 1909 0} Bind = | a | : | H IZ | re 3¢ | oo eed ae Se : | Sian | (b) ee | | | | Le | bee ps Ft ah ela ese T UIT c | e *] rE | - + a I ’ +4 x : bl ea = % = a | [ees ee Chart 12—Trend lines of employment and individual annual wages in the manu- facturing industries. Lines a and b show the remarkable similarity in wage trends for pre-war and post-war periods, in spite of the sudden shift to higher level in 1919. Scale reading is hundreds of dollars. Curve c shows the prob- able trend of employment in the manufacturing industries, expressed in hundreds of thousands of wage earners This was not and is not a battle in the once traditional war between “capital” and “labor.” It is a battle in the eternal war between the buyer and the seller, a war among the great occupational groups each of which is buyer from all others and seller to all others. No Group Can Receive Dispropor- tionate Share In this war capital and labor in any one industry are like army and navy of one nation, two branches of service united in one cause. All consumers, be they wage earners in or owners of enterprises, are buyers of the products of other industries. Neither wage earners nor owners of one group of enterprises can receive a dispropor- tionate price for their services or goods except at the expense of both wage earners and owners of other enterprises in other groups. To sum up, the number of wage earners engaged in the manufacturing industries would seem to be approach- ing or to have reached a stationary rate of increase. This fact does not of necessity imply an alarming and permanent condition of unemployment but may indicate, rather, an approach to an equilibrium with a stationary rate of increase in population growth. The average value of money wages per wage earner has shown a remark- ably uniform rate of increase through- out an 80-year period but including a remarkable offset without change in rate of increase. Current drastic reduction of an- nual earnings through unemployment rather than through reduction of wage rates in the manufacturing industry may, in considerable degree, be a re- sult of this extraordinary dislocation. Michigan State College, East Lan- sing, Mich., is planning to give a short course in foundry work on April 6, 7, and 8. Registration fee for the course will be $5. The Iron Age, February 23, 1933—307 = te COS ena RANSMISSION gears are re- quired to withstand heavier loads and more wear than per- haps any other part of an automobile. In view of these severe service con- ditions, the Packard Motor Car Co., Detroit, has used a case-hardening 5 per cent nickel steel for its gears, not- withstanding the fact that it entailed unusual gear manufacturing proc- esses. In the last five years the com- pany has made and installed in new cars more than 1,078,400 of these gears, and of this large number very few have had to be replaced. Helical transmission gears of the same hard steel are now made by the company. These require uncommon precision in machining, but transmis- sions equipped with them provide quiet operation with high strength and wear resistance. The helical gears require more than double the work necessary for producing the straight tooth gears. An important feature is the grinding of the interior splines to assure proper fit of the gears on their shafts. This is done on a grinding machine equipped with a special device by which the tops and sides of internal splines are ground within close limits with a single set- ting of the work. At the end of the power-driven re- ciprocating table are pedestals which constitute tool supports on which grinding devices are mounted. The work-supporting head, consisting of a pair of concentric sleeves, is mounted on a pedestal attached to a stationary bridge in the middle of the machine. Within the inner sleeve is a work- holding member having an axial bore and a counterbore in one end forming a circumferential ledge or shoulder 308—The Iron Age, February 23, 1933 against which the transmission gear rests. The gear, having a_ splined bore, is secured in the head by placing it on a mandrel, the protruding end of which is inserted in the bore. While thus centered, the gear is clamped against the shoulder and the center- ing mandrel removed. For grinding the top of the spline, a device at one side of the head is mounted on a pedestal which has a transverse slide carrying a bearing in which a high-speed spindle is mounted. The inner end of the spindle carries the grinding wheel. The spindle is moved in and out of the bore of the work by reciprocation of the table on which the device rests. When the top is ground within proper limits, the wheel for grinding the sides of the splines advances from the opposite side of the workhead which holds the gear. The machine is equipped with a special indexing mechanism by which the workhead is moved into pre- determined position and locked. After the sides of the homologous splines have been ground and the work has been rotated through one complete revolution, the gear is revolved again through similar steps during which the other sides. of the splines are ground. The special grinding wheel spindle acts as its own pulley. FTER heat treat- ment, transmis- sion gears are lapped to remove any dis- tortion. Lapping ma- chines are of the vertical type and gears are lapped an average of 50strokes. At the right of each operator is a sound- proof room where gears are mated and tested. Interior splines are ground to assure proper fit of the gears on their shafts. The tops and sides of the splines are ground within close limits in a single setting of the work. aaa Precision Required in Machining|Nic Be hard “sha' prox! the t oper powe press the | and finis! cont) erat push tabl head posi rack geal stro and ' is | a ge the ngiNickel-Steel Helical Gears Before gears are heat-treated and hardened, they are put through a “shaving machine” which removes ap- proximately 0.005 in. of stock from the teeth. The machine is hydraulically operated, high pressure being used for power on the table travel, while a low pressure cylinder mounted on top of the head raises and lowers the head and locks the machine to the correct finishing dimension. The machine is controlled by a single lever. The op- erator puts the gear on an arbor and pushes the lever, which starts the table traveling automatically, the head coming down to a predetermined position. On the table is a cutting rack with 62 teeth, which machine the gear teeth. After the table makes 50 strokes, the head automatically raises and the table stops. There are two arbors, one of which is loaded while the other is holding a gear for the finishing operation. At the side of the machine is a special EFORE heat treat- ment, the gears are burnished on a gear burnishing ma- chine. RIOR to being hardened, gears are put through a hy- draulically - operated “shaving” § machine which removes 0.005 in. of stock from the teeth. The gear is held on an arbor, and the table, with a cutting rack having 62 teeth, makes 50 strokes during the machining cycle. The machine handles three 25-tooth 10- pitch, %4-in. face gears per min. By BURNHAM FINNEY place to hold the arbor not in use and to drain the dripping oil into the oil tank in case the arbor has just been removed from the work. The head of the machine is fed into the gear under a certain pressure so that if a gear mounted in the machine has an ex- cessive amount of stock, the cutting rack will not break or the machine stall. The head of the machine can be set either right or left to 55 deg. helix angle. The production time for a 25-tooth, 10-pitch gear, with 44 deg., 24 min., 55 sec. helix angle, %-in. face, and 0.003 in. to 0.005 in. stock on the tooth thickness, is three per minute. From the shaving operation gears go to a gear burnishing machine where they are burnished and thence to heat treatment. After being heat treated and hardened they are lapped to remove any distortion caused by the hardening. Lapping machines are of the vertical type. Gears are lapped an average of 50 strokes, kerosene being used as both a coolant and lubri- cant. Each machine is individually motor-driven and has an exhaust duct above it to remove the abrasive dust and dirt arising from the lapping operation. To the right of the lapping machine operator is a special washing machine in which gears are cleaned and flushed of all abrasive material after being (Concluded on Advertising Page 12) The Iron Age, February 23, 1933—309 Practical Methods for Heating ¢, ‘NHE most simple and often most useful form of an _ inductively heated muffle is one in which a short or a long tubular muffle furnace is maintained at a desired tempera- ture by heating the tube inductively with an inductor coil which surrounds it. The tube may have any desired length and be of any cross-section as round, oval, square, rectangular or semi-circular with a flat bottom. The materia] of the tube may be nickel- chrome, chrome-steel, or even graph- ite when provision is made to protect this from oxidation. Whatever the material or form of the tube, it should be considered the susceptor on which the inductor acts. The inductor winding is usually made to conform to the contour of the cross-sectional area of the sus- ceptor. It is good practice to make the coupling between inductor and susceptor between 50 and 60 per cent. The space between the two is filled with any suitable highly heat-insulat- ing refractory. The material to be heated, such as strip metal, rods or tubes, may be in- termittently placed in the muffle un- til it has reached the desired tempera- ture or the same may be fed steadily through the muffle. All heating of the material results from heat radi- ated to it from the inside wall of the muffle. It is easy to maintain the heating chamber filled with a _ pro- tecting gas. The wall thickness of a_ tubular muffle would be chosen usually from 4g to %4 in. but it may be much thicker, if desired, without appreci- ably altering the efficiency of the muffle heater. A muffle heater of this character, with heavy duty to perform should be generally not less than about 2 in. in its least outside diameter. The reason for this is that a higher fre- quency is required to inductively heat a susceptor efficiently which has a small diameter than one of larger diameter. The highest frequency which may be economically supplied with a generator is about 5000 cycles and an inductor type of machine is re- quired for this. It takes about 5000 cycles to effi- ciently heat a susceptor as small as 2 in. in diameter. Wire wound al- ternators are now standardized to give 960 or 1000 cycles. This fre- quency is quite adequate to heat a 4-in. diameter susceptor but hardly high enough for one only 2 in. in di- ameter. The efficiency, uniformity of temperature, and the ease of control of temperature of an_ inductively heated muffle is surprisingly satis- factory. A manufacturer of razor blades has in service six inductively heated muffle furnaces which are giving en- tire satisfaction both from the tech- nical and economic viewpoints. Three of these mufflé heaters were photo- graphed before’ shipment and are shown in Fig. 3. One of the heaters appears with one end fitting taken off to reveal the sectional construction of the muffle, and the other heaters are shown partially and entirely com- plete. Following is a brief description of the power-generating equipment, the six muffle heaters, and their per- formance; also of the economies they have effected: The frequency changer set consists of a General Electric Co. 60-kw., three- phase, 550-volt, 4800-cycle generator, to- gether with control panels, instruments, voltage regulator and exciter. The gen- erator is directly connected to a stand- ard 125-hp., three-phase, 440-volt, 3600- rp.m. squirrel-cage induction motor. The overall dimensions of the muffle heaters are 10 in. by 10 in. by 14% ft. long The muffle is made of a welded nichrome tube 2 in. in diameter and flat- tened to an oval shape. Its wall thick- ness is 0.1 in. and it is also 14% ft. long. The entire length of the muffle is coy- ered with heat insulation and inserted in the inductor At the exit end of the heat- er, the muffle is held in a fixed position At the free end provisions have been made to allow for the expansion of the muffle which has been found to be 1% in. while the other end is free. The inductor consists of a thin wall copper tubing wound the full length of the muffle and is water cooled, Taps are provided in order to control the power input to each heater. Performance of the Heaters The heaters operate at a tempera- ture of 812 deg. C. (1500 deg. F.) and it is essential that a uniform tem- perature be maintained throughout the length of the muffle with the exception of 1 ft. from the exit end. The number of turns per inch are increased at this point to bring the temperature up to 870 deg. C. (1600 deg. F.). The heaters are electrically con- nected so that a balanced three-phase load is had when all the furnaces are on. In order to provide a bal- anced load when one or more fur- neces call for a drop in temperature (power off) a dummy load is used. This load consists of a water-cooled resistor and draws about the same amount of power as one heater. There is one dummy load for each heater. A temperature controller is con- nected to each heater and these are so arranged as to provide three con- trols, namely: (1) off (no power), (2) holding power (constant temper- ature) and (3) raising power (rise in temperature). These controls op- erate contactors mounted on_ the Fig. 3.—-Muffle heaters, in use in an American plant, for inductively heating razor blade steel. 310—The Iron Age, February 23, 1933 Ing Solids by Induction iffle ft. led at- ck- ng OV- in at- lon ree ow 1as all of ips yer ‘a- nd m- ut he id, he 00 se es re a fe te By DR. E. F. NORTHRUP Vice-President, Ajax Electrothermic Corpn. Trenton, N. J. BBs use of electric inductive heating in the manufacture of razor blade steel and chromium plated tubes is described in the current article, which is the concluding portion of an abstract of a paper presented by the author before the Pittsburgh district section of the Association of Iron and Steel Electrical Engineers. Wide application of inductive heating is en- visioned by Doctor Northrup, who states that its use will free metals and alloys from poisoning by deleterious matter and will prevent the forma- tion of hundreds of tons of surface oxides on metals during heating. switchboard in the generator room and function according to the tem- perature demands of the muffle. The thickness of heat insulation on the muffle was so designed as to give a reasonable efficiency and at the same time permit the tempera- ture of the muffle to drop quickly. This feature is highly desirable as it gives a precision control of the temperature. For example, after the working temperature has been reached, it is possible to raise the temperature 25 deg. F. in one minute with the “raising” contactor and also cause a drop of 40 deg. F. per minute on the “off” position. In the gen- eral operation of these muffle heaters, a temperature is held throughout the day at 1600 deg. F. with a varia- tion of less than 0.4 per cent of ab- solute. Some of the Economies The six heaters operate continu- ously with a power input of 62 kw. measured at the motor terminals. The total production of heat-treated strip steel for the six heaters amounts to 92 lb. per hr. giving a power cost of 5 mils per pound at 7.5 mils per kwhr. or $5 per 1000 lb. At this point the author fully dis- cusses details of many other applica- tions of inductive heating, some of which are: Muffles for batch heating, single sheet metal and continuous strip metal heating, radiation inductive sheet heater, ‘“‘semi-inductive sheet heater,” roll heating, heating fer- rous metals below decalescent point, and batch sheet heating by semi-direct induction. There is a section devoted to a dis- cussion of heating magnetic steels in- ductively. Direct Induction for Heating Straight Ferrous Tubes and Pipes oo application of induction to the heating of straight tubes and pipes is very simple and presents no complications, nor requires unusual vvyv equipment provided the tube or pipe has a diameter sufficient for effec- tive heating with the frequency available. For heating tubes three or more inches in diameter, very effec- tive heating can be done with cur- rent of the usual frequency used in metal-melting practice. Where mo- tor generator sets are used for this purpose, to supply the high-frequency power, a frequency of about 1000 cycles per second is now very gen- erally recognized as being most sat- isfactory from both technical and eco- nomic considerations. Any problem which involves the heating without melting of symmetrically shaped metals, solid or tubular, which are over three inches along the smallest dimension of a right section, may therefore be handled with what may be termed “standard high-frequency power equipment.” Owners of such equipment, and there is now installed about 12,000 kw. in the United States, may have at times an excess needed for their melting requirements. It is well for these to know the uses, other than melting, to which the equip- ment may be put. IG. 4. — Inductive heating of chrom- ium-plated tubes in use by the Heppen- stall Co. at Pitts- burgh. If the tubes are of smaller diam- eter, down to about 1% in., a higher frequency than 1000 cycles is indi- cated. High-frequency generators of the inductor type giving 4800 cycles are now in commercial use and the current supplied by these will heat efficiently tubes, rods, ete., of the smaller diameters. When a steel tube is chromium- plated, the cohesion and quality of the plating is greatly improved if the tube is heated by induction to 1000 deg. C. or over. The Heppenstall Co., of Pittsburgh, has been using the inductive method for heating chromium-plated tubes. The writer is greatly indebted to this company for taking and for allowing him to publish the photograph which is reproduced in Fig. 4. As plainly appears in the illustra- tion, the tube being heated passes through a relatively short inductor coil and is regularly and controllably rotated and advanced at an adjust- able speed by the revolving asbestos board disks on which it rests. This (Concluded on Advertising Page 10 The Iron Age, February 23, 1933—311 a Widening Scope structure in an aluminum alloy casting, the permanent mold process is usually employed. This process consists of pouring molten metal into a hot metal mold or die, but differs from the die casting process in that the metal is fed into a hot mold cavity by gravity rather than into a water-cooled die by pressure. In the best practice, the mold is de- signed and operated so that solidfica- tion takes place progressively; that is, the part of the casting farthest removed from the sprue and risers solidifies first, and it is this progres- sive feeding of the casting, together with the rapid chill the metal receives on coming in contact with the metal- lic mold, that makes possible a casting which represents the highest degree of metallurgical perfection in the cast- ing art. ik O produce an extremely dense The fine grain structure of a perma- nent mo!d casting and the fact that the casting is unusually free from the more common foundry defects, such as dross, air pockets, porosity and the like, insure the maximum physical properties that can be obtained in any type of aluminum alloy casting. Of equal importance, perhaps, is the ma- terial reduction in losses in the ma- chine shop and after polishing. This, of course, results from the general soundness of the permanent mold ~~ tt RIEFED in this article is the case of the type of casting that lies between the sand casting and the die casting. Applications include those where accurate and extensive machining are neces- sary, those where lightness and count, those where either or both dimensional ac- strength curacy and smooth finish are im- portant, and those where intricate shapes are desired, these some- times involving complicated cores. vrvyv casting. The amount of finish neces- sary for machining may be reduced below that required for sand castings, and since the surface of the casting is exceptionally smooth, for many pur- poses finishing operations may be eliminated. Permanent mold castings are not precision products. Their dimensional accuracy is midway between that of sand castings and that of pressure die castings. In specific cases, however, tolerances on certain important di- mensions have been held to 0.005 to 0.010 in. Typewriter segment as cast in a permanent mold and as machined, illustrating that now such a casting can meet extensive, accurate machining requirements. 312—The Iron Age, February 23, 1933 of Aluminu A number of factors are involved in determining the cost of permanent mold castings. The initial cost of the mold, of course, is greater than the initial cost of pattern equipment for many types of sand castings; but, where a large number of castings is involved, the mold cost becomes a practically negligible item. Asa rule, the foundry making the castings stands the expense of mold repairs or replacements and the customer has only the cost of the first mold to con- sider. Frequently, it is found that the smooth and accurate castings produced in permanent molds eliminate suffi- cient machining to more than offset the cost of mold equipment, even in the case of very limited production requirements. For example, in cast- ings requiring considerable etching and machining, it is often economical to employ the permanent mold process for a small number of castings be- cause of the saving in machining and finishing cost. Since the characteristics of the permanent mold process are such that Sideframe and base for Teletype machine illustrate that intricate shapes can be produced by the permanent mold process. a smaller amount of finish is required than for the sand casting process and the tolerances are smaller, less metal is required per piece. This saving in metal. cost often proves to be quite an CI imp¢ case go CE whe meti A pro mal ploy con: cor In pro pos ma int) per umPermanent Mold Castings d in nent of chan nent but, S is S a ‘ule, ngs $s or has ‘on- important item, as, for example, in the case of a quantity production run on a casting requiring 1.1 lb. of metal when cast in sand and only 1.0 lb. of metal when cast in permanent molds. A variation in the permanent mold process, commonly called the semi-per- manent mold process, is at times em- ployed. In this process, the mold is constructed of metal but one or more cores are of either green or dry sand. In general, the semi-permanent mold process is used to obtain as far as possible the advantages of the per- manent mold where the part is too intricate to be produced by the full permanent mold process. Fields of Application The physical and chemical proper- ties of a few of the more common per- manent mold alloys are given in the table. Some of these alloys depend upon heat treatment for their physi- cal properties, while others are not susceptible to heat treatment. The selection of alloy is governed by a number of different characteristics, such as strength and ductility, hard- ness, machinability, fluidity, casting properties, corrosion _ resistance, and/or the ability of the alloy to re- tain satisfactory properties at ele- vated temperatures. Few castings require as much ma- chining as the piston of an internal combustion engine, and the permanent mold process is employed for the production of this part almost to the complete exclusion of all other cast- ing processes. Not only do the sound- ness of the permanent mold piston and its freedom from foreign ma- terials insure better machining quali- ties, but its dense structure imparts to it better heat conductivity and greater hardness than can be obtained in any other type of cast aluminum pistons. The top plates and lower segments of typewriters also require a large amount of expensive and accurate machining and here, too, is found a definite application for the permanent mold process. The exceedingly fine grain struc- ture of a permanent mold casting makes it possible to obtain in this type of casting the maximum amount of strength with the minimum amount of weight per piece. This does not mean that permanent mold castings Bearing cap for airplane motor, which must have maximum strength with a mini- mum weight, was made by the permanent mold process