The Iron Age 1922-04-06: Vol 109 Iss 14

raga, ote x Z Gh W CC = 9 o~%, %, ; i & * GE New York, April 6, 1922 TABLISHED 1855 A PRB: Se eh Hardness Testing on VOL. 109, No. 14 a Commercial Seale The Various Testing Methods and Their Limitations —Results on Cast Iron Pistons and Steel Connecting Rods BY E. F. LAKE* have been invented and perfected for testing the hardness of materials. But it is still a question 1s to whether they are more or less valuable than the old fashioned file when sting metals on a com ercial basis. Accuracy ind reliability are the main requisites of any testing machine; but can we claim this for any of the hardness testers? Probably the oldest method of testing the hardness of materials was to drop a hard steel ball f a given size or weight from a given height above the piece to be tested, and then to measure the dis- tance of its rebound. The scleroscope greatly improved this principle. ais lessee A small portable instru- ,, ine Hardness of ment, doing the work an Automobile Pis more accurately than ton as Taken o1 ould the ball, with its ‘we Head by Sclero ruder rigging, has been ae. eee erfected. It drops a ham- LD we the past twenty years many machines ler of a given weight, Fig. 2—The Various Hardn¢ Double Connecting Ros vith a ball-shaped point f a given diameter, from i given height. It then measures the rebound of this hammer on an arbitrary scale of its own. This scale not comparable with the hardness factors of other ardness testers, unless it be through specially pre- ared tables that may err. The sclerometer differs somewhat from the scle- scope, but uses the same principle of dropping a ball, r weight, on to the test piece. It also has an arbitrary scale of its own which does not give a hardness factor hat compares with that of other hardness testing iachines. The Brinell machine was in use some years before he scleroscope and works on a different principle. It presses a hardened steel ball of a given size, with a ‘iven force or pressure, into the piece to be tested. By leasuring the diameter of the indentation made by his ball, or its depth, the hardness factor is obtained. or: ‘hus, to obtain the hardness factor, or degree of *Metallurgical engineer. Detroit. as Taken |} i Brinell Machine hardness, on this machine an arbitrary scale of its own must be used that does not compare with any of the ther hardness testers. [he Rockwell direct reading hardness tester has been on the market only a short time, but has become well known. It is in reality \ a Brinell machine that uses levers, instead of the more common hydraulic pres- sure, to force the hardened steel ball into the piece a being tested. It does not use air to operate the lev- ers, as does the Olsen horizontal Brinell machine. It has added a dial that gives a direct reading of the hardness factor, and this does away with the microscope reading that must go with other Brinell machines. The point where the pointer stops on the dial indi- cates the depth of the _ indenta- tion made by the Numbers on Different Surfaces of a ball. Therefore, to denote the de- gree of hardness, a special scale of its own is also required on this ma- chine. This does not compare with the hardness nu- meral given by other machines, unless it be the few that measure the depth of indentation made by the ball and use the same size of ball. The auto-punch has a given weight, dropped a given distance, on to an ordinary center punch that rests on the test piece. A ball, or a ball-pointed punch, can be used under the weight, if more desirable. This also requires an arbitrary scale of its own which gives a different hardness numeral from other machines. When the ball is used, this becomes in reality a Brinell tester. All Brinell machines give a different hardness nu- meral when the size of the ball is changed, or when the force or pressure is changed which causes the ball to indent the test piece. The diamond-scratch machine places a given weight, or pressure, upon a diamond, which is drawn across the 913 ote ag 4, ee oe DES BEETS Se 7 gy J ; 3 + ; * nee ° an 52 a ' h at t ” Fs ; Be 4 % 5 . . iy : ie * ie Ne By 4 £ > t: a+ ' > ¢ ue ; as : : . F ; F. : > b > B 4 Bt Ae ‘ ‘ i. : vy fy , ; a Ei » Pe . , a j «@ . A Bats 4 ' a. > %y Fi ‘4 i ” ® Pp iB ; if a a * a a 4 “ F 3 Ps Le e RS Pa Be ie - 4. te s cf - x be oe at Ph. is | * “ ” P ‘ x AS 914 THE IRON AGE surface to be tested. A microscope is then used to measure the width of the scratch or cut. The depth of the scratch could be obtained for the hardness fac- ter, but this would be more difficult and perhaps not as reliable. This also means that an arbitrary scale of its own must be used to express the hardness numeral, and this does not compare with the figure given by auy of the other machines. This machine has been varied by the use of a hard- ened steel cube in place of the diamond, for steels that have been tempered, or for materials softer than hard steel. One corner of the cube is used to cut, or scratch. the surface to be tested similar to the manner in which a diamond-point lathe tool starts a cut. This hard steel cube can be turned in its holder until all eight corners have been worn out. The magnetic hardness tester can be ignored here, as it is not used to any extent on commercial work. This enables us to divide hardness testers into three classes: Those that drop a ball or weight on the test-piece and measure the rebound; those that force, or press, a ball or center punch, into the test piece and measure the width, or depth, of the indenta- tion; and those that scratch or cut the surface of the test piece and measure the width, or depth, of the scratch, or cut. Difficulties Encountered With all of these and the different types of each there is a great deal of confusion as to hardness nu- merals because of each having a different scale from which to obtain this numeral. It is very difficult, if not impossible, to check one machine with another when we use two machines of the same type and more diffi- cult when they are of different types. This is because we cannot operate twice on the same spot without de- forming the test piece by the first operation in a way that will affect the result obtained by the second oper- ation, even though we use the scleroscope. It is clearly obvious that we cannot operate the Brinell or diamond- scratch machines twice on the same spot. The sketches in Figs. 1, 2 and 3 show how variations occur in metal even when hardness tests are made but a short dis- tance apart. To get absolute uniform hardness throughout a piece of steel which is large enough for checking pur- poses is more difficult than it might seem, outside of a laboratory. Balls under %-in. diameter that are used for ball bearings should be about as uniform in hard- ness as anything that is heat treated. Yet all makers of such balls have to reject many because they have soft spots in them. Sometimes these soft spots are no larger than a pin head, but they affect the wear of the ball. If one has to work on carbonized parts, this is all the more difficult, as soft spots of considerable size appear frequently, in spite of all that has been done to improve carbonizing methods and processes. Then, again, hardness testing machines are not so perfect but that they sometimes give different readings on two successive operations performed on the test piece as near each other as will prevent the deforma- tion of the metal of the first operation from influenc- ing that of the second. Hardness Versus Wear Probably the greatest difficulty comes from testing for hardness when wearing properties are what are really wanted. This is especially so of gears, cam- shafts, pistons and numerous other moving parts of machinery. It is a well known fact that bronze gears outwear steel gears when running in mesh with them. Yet hardness testing would tell us that steel gears were the best. Another example is Hadfield’s manganese steel. When subjected to the severe grinding action of min- eral dust, when used in stone crushers, gears made April 6, 19 from this manganese steel have worn three and times as long as gears made from any other kin steel, no matter how they were hardened and pered. Yet all of the above hardness testers would cate that the common carbon steels were harder, | better than manganese steel. Manganese steel] j toughest steel known, however, and it is this pro; of toughness that gives the greatest wearing qua Glass is admitted to be harder than any steel. it ean be cut and reduced to any desired size by { even though it be harder than a file. This is be hardness generates brittleness and the glass break into fine particles and crumbles away under the a of the file. We do not want this brittle hardness toughness, in parts that are to resist wear or st of any kind. It is a very difficult problem to show metals have the proper degree of toughness with kind of a hardness testing machine. It is probably possible and something is needed that will revea degree of toughness in materials on a comm basis. Cast Iron Pistons A concrete example of what hardness testers was brought out by some cast iron pistons that were testing for hardness to get their machina! before sending them to the machine shop. The sg; fications gave us a limit of from 35 to 39 on the sc! scope. Theoretically, pistons that were 40 hard wo not machine easy enough to pay for the scrap Even those that were 39 hard were given a spe mark and kept by themselves as doubtful. So many had to be scrapped under these limits that we sent some 40 hard to the machine shop without letting the workmen know the hardness. Not hearing any objections, we raised this to 41 hard; then we raised it to 42 and again to 43 hard. As none came back and we did not hear any complaints we were able to use up nearly all the pistons we had scrapped for hardness. To find the degree of hardness, we polished a spot, about 5% in. in diameter, near the outer edge of the piston’s head, and hit this twice with the scleroscop¢ Sometimes one reading would be five points hig! than the other, though but % in. apart. We resorted to the file for testing these pistons f hardness, and found we were able to satisfy the ma chine shop with machinable castings much better thar we did with the scleroscope, without throwing too mu scrap back on the foundry. For our own information we polished several | tons all around the outer diameter of the head and tried the scleroscope in several places with vary results. A piston that is typical of this lot is show: in Fig. 1. From the numerals, it will be seen that we could have accepted this piston on eight different readings if the scleroscope had happened to hit these places. If it had hit in three other places, however, we wo! have had to reject this same piston as being too hard or those showing 40, 41 and 43 hard. The 43 seen rather high for a piston that reads 36 and 37 at thé lowest place. But, such variations often occur in cast iron. The density might vary more than the hardness The scleroscope, Brinell machine or any other hard- ness tester would then show a greater variation b¢ tween the high and low readings than would be give! by the hardness of the piece. In commercial testing we can never feel sure that it is hardness, and no‘ density, that is affecting the results of the reading given by any type of hardness tester. But the file will bite into the metal in exact pro- portion to its degree of hardness or softness. Of course files get dull and lose their biting qualities, and in this case a little skill or good judgment is required ril 6, 1922 he file operator. Even with that weakness, it < better than blind devotion to a hardness tester might lead us astray. Density would not affect file. unless it be after it reached an extreme ap- hing porosity. Even then a file would break away etal in a brittle manner rather than in the smooth- of a cut into soft material. "This seems to be er place where the human element is not easily ed by a machine. Double Connecting Rods ‘nother concrete example is that of the double con- ng rod shown in Fig. 2. This is a fair sample of 10 rods that we Brinelled all over in trying to find tion for the many rods that were being rejected. of these were re-heat-treated three and four times e they came within the desired limits. e rods were heat treated as they were received the forge shop, before they were machined, to ise their strength and toughness and remove any that might have been set up by the forging itions. They were heated in a lead bath furnace 50 deg. Fahr. and quenched in oil. They were tempered in a lead bath heated to 800 deg. Fahr. ere is probably no better medium than molten for heating a steel piece of this size to a uniform erature throughout its entire length. The quench- was also kept cool by mechanical circulation. efore, we should get a uniform degree of hardness ver these connecting rods. We did not find much formity in hardness by the Brinell when we tested n all over with this type of machine. The limits allowed for the acceptance or rejection ese rods were from 410 to 440. A spot was always hed at A or B for the Brinell test. Thus we could » accepted this rod if it was tested on the A side, would have had to reject it as being 10 points too t if tested on the B side. There was a difference of ints from the A to B side about 1% in. through rod. Less than 2 in. along the stem there was a ition of 30 points on opposite sides or from 430 to This would wipe out our tolerance of 30 points ich testing. We find a much greater difference, gh not as close together, between the 410 on the kshaft end of the stem, on the A side, and the t the piston end of the stem; some 8% in. away. 80 point variation seems unreasonable, but we as great variations on other rods, and in two we found more difference than this. The flat of aring on the piston end gives a variation of 15 between two places but % in. apart. Through piston bearing, opposite the 460, the reading was 140, a difference of 20 points in a distance of Possible Causes he most common conclusion to jump at in such is a condemnation of the steel, the steel maker the forger, but these do not always prove to be lution. We might also argue that the bar from which such rging is made must be distorted so much that the ial strains would cause the steel to be more dense me places than in others. If that were so, why | the greatest variations occur in the stem where ast distortion takes place? If we found them in forked crankshaft end, such an argument would i more reasonable, as there the bar must be de- ed a great deal in order to form this fork. 'hen, again, we might say the steel was heated to emperature that was altogether too high when g forged. We know there is a great temptation all forge men to heat steels to a temperature that roaches the burning point on account of the : with which they forge at the higher temperatures. ' that very reason, however, the forging tempera- ‘©S are watched very closely, as a connecting rod is THE IRON AGE | 915 too important a piece to be weakened by overheaiing in the forging operations. Some of the causes for these variatiuns might be found in any of the above guesses, or a combination of all of them might give the reason. Then, again, it might be that none were responsible, as hardness testers are not so perfect but what they might err. Single Connecting Rods In Fig. 3 is illustrated the single connecting rod that pairs with the double rod. It is a fair average of the 10 rods that we Brinelled all over. Between the crankshaft bearing and the piston bearing we find a difference of 50 points. Too much reliance cannot be placed on the figures shown on the crankshaft end. The Brinell machine pressure of 3000 kg. causes the metal to flow for some distance away from the ball. The ring of metal on the crankshaft end is so narrow that this pressure may have caused the steel to bulge Fig. 3 4 Single Connecting Rod, Showing the Variations in Nume J Hardness Values Taken with a Brinell Machine This rod represents the one that pairs with the double mnecting rod of Fig. 2 out on the sides of the ring rather than resist the ball as it should to get a correct reading. But there is a variation of 20 points at C, in a distance of 1 in. on the piston bearing end. In two places on the stem, there is also a variation of 15 points in a distance of 1% in. These seem altogether too much for steels that are accurately heat treated in lead baths. Conclusions That the Brinell machine was working perfectly on these tests, was shown by the fact that the pump lever was pulled down just five times to get the necessary pressure to raise the weights for every indentation made. The indicator also pointed to a pressure of 3000 kg. each time. We do not know of anything bet- ter than the Brinell method for making this kind of a test or inspection, but the above results would indi- cate that one cannot work within a range of 30 points when working on a commercial basis. Probably a 50- point limit would be as low as is practical on parts like connecting rods and 40 points on smaller parts or simpler shapes. A 30-point limit might work to advan- tage on parts machined from bars or those where the original bar is not altered much in shape. Wear is not the important thing required of con- : Ora RM UN neni ee OP ee ee en ee” Ge ee eee oe et ——— SS ee ee Rn. y Or ’ bb sf z : * re a j ” as 3 t? eI Pd ‘ RS * * 4 a eo . . > : if i? Wey o ap te 2° 5 . . : ‘ zt Fy rity ty Ee} om % he a he + a cm . ‘ = is . , t > ie y ‘4 ‘ ‘ Ms - ia . 9 ” > Ls + ah f , f 7 ‘ He ; ‘ iy : Bt om) . a 4 . 2 4 - a a 7 ou +a 3S . i ty Bs bd o fy mm. ¥e ; ‘ t 4° 7 , oy . & ; . oe 4, 4 e 3 916 THE IRON AGE necting rods, but enough strength is needed to with- stand che strains it is subjected to in the engine. Ten- sile and vibrational tests show us that when we give these grades of steei the correct heat treatment, which means the correct degree of hardness, we can expect such rods to do their work without breaking or bending. On a commercial scale, these methods of testing or inspecting every connecting rod are impossible, as it would be altogether too expensive to carry special test bars along with each rod. Therefore, we have to resort to the next best test, which is the one for hardness. If we allow too wide a range in the hard- ness figures, we get too wide a range in the strength of the metal. On the other hand, if we do not have a wide enough range we get too many rejections. At the very best, we cannot claim much accuracy in inspec- Lion methods when we have to resort to such a round- about way of obtaining the strength of materials. Ball bearing makers have not found any better method than eyesight for the testing or inspection of their balls on a commercial basis. It surprises the uninitiated to see how quickly girls will sort the balls April 6, nS with soft spots no larger than a pin head, fr that are hard all over, by eyesight alone. <A of hardness testers have been used to check results and attempts have been made to disp human element. At the meeting of the Detroit Chapter of th ican Society for Steel Treating, on Nov. 10, 19: Freeland, of the Hoover Steel Ball Co., said: hands of a skillful operator, we have found tha will give us more accurate information as hardness than we are capable of procuring other instrument we possess or could possibly As balls have no flat surface, it would be extreme f ficult to get a correct reading from the ha testers now in use. The diamond-scratch n would probably give more reliable results tha Hardness testing machines were a great advance, and they have a field of usefulness not approached by any other sort of testing But to use them where they give misleading injures their reputation and everybody therewith. Cleaning Producer Gas Without Washing Gas Equalizer and Soot Collector Developed to Treat Gas from Bituminous Coal BY JAMES H. ERY little effective effort has been made to free the raw gas from dust, between the gas producer and the furnaces, and practically nothing has been done thoroughly to mix and equalize the gas from several producers. It is perfectly feasible, however, to clean raw producer gas and retain the sensible heat of the gas, in addition to saving the B.t.u. value of the tar and volatile hydrocarbon. This may be accom- plished without additional operating cost. WZ fy ALLL ELLER Ye Fig. 1—Vertical Section Lllustrating Gas Equal- izer and Soot Collector and Showing Flow of Gas from Producer to Distributing Main Referring to the drawings, the several passages leading from the producers communicate near the bottom to the uptake, as shown in Fig. 1. In each passage is a damper of the usual form, by which any one of the producers may be cut off temporarily from the soot collector. The downtake flue connects by means of a horizontal flue with the upper end of the uptake flue. From the downtake flue, gas flows across *Superintendent Tube Works, Reading Iron Co., Read- ing, Pa. MATHESON to a second vertical flue leading to the main flue shown in Fig. 1 below the ground level. The lower portion of the uptake flue forms a soot pocket in which is a discharge opening closed by a valve actuated by rod as shown. The bottom of the flue is tapered toward this opening so that, when the valve is opened, the soot will flow readily into the in- clined way and thence into the pit. A traveling bucket shown by dotted lines can be so located, when used, that the soot will flow directly into it. The lower end of the downtake flue forms a similar soot pocket. Each side passage has a soot pocket with bottom " * Fig. 2—Horizontal Section Above Producers and Plan of Layout of Producers and Gas Collecting Passages. This shows locations of dampers and of soot pockets beveled toward the opening, which is closed by 4 hand-operated valve. The uptake flue and downtake flues are rectangular in cross section, to provide broad and flat surfaces against which the gas impinges as it passes throug! them, thus more readily separating the soot and dust from the gas and thoroughly mixing the gas. When sufficient soot has collected in the pockets, the valves are opened so as to discharge the soot and dust from the pockets. By this construction the gas, when * ril 6, 1922 es the main flue, is found comparatively free ot, so that the cleaning of the main flue is ited. n Fig. 4 a modification is illustrated, in which the flue is above the ground. Otherwise the con- tion is the same. constructing the soot collector in this manner, vas of several producers is averaged and provides form quality of gas at all times and, consequently, ich better quality of gas can be produced. For nee, when one producer of a series is making lean ind another is making very rich gas, and the other 44 Fig. 3—Vertical Section through Uptake Flue, at Right Angles to First Section Shown lucers are making a medium gas, the gas from the ral producers becomes thoroughly mixed while iveling through the various flues, so that when it reaches the main flue it will be uniform in quality. \dvantages of this installation include a saving of » 25 per cent of the fuel, increased product, im- ved quality of product, saving in gas valves, less rick work and labor, less machine shop repairs, less labor in cleaning out soot at end of each week, more ntinuous operation of furnaces and more contented workmen. In addition to the above, the dust recovered has considerable value, as it can be briquetted and burned nder boilers or “in furnaces, or it can be used for ehold purposes. The analysis of the soot and dust "das to s as follows: Sulphur ... Slt Baidbioa bs.4%0 Made 1.00 per cent MOMIG. «tcc etks cx 6e nme 0.77 per cent Volatile combustible matter...... 0.18 per cent Fixed: carbon” ..%. o<. ~ i 78.71 per cent ASR Gace é 3 17.34 per cent 4 a ; 12,184 (he soot and dust recovered from one stationary ducer, 8 ft. inside diameter and 14 ft. high, equipped a Chapman mechanical agitator, gasifying 1500 Westmoreland gas coal per hour, is approxi- tely 4600 Ib. per week of 5% 24-hr. days, or in volume over 9 cu. yd. It is this soot and dust that the furnaces and producers “sick.” The Reading Co. has cured this trouble by the above installation by a thorough mixing, prevented the existence of ying strata in the gas. Varticular attention is drawn to the small diameter these producers, and to the large increase in gasification, after installing the Chapman agitators. irging the producers, we can increase the gasifi- n to 2000 Ib. of coal per hour, but do not advise it tantly. We have found it possible, by the use of gas equalizer and soot collector, to eliminate one lucer out of every four. Che life of the furnaces will be very much pro- ed, due to their steady working instead of frequent ing and starting as formerly, to clean the mains, es and checkerwork. The installation of the new ment makes it possible to use a cheaper grade of It will be noted by the illustrations that there complication, and the installation is not expensive. application to a battery of three or more pro- ducers can be made very quickly. he Reading Iron Co. is equipping all its producer 2 plants with this arrangement. At the tube works, ¢ — espns THE IRON AGE 917 in No. 1 gas house, there are now installed complete and in operation, two gas equalizers and soot col- lectors, each connected to a battery of three pro- ducers. In No. 2 gas house there has been installed one equipment, connected to a battery of four pro- ducers. In No. 3 gas house are two installations, each connected to a battery of four producers. At the Ninth Street rolling mills has been installed one gas equalizer and soot collector, connected to a battery of five producers. The balance of the rolling mills are to be equipped as early as possible. This equipment of twenty-three 8 ft. inside diam- 5 4 J i Downtake | take , bigin Five Sue f° Producer ———————— , ie a. Fas } ~sthes | , | ia 0} al | 7, i - ~ , +, | I eH at L H > r oma ae a) or “ar . her $4444 ha CAAA F~FEEAAA EAA 444 Fig. 4—Modification, with Main Above Ground Instead of Below the Surface eter producers with Chapman agitators is supplying raw producer gas to sixteen large Siemens regener- ative gas furnaces for lap-welded and butt-welded wrought iron pipe and to rolling mill heating furnaces and galvanizing plant. Bridgeport Factories Gaining In the week ended March 25, the metalworking in- dustries of the Bridgeport industrial district showed a further slight gain in operations, as reported by 31 manufacturing companies to the Manufacturers’ Asso- ciation of Bridgeport. The percentage of employees engaged, as compared with a normal basis, was 55.7, while the percentage of man hours was 50.5. In man hours, this was an improvement of 1.3 per cent over the preceding week. The gain in the number of men employed is 3746 since the week of Jan. 7, when the low point this year was reached. The estimated nor- mal number of employees for the 31 Bridgeport fac- tories is 25,318. Industrial Locomotives in Demand The American Locomotive Co., New York, has re- cently received an increasing number of orders for industrial locomotives. Among the companies which have bought one switching engine each are the Amer- ican Brake Shoe & Foundry Co., St. Lawrence Brick Co., Castner, Curran & Bullitt, U. S. Gypsum Co., John B. Smith & Son, Ltd., of Canada. Centrifugal castings are to be discussed at the spring meeting of the American Society of Mechanical Engineers at Atlanta, May 8, 9 and 10. The paper will be contributed by Leon Cammen, New York. The subject of welding will be made the topic of one session and papers are expected on the strength of mechanically welded pressure vessels, on forge weld- ings, this paper by F. N. Speller, metallurgical engi- neer, National Tube Co., and on tests of welded cylin- ders. a Sanne mw fe a * ; i} Rie ss ve » ie ; es BE) y ai Fi ref bes J ¥ 4 ay : *~ — = p 6 Sage pe a; e i ‘ 4 S 5 ° * " ; ; F i ' > . q 5: ‘ . > % : : * r Fy ” - i § i ‘ By « ~* > Fa ’ & iy , bi 1 . * . a. i. 2 : . ty a w ¥ er { 7 - ‘“ i . ' a 4," i 7 ~ = ® ‘NW i 7 > Be : 7 aM %. : 2, «3 % s 918 THE IRON AGE Fixture for Grinding Chamfer on Chasers The Geometric Tool Co., New Haven, has placed on the market a fixture intended primarily for grinding chamfers on the Geometric milled form of chasers of the various standard types. It may be used, however, for the tapped form of chasers, in which case the grind- ing is straight and does not conform to the contour of the threads on the chasers, being correspondingly more pronounced the smaller the diameter which the chasers are to cut. The fixture permits the grinding of left- hand thread chasers as readily as right hand. The table is graduated and may be set for grinding long or short chamfers. A narrow key at the top en- gages the keyway on the chaser and acts as a guide while grinding. An adjustable stop, which may be locked, governs the position of the chaser in respect to the grinding wheel, and the side of the fixture is graduated for tilting the table to the desired angle of Chamfer Grinding Fixture chamfer clearance. One fixture accommodates all sizes of chasers, but when used for the 5/16 in. size chasers it is necessary to remove the key at the top of the table and guide the chaser in the keyway. When the fixture is bolted to the grinder table it is necessary to slide the chaser forward by hand to the wheel and against the stop provided on the fixture. On machines that permit, it may be fed forward, by means of the machine hand wheel or lever , to a stop arranged on the machine. Another method is to mount the fix- ture on the machine slide and grind the chasers by pressing them back and forth across the edge of the grinding wheel. Too small a wheel, however, must not be selected for grinding in this manner. The chaser should always be held firmly against the grinding fix- ture key while grinding, and clearance should be al- lowed between the stop and the side of the chaser. Iron for Heating-Furnace Castings In response to a request for information as to the proper iron analysis for withstanding the frequent alternate heating and cooling to which domestic heating furnaces are subjected, Y. A. Dyer, Birmingham, Ala., has suggested the following: Percentage of Heavy Medium Light Silicon 1.75 2.00 9.95 Sulphur _ 0.08 6.06 0.06 Phosphorus ... 0.20 0.40 0.50 Manganese - 0.90 0.70 0.60 Total Carbon.... - 8.30 3.40 3.45 The Worcester, Mass., chapter American Society for Steel Treating held its initial banquet recently. Dr. John A. Mathews, president Crucible Steel Co. of America, and E. P. Gilligan, Hartford, Conn., president of the National Steel Treaters’ Society, were among the speakers. Dr. Mathews gave an interesting talk on the history of the iron industry from the early ages to the present day, outlining inventions responsible for recent developments. Mr. Gilligan confined himself to the society’s affairs. V. E. Hillman, retiring chair- man, acted as toastmaster during the dinner. J. E. Rogers was elected chairman for the ensuing year, and W. A. Bacon, secretary-treasurer. April 6, | 22 New Department of International Nicke! ‘0, Robert C. Stanley, formerly first vice-preside: recently elected president of The Internationa! Co. Mr. Stanley has ordered a reorganization sonnel, involving as its chief feature a new depa of development and research with headquarter 7 Wall Street, New York. The new department of the company is tl growth of a gradually maturing conviction that of its products and their successful use in th of the consumer is the key to the extension of distribution. Research departments at the min ers and refineries of the company, established the last few years, have now been crystallized headquarters organization in intimate touch bot its own plant developments and with outside e1 ing developments, touching on the use of nic! monel metal in all fields. This department under the direction of A. J. Wadhams as n Associated with him will be Dr. Paul D. Merica tor of research. Mr. Wadhams has hitherto be: ager of the Bayonne refinery of this company, i: of the manufacture of the company’s products. Portable Saw Bench for Pattern Shop; A portable saw bench for use in pattern shops and packing rooms, taking current from an electric t socket and designed to swivel and cut any angle from 45 deg. left to 45 deg. right, has been placed on the market by the Tannewitz Works, Grand Rapids, Mich. Both saw and motor are mounted on a tilting frame and belted together, as shown in the illustration. The frame in turn is mounted on a swiveling platform, a flexible spring serving to keep the saw raised when Portable Saw Bench. The snap switch on the handle starts and stops the motor not in use and furnishing the necessary resistance for accurate workmanship. The saw can be locked at any angle by means of the hand lever at the front. A stationary gage is provided at the axis of rotation and the saw hung so that all cuts are radial. Other features include the motor adjusting screw which permits adjustment of belt tension and the dust spout which delivers dust to the rear. The machin thoroughly pare with sntngwnnes, The Buckeye Land Co., subsidiary of the Youngs- town Sheet & Tube Co., Youngstown, Ohio, handles through its office 630 properties which are rent | or being sold to employees of the parent interest. If 4 worker purchases a home under the company’s terms, he is given a separate, paid-up insurance policy, £uar- anteeing full payment of the balance of the account in case of his death or permanent disability from any cause. Part of the premises occupied by the S. A. E. Corporation, 207-209 A Street, South Boston, A een leased for a term of years to the Angell Nail & Chaplet Co., East Seventy-ninth Street, Cleveland, to be ¥ sed as a distributing station. . tT By-Product Coke Tonnage Holds Up Well ia Shows Enormous Advantage Over Beehive Fuel i i in Year of Depression, Says United States Geological Survey ob BY R. S. MCBRIDE* t ¥ T was shown conclusively in 1921, according to the 1920. The output of by-product coke in July was 52 = United States Geological Survey, that by-product per cent of the average monthly output in 1920, and i coking of bituminous coal is continuing to super- the output of pig iron in that month was 30 per cent ‘ de beehive coking. This change in practice has been of the average in the preceding year. Thus it is evi- ; progress for some years, but the first convincing dent that the decrease in the metallurgical demand for ; monstration that the by-product branch of the cok- coke had the effect of greatly reducing the operation bE ¢ industry could maintain itself in a period of in- of beehive ovens, but that it caused only a relatively - strial depression more strongly than the beehive small decrease in the production of by-product coke. inch was made in 1921. Coke Output by States Pe In 1921 the output of by-product coke was almost T A , ie : F ables II and III are summaries of the production of § ,000,000 tons, and that of beehive coke was about beehiv . 9 9 Kf oe eehive and of by-product coke in 1920 and 1921, by - 00,000 tons, figures that show a striking contrast to ; : . states or groups of states. These summaries show FY se for 1920, when the output of by-product coke Semele thadk-the Gensenen te. ovedestion to. 1606 en ae is more than 30,000,000 tons, and that of beehive ae ae a me - / e was more than 20,000,000 tons. sarmenenairernsesrimenn nian sini sannnionsnnn * 4 6 i The output of beehive coke in 1921 was less tha “s r in any other year since 1885. One month of the 7@¥/e //——Production of Beehive Coke, by Groups of States, J y ir showed an output of only one-ninth the average ot SSOP ANS SS: LO ee SANS. t nthly output in 1920. The monthly average for Pa an ; ; 1921 was only about 27 per cent of that for 1920. Per . | The output of by-product coke in 1921 also showed aa song — Teas Com . 1 marked decline from that of 1920, though the out- ae Ghana a Nes "la oe a lanaes oe ; it in the minimum month was more than half of that ).hama. Tonnesece aad aes a ' the average month of the preceding year, and the Georgia ad 1,069,000 354,000 715,000 67 ' output for the entire year was practically two-thirds virginia and Kentucky 1,300,000 374,000 926,000 71 i that in 1920. This comparison of 1921 with 1920 be- Cooradc and N. Mexico. 512,000 119,000 393,000 77 ; mes still more striking if we remember that 192) Washington and Utah... 253,000 160,000 93,000 37 ; i was easily the “banner” year in the production of by- ea 7 ‘ United States 20,511,000 5,561,000 14,950,000 73 - *Final figures tEstimate ible I—Estimated Monthly Production of Beehive and Table III—By-Product Coke Produced in 1920 and 1921 by /-Product Coke and of ~~ dren in the United States States, with Increase or Decrease in 1921 (in Net Tons) é ra a — Increase (+-) or ’ Cone. . =" Pig Iron, 1920 — ~ Decrease (—) th Net Tons Net Tons Gross Tons* Ovens Output, Ovens Output, Tons Per —— — _ State Tons Tons Cent ly average 1920 1,708,000 2 569.000 3 077,000 labama 1,081 3,123,890 1,101 2,406,000 —718,900 —23 rp Gin bbeedes 1,137,000 2,278,000 2,416,000 Colorado 120 ° 120 . ° ° ee ee 865,000 1,888,000 1.937.000 Illinois 794 2,136,793 894 1,326,000 —811,000 —38 4 e's. atime 575.000 1.772.000 1,596,000 Indiana 1,216 4,553,697 1,216 3,030,000 —-1,524,000 —34 a ae 328,000 1.519.000 1.193.000 Kentucky 108 466,985 108 186,000 —281,000 —60 300,000 1,590,000 1.221.000 Maryland 300 682,132 360 293,000 —389,000 —57 ‘ 232,000 1,408,000 1,065,000 Mass 400 488,089 400 318,000 —170,000 —35 180.000 1,297,000 865.000 Michigan 389 1,393,445 389 778,000 —6§16,000 —44 me eee 248.000 1.383.000 954.000 Minnesota 220 674,801 220 431,000 244,000 36 mbee” SC. ae tace 289.000 1.423.000 986.000 Missouri 56 ° 64 e ° ° ew ick sor te 416,000 1,734,000 1,247,000 New Jersey 315 725,571 252 745,000 +19,000 + 3 mae" coe « 477,000 1,766,000 1.415.000 New York 732 1,040,192 732 815,000 —225,000 —22 NVGP. sins ledn 514,000 1.860.000 1.649.000 Ohio 1.558 5,614,877 1,558 2,964,000 2,651,000 —47 ee : Penn 3,006 7,730,256 3,154 5,439,000 2,291,000 —30 tad SORE vccwes 5,561,000 19,918,000 16,544,000 Rhode Isl 40) 40 ° . : Tennessee 24 139,121 24 58,000 —81,000 —58 *Figures for 1920 from American Iron and Steel Institute ; Wash 20 26,284 20 21,000 5.000 —19 for 1921 from THe IRON AGE. West Va 274 447,392 274 184,000 263,000 —59 Wisconsin 28 ° 238 ° ° e Combined States 1,590,426 924,000 666,000 42 duct coke in the United States. In other words, Total 10.881 30,833,951 11,164 19,918,000 10,916,000 —35 , pite the extraordinary slump in business, which greatly lowered the output of by-product coke, it was *Included in “(Combined States.” ; greater in 1921 than in any preceding year eXcept — mmmmmmmumnmennnnnnnsonnsinnnnnnin ; ssa \7, 1918, 1919 and 1920. or very generally distributed over the country, as there , Monthly Record of Coke Production was an increase in the output of coke in only one state, t Table I shows the production, by months, of New Jersey, where large plants that make by-product : h type of coke, and the output of pig iron. The Coke produce gas for municipal supply, and cannot tuations in the output of coke evidently followed ely the fluctuations in the output of pig iron. therefore be operated only for the production and sale of metallurgical coke. The increase in that state, } Table I shows that the lowest monthly output was however, is not sufficient to be really significant. . made in July, when only 11 per cent as much beehive Coke Ovens in Existence 4 © was produced as the average monthly output in Tables IV and V show the number of coke ovens in ‘nited States Geological Survey, Washington, D. C. existence at the beginning of 1921 and their status. 919 sO ne Vinee tet es ig g" Ce tryey =z 920 THE IRON AGE No new beehive ovens were constructed in 1920, but more than 6700 were abandoned. At the end of that year about 75,000 beehive ovens whose daily capacity was almost 200,000 net tons of coke, were in existence, and 332 new ovens whose estimated additional daily capacity was 1800 tons of coke, were in course of construction. The number of by-product ovens was increased in RRRTNNa vances pen ester ne saree vente veennnesecenes ten HTN HreeEEE EL cemeenTen crenecoennccercntte ‘ sstemeeeneancenancapnnnnense Table 11 Siatus of Beehive Coke Ovens at the Beginning of 1921 Ovens Under Construction, Jan. 1, 1921 Ovens in Existence Jan. 1, 1921 Ovens ——-_--—— — Abandoned Daily Daily State in 1920 Numb Capacity Number Capacity (Tons) (Tons) Alabama 253 8,482 13,571 Colorado 931 1,793 6,007 Georgia . phe 151 242 Kansas 2 ae nie Kentucky 272 855 1,684 New Mexico ; 1,030 1,710 Ohio 50 222 444 Oklahoma 300 335 ects sede Pennsylvania 2,412 14,569 140,638 302 1,750 Tennesse 180 1,848 2,033 Utah 819 819 as oa Virginia 135 3.906 7,270 30 60 Washington 25 407 545 West Virginia. 2,146 10,916 20,767 : Total 6,706 75,298 196,065 332 1,810 Deanne ennba nate ry vere nannne emnneneniiiriee serenenel SUUEORDOEDOESET LED OOOENELEDE. FUOOUTERDELEDOERDENIDEN | (0000000 | LO2REEDORORPRED ES LOESEOO CTRESERESRSRORERORSS SERRRRORERASERRAS 1920 by the completion of 757 ovens, whose daily ca- pacity of coke was 11,170 net tons. During the year 300 old ovens were dismantled or abandoned, so that at the beginning of 1921 there remained in operation 10,881 ovens whose daily capacity was almost 120,000 tons of coke. Of the 396 ovens being built at the be- a i an Table \ Status of By-Product Coke Ovens at the Beginning of 1921 Ovens Under Construction Ovens in Existence New Jan. 1, 1921 Jan. 1, 1921 — . be 2 5 be be 2 A 8 s32 2 = i Ee ga ssa & E Z, AD Oak Z. 7 Alabama 247 4,707 90 081 90 777 Colorado ..... 120 ; vas Illinois ie 1,600 794 100 1,500 Indiana 1,216 Kentucky ; 108 Maryland 300 Massachusetts 400 Michigan 389 Minnesota 220 néee Missouri 56 980 8 140 New Jersey ; 315 3,077 New York 210 2,454 10 732 7,822 CED bck aden 100 1,558 19,234 Pe fate Pennsylvania. 160 1,710 3,006 29,973 148 1,694 Rhode Island.. ‘ 40 456 Tennessee . 24 252 Washington .. 20 70 West Virginia 60 699 274 2,028 ; Wisconsin ... .. ; 40 228 2,357 50 583 Total 757 11,170 300 10,881 117,319 396 4,694 soenecesevnnnon neccancncxs0eveneeseertenseeasseseeseenseesseeeseseareeessanseenessonesesseareeetovseevseevveveenescancuvvseverseetanennensss ginning of the year, nearly all were completed before December 31, and at the end of the year only 85 ovens were under construction. Types of By-Product Ovens in Use Table VI summarizes, by types, the by-product ovens in operation at the beginning and at the end of 1921. This table shows a continued increase in the number of the two principal types of ovens built in the United States, the Koppers and the Semet-Solvay. Table VII shows the location, character, and number April 6, 19 of ovens installed during the year. At the end of year the Woodward Iron Co., at Woodward, Ala.. the Milwaukee Coke & Gas Co., at Milwaukee, were building 35 and 50 ovens, respectively, all of Koppers type. During the year 15) United-Otto and 40 §S, Solvay ovens were abandoned. Capacity of By-Product Ovens Table VIII shows that the annual production o{ product ovens in operation on Jan. 1, 1922, wou AvOUUUTTUEAGSAOLAUOOGEOENEDALEDAUGEONOOAAUANENOLODEOAENOGAOCAOOSNADORADELOUGEOEDEOELEL«IGEDELUAIGANNODEDEOSHORELODtOOcAROOOONRGsOSBORSOUSOREREBELCDERSNENEEONASEEDEALEPNeNnrs Table VI—By-Product Ovens in Use at Beginning and | 1921 and Under Construction Jan. 1, 1922, by Typ: Und In Existence In Existence Constr Type Jan. 1, 1921 Jan. 1, 1922 Jan. 1 ore coe | OSS 6,303 Semet-Solvay ...... 2,415 2,463 United-Otto ........ 1,513 3,413 PE wcasid e waes 282 282 a 206 206 Cambria-Belgian ... 250 310 Gas Machinery ..... 33 33 PE: cevataendicn a 42 42 eae aes 7% 104 104 Pee” GacwstéGeeseh ciwa 8 Teens | occas oni 10,881 11,164 ‘HULEF) LIRETREDSOTDSSOLURERREDERRSVORSTOROREREEPEIERULEREROLELOTEGEPUDOOENN NS /NTNUNE 11 TUDE TUES COERENEREG FLEE TROROECCEROORD FORDELGSILA LEC LENTFERELEBEE OWE ED Kr 44,275,000 net tons of coke if the ovens continued in operation without interruption 365 days in the year. As the average yield of coke for the industry amounts to 69.9 per cent of the coal used, this output repre- A LRDUADLADLE AAEDNELLUUOEUANATLVAEELOALOLOTOEULCODDONNADOARDOORRDUOOUCAUDLULOOLETUEORONOAEEAOOEN LOOTORODEOESO LET OONSCDRARERASSEROESOUESS DOT400 LER CRERSERONREDND 1eOtOTEOOOrIIREERE Table VII—New By-Product Ovens Completed and Put in Operation in 1921 Number and Type of Ovens Operator and Location of Plant New plants: Chicago By-product Coke Co., Chicago.. 100 Koppers St. Louis Coke & Chemical Co., Granite oP en ee ee $0 Roberts Additions to existing plants: Woodward Iron Co., Woodward, Ala... 20 Koppers Bethlehem Steel Co., Sparrows Point, et gaaare 6 5.288 b tae Renate se ees 60 Koppers Laclede Gas Light Co., St. Louis...... 8 Piette Camden Coke Co., Camden, N. J....... 37 Koppers * ria Steel C . > et-Solvay Cambria Steel Co., Johnstown, Pa...... ; = eae Milwaukee Coke & Gas Co., Milwaukee, WS -cacct caebubies ica 50 Koppers UGE. HEE cidica 100s OES Sah ee CUneeLennHLAAORLONeHHNAAAOANOLANNEDANABOUNONREDOUONEOUESAAUNEREESSHUNNLE VLUCCUULUUASUEUODUIAAROGUEDTAAEOLUESSORSS ODER SDAeUaRABOBONEGL ENOL 048s (RonenEERERELEnRORIN /entaaEIN sents a coal-carbonizing capacity of 63,340,000 net tons a year. However, operation at full capacity is impos- sible, even in periods of good demand for coke, and the figures for operation at 85 or 90 per cent of full capac- supnenvecennvenersacnoneneenseatesonnnennne evcsennennns uenenat VOC OUUCLUEEDE CT DEDEAOROEOET PET BRO EUNRE ONDE SOUBUBUNAA HHEGS neORESEEOAE ED LEDED RET ORNRRAEREBORN Er Table VIII—Annual Capacity of By-Product Coke Ovens Existence Jan. 1, 1922 Coal Coke for Charg¢ (Net Tons) (Net Ton: If operated at 100 per cent of capacity... 44,275,000 63,340,0' If operated at 90 per cent of capacity.. 39,848,000 57,006, If operated at 85 per cent of capacity.. 37,634,000 53,840,' | sUpNanECO SHU OOUEEUUCUENODCGUEAHOCOECHLOIOLUTUOTOTORENOUNHUOTORELEET CT ONEEOLESUOEYD ETT NSMAHOONORENREEDINROREAOEONEnORUEDOEEOERSORENAEAEOOEEENIOOOSIOOIOE ity really represent a normal maximum for any col- siderable period. In other words, the by-product coke ovens, operable Jan. 1, 1922, can treat only about 55,000,000 net tons of coal a year. The Morgan Construction Co., Worcester, 4!- nounces that new mills designed and built for A. D. Whitehead, Newport, Eng., have been started. The mills are of the continuous type, having approximately 2000 tons of steel strip and hoops per week capacity. pril 6, 1922 Portable Saw Bench for Pattern-Shop Use A small portable universal saw bench for pattern- ps, mill-wright work and for use also in the ship- ¢ department has been brought out by the Union ichine Co., Grand Rapids, Mich. It is designed to 2-in. stock. Although the company’s portable machines are ally offered without the base, using the base as wn, is recommended. A %-hp. motor is employed i can be connected to a lamp socket. The table is iron casting 18 by 20 in. and can be tilted and ked at any angle up to 45 deg. A graduated dial vided with a pointer serves to indicate the position the table and a stop facilitates quick return to hori- tal. The cross-cut gage can be used on either of the saw, two slots being planed in the table, on either side, for this purpose. This gage can Portable Saw Bench for 2-In. Stock. The table can ve tilted or locked at any angle up to 45 deg. When » handle is pulled forward a cam raises the feet from the floor set quickly at any angle and clamped. Holes are vided for mounting an auxiliary wood face piece. ripping gage is machined on both sides and can so be used on either side of the saw. Tightening of » lever-head screw locks the gage in position and tomatically lines up with the saw at the same time. The saw is 7 in. in diameter, runs at 5000 r.p.m. d may be either of the ripping, cross cut or combina- n type. A guard is provided which sets down over ’ Saw and can be instantly set to allow for thickness f stock cut. A splitter guard to keep the stock from nching the saw is also provided. The saw arbor is also mounted in ball bearings and the saw-arbor yoke is hinged so that it can be raised and lowered instantly, permitting the saw to project above the table high enough to cut stock from - in, down. Dado heads up to % in. wide can be car- ried and should be 6 in. in diameter. Belt guards are THE IRON AGE 921 provided and belt tension maintained by means of a turn-buckle device. A metal throat plate, which is removable, permits replacement with a special hard- wood throat plate for use with dado or grooving heads. The motor switch is located at the front as shown. The base has two rollers at the back and two *sta- tionary feet at the front to give firm footing. When the handle is pulled forward for moving the machine a cam raises the feet from the floor, shifting the weight of the front of the machine on to a third roller which is carried on a swivel bearing, moving with the handle. The height of the machine with base is 36 in. bench type, 10 in. The weights are 268 lb. and 133 lb. respectively for base and bench machines. February Output of Automobiles According to Automotive Industries, the February production of automobiles in the United States amounted to 129,500 cars, compared with 90,486 in January, 78,995 in December and 116,349 in November. A much larger proportion of these were shipped from factory by railroad than in the previous year. Drill and Tap Crib with Gage Feature A drill and tap crib and gage, the general construc- tion of which is indicated in the accompanying illus- tration, has been placed on the market by the Victor R. Lawson Co., Boston. It is a drill crib with a drill gage built into the body. The container is 41 in. by 9% in., and has 60 semi- circular pockets 1% in. wide, a