The Iron Age 1907-06-13: Vol 79 Iss 24

THE IRON AGE Published every Thursday Moraing by David Williams Co. 14-16. Park Place, New York. $3.00 a Year, anemos Postage. Vol. 79: No. 24. New York, Thursday, June rf, 1907 Sitighe Copied, 10 Gan Reading Matter Contents page 1837 Alphabetical Index to Advertisers ‘‘ 193 Classified List of Advertisers “ 184 oe meen Wi iT FIRES FIVE SHOTS Advertising and Subscription Rates‘‘ 192 Wy TPT ce KB IN ONE SECOND AND : 4 CAN BE RELOADED IN i HALF SECOND Se =~ Remington The American Mfg. Co. TOLOADI IG RIFLE Ropes and Twines Al 65 Wall Street, New York ig game hunters have ne tpid fir , “big enous - for the Keves game.” The NET R. m orcas ‘Autole vader meets the requiremen is it zs extremely accurate and delivers F SmasAin } > , , , own blows inone second. Big game hunters have judged THE BRISTOL COMPANY ‘emington SUPERIOR .35, .32 and .30-30 Remington calibres Ct ta go: 759 Mo ad ae Bldg e ‘ . Bris! ¢ Recordi os Ral | Remington Arms Company, -_ Ilion, N. y. ristol's necor ing nstruments | Agency, 315 Broadway, New Yerk City. For Pressure, Tempera- tare and Electricity. Simple, Accurate, Reliable. All Ranges, Low Prices, and Guar- anteed. Send for Catalog R. —————— earns SPOT CORD WATER TUBE O64e Babcock @ Wilcox Co. BOILERS 5. pege57 9 SSE Also Linen and Italian Hemp Sash Cord SAMSON CORDAGE WORKS, Rosin, wass.|| HE LARGE AND STEADILY IN- nig Za Su Se eee Cleveland City Forge and tron Co.. + Cleveland, O. 66 THE CAPEWELL dd HORSESHOE NAIL Is attracting wide attention among HARDWARE DEALERS who appreciate the fact that a large demand results in QUICK SALES; quick sales in more frequent DIVIDENDS, and a higher annual RATE of INTEREST upon every dollar invested. MERRILL BROS. BROOKLYN LY DROP HAMMER Made by ‘BASIC. PIG. The Capewell Horse Nail Co., “eonst* Girard Building, Phila. Pilling & Crane Srv Pe rene JENKINS 96 SHEET PACKING There may be some sub- "a ; ‘ % , Flexible enough to bend easily without cracking, soft enough : a) €6to fill up any slight unevenness of surfaces, but sufficiently stitute for 1), oe ac strong and tough to resist all pressures of steam. Has been IF used for years under all conditions, and has proved its merits in thousands of plants. HIGH QUALITY | Sb evans tsi is iy, do —but so far it has not been “We” Cold Rolled sa ait and ‘ani discovered. THE AMERICAN TUBE & STAMPING COMPANY poke. Water and Rail Delivery) BRIDGEPORT, CONN. MAGNOLIA METAL See VY j Best Anti-Friction Metal for all Machinery Bearing. AMERICAN Na | Fac-Simile of Bar. Beware of SHEET & TIN PLATE y imitations. COMPANY’S SSS MAGNOLIA METAL CO. San Francisco, Montreal and Pittsburg. Owners and Sole Manufacturers. 113-115 Bank Street W Ad. on Page 16 Chicajo, Fiscner oldg. NEWYORK. ° ° Metals at compediive prices. AGE SHEET THE IRON Bip pn acc mit, |e Plame & Atwood ie Co, WIRE SHEET TIN PLATE od : SHEET - GERMAN |S" STEEL SILVER | we DEEP STAMPIN DRAWING LOW BRASS, SHEET BRONZE, SEAMLESS BRASS AND COPPER FOLLANSBEE BROTHERS TUBING, BRAZED BRASS AND COMPANY BRONZE TUBING : + PITTSBURGH WATERBURY BRASS C0., WATERBURY, CONN. 99 John St., New York. Providence, R. I. Bridgeport Deoxidized Bronze & Metal Co. BRIDGEPORT, CONN. Phosphor and Deoxidized Bronze Composition, Yellow Brass and Alumi- , num Castings, large end small Matthiessen & Hegeler Zinc Co., LA SALLE, ILLINOIS. SMELTERS OF SPELTER AND MANUFACTURERS OF SHEET ZINC AND SULPHURIC ACID. Special Sizes of Zinc cut to order. Rolled Battery Plates. Selected Plates for Etchers’ and Lithographers’ use. Selected Sheets for Paper and Card Makers’ use. Stove and Washboard Blanks. ZINCS FOR LECLANCHE BATTERY. TNS amr PALL GARE 105 -109 So. Jefferson St. Chicago. Best Bronze, Babbitt Metals, Brass and Aluminum CASTINGS NICKEL ANODES GERMAN SILVER | onccs, snore. cat corres The Seymour Mfg. Co., - - Seymour, Conn. HENDRICKS BROTHERS PROPRIETORS OF THE Belleville Copper Rolling Mills, MANUFACTURERS OF Braziers’ Bolt and Sheathing COPPEF COPPER WIRE AND RIVETS, Importers and Dealers in Ingot Copper, Block Tin, Spelter, Lead, Antimony, etc. 49 CLIFF ST., NEW YORK. MANUFACTURERS OF Sheet ana RollBrass WI R E. Printers’ Brass, Jewelers’ Metal, German Siiver and Gilding Metal, Copper Rivets and Burrs. - - = Pins, Brass Butt, Hinges, Jack Chain, Kerosene ®urners, Lamps, Lamp Trimmings, &c. 279 Broadway, NEW YORK. Room 508 Heyworth Building, East Madi- son St., CHICAGO, ILL Rolling Mill Factories : THOMASTON, CONN. WATERBURY, CONN. SCOVILL MFG. CO. MANUFACTURERS OF BRASS, GERMAN SILVER, Sheets Molls, Wire Rods, Bolts and Tubes, Brass Shells, Cups, Hinges, Buttons, Lamp Goods. Special Brass Goods to Order. FACTORIES: WATERBURY, CONN. DEPOTs: NEW YORK. CHICAGO. BOSTON. Henry Souther Engineering Co, HARTFORD, CONN. Consulting Chemists, Metallur- gists and Analysts. Complete hnyaical Testing Laboratory. Expert Testimony in Court and Patent Cases. Arthur T.Rutterd Co. 256 Broadway NEW YORK Small tubing in Brass, Copper, Steel, Aluminum, German Silver, &c. Sheet Brass, Copper and Ger- man Silver. Copper, Brass and German Silver Wire. Brazed and Seamless Brass and Copper Tube. Copper and Brass Rod. “ Search-Light’”’ GAS Bicycle Lanterns Send for Circulars and Electrotypes. The BRIDGEPORT BRASS CoO. BRIDGEPORT, CONN. Postal Telegraph Building, Broadway and Murray Street, NEW YORK THE RIVERSIDE METAL CO. RIVERSIDE, W.). THE IRON AGE New York, Thursday, June 13, 1907. An Interesting Foundry Problem. Molding an Acid Egg or Still. BY L. N. PERRAULT, WATERBURY, CONN. An acid still, or an acid egg, as it is usually called on account of its shape, is a hollow iron casting, the making of which is something of a problem for the founder. A cross section of such a casting is shown in Fig. 1, and the following describes how it was molded in a foundry with which the writer was connected : The pattern, as shown in Fig. 2, was about 72 in. long dddddddddddddddddddddddddddaadaaaadl D> ’ 7» y > YU LLL YUL, Yf ii Uy VP KY a ig. 1.--Cross Section of an Acid Egg Casting. by 36 in. in diameter, outside, and had hemispheroidal ends. The core was about 69% in. long by 33% in. in diameter, thus allowing for an even thickness of metal of 1144 in. This had to be supported by the prints A B and C, Fig. 2, of which A and C were each 4 in, and B 6 in. in diameter, while all were 214 in. long. These prints were required to hold the core in exact position in the mold during pouring and setting of the casting, without the aid of chaplets or studs. The use of these was pro- hibited, because acid readily attacks an imperfection in a casting, even finding its way between the bond of a stud or chaplet and the iron surrounding it, and soon renders it porous and unfit for use. The first thing needed was a core arbor sufficiently strong to hold the core and resist the crushing pressure of the molten iron. As the core would be almost entirely surrounded and a quick free exit for the gases was abso- Fig. 3.—Pattern for the Core Form. - lutely necessary it was decided to make the core hollow with walls about 6 in. thick, and in halves bolted to- gether on the inside. To allow the workman to get his arm inside to bolt the halves together and insert the bolts for fastening the core down through the flask bot- tom plate, a hole was provided which also served as a main vent pipe for most of the gases. As the largest cored opening in the casting was to be only 6 in. in diameter the arbor had to be sufficiently light to permit it to be readily broken after the removal of the core so that it could be picked out through the holes A, B and C. Pattern for the Core Arbor, A pattern for a form corresponding to the joint line of the core box was made of 2 x 1% in. stock, allowing for 1 in. of sand around the periphery. This shown in dle in. 91 Fig. 3 was stiffened by a center piece 5W in. long by wide by 2% in. thick, with six lateral braces 24% x in. Core prints 1% in. in diameter were set as shown at A and B, Fig. 3, for bolting on castings A, Fig. 4, which were to act as a guide for the core to its exact seat, and also allow the passage of bolts to fasten the core to the bottom plate. A 2-in. core print was placed at C, Fig. 3, to later allow inserting a gas pipe through this point, thence down through the bottom plate. To this an elbow was to be screwed on and connected with another pipe sufficiently long to clear the flask, so that the vented gases might be quickly lighted and allowed an unobstructed flow. Three half-hoop patterns were next made of stock 14 x % in. Fig. 2.—Pattern for the Acid Egg. and varying in size, the use of which will be described later. A pattern for Fig. 4, was made, castings to go over the core prints A and B, Fig. 2, and the holes to be an exact fit for the leg castings, A, Fig. t, thus furnishing a metal print which would not crush or sag when the core was bolted down. castings, B, these Molding the Core Arbor. The frame pattern, Fig. 3, was laid upon a level bed struck in open sand and sand was rammed all around it and swept off level. On another and similar bed close Fig. 4.—-Leg and Socket Castings for Supporting the Core. by, the molds for the hoop patterns were made in the same way. These hoops were quickly poured and while still hot were stood in place in the frame mold, care being taken to leave sufficient space for the legs and gas pipe to project through. The hoops were held in place by setting bricks between them and across the mold, and the frame was poured as soon as possible. The hoops being still hot were firmly held in the casting, the finished form of which is shown in Fig. 5. The other half of the arbor was made in the same way but with the addition of two l-in. iron staples cast into the center bar for lifting the cope half of the core from the core plate and for setting the whole core into the mold. On the day following, two of the leg castings, A Fig. 4 (which were so simple that their molding needs no de- 1794 THE IRON AGE scription), were bolted on to the nowel or bottom half of the arbor. The whole arbor was then thickly coated with claywash and set into the core box, the taper prints of the leg castings supporting it in its exact position with re- spect to hight, while a few wooden wedges driven between it and the core box on each side held it central. A wooden sprue 2% in. in diameter was next put through the hole at C to leave an opening for the vent pipe, after the core was fastened in the mold. The core sand, a mixture of equal parts of fire sand and coarse sharp sand with 1-12 ltt Fig. 5.—One-Half of the Core Form Completed rye flour, was thrown into the box and carefully tucked around the leg castings, the sprue and all the hoops, and firmly rammed about 2 in. deep. Then %in. rods were laid lengthwise above the hoops to resist the upward pres- sure during pouring. Another course of sand Was then put in, well rammed and followed by a layer of coke (egg size) and a few more rods*laid on the quarter sides. This was covered with hother course of sand, well rammed, and a 4-in. course of coke. The core was then rammed completely without additional rods or cinders, June 13, 1907 support it in place; was not set in the box until after a course of sand about 3 in. deep had been laid on the bottom and partly up the sides of the core box. This done the arbor was centered and rapped down about 1% in. below the joint of the box. The sand between the hoops, directly under points A and B, was cleared away and a wooden block 6 x 8 in. laid against the core box to allow for hand holes, through which the chains for handling the core might pass and the bolts fastening both halves of the core together be set in place. The sand was then rammed hard and about 6-in thick all around the outside of the arbor and the hollow center packed with coke, which was to be removed after the core was baked, leaving the cope half just a shell about 6 in. thick. The box was next clamped, rolled over and drawn as before. The two wooden blocks were withdrawn and two cores of the same size were made, to be fitted and pasted into the holes in the main core after everything inside had been secured. A hole large enough to admit the vent pipe was next cut through over C, the core care- fully slicked, blacked and put on to the carriage and both run in to be baked. Making the. Mold. # It was decided to make the egg in dry sand to avoid all risk of cutting or scabbing, for as before stated any sand holes or other imperfections would cause the casting to be rejected. The nowel half of the casting was laid on the mold board and the nowel half of the flask, which was about 10 in. deeper than the pattern, was next lowered on the board and the pattern centered in it. Two gates, one of which is shown in Fig. 6, were set in place, allowing the iron to enter the mold under the core with as little fric- Fig. 6.—Section of the Lower Half of the Flask and*the Casting. the joint struck off and the sand cut away from the arbor at A and B to admit the chain’hook for handling the core after it was baked. The sprue was next withdrawn and the whole core carefully vented with a 5-16-in. wire forced through the cinders, which were also exposed in several places to allow the escape of steam while baking. A light bed of black sand was next sifted on and the core plate rubbed firmly to place, the box and plate clamped together and turned over by the aid of the crane, clamps removed, box well rapped and drawn from the core. The core being carefully inspected for soft spots near the hoops, slicked over and given & heavy coat of blacking, was then placed in the oven. The half arbor for the cope having no leg castings to tion as possible. Dry sand facing was then shoveled in, covering the bottom board, gates and pattern to a thick- ness of about 6 in., and heap sand, wet with thin clay- wash used as a backing, and rammed in the usual way until the pattern was covered. The sand was then brushed from the bosses A, B and C, Fig. 2, and castings B, Fig. 4, were fitted over prints A and B to provide a bearing for the core and hold it (when placed) in its exact location. A 2%-in. sprue was set on the center of the core print B, Fig. ‘, and the flask filled up and rammed to the. top, struck off, vented and the bottom plate which had holes in it to correspond with prints A, B and C was put on and struck several blows with a sledge to give it a solid bearing. It was then firmly | | | : | Smee eae es ta eae ANGRY aE RA mNNND aR | | June 13, 1907 clamped to the bottom board and flask and turned over on a level bed, after which the clamps were removed, the board lifted off and the joint made ready for the cope pattern. The ramming of the cope needs no description other than to say that the sprues were located between the bottom gates so that they could be connected to them by gates cut in the sand at the joint, thus breaking the fall of the molten iron before it enters the mold, which if not done might throw the iron against the core with sufficient force to cut the sand and cause the casting to be dirty. A riser was set on top of the pattern at the end opposite the gates to act as a flow off when the mold was full. After lifting off the cope the mold was finished in the usual] way, the sprue at B removed and the mold put in the oven to be baked. The core when dry was taken from the oven and the plate with the bottom half was carried by the crane to a soft heap of sand, turned over and the plate removed to the yard. That half of the core was then set on blocks on the floor, the cope core lifted from its plate (by chains inserted through the holes at A and B, and hooked into the staples, which it will be remembered were cast into the arbor for that purpose) and set in place on the bot- tom core, calipered and rubbed to exact size. The cope core was then lifted off and a heavy coat of flour paste put on the joint of the bottom core, the cope core lowered and rubbed into place, the hooks removed and the crane swung out of the way. By inserting his arm in holes A and B the coremaker was enabled to bolt the cores to- gether, by putting in a 1-in. bolt through holes cast in each end of the arbor near A and B. These were tight- ened by a long socket wrench. This done the joints of the core were carefully filled with slurry and given a coat of blacking, and after being thoroughly dried with a torch the core was ready for the mold. Setting and Securing the Core. The mold was taken from the oven and the bottom half set on horses, as shown in Fig. 5. The core was then lifted by the staples and lowered in place in the mold (which had been carefully cleaned and brushed out), and blocks were set around the sides between the core and mold to insure an even thickness of metal. Two bolts were then put through the arbor at A and B, passing down through castings A, Fig. 4, and the holes in the bottom plate, washers and nuts put on and the latter tightened until the core was immovably held in place. The blocks from the sides of the core were removed and cotton waste packed between the core and mold to prevent dirt from falling in; the gas pipe was passed through the hole In the center at C, down through the core and bottom plate and firmly wedged in place. The cores made for that purpose were then pasted into holes A, B and C and thoroughly dried with the torch. This done the mold was lifted from the horses and carried to the place where it was to be cast and a section of gas pipe 3 ft. long was screwed into an elbow which was in turn connected to the pipe fastened in the core. This afforded a vent pipe long enough to clear the flask and be removed from any danger ‘should there happen to be a run out. Finally the mold was lowered on a bed of soft sand, the crane re- moved and the waste carefully picked out from between the core and mold; a coat of paste put around the joint of the mold, the cope lowered on, clamped, the runner and overflow built up and the shavings heaped near the end of the vent pipe, when the mold was ready to cast. After being cast the mold was left undisturbed for 36 hours, to allow a gradual and uniform cooling. Then it was taken apart and the casting taken to the chipping room and laid on two 8-in. skids. A few blows from a sledge broke off the castings, A, Fig. 4, the gas pipe was also removed and two men with pointed bars dug out the core sand, rolling the casting back and forth on the skids until the loose sand and coke were all removed. A heavy iron bar was next held against the arbor by one man while the other struck it with a sledge until the arbor broke, the broken pieces were removed and the bar and sledge again used until the arbor was entirely broken and removed. The casting was then chipped and cleaned and was ready for delivery. THE IRON AGE The Ajax Grapple Hook. A simple and inexpensive device for holding boiler plates or other metal sheets while being punched or other- wise worked, or carried from place to place, is the grapple hook herewith illustrated, and made by the Ajax Grapple Fig. 1—The Ajax Grapple Hook as Used for Handling Plates and Sheets. Hook Company, 136 Liberty street, New York. Fig. 1 shows the ordinary way of using the hooks, two being necessary in each case, and Fig. 2 is an enlarged view of one of the hooks. Each hook consists of an outer burden bearing hook, an inner holding hook and actuating lever and a cotter pin shackle. As Fig. 2 clearly shows, the inner hook Fig. 2.—A Larger View of One of the Hooks, Showing Its Principle of Action. travels up and down on the flange of the outer hook, actuated by the lever, which is long enough to provide a handle for quickly releasing or engaging the plate, as well as guiding it while being punched or worked. As a safety time saving device it will no doubt commend itself to boiler and tank makers, locomotive builders, contrac- tors and manufacturers in general. 1746 THE IRON AGE Atlantic City Meeting of the Testing Society. Announcement is made of the programme for the tenth annual meeting of the American Society for Testing Materials, to be held at the Hotel Chalfonte, Atlantic City, June 20, 21 and 22. Beginning on Thursday after- noon there will be two sessions each day. In the morn- ing sessions of Friday and Saturday the membership will divide between the cement and the preservative coatings meetings on the former day and between the iron and steel and the concrete, fireproofing and waterproofing meetings on the latter. At the opening session on Thurs- day afternoon will come the report of the Executive Com- mittee, the election of officers and the transaction of mis- cellaneous business. The following reports and papers are also scheduled for that session : The Raw Material Supply. P. H. Knight and C. E. Skinner. Report of Committee J, on Standard Specifications for Coke. Cc. Hi. Zehnder, chairman. The Purchase of Coal Under Specifications and Method of Testing. S. 8S. Voorhees. Cast Iron— Some Causes of Failure in Service. Robert Job. Notes on Brick Pier Tests. J. E. Howard. Report of Committee Q, on Standard Specitications for the Grading of Structural Timber. Hermann von Schrenk, chairman. The Effect of Moisture on the Strength and Stiffness of Wood. Hi. D. Tiemann. The papers to be presented at the following sessions are given below. Friday evening is left open for recre- ation. THURSDAY, JUNE 20, 8&8 P.M. Annual Address by the President: The Enforcement of Speci fications. Tests of Concrete Columns. A. N. Talbot. Some Additional Notes on Tests of Concrete Columns. I. E. Howard. The Testing of Wooden and Reinforced Concrete Telegraph Poles. IK. A. Cummings. The Corrosion of Iron. Allerton S. Cushman. Lhe Influence of Stress Upon the Corrosion of Iron. W. H. Walker and Colby Dill. FRIDAY, JUNE 21, 10 A.M. Cement Section. Report of Committee C, on Standard Specifications for Ce- ment. George F. Swain, chairman. Avoidable Causes of Variation in Cement Testing. E. B. McCready. Some Problems of a Cement Inspecting Laboratory. R. 8. Greenman. The Specifie Gravity of Portland Cement. R. K. Meade. The Control of Physical Test Results in Portland Cement. W. A. Aiken. The Effect of Oil on Concrete. R. C. Carpenter. The Structural Materials Testing Laboratories, U. S. Geolog- ical Survey: Progress During the Year Ending June 30, 1907. R. L. Humphrey. Preservative Coatings and Lubricants Section. Report of Committee FE, on Preservative Coatings for Iron and Steel. S. S. Voorhees, chairman. Priming Coats for Metal Surfaces. Linseed Oil versus Paint. Fk. P. Cheesman. Deleterious Ingredients in Paints. L. S. Hughes. Discussion to be opened by G. W. Thompson and F. P. Cheesman. The Physical Properties of Paint Films. R. S. Perry. Physical Testing of Varnishes. J. C. Smith. Discussion to be opened by C. B. Dudley, Robert Job and A. H. Sabin. Paint Legislation. E. F. Ladd. Report of Committee H., on Standard Tests for Lubricants. A. H. Gill, chairman. FRIDAY, JUNE 21, 3 P.M. Tron and Steel Section. Report of Committee A, on Standard Specifications for Iron and Steel. W. R. Webster, chairman. General Discussions on Modern American Steel Rails. This discussion will be opened by representatives of leading railroads, rail manufacturers, locomotive builders, car wheel manufactur- ers and inspecting engineers. Segregation in Steel Ingots. H. M. Howe. Effect of Conditions of Rolling on Certain Properties of Steel. H. M. Howe and William Campbell. The Heat Treatment of Steel. William Campbell. Mechanical Experiences with Limber and Stiff Rail Sections. P. H. Dudley. : SATURDAY, JUNE 22, 10 a.m. Tron and Steel Section. Keport. of Committee on Standard Specifications for Stay bolt Iron. H. V. Wille, chairman. Results of Tests of Staybelt Iron. E. L. Hancock. Report of Committee T, on the Tempering and Testing of Steel Springs and Standard Specifications fom Spring Steel. J. A. Kinkead, chairman Report of Committee O, on Uniform Speed in Commercial Testing. Paul Kreuzpointner, chairman. June 13, 1907 The History and Development of the Alloy Practice in the United States as Applied to Railway Bearings. G. H. Clamer. Compressive and Transverse Tests of Steel Connecting Rods for Locomotives. Gaetano Lanza. Tension Tests of Steel Angles. F. P. McKibben. Conercte, Fireproofing and Waterproofing Section. Report of Committee I, on Reinforced Concrete. F. BK. Tur neaure, chairman. Report of Committee P, on Fireproofing Materials. Ira H. Woolson, chairman. Investigation of the Thermal Conductivity of Different Con- crete Mixtures, and the Effect of Heat Upon Their Strength and Elastic Properties. Ira H. Woolson. Effect of Time Element in Loading Reinforced Concrete beams. W. K. Hatt. Tests of Twisted Steel Square Rods for Concrete Reinforce ment. J. J. Shuman. ‘rests of Bond Between Steel and Concrete. T. L. Condron. Report of Committee S, on Waterproofing Materials. W. A. Aiken, chairman. SATURDAY, JUNE 22, 3 P.M. Testing Machines and Apparatus. Report of Committee K, on Standard Methods of Testing. Gaetano Lanza, chairman. The National Bureau of Standards. 8S. W. Stratton. The White-Souther Endurance Machine. H. Souther. Notes on the Endurance of Steels Under Repeated Alternate Stresses. J. Ef. Howard. Effect of Combined Stresses on the Elastic Properties of Steel. E. L. Hancock. A New Impact Machine. L. W. Page. The Development of the Penetrometer as Used in the Deter- mination of the Consistency of Semisolid Bitumens. Clifford Richardson and €. N. Forrest. Multiplying Dividers for Locating Yield Point. J. A. Capp. An Instrument for Measuring Deformation in Tests of Mate- rials. H. F. Moore. ———__ +e —_____ A Remarkable Machine Shop Record. An unusual instance of the efficiency of American ma- chine shop method and system is afforded by the record recently made by the Hobbs Mfg. Company, Worcester, Mass., in turning out a rush order for envelope ma- chinery. Forty-two envelope folding machines, each con- taining about 1000 parts and having 17 or 18 motions, and in five different sizes, were built between April 13, the date of receipt of order, and June 7, when the last was shipped, in a department employing 130 men. All had been tested, adjusted and accepted by the purchaser be- fore they left the shop. An envelope making machine is not only a complex one, but requires nicety of workman- ship and adjustment equal to that of the high class ma- chine tool. It is automatic in action, and each of the lot of 42 is designed and tested to produce 6600 completed envelopes hourly. When the Mercantile Corporation, New York, secured the contract for the stamped envelopes of the United States Post Office Department, the company was not equipped to manufacture a single envelope. The contract was awarded early in April, and became operative July 1. The company had a very serious problem on its hands if it was to keep the contract. A plant was estab- lished at Dayton, Ohio; some machinery was found in the market, but the greater part of the equipment had to be produced. The Hobbs Mfg. Company accepted an order for the 42 machines of the type referred to, and for five other machines. The contract originally called for completion within four months, was afterward reduced to 90 days, and 42 of the machines are delivered in 54 days, or in 46 working days, taking out Sundays and holidays. It is claimed that it was the largest single order for envelope machinery ever placed. The remain- ing five machines of another class will be shipped well within the 90-day limit. The 42 machines will manufacture at the rate of 2,772,000 envelopes a 10-hr. day. So urgent was the de- mand that every machine was shipped from Worcester to Dayton by express, costing about $40 a machine. _ o-e—__—_ A. R. Hunt, general superintendent of the Homestead Steel Works of the Carnegie Steel Company, at Home- stead, Pa., has issued new orders at the plant which more fully regulate the working hours. Going on turns are 5.30 to 6.50 a.m. and 5 to 6 p.m. Coming out hours are 6 to 7.10 a.m. and 5.30 to 6.20 p.m., with some devia- tion for Saturdays and Sundays. [aieti tinct camnssemmnntennnietiditiemenns. £ethcaeidieh Deena. ainda ean pe ee ra 2 are ISH ee ores eo June 13, 1907 A Loose Ring Riddle. BY E. H. MUMFORD. In all horizontally operated sand riddles, whether driven by hand or power, the sand which lies next the screen itself moves little, if at all, on the screen, especially when the sand is damp, as in ordinary foundry practice. The sand above slides somewhat upon this nearly sta- tionary sand, and the top layers move nearly the whole stroke of the riddle, with reference to the riddle itself, in other words, remain comparatively stationary with re- spect to the ground. Thus it happens that a large part of the work expended in riddling sand in the old way is consumed in simply moving the sand upon itself, which, of course, produces no riddling effect. It is an old trick to carry scrap in the riddle, which. sliding back and forth on the wire, scrubs the screen clean, The writer was operating a power driven riddle in connection with some very particular work on a Rath bone multiple machine, which required that a flask which, with its sand box, measured inside 1444 x 16% in. and 6 in. deep, should be filled completely with sand from rhe Mumford Loose Ring the power riddle in 6 see. The shaker in use was of the double screen type, the upper screen being of *4-in. mesh, and located 2 in. above the lower screen, which was a No. 4 and about No. 14 wire. About 50 molds an hour were being handled, and although plenty of scrap was used the bottom screen fouled very rapidly with clay from the sand, which was a little heavy, and on a damp day the riddle clogged rapidly. With an idea of cleaning this bottom screen with scrapers driven by the inertia of the sand itself No. 20 galvanized iron was cut into 2-in. strips and bent by hand into rings approximately 5 in. in diameter. The writer half expected that not only would the bottom screen be effectively cleaned, but that some novel effect would fol- low. Stiil the result surprised him, for not only were the patches of clay at once swept off the lower screen, but the inertia of the entire body of sand, or what is the same thing, the entire work of the power shaker, was de- voted to moving all the sand in the riddle at a uniform speed over the lower screen. It was apparent at once that the efficiency of power shakers for sand was ¢apa- ble of being greatly increased. The galvanized rings did not last long and egg crate structures of sheet steel were tried, but proved less satisfactory than the loose rings. The results of careful working tests under practical foundry conditions of a power shaker, fitted with loose rings and double screen, were as follows: A No. 4 THE IRON AGE 1797 screen, 20 x 30 in., handled 195 shovelfuls in 9 min. A double screen of same dimensions, with the upper screen consisting of %-in. mesh, No. 10 wire, and the lower screen No. 4 mesh, No. 14 wire, handled 150 shovelfuls in 5144 min. The same double screen, with rings added, handled 281 shovelfuls in 8% min. The sand for all tests had been run through with the rings to get the scrap out, in order that what went through afterward might be of uniform quality, so that all the sand used had first passed through the No. 4 screen. Pains were taken to make the shovelfuls as uni- form as possible, and conditions of shoveling. working pressure, &¢., were practically the same for all of the tests. The gain in efficiency produced by the loose rings over the double screen was 21.4 per cent. The gain in effi- ciency of the double screen over the single screen was 25.8 per cent., while the gain in efficiency of the double screen with the loose rings over the single screen was 52.8 per cent. It is a familiar experience among builders of labor saving machinery that as efficiency increases improved methods find a constantly narrowing field in which to make a showing, so that a gain of 21 per cent. over the Power Driven Foundry Riddle. fastest sand riddling previously known becomes quite in- teresting. Another way of showing the gain in efficiency is the following: Total Time. Shovelfuls shovelfuls Riddle. Minutes. per minute. Efficiency. 195 Single, No. 4. 9 21.67 1 150 Double. nl, 27.3 1.258 281 Donble, with rings Si, 33.1 1.528 From which it will be seen that the rings add again as inuch speed to the riddling as does the double screen. In the illustration, which shows receiving and dis- charge hoppers, with a power shaker between them. forming an equipment such as would be attached to the overhead structure of a sand conveying apparatus, the loose rings are just under the coarse screen seen under the upper hopper. The machine is built by the E. H. Mumford Company, Philadelphia, Pa. The first annual dinner of the Wellsburg Board of Trade of Wellsburg, W. Va., was held June 6, and about 250 persons were present. Congressman William P. Hub- bard was the principal speaker. Addresses were also made by H. C. Franzheim of the Wheeling Potteries Com- pany, R. B. Naylor, secretary of the Wheeling Board of Trade, and William Banfield of the Follansbee Brothers Company, which has sheet and tin mills at Follansbee, Ww. va. Sentinel Pyrometers. Their Application to the Annealing, Hardening and General Heat Treatment of Tool Steel. A paper on the above subject was read at the recent meeting of the Iron and Steel Institute by H. Brearley and F. Colin Moorwood, which is particularly interesting to all makers and users of tool steel. Most authorities state that to anneal or harden a piece of tool steel with perfect success, one needs merely to heat it uniformly in all parts to a temperature slightly KCl. 20 40 60 80 THE IRON AGE Fig. 1.—Diagram Showing the Lowering of the Freezing Point by the Gradual Addition of Potassium Chloride to Sodium Chloride or Vice Versa. beyond its change point; then for annealing, to allow it to cool slowly; or for hardening, to quench it in water before the temperature falls again to the recalescence point. A suitable outfit for this purpose, however, is generally costly ; and not alone the cost of the particular form of pyrometer chosen, but also the delicate handling required (and it should be noted that repairs mean delay as well as expense) forces many responsible persons to choose readier, if rougher, means of determining the change point. Determining Changes by the Eye. The readiest of all means—that is to say by the eye— has been much despised since instrumental methods be- came possible. Whatever claim to accuracy it may possess should not, however, be overlooked. If a chisel or other form of taper instrument be heated in an ordi- nary smith’s fire, the change point is reached first at the cutting edge and then gradually farther and farther back, as the mass of metal increases. This reaction ab- sorbs heat, and a dark band slowly traveling upward shows precisely where the change is actually taking place. Below this band the change is completed; above it the change has not yet commenced. When the dark line has traveled the desired distance beyond the cutting edge the tool is quenched and found to be hardened up to the line, but not beyond it. This observation can be applied to the hardening of chisels, sets, mill picks, axes and numerous tools used by the smith, the bridge builder, the miller or the wood cutter, in regions where any form of pyrometer would be use- less. Determining Changes by a Fusible Material. The more useful, though less accurate, forms of tem- perature indicators, so far at least as the control of fur- nace operations are concerned, are those depending on fusibility. The substances which have generally been applied are metals, metallic alloys, or earthy silicates. The most successful instances of this class are the Seger cones, which enjoy a reputation over a range of 600 to 1900 degrees C. (1112 to 3452 degrees F.) wherever pot- tery and general refractory substances are fired. For noting the temperature of regenerative chambers, and as standards of fusibility for refractory materials, Seger cones have also been found distinctly useful in iron and steel works, but for the more precise annealing or harden- ing operations they leave much to be desired. Their de- fects are inseparable from those of any form of indicator THE IRON June 13, 1907 AGE which gradually softens without actually melting. The deformation is not entirely dependent on the degree of softening, and the softening of the cone itself is, to some extent, influenced by the rate at which the tempera- ture increases. Salts of Metallic Oxides as Temperature Indicators. In spite, however, of these defects, a “tell tale,” which can in most cases be placed exactly on the spot of which the temperature is required to be ascertained, may occasionally be more serviceable than a _ fixed pyrometer of a more expensive type, which registers the temperature of its immediate vicinity only. By making the “tell tales” from materials which clearly melt at or above a definite temperature, and, after melting, also continue to show when the temperature falls to or below that point, a simple form of indicator is discovered, which may claim a place among instruments of precision. The authors would suggest that for this purpose no more suitable material could be adopted than well chosen salts of the metallic oxides. Some metallic salts have solidifying points so easily observed and so reliable that they are used for calibrat- ing the most delicate thermo-electric pyrometers. Sodium chloride, for example, whether in the form of the purest native rock salt, ordinary granular kitchen salt, or spe- cially prepared from chemically pure bydrochloric acid and sodium carbonate, melts at 770 to 775 degrees C. (1418 to 1427 degrees F.). Within a range of 2 or 3 de- grees one may also obtain potassium chloride to melt 1050° CENTIGRADE K2 S04 (1042) 1000 900 (860) Na2 SO4 $50 (818) Na2 CO3 800 (770) Na Cl, 750 (740) KCI, i ,aNa2 CO3 Ko 3*—|CO3 (662) 400 MOLECULAR 80 60 40 20 0 _PROPORTIONS 20 40 60 80 100 Fig. 2.—An Illustration Showing How, with Comparatively Few Salts, a Range of 500 to 600.Degrees C. Can Be Covered at Every Point. at 740 degrees C. (1364 degrees F.) ; barium chloride to melt at 930 degrees C. (1706 degrees F.), and so on. How Such Salts Are Used. Assuming that these salts are stable on prolonged heating, they are, for the purpose in view, cast in cylin- ders and placed in small porcelain saucers on the floor of the furnace or oven to be controlled. If the tempera- ture be gradually raised, the potassium chloride cylinder r ) t : ; t / a a eeeern ee June 13, 1907 remains erect and retains its form absolutely until 736 to 738 degrees C. is exceeded and 740 degrees reached, when it melts down to a clear liquid in the saucer. In the same way sodium chloride and barium chloride melt in their turn. Should it be desired, therefore, to control and maintain any particular area within certain tem- perature limits—say A and B—a salt cylinder, or senti- nel, as it is called, whose melting point corresponds to A, must become molten, and a sentinel whose melting point corresponds to B must remain solid. If the tem- perature exceeds the melting point of B, both sentinels become molten; if it falls below A from a higher tem- perature, then the mirror faced liquid freezes, and at once indicates the fact. If a variety of pure salts be ) arranged according to their Yt melting points, the list shows inconvenient gaps. Not only roel; so, but many salts volatilize CA or partially decompose, or are Z corrosive in the molten state, he while others are hygroscopic, strongly crystalline, expensive, or otherwise unsuited for the purpose, so that the list of really eligible materials is a small one. To make up the deficiency two or more sub- stances must be combined in varying proportions. An Illustration Given of Cov- ering'a Bange of Tempera- ture, —+— ith oy Sosy rpperenes As an illustration of what ean be done in this manner, the temperature range of 630 to 770 degrees C. has been chosen, which is covered en- tirely by various combinations of sodium potassium chloride. The gradual addition of potas- sium chloride to sodium chlo- ride, or vice versa, lowers the freezing point, until with one molecule of each salt the low- est attainable point is reached. A diagrammatic representa- tion of this phenomenon is given in Fig. 1. It is at once obvious that if it is required to make a sen- tinel of any desired melting point between 630 and 770 de- grees C., it is merely necessary NT} {i to glance along the tempera- Ni ft ture, note where the line from the required temperature cuts the curve, and from that point drop a line to the composition ordinate, to discover precisely what molecular proportions of potassium and sodium chloride ow should be used. This the au- neuen eaiaenansos thors have done hundreds of an Annealing Box or a times, and always with most Closed Furnace. satisfactory results. As the salts in question are somewhat volatile in their molten state, it is important to know whether the respective rates at which the separate salts may volatilize at temperatures beyond their melting point make any considerable difference to the reliability of the sentinels. The Rates at Which These Salts Volatilize. > ‘ ' Liedtke hbk lddbbdathiddlibdbigiiibiaiaiadiiididbidibibdadbddd Four separate lots of sentinels solidifying at the tem- peratures shown in the first column of the subjoined table were exposed for two consecutive hours—in the one case in platinum and in the other case in porcelain crucibles—to temperatures respectively 100 and 200 de- grees C. above the designated melting point. The re- determined solidifying points given in columns 2 and 3 THE IRON AGE 1799 show that differential volatilization does not take place to any appreciable extent. Original After keeping 1 hour successively. melting point. At. 750° C. At 850° C. 656 655 655 683 681 680 At 800° C. At 900° C. 710 710 712 734 734 734 Without dealing further with a detailed description of other melting point curves, a selection is added in Fig. 2 which shows clearly how, with comparatively few salts, a range of 500 to 600 degrees C. can be covered at every point. The portions of the curves shown in broken line indicate mixtures which give less distinct solidifying points than usual. The practical utility of the sentinel has already been referred to in general terms. A common, but none the less important, operation in many steel works is the an- nealing of blanks, which have subsequently to be ma- chined. They come from the forge or mill, some finished too hot, some too cold, but all requiring to be made soft, and at the same time to possess a uniformly fine structure. This result can be attained ina reverberatory furnace by placing a sentinel, whose melting point is, say, 780 degrees C. (1486 degrees F.), in front of the blanks and nearest the fire grate, and a further sentinel melting at 750 degrees C. (1382 degrees F.) some dis- tance away, leaving one or two rows of the blanks in between. The latter can only become molten after the blanks have attained a temperature of 750 degrees C., and so long as the former sentinel remains intact, the maxi- mum temperature is certainly below 780 degrees C. In this way, providing suitable arrangements are made for depositing the blanks where they can cool slowly, an unbroken series can be continuously raised beyond the critical range to secure the fine structure, and cooled slowly to confer the desired softness. In a similar manner two sentinels can fix any area of a hardening furnace between certain extreme tempera- tures. The distances can be made as small as desired, so long as the mode of fixing can be regulated with corre- sponding nicety. It is possible to keep a hardening furnace at so con- stant a temperature that it may easily be divided by means of suitably selected sentinels into areas comprising some 770 to 750 degrees C. and 750 to 730 degrees C., and so delicately may this be done that occasionally, after being melted down in the saucer, one-half of the sentinel is observed to be liquid and the other half solid. To indicate the temperature inside an annealing box or closed furnace or any place where the sentinel cannot be kept under direct observation, the instrument shown in Fig. 3 is in use. It consists of a wrought iron tube partly closed at the lower end; inside is a-rod to which one of the small sentinels can be attached, and at the upper end of the rod-is a spring which causes pressure to be exerted on the sentinel. As soon as the desired temperature is reached the sentinel melts, the inner rod _ falls, making electric contact and causing an electric bell to ring. a. B. W. ee The Legislature of Wisconsin has enacted a law by which the Milwaukee School of Trades will become a part of the Milwaukee public school system and be sup- ported by funds from the school tax levy. The law, as passed, provides that the School of Trades shall be rep- resented in the School Board by an advisory committee, and this advisory committee will continue to have the active management of the school under its personal atten- tion. The Milwaukee school was opened a year ago and has been supported by public spirited manufacturers of Milwaukee. Wisconsin is one of the first States in this eountry to recognize officially such an institution and make it a part of the public school system. British stocks of pig iron have been heavily reduced this year. At the beginning of this month the total stocks in Connal & Co.’s stores of Middlesbrough and Scotch pig iron was only 328,156 gross tons, against 688,- 931 tons a year ago. For a long time every week has shown a substantial decrease. The Scoville Check Valve. To overcome the troubles that are experienced with the ordinary type of check valves, due to their tendency to stick or bind, C. L. Scoville, Ashtabula, Ohio, designed what is known as a self-grinding check valve, and it is now being manufactured by the Dolph Valve Company, Buffalo, N. Y., and Erie, Pa. The principal feature of the valve is the provision of means to cause the disk to rotate as it seats, which has the effect of grinding the seat and keeping it smooth and decreasing the tendency for the valve to cock. Other features of the valve are an arrangement whereby the lift of the disk may be Fig. 1.—The Scoville Check Valve, Made by the Dolph Valve Company. limited and the value locked and closed while cleaning. The accompanying illustrations show the exterior of the valve, Fig. 1: a central vertical section of the valve, Fig. 2, and the disk removed from the valve, Fig. 3. It may be seen from Fig. 1 that in exterior appear- ance the valve resembles the ordinary type, but a modifi- cation in the form of the body, consisting of a dropping of the bottom, gives an interior chamber so shaped that « direct pressure is obtained on the bottom of the disk, assuring, it is claimed, greater action and less liability to bind when unseating. The interior of the valve, as shown in Fig. 2, has the ordinary diaphragm, containing the valve seat. The valve disk differs principally from the common type in that it has near its upper face several radial curved blades, produced by recessing or chamber- Fig. 2. Fig. 3. Section of the Valve and the Disk Removed. ing the upper part of the disk. In these pockets the fluid gets a bearing on the blades, causing them to rotate as the valve seats. The screw shown at the top in Fig. 1 is used to regu- late the lift of the valve and to lock it. This screw is threaded through the top of the casing and its lower end limits the movement of the valve stem. The adjustment of this screw is fixed by a jam nut. By loosening the latter and running the screw clear down the valve disk can be locked or closed when such a condition may be desirable, ‘ When this valve is in operation the pressure brought to bear upon it by water, steam or other fluid exerts the THE IRON AGE June 13, 1907 ordinary downward pressure on the top of the disk and at the same time produces the rotary motion while the valve is seating by contact of the fluid with the wings. This rotary motion is produced by the back pressure, so that the valve is caused to turn as it is seating, and any foreign substance which may adhere to the valve seat or disk is wiped or ground off, so that a perfect closure is secured. Part of the good effect is a consequence of the fact that the valve disk does not seat in the same place each time. The manufacturer claims a considerable saving by use of the Scoville check valve, inasmuch as it is very seldom necessary to regrind it. Regrinding can, however, be accomplished by loosening the set screw and grinding the disk in the same way as is done with an ordinary valve. Demonstrations from use of the valve, it is claimed, bear out the contention that it will never stick and will maintain a tight joint by its self-grinding action. pee The Extent of the Lapland Iron Ore Deposits. The Swedish Government, in view of its important acquisition of Lapland ore properties, with options on additional interests that may be acquired in 1952, as al- ready detailed in these columns, has had an expert report on the extent of the Kiirunavaara and Luossavaara de- posits in Lapland. Some questions had arisen as to the ore in areas which are the exclusive property of the Luossavaara-Kiirunavaara Company and where the State has no ground rights. London Hngineering gives the following as to the findings of the Government expert, which do not touch at all on the third, or Gellivare, field in Lapland: A previous Government proposal was based upon the calculation that the Kiruna deposits above the level of the Luossajiirvi (Luossa Lake) amount to 265,000,000 tons. and under the level of the lake to 1000 ft. depth to 510,000,000 tons, making an aggregate of 775,000,000 tons. The Luossavaara deposits were calculated to con- tain 18,000,000 tons of iro