The Iron Age 1909-08-05: Vol 84 Iss 6

THE IRON AGE New York, Thursday, August 5, 1909. The Westinghouse-Leblanc Condenser." BY EDWIN YAWGER.+ " a The Westinghouse-Leblane condenser bears the same relation to the familiar type of condensing apparatus that the steam turbine does to the reciprocating engine. It is in fact a turbine type condenser. Like the turbine, it occupies only a small fraction of the space formerly al- lotted and develops superior efliciency through a simple i. +o cael | ee] BERG ca aie ts eee Fc en ee ee Fig. 1.—A Motor Driven Westinghouse-Leblane Condenser Serving a 1000-Hp. Turbine. application of rotary mo- tion, with no reciprocating or rubbing parts and no valves. At the time of the in- troduction of the steam turbine it was. announced that a very high vacuum would improve _ turbine economies to an _ extent hitherto impossible when applied to reciprocating en- gines. It became evident at once that the old types of condensers, good enough for 25 and 26 in. vacuum, would be practically useless for a vacuum of 28 or 29 in. While many refinements have been made in all features of condenser design, they have been generally along the lines of former practice. The principal improvement has been to apply a separate dry vacuum pump for the removal of air and noncon- densable vapors. The dry vacuum pump, as commonly constructed, is a direct steam driven reciprocating -unit, with its air cylinder and valve mechanism designed to reduce as far as possible the return to the condenser of the compressed air from the clearance spaces. When x From an address before the Association of Iron and Steel Blectrical Engineers, at Buffalo, N. Y.. June 24, 1909. t+ Westinghouse Machine Company, East Pittsburgh, Pa. SECTION M-M. THROUGH WATER PUMP. Fig. 2.—Details of the Condensing Chamber and Water and_Air Pumps. it is realized that the air following back from the clear- ance will exceed many times the original volume it be- comes evident that the ideal vacuum will never be reached by the reciprocating type of pump. In the effort to overcome these inherent defects build- ers have resorted to numerous refinements. Air cylinders are water jacketed to prevent overheating. Mechanically operated air valves are introduced to prevent the building up of a back pressure in the condenser sufficient to lift voluntary valves from their seats. Two air cylinders are sometimes put in series, which manifestly improves the efficiency. An additional set of flash parts is some- times introduced, which permits the air compressed in the clearance to be almost instantaneously discharged into the opposite end of the cylinder, just before the suc- tion valves open. This last would largely remove the bad effect of clearance if it did not in a measure defeat it- self. The sudden expansion resulting causes a re-evapo- ration of the moisture on the cylinder walls, and hence no air can enter from the condenser until the piston has traveled far enough to equalize the pressure. The net result of the combination of such expedients is to im- pose a burden of first cost and maintenance that will rp -- he SECTION N-N. THROUGH AIR PUMP. overbalance the doubtful benefits to be secured by ex- treme complication. Essential Features, The most striking feature of the Leblanc condenser, as may te judged by the view of an installation given in Fig. 1, is its compactness and simplicity. While it em- ploys the excellent feature of separate removal of water -and air, its functions are performed by a pair of small turbine type rotors ona common shaft, in a single unit casing, which is integral with the lower portion of the condensing chamber. The condensing chamber is of small diameter, being but slightly larger than the exhaust open- ing of the engine. The pre-eminent superiority of the DISCHARGE. I | | ; | ere 384 system lies in the practically perfect removal of air and noncondensable vapors. The detailed description of the air pump, given later, shows how this result is obtained by mechanism that is practically indestructible. A general sectional view of the standard Westing- house-Leblanc condenser is given in Fig. 2. Exhaust steam enters at D and cooling water, entering through pipe A, is projected downward through spray nozzles B. The injection water and condensed steam flow to the centrifugal discharge pump M under a head of 2 or 3 ft., which insures positive filling of the pump. The exhaust steam is drawn downward and condensed by the water spray. The space E above the water is occupied by wa- ter vapor and the air released from the injection water Fig. 3.—The Pump Impellers. and from the exhaust steam. This space communicates with the air pump N through pipe K. The Air Pump. The air pump is entirely new in principle and differs from all ejector type pumps which depend on friction for the entrainment of air. It projects a series of wa- ter pistons through the discharge nozzles, each one of which forces ahead of it a small pocket of air. The air mingles with the water in the lower portion of the nozzle, but the speed is such that none of it finds its way back to the condenser—i. c., there is no leakage past the pistons. successive layers of water is positive and the neutralizing effect of clearance is entirely eliminated. The water supply for the air pump may be taken from the main water inlet or a supply may be placed in a tank and used over and over in the air pump. Since the air pump water is in communication with the condenser it is drawn by suction into an annular chamber, G, which is over- hung by the buckets of the pump rotor. The water passes out of the chamber through the ports H and is projected downward in a rapid succession of water pistons. At the lower end of the air pump nozzle is an auxiliary ejector nozzle, L, to which is connected a steam pipe. In starting up the condenser steam is turned into this auxiliary nozzle for a few moments to create suffi- cient vacuum to start the regular flow of water through the air pump. Where the level of the cold well is 3 or 4 ft. above the basement floor the air pump may be started without the use of steam. The air pump rotor and the main pump runner are in- closed in a common casing and mounted on the same shaft. There are only two bearings and the shaft glands are made air tight by water seals. The impellers and shaft removed are shown in Fig. 3. Power Requirements. The pumps are usually driven by a Westinghouse steam turbine and under ordinary conditions require from 2 to 3 per cent. of the power generated by the main engine, The exhaust from the condenser turbine is util- ized for heating feed water and when combined with the exhaust of other plant auxiliaries the quantity is just about sufficient to maintain a feed temperature of 212 degrees F. Where economizers are used, or there may be extra sources of exhaust steam, it would be advisable to operate ‘either the condenser pumps or the exciter by an electric motor. The main pump is commonly designed to dis- charge against only a few feet head, sufficient to over- THE IRON AGE The initial pocketing of the air between the - August 5, 1909 come friction in the discharge line. If it is desired to elevate the water to the top of cooling: towers, or other moderate elevations, the pump can readily be modified to meet the additional duty. Counter-Current Principle. This term is often used in connection with various apparatuses whose functions involve a transfer of heat. Aside from its application to surface condensers, it is generally ignored by jet condenser builders, although sometimes vaguely referred to. In general it may be said that counter-current principle as applied to a cool- ing process consists in so disposing the cooling medium that the substance being cooled will at the instant of withdrawal be subjected to the-full effect of the lowest temperature. Thus, in a surface condenser the water is introduced at the top and the steam at the bottom, which, rising to the top, is exposed to the entering cold water. The air, which is always present, being noncondensable, is little affected by this final cooling, but the effect on the final volume of steam is remarkable, a much greater proportion of it being condensed in the cooler region, and — the air pump, instead of handling a certain volume of air plus a relatively large volume of steam, is enabled to draw out a mixture from which a large part of the steam as such has been eliminated. This law may be illustrated as follows: Case I.—-Slight Counter-Current Effect. Assume initial temperature injection water —............. 70° Temperature at which air is removed =.............-+000. 90° Vaetumn {temperature 101.3°) eis ier oes ole og cee ie wes 28 in. Weight of air entering condenser per minute —........... 1 Ib. In this case, owing to an excess of cooling water as ordinarily supplied, the mixture of air and vapor is taken . off in a partly cooled condition—i. e., from 101.3 degrees (the hottest point) to 90 degrees. At this temperature and pressure the volume of the pound of air alone is 221 cu. ft., while the volume of the steam in the mixture is 589 cu. ft. Therefore, the air pump, to extract a pound of air per minute, must have an effective displacement of 221 + 539 = 750 cu. ft. per minute. Case II.—Full Counter-Current Effect. Assume initial temperature injection water =............. 70° Temperature at which air is removed =—...........05--008 70° Vaeunm: (temperature 201.8" eso. . hedicnic esa sed ceases 28 in. Weight of air entering condenser per minute —........... 1 Ib. In this case the full counter-current effect is realized, the mixture of air and steam being taken off at 70 de- Fig. 4—A Westinghouse-Leblane Ejector Condenser as It Would Be Operated in Connection with a Westinghouse Vertical Compound Engine. . grees (a cooling of 31.3 degrees below the hottest part). At 70 degrees the volume of 1 Ib. of air alone is 213 cu. ft., while the volume of steam in the mixture is only 125 cu. ft., making a total of 338 cu. ft. for the air pump to handle, or less than half the size required for case I. These relationships remain the same whether the cooling is done in a surface or a jet condenser, and the Leblanc air pump as applied to either type combines in its cold circulating water both the means for expelling the August 5, 1909 rr WATER y Tete a ; air and simultaneously cooling the mixture to the point of minimum volume. Small Sizes and Separate Air Pumps, For units smaller than 300 hp. it is customary to — eliminate the main condensing chamber and pass all the exhaust steam, as well as the air, through the air pump only.. For this service the air pump is slightly modified, a relatively greater amount of water being used, which serves both to expel the air and condense the steam in one operation. The same high efficiency is maintained, and the apparatus occupies scarcely more space than that required for the exhaust pipe alone. The perform- ance of this type is shown by tests recorded later, and’ Fig. 4 illustrates a vertical steam engine equipped with one of these small condensers. For use with surface condensers, both stationary and marine, and for application to barometric and other types of jet condensers, evaporating pans, &c., the air pump is furnished separately. Embodying as it does the vital element of the Leblanc system, its application in any situation requiring an efficiggt vacuum will insure a marked improvement in the effectiveness of the entire equipment. In the case of new installations of surface condensers the air pump and the circulating pump may lhe combined in a single compact unit substantially as shown by Fig. 5. The First Year’s Showing. At the present writing there have already been con- tracted for in this country over 60 Leblanc condensers, aggregating 75,000 hp. Most of these serve turbines of various types, while a few, especially small ones, are used with reciprocating engines. Results obtained from some of these plants are set forth in the following tables: Shop Test.—East Pittsburgh. No. 12 Condenser.—Capacity and Efficiency. Vacuum referred Per cent. toa 30in. of ideal barometer. vacuum. Steam condensed. Pounds per hour. - -~—Temperatures.—, Injection. Discharge. 79 BEBO ei Rai hie or 65 28.76 99.5 MOBO SOS i NR 70 92 28.09 98.8 PEDO bic sii GSLs oe 71 97 27.96 99.3 SPA sia careening 70 104 27.59 99.3 GIO 6 isis ASS cRaS 70 112 26.81 98.8 RPO. 4:45 Cewita sean 55.3 72 29.06 99.5 E400 Fi656,5 he sae weeaeey 56.5 86.3 28.44 99.4 Sy 55:55: sag Aisi cae eae 62.3 94 28.21 99.7 DAUD b 5:10 RAR MRE’ 65.5 103 27.51 99.0 ENO 4 sik Scoisnial 59 nie woe 54.0 102 27.56 99.0 Jersey Central Traction Company, Keyport, N. J. No. 5 Condenser.—Capacity and Efficiency. (Rated capacity 8380 Ib. steam condensed at 27-in. vacuum and 90 degrees injection temperature.) Vacuum a referred Per cent. -—Temperatures.—, toa30in. of ideal Injection. Discharge. barometer. vacuum. Steam condensed. Pounds per hour. DOO k shasta ieee Sy 85.5 100.5 27.8 99.1 DBO Sk ws yt 87.0 108.0 27.3 99.1 SAO a4 scan 030 eece see 88.0 120.0 : 26.4 99.4 END hoi bins Sone sake 87.0 124.0 26.1 99.7 EWG O Ss pce ae cea anaes 87.0 139.0 24.0 98.9 IRON AGE 385 WATER GUPPLY FROM cao - . SAW ic em AT | Fig. 5.—Side and End Elevations of a Surface Condenser Equipped with tie Westinghouse-Leblanc Water and Air Pumps. WOT WELL fi ‘ 0 ~~ Union Sand é€ Material Company, No. 5 Condenser.—Efficiency Only. Temperature Vacuum referred to Per cent. of discharge. 30-in. barometer. ideal vacuum. We was 6 bed eeels 5 epee ees oa 28.9 99.2 OEP RCE 1 OI et 28.7 99.1 DG Sha ree ran cea ate ee ee 28.7 99.3 WO hme a sied Hake Ch Teed ee ae eo 28.6 98.5 ings—viz., from 450 to 525 kw. on the turbine.) Jacksonville Oil Mill Company, Jacksonville, Ala. No. 1 Condenser.—Efficiency Only. Temperature Vacuum referred to Per cent. of discharge. 30-in. barometer. ideal vacuum. ROBE. do 6 SitiW Sas Ais Be 27.86 99.6 SDs Bild 68s i ee es CORA 27.66 99.8 OD ish cs seictdicin Gia 3S Oa sie ie 28.18 ; 99.5 Relative Volumes of Air and Steam in a Saturated Mizture at Various Temperatures. a% ¢e2 mm Sy 4 3 oS §s ‘ Pounds Per Square Inch Absolute. Br cae ok 9-49 0.98 1.47 1,96 2.45 2.94 14.697 Da, o. ghee Inches Vacuum Referred to a 30-In. Se Bo > g = Barometer. -. £ atse, 29. 28 27. 26 26 24 0 s2 eZt Se &s Per Cent. Volume of Saturated Air Pres- TS oao Sisa ent in a Mixture of Air and Vapor of a os = Water. 0.2545 29.48 48.0 74.0 82.6 87.0 89.5 91.4 98.4 70...0.3602 29:26 26.5 63.0 75.5 81.5 85.5 87.8 97.6 80...0.5027 28.97 ... 486 65.8 74.3 79.5 82.9 96.7 90...0.6925 28.59 ... 29.4 52.9 646 71.8 76.4 95.4 95...0.8090 28.35 ... 17.4 45.0 58.7 67.0 72.5 94.6 100...0.9421 28.08 ... 38.7 35.9 51.9 61.5 67.9 93.7 105...1.0988 27.77 ... ... 256 44.2 55.3 62.8 92.6 110...1.2683 27.42 ... ... 188 35.4 483 56.9 91.4 112...1.8416 27.26 ... ... 8&7 3816 453 544 90.9 114...1.4207 27.10 ... ... 388 27.5 42.0 51.6 90.4 116...1.5089 26.93 ... ... ... 283 88.6 489 898 118...1.5912 26.75 ... ... ... 188 385.06 45.85 89.3 120... 1OEG DRT oo. hs, ese. Omen Gaee eae Gaee 122... 0.7760 GGOt ek eee. HEC ee ee ee ee 128 27 TRIO AGE a ee eee ee oe 126...1.9852 25.95 BO eae hae Se 128i. : 20050 BIB ee ow ee ae 180.) D:2220! BB.4B a ON OS OTP SES S60 18H BBBSS BME a a a PON, OS See 18856 REGO 3: BEOO iO A es, SESS 186 SBOGR* GATE 288 5s RAS SS ES 188i: 057OR1) BOAR en ual de at See FO 140...2.8774 24.13 ia tide Gl RAE BSS Efficiency is here expressed by the percentage of an ideally perfect vacuum actually obtained. For instance, if the discharge temperature is 100 degrees F., the corre- sponding ideal vacuum would be 28.08 in. If, however, the observed vacuum is 27.75 in., the efficiency percentage = 98. t. 58.08 98.8 per cen would be ——+o—___ The Bethléhem Steel Company has again been de- feated in its suit against the Niles-Bement-Pond Com- pany for infringement of patents covering the former’s process of treating high speed tool steel. The adverse decision of the Circuit Court of New Jersey was ap- proved July 21 by the United States Circuit Court of Appeals. cece sislicietaaittnaescatiiitascn naintatineitpiialamnintintet ia tT, sisters annette 386 The Electric Furnace for Steel Castings. ° The Importance of Heat Treatment. The possibilities of the electric furnace in steel foun- dries were the important feature brought out in a dis- cussion on steel castings at a recent meeting of the me- chanical section of the Engineers’ Society of Western Pennsylvania, as reported in the July Proceedings of the society. This phase of the subject was alluded to by J. S. Unger, manager of the research laboratory of the Carnegie Steel Company. Referring to what the electric furnace has done in making it possible to produce in- © tricate shapes and very thin walled steel castings, Mr. Unger said: I saw a steel casting recently which was a fork for a bicycle, 6 in. long, with walls not over % in. thick at any point, cast from steel made in an electric furnace. We had the specimen sawed longitudinally and it did not show a single cavity. It was absolutely as sound as though forged or drawn from a piece of tubing. I do not believe that a casting with walls as thin as that could be made in an ordi- nary steel furnace. It is possible to raise the temperature of the steel in an electric furnace to such a high point that it will flow in as thin a wall as that, while the reducing ac- tion of the electric furnace prevents many of the gases from being occluded. There is not any bubbling or motion at all while the casting is being poured. While I do not wish spe- cially to recommend electric steel casting, I bring this up to show that very small steel castings can be made in an elec- trie furnace. Most steel casting people say that sound cast- ings with very thin walls cannot be made in a stéel furnace, and I agree with them. But there is another method of doing this. . When you consider that the temperature of an open hearth furnace is not over 150 degrees above the melt- ing point of steel, you realize that the steel foundryman’s means are limited. The steel is being made under oxidizing conditions and absorbs and holds large amounts of gases in solution, giving up a portion of these on cooling, producing sponginess. If he ‘could raise the temperature of that steel by means of the electric furnace to almost 1000 degrees above the melting point and make the steel under reducing conditions, he could make sound castings that under ordi- nary conditions he cannot secure. Say the melting point of steel is 1600 degrees, a good open hearth furnace will he about 1750 degrees, while an electric furnace will aa at 2600 degrees. Heat Treatment of Special Steels. The questions of design and of heat treatment were discussed by a number of speakers at the same meeting. On the latter head Mr. Unger gave interesting comment and experience, as follows: We should not forget something that is very important in steel casting, and that is treatment. I know that the price at which we buy ordinary steel castings to-day will not | justify a treatment that is at all complicated or that will cost much money. However, one is almost compelled to use steel castings in some cases, owing to difficulties in prepar- ing a forging of the design suitable for the purpose. At the present time in some classes of work, automobiles, for in- - stance, they are making special] castings of alloys which con- tain such metals as nickel, chromium, vanadium, titanium and quite a number of other metals alloyed with the ordi- nary constituents of steel. Some of these alloys have given very excellent results. One finds that when using a special alloy of any kind, in order to get the best results and develop the best qualities in the steel, it is necessary to treat it, not simply to anneal it, but to give it treatment more or less complicated, depend- ing on the results required. Ordinarily treatment by heating the casting to the proper temperature, then plunging it into oil and chilling it, afterwards reducing the hardness pro- duced by the oil, is sufficient. There are special alloy cast- ings in which ordinary treatment in oil is not sufficient, but water must be used to obtain the results, the oil not being sufficient to break up the coarse grain or structure. It is possible by treatment to make these special castings the equivalent of a forging. The effect of treatment is especially noticeable in manganese steel castings. It is a common and necessary practice among those who manufacture manganese steel to heat the article to approximately 950 degrees, plunge it into cold water and leave it until it is absolutely cold. This develops the characteristic toughness and increases the — to a higher point than it was prior to this treat- men I believe “flask annealing is practiced to some extent. _—_ annealing may be a benefit, but I do not believe it is as removing the casting from the flask and re- baat to the proper temperature, allowing it to cool slow- ly. In some experiments I had knowledge of, it was neces- THE IRON AGE August 5, 1909 sary to determine what was the best method of treatment to give a large mass. There were prepared from one and the same heat four or five solid cylinders measuring abgut 4 in. diameter and perhaps 6 ft. long. They were cast one right after the other in sand molds. The first one was laid aside and tested in its ordinary condition; the second was an- nealed at a temperature of about 900. degrees, cooled in the furnace until it had lost all its color and then removed and allowed to cool in the air. The next was annealed at 900 degrees and reheated to 750 degrees and then allowed to cool in the air. The next was annealed twice at 900 degrees. The next was annealed twice at 750 degrees. After they had been prepared in this way they were put in a large lathe, a nick cut around the outside, and broken and the structure examined. In annealing a steel casting you rarely reduce the tensile strength. If there is any change at all, unless it is a defective specimen, the tensile strength is al- most invariably a trifle higher than in the original casting. Annealing does make a great difference in the elongation. Approximately the elongation is increased from two to three times, speaking now of large castings, not small ones. The Breaking Up of Coarse Structure, The cylinder in its unannealed condition, just as it was taken out of the sand, was broken and the structure over the entire surface was examined and found to be very large _ grains, octahedral in character, from 3% to % in. through the crystal. One blow of the drop broke this cylinder. The next one, that had been heated to 900 degrees and cooled in the furnace until it lost color, also broke at one blow, the differ- ence in structure being noticed for about 10 in. from the out- side. The coarse grains had disappeared, but the extreme center was still coarse, showing that the treatment had not affected it throughout, and there was still a central core that seemed to be just as it was in the original casting. The next cylinder, which had been treated at 900 degrees and then at 750 degrees, showed practically the same appearance, the ef- fect of the treatment having been felt for about 12 in. from the outside, but the grain was finer than that produced by annealing at 900 degrees only. ‘The next cylinder, which had been annealed twice at 900 degrees, showed that the ef- fect of this treatment had been felt almost to the center. There was a small portion, perhaps 6 in., in the center still! coarse grained. The next one, that had been treated twice at 750 degrees, was very fine for 4 to 6 in. in from the out- side, the remainder being coarse. The object of using these two temperatures was this: If one can break up the coarse structure produced in a cast- ing or forging at a low temperature you will get a finer and stronger grain. But the coarse fracture that is produced in the large casting that cools slowly in the sand is not read- ily broken up at a temperature of 750 degrees. If one could anneal often enough at 750 degrees I believe we would have a stronger casting than when annealed at 900 degrees, but the amount of work and cost would be excessive. It is not practicable, so most manufacturers try to do their annealing at 900 degrees in order to break up the coarse structure and get the desired effect at a minimum cost. We afterward decided that for ordinary large castings, somewhat repre- senting the cylinders I spoke of, there was not enough good effect produced by a first annealing at 900 degrees and re- heating at 750 degrees to justify the adoption of that treat- ment, and the treatment since that time has consisted of re- moving the casting from the sand, heating it up to 900 de- grees and holding it there when that temperature is reached to allow the casting time to lag. By that I mean to give the grains time to rearrange themselves and allow for that chemical change in the carbon which occurs to a greater or less extent. After the casting has reached 900 degrees one should maintain that temperature for approximately 144 or 2 hr. to be sure that the change has taken place. Then you may begin to reduce the temperature in the furnace. I do not believe there is any real benefit to be derived in allow- ing a casting to soak a long time in the furnace. When it has reached the proper temperature and is of the same tem- perature throughout no further good can be accomplished by allowing it to cool down very slowly. After it shows no visible color in the furnace, I would recommend removing the casting from the furnace and allowing it to cool in the air. The additional cooling in the furnace is of no benefit to the casting and it only holds the furnace back from fur- ther use. I believe those people who make large castings, say from 15,000 lb. up, now make about as many castings from the basie furnace as from the acid furnace. I believe both make good castings. The basic furnace is a little more difficult to handle than the acid. In addition to being an oxidizing it is a purifying process as well. As the steel casting business. is difficult at best, one tries to use the easiest method to ar- rive at results, and therefore they prefer to use the acid method, it béing very much easier to operate. A Stucki, in speaking of the high shrinkage of steel castings, often twice as great as that of cast iron, said that shrinkage cracks are the result of an improper dis- tribution of metal which in many cases might have August 5, 1909 been avoided in the design. At times, however, it is almost impossible to design a casting for a certain pur- pose so as to comply with the conditions as to machining or fastening and yet avoid the danger of shrinkage eracks. In the case of castings open at one end, for ex- ample, U-sections, he said that the open ends are apt to spread, as the intervening sand will hold them apart. This trouble is remedied either by allowing for such spread in the pattern, by closing in the free ends under a press, or by connecting them by extra metal which will be removed as soon as the casting is cool. The latter method is often used to hold the pedestal legs of a locomotive frame in place. In driving wheel centers the spokes cool more quickly than the rim if not prevented by covering with sand, and will pull inward from the rim, producing cracks in the spokes or arms. The piping of castings which cool and reduce in volume may be prevented by large gates and large heads. Troubles from Bad Design. C. B. Albree, president of the Chester B. Albree Iron Works, spoke from the standpoint of the customer of ihe steel foundries: I know very little about the process of making steel castings, but I have been a user of them for riveting ma- THE TRON AGE 387 there is a finished surface to allow ample material every- where for finishing. Sometimes we had to make a consid- erable change in measurements of machine parts to make use of the steel castings received, and it is difficult to keep track of such changes when making a duplicate part later. Another difficulty we meet is in getting castings with solid trunnions. In this particular case we cannot help our- selves, as we have to have trunnions. The first thing we do with these castings is to drill a %-in. hole through the center of the trunnions, and about one out of every four castings has a blowhole or cavity in the center. The other portions of the casting may be perfect, yet if the trunnions are defective the casting is absolutely useless, as the great- est strain comes on them. We have taken up this design with almost every steel casting man in Pittsburgh and many outside and practically every one of them has had trouble. These shrinkage cavities have caused more loss, both to us and to the steel casting men, than all the other defects com- bined. And the peculiar thing that we cannot understand is why we should get three good trunnions and one bad one. It was suggested to me by the president of one of the largest steel casting companies that the trouble could be obviated in a very simple way by putting a small core hole through the center of the trunnions, so the interior would cool as rapidly as the exterior. We have not tested this scheme, but expect to do so soon. Another point about annealing steel castings. We have received annealed castings from many foundries which came out all right and have received others which from a front “HTT | El HEAVY FILLETS SECTION A-A Shrinkage Cracks on Steel Casting for Riveting Machine.—Section of the Casting, Also Section of Trunnion Showing Cavity. chines, and in that line of work we have all sorts of troubles. The majority of machinery manufacturers who use steel castings in place of cast iron on account of strength and reduced weight are much more familiar with cast iron than with steel castings. A drawing of the part required is made and the pattern sent to the foundry, but when the casting is received it is often full of blowholes and out of shape. Surfaces requiring planing do not clean up and blow- holes and shrinkage cracks make it impossible to use the casting, necessitating a-wait of from one to four weeks for a new one. Our principal experience has been with riveting ma- chines, as shown in the illustration, which consist of U- shaped parts having a heavy tension section, a lighter com- pression section and a comparatively thin web. We de- cided on a very thin web at first and as a result we got a casting that cracked all around the web close to the flanges. The foundryman told us we must have a big fillet to avoid such cracks. Sometimes we found that the foundries were not annealing the castings and we had them annealed, which saved a little trouble. We tried putting in ribs to stiffen the web and that made matters worse, because we got cracks along the ribs. We did away with them and now make the web very heavy without ribs. It took us some time to learn these details and we found that the fewer the stiffening ribs and the heavier we made the web, the better the castings. I think it cost the steel casting people a great deal because we did not know how to design steel castings. If they had told us that we were making a mistake in our designs we would have changed the patterns. In the matter of cores it has been said that the cores shrink, making the holes smaller than the core. This is certainly true. We have to machine out these core holes and we have found it very decidedly to our advantage to pay for % in. more metal and avoid trying to scrape out scale. In steel castings: we cannot rely on as accurately cored holes as in iron castings, and it is much better where view present a badly warped and crooked appearance. The whole thing would be so out of shape that, while it was a ‘perfectly good casting otherwise, we could not possibly use it. My conception of this is that they are not careful enough in supporting it in the furnace when they reheat it, and there is a bending due to the unsupported weight of part of the casting, which sags when it is heated. We had a machine to build for the United States Gov- ernment that called for a 12-ft. gap. It made a very heavy casting, about 30,000 lb., and as the pattern was pretty long and difficult to handle the foundrymen, unknown to us, put a piece across the open end of the jaws, 8 in. wide and about 3 in. thick. When we received it this piece was cast in solid and we had to cut it out, which was a rather ex- pensive operation. After cutting it out, the jaws, which were originally 20 in. apart, closed in until the distance was only 17 in. The Government threw it out, the steel foundry lost the casting, and we lost the work on it and had to pay a penalty for delayed shipment. When it comes to small castings our experience has been very bad and we use forgings throughout. Some of our competitors use steel castings, which are considerably cheaper, and we thought we would try it. We had pat- terns made of perfectly straight pieces without cores or other troublesome features, and there was no reason why they should not come out right. We tried them on ten dif- ferent machines, and in every case the castings had such blowholes that they were absolutely useless. We have never - yet had a yoke break through where the greatest strain is. under loads of from 50 to 150 tons, on the outer ends of the jaws. We use a fiber stress of 10,000 to 12,000 Ib. in designing sections. add It seems to me that the success of a steel casting de- pends first on its design and second on the foundry; so the customer and the foundryman ought to get together in order to secure successful results.” epi ra <n cnesineiinnarilaieegagetattaaen tO ‘ q ? ' a : U i | ae The Metallurgy of the Electric Steel Paroace. While technical literature contains a great deal. in recent years on different furnace types and on the elec- trical side of electric steel manufacture, the metallurgi- cal side seems to have been studied very little. The clearest summary of what has thus far been developed has appeared in a paper read before the Verein zur Befoerderung des Gewerbfleisses at Berlin by Professor E. R. Eichhoff, the successor of Dr. Wedding, who has - been closely connected with the work at Remscheid. At first the possibility of obtaining high temperatures and the neutral atmosphere were considered very im- portant features of the electric furnace, and the removal of gases was regarded as attained, since the presence of gas in steel is a function of its temperature. Still, steel saturated with a certain gas may stand for hours with- out changing its gas contents in the least, provided the temperature’ remains the same. The deoxidation of the bath is the vital point. Professor Eichhoff reasons as follows: The entry of the oxygen in the gas is due to the inter- vention of the protoxides of the slag, which must be removed. The point, therefore, is to produce slags free from iron. It is in this that the high temperature at- tainable in the electric furnace is of some utility, since slags free from iron melt with greater difficulty. The question is, How can such slags be produced in the basic process without carbonizing the bath? Simply by the use of fine coal on the slag in the furnace. This, how- ever, worked too slowly owing to the large quantities of slag which were then used and which were thought to be necessary for desulphurizing. A trial showed that carbide of calcium did not carbonize the iron, in spite of previous assumptions. Carbide of calcium was used as a reducing agent, but’ the method proved too expensive. Then the quantities of slag were reduced, which caused the electric are to be much longer, and with the carbon an adequate quantity of carbide of calcium was formed directly on the bath. It was believed that all difficulties were overcome, since there was available a hearth which did not affect the steel and a cover of slag which did not influence it, and besides cut it off from the effects of the atmosphere. The disappointment was great when it was discov- ered that under these conditions the steel contained as much oxygen as before. A series of further experi- ments were undertaken, which showed that at higher temperatures a combination of oxygen and iron is formed which differs from protoxide of iron, and. which is then stable; that the affinity for iron and oxygen rises rapidly with rising temperatures; that the affinity of carbon and iron also rises rapidly with increasing temperature, and that the affinity between oxygen and carbon rises very little relatively with a rising temperature, and probably even declines relatively so long as iron is present. Ob- servation led to the assumption that oxygen combines in some form with iron at temperatures higher than its point of fusion, and that the quantity increases with riSing temperature. In this form it cannot be attacked by carbon. When the temperature declines below that corresponding to the point of saturation then oxygen may possibly be expelled in the form of protoxide of iron, and this combines at once with the carbon of the carbide, forming carbonic oxide. , The question arose as to how these observations could be made useful for the deoxidation of the steel below the slag freed from iron. First of all it was found that oxygen could only be expelled where the tempera- ture was lowered close to the point of chilling and reheating followed under the slag free from iron. The small amount of carbon in the steel sufficed to destroy . the particles of protoxide of iron, which had not risen to the surface. It was shown Very soon, however, that this process worked too slowly; therefore a special ad- dition of carbon was made to hasten the reaction and was combined with the duplex process to be described. Success crowned the efforts to produce carbide cheap- 388 ‘Tue IRON AGE | August 5, 1909 ly in the electric furnace and thus to reduce the pro- toxide of iron in the slag. Success had crowned the deoxidation to the point that no manganese was lost when ferromanganese was added at the end of the heat. The effort was then made to produce cheap manganese directly in the furnace by adding manganese ore directly to the slag. frag This represented important. progress in three direc- tions: To begin with, all losses. of manganese were avoided; the use of expensive ferromanganese was ul- necessary, and, finally, the deoxidation by means of carbon and the cooling connected therewith in order to hasten the process. could be restricted, because the me- tallic manganese produced entered the bath, combined with the oxygen of the iron, entered the slag, was again reduced, then to re-enter the bath and continuing this course until all the oxygen had been eliminated and all the manganese had entered the bath. Until then the practice had been to carry out the melting, the purification and the making of the steel in the electric furnace. Until then it had been assumed that in this way the raw materials would be freed of sulphur and of phosphorus during the middle period. Experiments and observations in melting with slags free from iron showed that desulphurization proceeds inde- pendently of the quantity of sulphur very rapidly toward the very last stage of the process, provided the slag be entirely free from iron. Since it was possible to elimi- nate phosphorus down to minimum quantities in the open hearth or in the converter by overflowing, therefore the same quality of steel could be made in the electric furnace from impure raw materials, provided the last operations—the deoxidizing, the desulphurizing and the carbonizing—be carried out in the electric furnace. The duplex process was therefore adopted. The en- tire oxidizing process was consigned to the ordinary fur- naces, the dephosphorized material was transferred to the electric furnace, and it was only then that the greater part of the oxygen was removed by carbon and the process was carried to a close in the manner indicated. A peculiar method of carbonizing has been adopted in the electric furnace with the aid of carburite, a mixture of 50 per cent. of carbon and 50 per cent. of iron, which is so solid that it is not mechanically destroyed by the slag, nor crumbled by the heat, and is so heavy that it penetrates through the slag. This permits the car- burizing to be done with extraordinary accuracy. The process makes it possible to utilize the cheapest impure raw materials, even copper and arsenic having lost their terrors, since they are injurious only in the presence of sulphur. The steel is as thoroughly deoxi- dized as in the crucible furnace, without being affected like it by the character of the walls, and much larger quantities, uniform in composition, can te produced at a cost half that of crucible steel, or even less, under certain circumstances. —_3--@—_—_- The Union Malleable Iron Company, East Moline, Ill., which shut down July 1 for three weeks for neces- sary repairs and some needed improvements, has re- sumed operations and is running 50 per cent. heavier than at this time last year. During. the shutdown the plant was equipped with an industrial railroad for the economical handling of material, and a complete new equipment of 28 mills was installed in the hard milling department. The outlook for an increasing volume of business is very promising, as, in addition to the heavy specifications already entered, contracts are being re- newed at a considerably increased tonnage... It is ex- pected that by September 1 the plant will be running at full capacity. The New York Central negotiations for 2500 box cars, which it is expected will be divided between two car companies, are to fill requirements noted some time ago. The Buffalo, Rochester & Pittsburgh has bought 500 steel underframe box cars and 1000 all-steel hopper cars. Of the Burlington requirements of 3000 cars, 500 refrigerator and 2000 box cars are reported to have been placed, while 500 steel gondola cars are said to be under negotiation. August 5, 1909 A Portable Gasoline Engine Pumping Outfit. A novel portable pumping outfit, upon which patents are pending, has recently been brought out by the Water Works Equipment Company, 50 Church street, New York City. The outfit consists primarily of a steel wagon truck with metal wheels having tires 5 in. wide, and provided with a drop tongue to which a team of horses may be attached. In the middle third of the bed and under slung, to keep the center of gravity low, are fixed two No. 4 Edson diaphragm trench pumps, each of a nominal capacity of 6000 gal, per hour. At the rear end of the bed is mounted a 5%-hp. Fairbanks vertical gaso- line engine, and upon the inner end of the crankshaft is a pinion which meshes with a large gear wheel. The latter transmits the power of the engine to the pumps through a horizontal shaft, which operates the two eccentrics shown in the illustration. A tank is provided ' for holding cooling water for the engine and is held firm- ly in position in a frame of strap steel by a novel tight- ening device. The battery for supplying current to the ignition ap- paratus of the engine is under the driver’s seat, which THE IRON AGE 389 These outfits have already been supplied to several water departments, among them being Trenton, N. J.; Port- land, Ore.; Schenectady, N. Y¥., and numerous others have been ordered. A modification of the outfit is now being constructed, having in addition to the pumping plant an air com- pressor driven by the same engine for operating two calking tools, two riveters and a baby rock drill. ——_»-+e—__—__ The Lake Superior Mining Institute. The fourteenth annual meeting of the Lake Superior Mining Institute will be held on the Marquette range, with headquarters at Ishpeming, Mich., August 25 to 28. For the trip over the Marquette range the itinerary is as follows: Wednesday, August 25, will be spent at Ish- peming, with business session in the evening. Thursday, August 26, at Negaunee, Marquette and Munising. Busi- ness session to be held at Munising. Friday, August 27, at Munising, Grand Island and the new mines in the Swanzy District, returning to Ishpeming in time to meet the south and west bound trains. Members desiring to go underground at Ishpeming on Saturday can make A Portable Gasoline Engine Driven Pumping Outfit Built by the Water Works Equipment Company, New York. latter is made in box form and provides a convenient place for small tools. The wires are led from the bat-. tery to the engine through iron pipes which protect them from injury. Each outfit is provided with two 20-ft. lengths of suction hose complete, with couplings and strainers, and these when not in use are carried on side supports, as shown, being looped around back of the engine. These pumping outfits will handle any quantity of water up to their capacity, containing sand, gravel or sewage which is liquid enough to flow, and they can be used for pumping sewers and for unwatering trenches, foundations, excavations or any other similar class of work. They give a much more efficient and reliable service than can be obtained by the use of hand pumps. With four men the maximum amount that can be pumped in 1 hr. with a No. 4 pump is 4000 gal. Ona recent test, the results of which were by no means un- usual, made at Trenton, N. J., in the presence of the city engineer and the superintendent of the water works, Alvin Bugbee, one of these outfits pumped 14,500 gal. per hour against an 18-ft. head. The test lasted for 5 hr. and the total amount of gasoline consumed by the engine was but 1% gal. These pumps are capable of an ex- treme lift of 28 ft. The outfit as described weighs 3400 Ib., and does not require an engineer nor other skilled labor to successfully operate it and keep it in order, arrangements by applying to the secretary on Wednes- day. A partial list of the papers to be presented at the meeting is as follows: “ How Reforestation May Be Ap- plied to the Mine Timber Industry,” by Thomas B. Wy- | man, Munising, Mich.; “Mine Accidents,” by John T. Quine, Ishpeming, Mich.; “The Sociological Side of the Mining Industry,” by W. H. Moulton, Ishpeming, Mich. ; “ Biographical Notes,” by J. H. Hearding, Duluth, Minn. ; “ Reminiscences of the Early Days on the Marquette Range,” by George P. Cummings, Marquette, Mich. ; “ Historical Sketch of Copper Mining on Lake Superior,” by Alfred Meads, Marquette, Mich.; “ Code of Mine Sig- nals, Cleveland-Cliffs Iron Company,” by O. D. McClure, Ishpeming, Mich.; “Capillary Attraction in Diamond Drill Test Tubes,” by J. E. Jopling, Ishpeming, Mich. ; “ Sinking Reinforced Concrete Shafts through Quick- sand,” by Frederick W. Adgate, Chicago, Il. ee nr rr Among recent contracts for heavy electrical machin- ery taken by the Allis-Chalmers Company, Milwaukee, Wis., is one covering two three-phase 25-cyele, 6600-volt alternating current generators, with a combined capacity of 8000 kw., to be direct coupled to gas engines, in the power plant of the Carnegie Steel Company’s Carrie furnaces, at Rankin, Pa. These are the largest machines ever built for service of this character. : { 390 , THE IRON AGE A New Process of Making Chilled Car Wheels. BY THOMAS D. WEST, CLEVELAND, OHIO. At present considerable interest is being manifested in how far steel and wrought wheels may displace chilled cast iron car wheels. If the opportunities that exist to raise the standard of chilled wheels are improved there is no likelihood of cast iron wheels being abolished. The wrought and steel wheels are still largely in the experi-. mental stage, but have already proved that they have their disadvantages as well as their advantages. Brakes are not as affective on them as on chilled wheels. Great- er power is required to pull a train having wrought wheels, and wrought wheels can bend if a car runs off the track, making them unreliable when replaced on the tracks. A cast iron wheel never bends, it must break if overstrained, and can do no further damage. Brake shoes are more likely to stick to wrought ; 2 wil din ma ; NY N = . YW euuvant- \= : *° 3 J ‘ " q Ne ZER wees Me [NW SY : bse NSS “NS —— SSO Y 2 Fig. 3. The Mold just after being poured. August 5, 1909 wheels or the track work would be more likely to fail. An uneven depth of chill in the throat or through the line s and t, Fig. 6, is the most prolific cause of internal strains or weakness. In a very large number of car wheels broken, after being taken from under the cars of one of the leading railroads, 70 per cent. showed variations in the depth of their chill. If the chill is uneven in thickness, the thin side will wear away faster than the thick side, not only making the wheel untrue, but weak. This is be- cause the hardness cannot be uniform if the thickness of chill is not. Here then is another importance of an even depth of chill, for wheels that are not round initially or do not remain so, cause vibration that is not con- ductive to smooth running and track endurance. Cause of the Defects. The processes in use at the present time are not posi- tive in producing perfectly round wheels. Some are so untrue that they have to be ground after being pressed on their axles, which produces nonuniform hardness of the tread, so that they are soon untrue again and often i == | Ee =i = —— SS Fig. 7. Mold seven minutes after being poured. A New Process of Making Chilled Car Wheels. wheels than to chilled cast iroa wheels. In braking, it is not wise to entirely stop the wheel’s rotation, which gives the chilled wheel another advantage over the wrought wheel, for if the wheel slides on the track, not only is the rail. worn, but a flat place is ground on the wheel, making it pound thereafter and sometimes sufii- cient to break the rails. : The writer has lately perfected a process of casting chilled wheels whereby perfectly round wheels, with an even depth of chill all around the tread, throat and flange are obtained. This process is p