The Iron Age 1925-12-17: Vol 116 Iss 25

ESTABLISHED 1855 THE IRON AGE New York, December 17, 1925 VOL. 116, No. 25 The Future of Merchant Pig Iron High Freight Rates, Large Imports of Foreign Iron and Economic Changes Have Put Many Furnaces on Inactive List—By- Product Coke Ovens Considered as Possible Remedy BY CLARENCE E. NLY 10 of the 44 so-called merchant blast furnaces () supplying pig iron to foundries in the Atlantic seaboard States were in operation on Dec. 1. The 44 furnaces are located in New York, eastern Penn- sylvania and Virginia. Some of them have not been operated for a year or more; others have been idle most of the time since the war. Large amounts of capital invested in such furnaces are yielding no return, and the future of the domestic merchant pig iron in- dustry, particularly in the eastern States, where it is seriously affected by imported iron as well as by high railroad freight rates and other economic changes, is giving concern to all who are financially interested. About 400,000 tons of foreign iron has come in this year, much of it at Atlantic ports. This is approxi- mately double the amount imported in 1924. Num- bered among the idle domestic furnaces are some high- cost plants which cannot successfully compete with the low-priced iron from abroad. Several of the idle furnaces may never operate again, while the problem of the others, a few of which have been modernized at great expense, is to find ways and means of meeting the new situation the pig iron industry faces. In the minds of some who are vitally interested in the prob- lem the construction of by-product coke ovens in conjunction with existing blast furnaces is believed to be the ultimate solution. The construction of new capacity such as the Mystic furnace, near Bos- ton, and the Troy, N. Y., furnace of the Hudson Valley Coke & Products Corporation, both of which are linked’ to by-product coke plants, is being watched with interest by other pig iron producers whose furnaces have not been able to operate profitably within the past year or two. WRIGHT The Virginia pig iron situation is probably the most serious. Only two of the 17 furnaces in that State have been in operation for any considerable part of this year. High railroad freight rates to most of the districts in which Virginia iron was formerly dis- tributed have effectively kept that iron out of its ac- customed channels and relief which was asked for from the railroads through the Interstate Commerce Com- mission has not been forthcoming. The Case of Virginia Furnaces In Virginia the subject of construction of a by- product coke and gas plant to serve all of the furnaces is now being seriously agitated. The effort made about two years ago to obtain a gas contract from the city of Richmond is being revived and a more hopeful atti- tude prevails as to the possibility of success. The stumbling block is that the Richmond city charter pro- vides that its gas contract must be renewed from year to year. Capital could not be found to finance the con- struction of a by-product coke plant there unless a 20-year contract with the city could be entered into, and this would involve an amendment to the city char- ter by the State Legislature. However, some of the Virginia pig iron producers are bending their efforts to bringing about that result, and if the way should be paved for a Richmond coke plant it is considered likely that it would serve the Virginia furnaces with fuel on a basis permitting them to operate more economically and meet the competition of producers in other districts. New Jersey and Delaware River Coke Plants Although it has not yet gone beyond the discussion stage, there is a possibility that a by-product coke The Mystic Blast Furnace Near Boston, Which Will Be Completed and in Operation Early in 1926, Has Been Built in Conjunction With the Coke Ovens of the New England Coal & Coke Co., Situated About a Quarter of a Mile from the Furnace. The furnace will operate on foreign ore which will be unloaded at its own dock 1657 1658 THE IRON AGE December 17, 1925 on Pig Iron Trade Is Watching With Interest the Development of the Troy, N. Y., Project of the Hudson Valley Coke & Products Corporation, Which Combines By-Product Coke Production With a Merchant Blast Furnace. The illustration shows in the foreground the coke ovens and auxiliary buildings, but the blast furnace, which is being erected in the area at the extreme right of this view, has not yet risen high enough to become visible from this point of outlook plant may be built in New Jersey to supply gas to one or more of the New Jersey municipalities and to furnish fuel to the furnaces of the Replogle Steel Co. at Whar- ton, N. J. rebuilt a few years ago, have been in operation only a fraction of the probably These two furnaces, which were completely time. The coke project, if it goe the Replogle Steel Co., but by other interests cooperating The Wharton advantageously located with respect to supplies of New through, will not be undertaken by with that company. furnaces, although Jersey ores and also with respect to distribution of iron to a large consuming territory at favorable freight full low selling prices for have not attained the advantage of then rates, location because of pig iron and high cost of coke. \ project for the the Delaware River at Philadelphia has been construction of a by-product COKE plant or discussed from time to time and is now being revived. The has proached viewpoint of Philadelphia Foundrymen’s Association ap- the both the bringing a new industry to Philadelphia and of pos question from sible savings on pig iron through more economical pro- duction. This project aroused active interest not many months ago through a proposal to build a coke plant and blast furnaces on the Delaware River adjoining a steel plant located not far from Philadelphia. The latest proposal, which was brought out at a meeting of the Philadelphia Foundrymen’s Association at the Club, Philadelphia, last Tuesday even- interest capital in the construction of blast Manufacturers’ ing, 1s to furnaces in conjunction with a coke and gas plant, the supplied to the city of Philadelphia. At this writing the city of Philadelphia and Co., which vas to be the United Gas Improvement now supplies gas to that city, are negotiating a new gas lease. Some of the city officials have been quoted as favoring the taking over of the city-owned gas plant and leasing it to another company to operate, and the subject is being gone into thoroughly,.the city having employed connection Hudson William H. Blauvelt, who was consultant in blast furnace Products Corporation at with the coke plant and of the Valley Coke & Troy, N. Y Whatever the outcome of these Philadelphia gas nego- tiations, it is fairly certain that efforts. will be made to build a coke plant to supply gas to the company which contracts to furnish the city stated that some of the capital necessary for the coke plant and gas. It is blast furnaces has already been assured Walter Wood of R. D. Wood & Florence, N. J., favor Co., cast iron pipe manufacturers at was the spokesman at the fovndrymen’s meeting in of the construc- tion of blast furnaces at Philadelphia. It was admitted that discussion of the matter was in the nature of propaganda, with the hope that capital might be at- tracted to the proposition. Mr. Wood, who has been for the past year or two one of the largest importers of pig iron, called attention to the economies which would be possible if blast furnaces were built at tidewater. He cited the case of a furnace company located not far from Philadelphia which has a total cost on shipment of foreign ore from docks to its plant and on shipment of the pig iron back to Philadelphia of about $2 a ton, and he said this could be saved by a plant having ore docks adjacent to its plant. He further cited the low freight rates available on pig iron when transported by water instead of by rail. A considerable part of the iron made at Philadelphia, he said, could be shipped to plants having to water transportation. Mr. Wood expressed the belief that the manufacture of pig iron at lower costs is the best possible means of making it difficult, if not impossible, for foreign producers to ship so much iron to this country as has come in during the past year. access Capital Available for By-Product Plants It is freely stated by those in the pig iron trade who have considered the subject that capital is readily available for the construction of by-product coke plants if a profitable outlet for the gas can be found. The place which coke has taken in the public mind during the anthracite coal strike is given as a reason for the willingness of investment companies to undertake the financing of coke and gas plants. Moreover, by-product coke has been supplanting beehive coke in recent years to such a large extent that the almost complete absorp- tion of the coke business by by-product plants is not very remote. In 1913 by-product coke production was less than 30 per cent of the total; in 1918 almost half of the total coke production was by by-product ovens, and the gain of by-product coke has been steady ever since. In 1923 beehive coke made a temporary recovery of a part of its lost tonnage, producing more than 30 per cent instead of only about 20 per cent as in 1921 and 1922; but in 1924 by-product coke shot farther ahead, bee- hive ovens producing only 22 per cent. In 1923 the total coke production was 56,977,534 tons, of which 37,597,664 was by-product and 19,379,870 was beehive. In 1924, a year of low production, the total was 43,- 663,000 tons, of which 33,995,000 was by-product and By-product coke production this 9,668,000 was beehive. j December 17, 1925 THE IRON AGE 1659 year will probably be well in excess of 35,000,000 tons, the total up to Oct. 31 having been 32,650,000 tons, while beehive production for the year will be 10,000,000 tons or more, the aggregate reached on Nov. 28 having been 9,357,000 tons. By far the largest part of the coke production is charged in blast furnaces, but the increased use for FAVORS ARBITRATION Mining Congress Indorses President’s Sugges- tion for Settling Industrial Disputes WASHINGTON, Dec. 15.—The American Mining Con- gress at the close of its twenty-eighth annual conven- tion here last Friday, adopted resolutions which, among other things, indorsed the principle set forth by Presi- dent Coolidge in his message to Congress for the settle- ment of industrial controversies by arbitration. The Mining Congress also approved the transfer of the Bureau of Mines from the Department of the Interior to the Department of Commerce and commended Sec- retary Hoover for the high type of men chosen to ad minister the work of the bureau. Among other resolutions was one which commended the Ways and Means Committee of the House of Rep- resentatives for its “non-partisan and fair manner in drafting tax legislation,” but called attention to the fact that recommendations are to be made for changes in the provisions of the law with reference to Federal taxation of mines. The mining industry, according to the view of the Mining Congress has not been afforded an opportunity to present facts to refute attacks against it. In addressing the congress, Secretary of Labor Davis touched upon controversies in the coal mining industry and declared that the principal trouble in the bituminous fields was over-development. He suggested that both the anthracite and bituminous industries set up a fact finding commission made up of both operators and miners with a neutral representative as chairman in order to prevent strikes and to keep industry stabil- ized and operating smoothly. Gilbert Montague, of the New York bar, told the Mining Congress that 25 years of experience with the anti-trust laws has conclusively established that it is folly to apply them to the natural resources of the United States. Prusecutions against producers and dis- tributers of natural resources, he said, have been spectacular and well advertised. As a result of the fear of anti-trust laws, he asserted, employers are re- luctant to engage in collective bargaining, operators are prevented from going into consultations or co- operative arrangements which would prevent demorali- heating purposes—a market which some producers will make strenuous efforts to retain—has awakened interest on the part of a number of pig iron interests in coke production as an auxiliary to the blast furnace. Well- informed spokesmen for the pig iron industry consider that this may be the most important development in connection with the blast furnace in the next few years. zation by high-cost operating units, and distributers are afraid to join in combinations that might prevent duplicated, high-cost distribution. Mr. Montague cited a number of laws which granted exemption from anti- trust prosecution, and declared: Should Congress delay its discharge of this clear duty, the Federal courts may in time. as in the past, through the ever-growing “rule of reason” in their interpretation of the anti-trust laws, accomplish a certain degree of relief Recent Supreme Court decisions carry a strong intima- tion of this trend. Consolidations and combinations, bringing about less waste, greater stability, and real economies in the production and distribution of natural resources, are likely in the future to be more sympathetically viewed by the courts. The Mining Congress also was addressed by William E. Humphrey, of the Federal Trade Commission, who spoke of benefits from the new rule of stipulation under which the commission now operates. He said that by this new rule the commission has settled more than 75 per cent of the cases promptly, saving time for the commission and money for the taxpayers, pre- venting wrong by unjustified publicity both to the pub- lic and to the accused, and more fully protecting the public’s interest. The following officers of the American Mining Con gress were elected: President, Col. Daniel B. Wentz, Stonega Coal & Coke Co., Philadelphia; vice-presidents, Robert L. Tally, United Verde Copper Co., Jerome, Ariz.; B. L. Doheny, Los Angeles, Cal.; W. H. Lindsey, Napier Iron Works, Nashville, Tenn.; secretary, J. F. Call- breath, Washington, D. C.; member of the board of directors, George B. Harrington, Wilmington & Frank- lin Coal Co., Chicago. To Discuss Labor Turnover Labor turnover will be discussed at the Tuesday morning, Dec. 29, session of annual meeting of the American Statistical Association, to be held at the Hotel Pennsylvania, New York, Dec. 28, 29 and 30. Papers will be presented on the following phases of the subject: The measurement of labor turnover as a prob- lem of management; the best measure of labor turn- over; the significance of statistics of labor turnover; and the place of statistics of labor turnover’in a pro- gram of governmental labor statistics. pM OE NS SAD ROTI OMISSION if oe | i" i a i- esting Metals for Aircra Selecting Representative Samples—Proper Design of Test Specimens and Methods of Loading Them = HE standardization of pnySsical tests used in pre- . 2 the characteristics offers many Chief among these are selection of representative sam ples, design of dicting metals will display 1 service problems for investigation specimens and methods of loading. Much consideration has been given these problems by I government bodies, technical societies and in- i Oy in 60 V Notch =F > a, 2 ype + “1% ng = —7—————— K = -> Types of mens Used Geometric lustrial organization with the view of establishin standards to meet the general needs of collective in daustries. There are, however, certan ni nsider tions to be found in individual industries which gover! t ree of refinement extended to the control of val é ‘hese conditions do not minimize the value yf general standards for physical tests but rather em- phasize the importance of their being fundamentally sound and capable of giving results that will meet the most critical needs. If any leniency is to be allowed it hould indeed be left to the individual industries. It is well to remember that, in order for stan lards to remal healthy, the maximum cumulative error that may occu under thi ise must be well within 1 nable working n I juired | e industry in which the ure et ut B ause oT tft eve pr el iemand tor \ aving in alr raft, there is pe rhaps no other ndustry requiring more accurate knowledge of metals. Sin plicity and ease with which results can be interpreted al and applied to design have rendered physical testing the most agent in supplying this The important factor in aircraft design is cost transporting pay load In order that this may be decreased it is neces- sary to keep structural weights at a minimum, which active knowledge. per ton- reasonable mile for over a ourse. *Engineer of Test Naval Aircraft Factory, Philadelphia OTEY* done only through an accurate knowledge of and conditions to be met in performance. materials Selection of Representative Samples y the test results obtained from a metal specimen represent only that particular piece at point of However, there are advantages to be gained by saving some stock for construction, which certain and there our troubles begin. The major responsibility for the selection of repre- speaking, fracture. concessions makes necessary, sentative samples rests with the manufacturer in the ise of uniform raw materials and maintaining uniform melting and rolling practice. Unless the metals sup- plied are uniform within specified limits, no amount f physical testing short of a specimen from each bar will give representative results. Even under the best mill control, there is consider- danger of impurity segregations and piping at the center of large bars, forgings and sheets. This ondition is more prevalent in aluminum alloys than n steel, which is partly due to the newness of the art. lhe location of specimens then, with respect to longi- tudinal axis of bars, forgings and sheets, will vary with naterials and sizes. The important actions to be taken may be summar- zed thus: abDle . Maintaining uniform mill practice Maintaining complete historical information from terials to finished metals. stablishing standard methods of sampling which ure re¢ ble knowledge of the metals furnished. Design of Specimens In general the results of physical tests are most iffected by the geometric shape of the specimens used. Reduction of Area, per cent = 9 } -—Charted Physical Test Results of the Sir Type 8s of Té st Bars Dar — +. ‘ +] . : “or exampie, two tensile specimens of each type shown in Fig. 1 were prepared from a single bar of mild car- bon steel and tested under identical conditions with re- sults shown in Figs. 2 and 3. The reduction of area (waisting) in types 1 and 2 was negligible, due to the supporting shoulders being near the point of failure, whereas type 6 showed 70 per cent reduction of area. This is an extreme case and would have shown less ap- parent variations in properties if a harder metal had 1660 December 17, 1925 been used. However, it will serve to emphasize the possible discrepancies that may be incurred through variations in geometric shape of the specimen. Again, it is common practice to specify that tubing shall have certain tensile properties when tested in either full section or flat specimen. Ordinarily tubes which have a tensile strength within the capacity of the largest testing machine available are pulled in full section, while larger tubes are flattened and speci mens prepared similar to those used for sheet metals Tables I and II show some of the results of this prac tice, where it will be noted that per cent elongation is primarily affected. Abrupt changes in contour of a stressed member introduce components with consequent resultant stresses of higher intensity than the loading would otherwise produce. The intensity of these, resultant stresses is directly proportional to the abruptness of change in contour, whereas the area affected is greater for more gradual contour changes. Such a condition is found at the shoulder radius of the ordinary round and fiat tensile specimens and many failures that occur at this point can thus be accounted for. Because of Fig. 3—Appearance of the Broken Test Bars of Six Different Types or Shapes the complexity of these stresses, they have almost com- pletely eluded mathematical analyses, but by photo- elastic methods’ they have been clearly isolated and measured. For design information Fig. 4 shows approximatel) how these complex stresses occur in tensile specimens. It will be noted that the region of high stress extends into the parallel gage portion of the specimen and, if the true tensile properties of the material are to be determined, it is necessary that these abnormal stresses be eliminated from that section of the specimen shown as the gage length. This can be done by increasing the parallel length sufficiently to allow a gage length of ma- terial which will be subject only to pure tensile stresses. The amount by which the parallel length must be in- creased is directly proportional to the shoulder radius, and for this reason it is desirable to keep the shoulder radius as small as practicable without incurring ab- normal stresses of sufficient intensity to cause failure at that point. Referring to Figs. 1 and 2, it will be seen that, even with a %-in. shoulder radius, this re- gion of high stress extends more than % in. into the parallel gage section, otherwise type 4 specimen would have shown the true strength of the material. THE IRON AGE 1661 Another factor which contributes to these abnormal stresses at the shoulder radius is the relative enlarge- ment of the gripping end of the specimen. It has been found that failures near the shoulder are practically eliminated if the enlarged ends of specimen are not more than 13 per cent larger in diameter or width than the gage section. This simply reduces the abruptness Jif |. Parallel Length \ et | Gage Length Region o FHigh Stresses due to Change in Contour Fig. 4—How Complex Stresses Occur in Tensile Specimens of change in contour. For a more detailed explanation and quantitative analysis of these conditions the reader is referred to the very creditable work of Professor Coker’, and of H. A. Anderson’* of Bell Telephone Lab- oratories, Inc., New York. When preparations were being made for the con- struction of the American airship “Shenandoah” at the Naval Aircraft Factory, the question was raised as to the best type of tensile specimen to use for testing duralumin sheet metal. The results of this investiga- tion are timely for showing some of the effects of variations in gage width, parallel length and gage width-thickness ratio. This investigation covered a range of gage widths from 0.185 in. to 1.250 in., sheet thicknesses from 0.015 in. to 0.250 in. and parallel lengths from 2.250 in. to 3.500 in., the gage length in all cases being 2 in. Space will not permit a complete report of the investigation. However, the results of two series will cover the important points, since they are typical of the lot. Series A This series represents the results of check tests made on three types of specimens found to be the most favorable for the range of sheet thicknesses covered. These specimens are shown in Fig. 5. Material—The material used was an aluminum alloy, duralumin, having the following specified com- position: Per Per Per Per Per Per Per Cent Cent Cent Cent Cent Cent Cent Al Cu. Mg. Mn. Pe. Si. Ca. 92 Min 3.5-5.0 0.2-0.75 0.4-1.0 0.6max. 0.4 max. 0.4 max. All the material used was made by the same manu- by Rj | ¥ chee 4 x f ¥ ~ p ~ Fig. 5—Three Types of Specimens Found to Be the Most Favorable for the Range of Sheet Thick- nesses: ——_—Inches—— Type Symbol H r Ww A os 0.67 2.25 0.500 B O 0.67 3.50 0.500 C A, 1.00 3.50 0.750 facturer and under strict supervision and inspection. Preparation—All specimens of a given thickness were cut from a single sheet and the three types taken “Tension Tests of Materials,” by Prof. E. G. Coker, Engineering.” Jan. 7, 1921, 35 and 36 Bedford Street, Strand, London, W. C. 2. ?“"Tension Tests of Thin Gage Metals and Light Alloys,” by H. A. Anderson, Proceedings American Society for Test- ing Materials, page 990, Vol. 24, Part II, 1924, 1315 Spruce Street, Philadelphia PLN RIPEN ST ne ae FRE Rt HT sam" * P; ca a Pa Ss 6 j , TET EET IR re alternately, so that the distribution over the sheet was | | the same for each type. The direction of rolling was \c i } parallel to the longitudinal axis of the specimens. 24 ies } eae erence een +—— Heat Treatment—All specimens were packed (dif- ferent types alternating) and heat treated at the same time, as shown in Fig. 6. During this operation they 25 were so arranged that the spacing was equal to the sheet thickness, to insure uniform cooling when quenched. Heating was done in a 50-50 salt bath of o “S/ sodium and potassium nitrates to a temperature of 500 ~< deg. C. They were quenched in water at 21 deg. ‘ — and allowed to age in still air at room temperature for ’ 20 days before testing. No care was spared in prepar ing, heat treating and testing these specimens, to in- c 20} sure uniform procedure. The results of these tests are compiled in Table Il. Figs. 7 to 9 inclusive show the elongation c Id values for comparison of the three types of specimens. 4n explanation of one of these curves will serve for the jat uw «= RL = o 6 c \ ; S \ = Sy — + —___—_—_——| Ss rv Si.8z RBS iC. | i A ce cn Oh eee 16 7 TT ss ss a Se lb om aes PAR St} D qh So Fe ~ | 2X RBSE BES SSEke | 5 s ss Ssss_ sss SSSSS Sheet Thickness in Inches Fig. 7—Results of Tests of Type A (Fig. 5) Plotted on the Basis of Data in Table III The following observation can be made from these curves and Table III: = Fig. 7 and Table III show that Type A specimens give lower elongation and higher ultimate stress values than other two types, because of the shorter parallel length. P Fig. 8 and Table III show that type B specimens give values which are a fair average of the three types. Fig. 9 and Table III show that type C specimens give uniformly high elongation values and low ultimate stress values, due to the greater volume of material which in this shape allows greater deformation (waisting). This will be further demonstrated in the next series. Series B This series represents the results of tests intended to show the effects of varying the gage width-thick- Fig. 6—Packing of Types of Test Specimens for Heat Treating entire series: In Fig. 7 the elongation values for type A specimens (see symbols in Fig. 5) are plotted to the ordinate scale and along a straight line curve hav- ing a slope of 45 deg. In this way the curve is made to represent every elongation value for type A speci- mens. This curve is used as a basis for comparison and for convenience will be called the base curve. Next pees eee Penne beeen Pe pce [ + _ Per Cent Elongation in 2 Inches x S ~ S ‘ ie the elongation values for types B and C specimens are Sg x plotted to the ordinate scale and on vertical lines corre- Sol! S sponding in sheet thickness to the points for type A S23 Ss 8 specimens. Thus it will be seen that all points having the same abscissae represent a single thickness of sheet and are directly comparable on the ordinate scale. Sim- 15 ilarly in this series of curves each type of specimen is compared with the other two types for percentage elongation in 2 in. Ultimate stress values can be Fig, 8—Results of Tests of Type B (Fi - 5) Plott plotted in the same manner. the Basis of Data in Table wil ed on 2 0025. 008 003 —_— 2 0.0468 ~0.078 Sheet Thickness in Inches ; December 17, 1925 24 23 ae" * e ~ © S © ?ltr = 2 ou L j c 20} o — 5 >, c i £ du +> c 5 2 + . S | % a. nS Li Sn Sy | wm tic LH Li tt dg fees Sh = Ss _|isss s lf + ,S ~|\ss8 ae IR rs ae > he 1S4.' Bo \ = is | VSS Res ssRe & S SS! S$ |jss SSssscs & 15 S S Ss S isS SsssososS SG Sheet Thickness in Inches Fig. 9—Results of Tests of Type C Plotted on the Basis of Data in Table III ness ratio. All of these specimens were made from two sheets of duralumin 0.065 in. thick and 0.035 in. thick respectively and to the shapes shown in Fig. 10. The selection of material was made as follows: Five sheets of each thickness were examined for tensile properties, using geometrically similar speci- mens. From these tests, one sheet of each thickness was selected, which two sheets showed the same physi- cal properties, within close limits. These two selected sheets were then cut into pieces 10 in. square and three geometrically similar tensile specimens tested from each square. From these test results, pieces were selected from each sheet which had practically identical properties. The sam- ples thus selected were very uniform and the two thicknesses permitted a wide range of W/T ratio. The same methods of alternate cutting, packing, heat treatment and aging as described for series A Table I—Comparative Tests of Full Section and Flattened Specimens. Carbon Steel Tubing (1025) Tested as Received Elonga- Elongation Tensile tion Ratio Yield, Lb. Strength, Lb. in 2 In., (B)to(A), Specimen per Sq.In. perSq.In. Per Cent Per Cent 1% in., 16 ga. tubing ‘ Tube (A) .... 76,700 84,900 24.0 49.5 Fiat (B) .... 73,200 90,900 11.88 : 1% in., 16 ga. tubing Tee CI» secs Sheen $1,200 21.4 44.0 Flat (B) .... 80,800 82,900 9.44 ae 1 in., 16 ga. tubing : Tube (A) .... 61,600 74,500 25.5 59.5 vue (B) ase 69,900 77,600 15.2 in i. 2 . tubing ; * Tube fA) coce One 73,800 23.1 60 Flat (B) .... 67,200 72,100 13. All of above values are average of two tests. All flat specimens cut adjacent to tube specimens. Table I1.—Comparative Tests of Full Section and Flat- tened Specimens of Duralumin Tubing—Heat Treated Elonga- Elongation Tensile tion Ratio Yield, Lb. Strength, Lb. in 2 In., (B)to(A), Specimen per Sq.In. perSq.In. Per Cent Per Cent 1% in., 0.059 in. ga. tubin : Tube (A) .... 36,700 60,400 21.5 86.5 Fiat (B) .... 46,100 62,100 18.6 oe 1% in., 0.041 in. ga. tubing 2 Tube (A) .... 38,400 55,200 25.0 71.2 Flat (B) .... 40,600 56,100 17.8 1 in., a in. ga. tubin Tube (A) seco re 59,300 25.8 66 Flat (B) .... 47,200 60,250 17.0 en All of above values are average of two tests. All flat specimens cut adjacent to tube specimens. eamnenaversesnsenenssennvesnsesnnsetnnssauesenssseEstOnsOtttES8/#1000SURESUNSUSDOORSESESOS SESSIONS SSIES ERINELERETESSESTNERES SEES SENSUSRIST SENSIS THE IRON AGE 1668 were used. The results of these tests are shown in Table IV and Fig. 11. So far as the writer is aware it has been generally believed that the more nearly a rec- tangular specimen approached a square cross section the greater would be the deformation under tensile loading. These results clearly show that this is not the case. Further demonstration of increased deforma- tion with increased width-thickness ratio between the limits of 1 and 30 is shown by comparing Fig. 8 with Fig. 9. It is evident that even slight variations in geometric shape of test specimens have effects that will be readily reflected in the test results and that much care should be used in selecting standard test specimens for air- craft work, where specifications are necessarily very rigid. Barba has pointed out that “Geometrically similar bodies of the same material, under identical conditions of stress, undergo similar deformation.” The use of this fundamental law in the design of round tensile specimens is especially important. Since they have to be machined separately, they can be made geometri- cally similar and in various sizes without causing va- riations in test results. In the case of flat specimens, however, the solution is not so simple, as individual machining would entail prohibitive expense. The al- ternative here is so to design and standardize the speci- men that the true properties of the material will be measured and then calibrate it for the various thick- nesses of sheet and materials it may be called upon to represent, The writer is conducting a series of tests on carbon steel, alloy steel and duralumin tubing to determine the comparative results to be expected from use of flat and full section tensile specimens and if possible to establish a fundamental basis fér their design. Methods of Loading The laboratory technique best suited to the various industries will depend largely on the size and kind of OHDeDeEEONLAHOEEAEDRERDHVERUDEESERENNNHNE® (TRYT EE: HTNERUOTT TORN L OU OREDEREETURERT ON: /9UESEEPHEOELY //UREERODENOETY!/r0/ 1040) °F F700 KE) LHREHSOEDOERED ES PRO ORH FY S#DES\sPEREROREBEBER EFAs» Table III.—Results of Tests on Duralumin Sheet Using Three Types of Specimens Shown in Fig. 5. Ali Values are the Average of Three Tests Sheet Elongation Ultimate Thickness in 2 In., Per Cent Stress, Lb. per Sq. In. in aerate, baker —_ Inches Type A Type B TypeC TypeA TypeB Type 0.0155 14.5 14.5 14.0 63,000 63,500 61,700 0.0175 18.2 19.0 18.3 63,700 64,700 59,200 0.0195 17.2 16.8 17.3 64,100 64,100 62,300 0.023 17.8 17.2 18.8 64,300 63,000 62,400 0.0255 19.8 20.3 22.5 61,300 64,100 61,100 0.029 19.5 19.2 19.7 59,300 58,400 58,700 0.0325 20.3 20.5 21.3 66,700 60,900 60,400 0.033 19.7 20.3 21.2 62,200 65,800 66,200 0.037 20.0 20.7 20.6 61,700 61,100 60,800 0.039 20.2 20.7 21.5 63,500 63,300 61,600 0.0468 19.5 20.8 20.3 66,100 65,030 64,500 0.0472 22.2 21.0 21.8 60,900 63,000 63,000 0.051 20.8 20.7 21.7 60,700 60,900 59,800 0.0565 22.7 21.2 22.0 62,900 61,400 60,900 0.057 21.7 21.8 22.2 61,500 61,200 60,800 0.067 21.2 20.3 21.5 63,100 62,800 62,800 0.072 21.7 21.3 21.5 63,200 61,700 62,500 0.078 21.0 20.8 20.5 63,300 63,100 68,300 0.084 20.8 20.7 22.2 61,100 60.800 61,400 0.090 22.0 21.5 22.7 62,300 60,500 60,800 0.104 19.8 19.8 21.8 63,500 62,400 63,000 0.1215 22.5 21.8 23.3 59,000 58,100 59,100 0.160 19.5 20.5 21.8 57,500 58,300 58,700 0.190 22.2 24.3 24.7 62,300 62,100 62,100 0.256 24.5 25.3 26.2 60,500 59,800 60,000 Table IV.—Results of Test on Duralumin Sheet Using Types of Specimens Shown in Fig. 10. All Values Are Average of Five Tests Elongation Tensile Sheet Gage W/T in 2 In., Stress, Lb. Thickness Width Ratio Per Cent per Sq. In. 0.065 0.19 2.39 18.0 59,300 0.065 0.33 5.0 18.6 60,400 0.065 0.52 8.0 19.8 60,200 0.065 0.65 10.0 20.5 59,200 0.065 0.65 10.15 20.6 59,900 0.065 0.75 11.7 21.4 59,900 0.065 1.02 15.6 21.5 60,100 0.065 1.31 20.1 22.4 60,400 0.065 1.99 30.5 22.8 59,000 0.036 1.19 34.1 21.8 60,860 0.035 1.37 39.1 21.4 60,100 0.035 1.52 43.3 22.4 60,400 0.035 1.71 43.8 22.9 60,300 0.035 1.83 52.1 22.0 59,500 0.035 2.05 58.5 21.5 60,500 507) 080-POGSEEESENRESURNPURPEETRONSUNASRASRAAAAASEEEROAAEES 507009 (0°08 2 7 ESTEE RIED 98 ETTNLESRENESEESTENS SS TRESEN TSIEN NARI PER TT ras 5 eT * —— é \ Sci pen TARE ES IRE ie A ln i NGS tel ONE SEAN ct nants . , < 55 : 1664 metals used. metals used in non fer rous Most of the structural construction are either ferrous 01 alloys and are furnished in comparatively small sizes, such as wire and rod diameters from 0.010 in. to 2 In., sheet thicknesses from 0.008 in. to 0.250 in. and tubing alrcratit rc t from 7 to 35 DD _ ratios. The requirements fo. liable loading conditions are further complicated Dy these alloys being tested in the heat-treated condit Errones ] test results traceal e to meti ods ot | ; x look i Ww ¥ essen ase oe > >] ~ } , i é ] t Bar Used in Study ng the Effect } the Gage Width-Thickness Ratio Whe H —1.13 W With the Values of T, in Table Ty W and W/T Show re y tally 4 itty ited one OY more < I wing n ons , (; 2 port t t d ré = 4 Té met d condu I ne I x ive i > é aging It specimens occurs versa on t arrangement is used T the } é monls r rTrip] iz£ Technical Control in the Foundry Dis- cussed at Boston Ohio Brass Co., strides in During the past ars the Mansfield, Ohio, has notable control in the foundry as an aid to the foundrymen director of the laboratories of that A. A. Grubb, gave an outline of what has been accomplished in few yt made technical com- pany, that direction to the members of the New England Foundrymen’s Association, Dec. 9, at the Exchange Boston. To have proper technical control in the foundry it have control of proc- product and losses and to develop and improve Mr. Grubb The is to adapt supplies as nearly as possible to determine if is securing full value for money paid out and to hold producers o materials accountable. Adjustments in the matter o composition ingots alone has paid the company for. the raw materials. By control of fuel sup- meltir malleable Club, Ss necessary to raw materials, esses, said. purpose of raw mate processes ations SO as one spe iC i ¢ i analysis of all con pany’s y reduced. The company one-third the of binder for its Sanas as a of its control of sands The technical control of a product calls for five in spections for testing its fitness for service. By a series of charts maintained on the foundry floor for the mold o time for the materially plies the furnace has been also has reduced by cost core result ers’ inspection, keen rivalry has been built up betweer the molding who strive to keep their charts as nearly perfect In the development and in company has given careful has three types of gangs, nossible processes the furnaces It provement study to its electric furnaces in operation and with its control of raw ma terial finds it can make as good metal with one as witl the other. Th . n: ‘ h; ‘ strides in the recover The company also has made strides in the rec J A saving of at least half the cost of technica] practice at the plant has been made by The company discards no sands in its brass a proper application of rebinding, used of waste materials reclamation foundry. By sands are reconditioned. THE IRON December 17, 1925 AGE a 23 —— ve g : ; |} we 5 ° o| » > = oc?! 460 2 & =e: | os Ho oc} | . - —58£ 3 ot Ss 94 | £s r= 99 a | i156 = + ‘ 4 + + + : peat T pls | ~,£ 4 | o+ | i | } * +4 -o 7p ayo \Z + J) 50 55 60 Fig. Charted Results of Data in Table IV res t compensate for variations in s ¢ l I l Witl the range of ordinary Z I the speed of loading has f pl il determinations. What \ is that the machine be will permit accurate balanc- d ll times Experience has st a I etals should not be loaded i compression than s s ver speeds are / } s I is 1ecessal that the testing machines be known at all times, nsiderable error often results from this source. load determinations be well man factors which may d be eliminated through com- I proceduré Too often the igment f the per r is left unchecked, and de- a ompletely defeat the purpose of - : sussenescassnssacnne® (To be concluded) R. F. Harrington, Hunt-Spiller Mfg. Corporation, Boston, president of the association, presided at the meeting. He appointed a nominating committee, con- sisting of B. M. Shore, Walker & Pratt Mfg. Co., Watertown, George P. Aborn, Worthington Pump & Machinery Corporation, Cambridge, and E. H. Ballard, General Electric Co., Everett, which will select nomi- nees for election as officers at the annual meeting next month. The Superior Sizing Co., Lockport, N. Y., W. Scott Thomas representative, was admitted to member- ship in the association. Foundrymen to Conduct Symposium on Permanent Molds Of especial interest to foundrymen planning to at- tend the Second International Foundrymen’s Congress, to be held in Detroit the week of Sept. 27, 1926, will be the symposium on permanent and long life mold castings. This phase of foundry practice which has assumed such importance in the past few years will be thoroughly discussed at the meeting, the discussion being planned to cover non-ferrous, iron and steel prac- tice, and to be international in scope. The Cothias and similar processes used in England and France will be explained. The American practice developed in molds for pistons, the process of using zinc-impregnated molds and the many = special developments in the field will be consid- ered. = i mM on oil-cooled using mittee organizing this symposium, under the « Jesse L. Jones, metallurgist Westing- house Electric & Mfg. Co., East Pittsburgh, will wel- ‘ome information from any one who has been instru- mental in developing or carrying on work in this field of casting procedure. airection of The Republic Iron & Steel Co. is preparing to con- vert its plate mill at the Youngstown, Ohio, works into a universal mill, so as to diversify its output, and permit the rolling of skelp as well as the usual sizes of steel plates. Economizes Fuel in Tinning Producer Gas Equipment Reduces Tin House Fuel Consumption, Improves Quality and Increases Output BY ROGERS A. FISKE* DECREASE in fuel consumption for tinning to A an equivalent of 3.52 lb. of coal per base box of tin plate was effected through producer gas equipment recently installed at the Canonsburg, Pa., plant of the Standard Tin Plate Co., a subsidiary of the Continental Can Co., New York. The gas producer plant was so designed as to be at once flexible in its operation and economical in over-all fuel consumption, and at the same time to assure an increased quantity and better quality in tin plate out- put. Natural gas had been used and proved highly desirable except for its limited supply. Various methods of direct coal burning had been tried and found wanting, both from the standpoint of economy in fuel consumption and definite control of furnace heat. In view of the demonstrated superiority of gas, it was but natural that it should again be used despite the necessity of manufacturing it. The coal available for gas production is of the ordi- nary run of mine as obtained in the Pittsburgh district and such as the company had been using for some time in the boiler plant. The normal supply all comes from one mine and the quality is therefore uniform. It yields an average of about 14,000 B.t.u. per lb. as re- ceived. One engineering phase of the problem was to make available, so far as possible, the coal handling installa- tion which already served the boiler house. This con- sisted of a track hopper, an inclined belt conveyor dis- charging coal to a crusher near the ground level, and a long inclined belt conveyor extending from the crusher to the boiler house bunkers. Advantage was taken of this situation by laying out the gas generating plant to have the long belt con veyor between the producer and cleaning buildings and thus make the old coal handling equipment serve the gas plant as well as the boiler house. Accordingly two *Western editor, THe IRoN AcE, Chicago Coal Conveyor to the Boiler House Passes Be- tween the Gas Producer House and the Gus Cleaning Building producers were located north of the conveyor and the gas cleaning house was placed to the south. To make room for the new gas plant it was merely necessary to tear down a small pipe shed. Only Coarse Coal Used in Producers Coal is now unloaded into the track hopper, crushed and plowed off the long belt at a point about midway between the producers and at a height of 60 ft. above the yard level. It is spouted down to an automatic skip hoist loader which is below the yard level so that it ean be filled by truck should occasion demand. The skip hoist, motor-operated and of the unbalanced type, elevates coal to the skip house head, where it is auto- matically dumped into a steel plate hopper which is provided with a feeder. The coal then passes over a disk-type grizzly. The fine coal passes through and is spouted back to the belt for delivery to the boiler house bunkers. The coarse coal is carried over the grizzly and is spouted two ways to two automatic weighing machines, furnished by the Richardson Scale Co., Pas- saic, N. J. It is then dropped into passageways known as retarders. These are essentially elongated steel boxes set up on end and provided with a series of in- clined baffles. They serve the purpose of retarding the downward velocity and thereby reduce the momentum of the coal in its descent to the gas producer bunkers below. The reason for this precaution is that the final cleanliness of the gas and tar is somewhat dependent upon the initial condition and form of the coal fed to the producers. Therefore, after once having separated out the fines, it is highly desirable to avoid producing rama meh a me go écnaae eat A A OEE IS ES SE A — . ay ag ago : F , DS AE SE ik bors. soa a ses pau 1666 more fines in subsequent handling of the fuel. Each of the retarders unloads into a 120-ton self-trimming, Gunite-lined steel bunker, which feeds the coal by grav ity to the gas producers. The new coal handling equip- ment and structural material were furnished by Phil lips, Lang & Co., 431 South Dearborn Street, Chicag: Ash handling is effected very simply by means of an industrial car, loaded by hand and moved on rail which terminates at the boiler house ash skip hoist The gas I a mono- generating equipment comprises two fu producers furnished by Wellmar Seaver-Morgan Co., Cleveland. These ars mechanical gas placed on housed in bunkers are direct! there is an open } allel to the coal conve yo! steel and brick building. The coal erhead and are so designed that petweel ine pare and art + the building walls and the vertical sid space of the and serve as severa! a seal to prevent coal dust When coal is not bunkers. trap doors bridge this ga from entering the producer room. being discharged nto tne bunkers these trap doors are opened and thus provide ventilation over the gas producers. It is notic« : ’ ible tha pace was not wasted and still there is ampli room for yperators to move free] about all parts ot the equipment. Low Velocities Provided in Gas Flues Raw gas is taken from the top of each produce! 1-ft. fire brick-lined, steel flue. The flue fi t right angles and then downward at an anglk of about 45 deg. to the base of a tangential type dust atcher. The downcomers ar: 4 ft. in diameter, and the dust catcher diameter. The gas passes upward in the catchers to a main flue, which horizontal and 6 ft. in diameter. The east end of this flue is connected to the upper end of a which joins the base of a burn-out stack dust catcher. The gas leaves the burn-out stack just below the burn-out valve; it then travels horizontally, and after making tw right-angle turns enters the bottom of a cooler All gas flues up to the cooler are steel with fire brick. The flues were designed to main- tain a low gas velocity with frequent changes in the direction of flow, in order to trap a maximum amount of dust before the gas reaches the cooler. Adequate clean-out and burn-out valves are used so that it is practicable to operate either or both of the pro- ducers as occasion demands. The cooler consists of a cylindrical steel tank through which the gas passes, intermingling with water tnrougn a each is 8 ft. in downcome! The stack also acts as a passes up and and are lined doors THE IRON AGE December 17, 1925 T—— = The Tar Ea Passages. tractors Are Made with Five Concentric Gas Raw gas enters above the plug and is drawn through the extractor by the exhauster suction sprays, until its temperature is reduced to the required point for entering tar extractors. In the cooler, much of the soot, dust and heavy tars are separated from the gas as it passes upward. Water is required in the cooler, the tar extractors and a final scrubber, the amount used being about 3 gal. per Ib. of coal gasified. Four Batteries of Tar Extractors Were Installed. Only three required to op- erate the plant at capacity are December 17, 1925 Water from the final scrubber is recovered and used in the cooler, From the cooler the gas passes into a manifold header supplying the tar extractors. Specially con- structed hot gas valves are used for cutting in and out the tar extractors. A Flinn & Dreffein type of tar extractor (patented) is used. It operates on the impact principle. It con- sists of a tapered plug, heavily threaded, fitted inte a female similarly threaded. The plug is mounted on a vertical shaft, which has a thread the same pitch as the plug. The threaded part of the shaft passes through an outside yoke. The plug may be screwed up or down and remains in mesh with the female thread, but not in contact with it. The width of the passages, or chan- Mize Burtodine THE IRON AGE 1667 object of the scrubber is further to cool the producer gas and remove the remaining dust particles that it may contain. Gas Leaks Are Guarded Against Gas leaves the top of the final scrubber, passes to two Wilbraham-Green Blower Co. (Pottstown, Pa.) ex- hausters, which are motor-driven through Lenix drives made by F. L. Smidth & Co., 50 Church Street, New York. Either blower will operate the plant at full capacity. It should be noted that the entire clean- ing plant is on the suction side of the exhausters. This is an assurance that no outward gas leaks will develop in any of the apparatus and thus the cleaning house is made a desirable place in which to work. Forty- _Redtroad | a? : lar Tartks.. ir Seep Ser bher +Exhaus ters % } | Le Lt % Tar..-45, $ Separa , a = _— E ‘ Sootleg “ae 8 - . ? | 4° \ , : i { = q Burn Out Stack-4 b i CL. of Conveyor Belt oa! Crusher\ ER ( Od nS _ ey ~ x & 7 “ ae | Dust "| | a SRS ss Ollectors| | __ , “SS Compressor if ARAN |! thal. House ‘i \ Hi Wy 3 i Oo a/ ' 4 ti Producers <Q a : > ‘Coa . = Pa 2a a Hopper Q 10. 2' oe ' Burn Out | Stack > ; e » a No. 6—General Plan a of Producer Gas Plant iy Gon, with Lateral Cross- py, Section Projected at é Right and Longitudinal Cross-Section at Bot- Pa } tom Producers with Dust Collectars directly it then rear nels, between the threads is varied through changing the position of the plug. Tar-laden gas enters the tar extractor above the plug and is drawn through by means of the suction produced by the gas exhauster. In the upper chamber of the extractor, above the plug, are water sprays. Water is used to further cool the gas and also to act as a vehicle for maintaining the flow of tar and soot removed from the gag. The position of the plug is determined by the drop in pressure necessary to overc