The Iron Age 1912-04-04: Vol 89 Iss 14

OZ HE | Established 1855 New Téex. April 4, 1912 Vol. 89: No. 14 Production Betterment by Time Studies Instances Showing How So-called Differential Time Observations Plant Operation, Have Developed Improvements in Equipment and Arrangement BY N. E, ADAMSON, JR. In a previous article dealing with time study observa- tions which appeared in The Jron Age of November 16, IgII, the author showed the various types of observations and described the method of obtaining them. As stated therein, short observations are divided into three types, any one of which is adapted to timing any process, the type to be used being determined by the nature of the information desired, Type A was described as being one where the starting and stopping time and the number of units performed are recorded. From this information the unit direct labor cost is cal- culated. This type is used for two purposes, viz., when only rough information is wanted before a complete time study is made of the operation, and when the efficiency of an oper- ation or operator is wanted after a complete time study has been made of the operation. No attempt is made to show how the time is expended nor the method by which the oper- ation was performed. Type B was one where the operation divided itself into elementary movements. The stop watch was started at the beginning of the first movement and was read at the end it can be successfully used in determining which of the two apparently equally good methods is the better. It is ap- parent that in order to fully standardize the manufacture of an article each operation should be investigated by making observations of type C. The object of this paper is. to show how these obser- vations are grouped in order that they may become useful and to describe the various steps to standardize the manu- facture of an article. First, it is necessary for the observer to be familiar with the successive steps of the manu- facture from the time the raw material is first handled until the finished product is ready to leave the shop. It is not essential that he be a skilled mechanic in the trade, but the more he knows about the trade which he is observing, other conditions being equal, certainly the more valuable will be the observation. There must be co-operation between the foreman of the shop and the observer. The foreman often can furnish an outline of the manufacture which is quite helpful in the preliminary study. This outline can be more clearly described if a definite manufacture is considered, and the author again discusses the manufacture of a galvanized iron 10-gal. paint drum. Fig. 1—The 10-gal. Paint of each movement, the. watch being run Deus Standard Operations for 10-Gal. Paint Drum Taken from Book of Standards. — | . | ; Stand. = Operation | Machine | Pay | | Stand. Output, Description of the Best Method. NO | No | per Diem. | Unit Cost.| per Hour. A-7 | Rivet Collars hale 2.00 | 0.001695 147 Stack on bench in front of operator, strike rivet one blow, | rivet set one blow. Drop in receptacle to left. } } ? | Shear Handles. . 58-44 | 3.04 | 0.000375 1600 Shear from binders. Place binders on truck behind oper. 1.6 24 ators. Each operator handles two strips, cutting four handles off at one time. Let handles drop to floor be- | hind machine. Move man pick up and place in truck. B-2 |Punch Handles.... ..| 58-93 | 2.00 0.00033 | 758 Place truck with handles to be panabel to left of operator as Remove hand full and place one on each end.of ma- : = ‘ | | chine. Punch both ends of two handles at one time. B-3 | Forming Handles.... ..| 58-27 2.00 0.000312 800 Place truck containing handles to be punched on the left of operator. , About 15 are picked up at one time and one placed on each end of machine. top formed han- . . She ‘es dles in a receptacle on operator's right. C-1 | Shearing Tops and Bottoms} 58-3 76 0.000742 634 Assistant stands behind machine. ile stock“on bench 1 | 2.00 behind machine ndicular to blade. Adjust gauge for assistant to Reed t , so that operator will receive sheet at a point where he can conveniently grasp the sheet. Stack blanks on machine until all of one sheet is cut. While assistant is placing new sheet, operator will } remove blanks to truck. throughout the observation. Deductions from this obser- vation showed the unit direct labor cosi, and also the method used. It is useful to a limited extent in deducting From Fig. 1 it is apparent that the drum consists of several parts. Each of these parts must be manufactured before they can enter into the construction of the whole. lasses of time, due to evils which usually can be corrected. Type C was the differential type. By use of it the manufacture of an article can be thoroughly investigated, for it shows all that the above types show and in addition These parts are: A, collar; B, handle; C, top and bottom; D, bands; E, body. The collars are made from the scraps left after the body, top, bottom and center bands have been cut from a 835 CHARACTER OF WORK PEKFORMED THE IRON AGE April 4, 1912 and the scrap would follow the punch and had to be pried off. A single stripper was at- tached and now the duration of every movement of the operator is almost constant. Many think that any foreman should see the necessity of a stripper without having to see an observation card. Our fore- | man saw it, but when he saw the amount in cash that the absence of a stripper was costing he didn’t delay getting oe Starved) S'©p-] Time | Units 7 Jo lala] a ome fm | we 2 = jo |e eee o is Assistant's Inspected by ____Not __Workman’s Signature. 36 x 96-in. sheet. As this scrap comes out in a desirable shape, it is not necessary to trim it before entering it in the press which stamps out the collar. It may be, therefore, said that the first operation on collars is “Punching out Collars,” and this operation is designated as operation A-I. The next operation, A-2, is punching collars in a gang punch. The following operations are: A-3, turning edge on collar; A-4, cutting wire for collar; A-5, wiring collar; A-6, forming up collar; A-7, riveting collar. When the seven operations have been performed, the collar is ready to enter the construction of the drum. The handles are made from binding strips which hold the sheets of metal together when they are shipped from the sheet metal manufacturer. It takes only three operations to make the handle. They are: B-1, cutting the handles in square shears; B-2, punch- ing handles; B-3, forming handles. The first three operations of group C are the same for both tops and bottoms. Beginning with C-4 and ending with C-11. The operations apply to the tops only, for the bottom is complete after C-3. The group of operations is: C-1, cutting out tops and bottoms in square shears; C-2, cutting out tops and bottoms in circular shears; C-3, turning flange on tops and bottoms in press; C-4, cutting holes for collars; C-5, flanging holes for collars; C-6, enter collar in hole; C-7, punching holes around collar and flange; C-8, riveting in collar; C-9, punching tops for handles; C-10, riveting on handles; C-11, soldering around collars and rivets of handles. Thus we see that collars and handles enter into the construction of the drum by first becoming a part of the top. In like manner, group D is divided into eight opera- tions and group E into 17. There are 46 separate opera- tions in the manufacture of a drum and each of these operations receives careful study. Probably the first ques- tion to consider is, “Isn’t there a better method than the one pursued?” This question was asked on operation A-1, and was profitably answered. When time study began on drums, the collars were cut out on slitting shears. It was necessary to scrive around a pattern, and after the collar had passed through the shears a notch was cut out of one corner with snips. Now one stroke of the press takes the place of the three former operations. The question was asked concerning operation A-3, and now, instead of the slow process of knocking the edge over by hand, a machine does the work neater and faster. Operations C-4 and C-5 should be com- bined by having both per- formed at the same _ time, using, of course, a_ special punch and die. Fig. 2—Instruction Card for Workman. Flexible manila card, 4 x 9 in. CHARACTER OF WORK ........_.............-------- a stripper. ns Interesting methods were — adopted during the accumula- tion of the first set of stand- ards for this article. Opera- tion D-7, the forming of top and bottom bands is per- formed on 6 ft. power rolls. Because of the fact that the top and bottom bands are punched alike, that is, with the same setting of the gang punch, it is necessary that the top band be formed the reverse of the bottom band. This is done by laying two bands one on the other, having the burred sides turned out, and passing them through the rolls in this relative position. The method the operator chose was to pick up two bands, place them on each other, start them in the rolls, wait for them to come around, and then take them out and throw them aside. The observa- tion card showed that the time consumed in waiting for the bands to come through was in excess of the time re- quired to pick up the bands, to start them in the rolls, to take them out and throw them aside. This showed that there was time for the operator to start in two more bands while the first two were coming through. The new method caused the operator to have to work faster, and for that reason was detrimental. Working at this speed and having to remove the bands at exactly the right time increased the danger of the operator’s fingers being caught between the rolls, therefore the helper, who in the above two methods was employed in arranging the bands so as to be easily reached by operator, was called on to remove bands from the machine while the operator started them in. By this method the whole length of the rolls was utilized instead of only 2 in., as in method No. 2. The direct labor cost per band in the three methods was: First method, $0.00093; second method, $0.00057; third method, $0.000328. Special attention is paid to the placing of material to be worked at a position which the operator can most easily and speedily reach. The disposition of material after an operation also receives consideration. The use of flat and box trucks has in most cases given satisfaction because it elevates the material so that the operator does not have to stoop, and it affords a means of transportation from one operator to the next. In a shop where 1000 drums were being constructed at one time, one man was sufficient to keep the material moving from operator to operator. Together with the consideration of the best method of performing an operation should be the consideration of the advisability of purchasing more efficient machines and the changing of the old machines to increase the output. Operation E-16 is soldering drums. At first, ordinary sol- dering irons were used, while the drum rested on a flat | o wn Having made such changes, the next step is to determine what elementary movements are unproductive and can be eliminated. This is where the observation of type C becomes useful. An excellent punch and die had been installed to punch out collars, and appar- ently everything ran satisfac- torily, but observation showed an occasional elementary movement of “knocking off scrap.” There was no stripper Fig. 3—Original Type of Shop Observation Card. White, 4 x 9 in. April 4, 1912 bench, Now a much heavier iron is used having a curved nose to correspond with the curvature of the flange around the top and bottom where the solder is applied. At a small cost adjustable stands were constructed from iron pipe to carry the drum while being soldered. The drum allows the stand to be revolved while the solder is applied and also allows it to be swung end for end, so that either end of drum may be presented to the operator in its best work- ing position. Because of this equipment the solderers have been able to create a standard output of 131/5 per hour per man and to maintain an efficiency of between 90 and 95 per cent. Operation C-2 is the cutting out of tops and bottoms on circular shears. 6.18 r.p.m. Formerly the speed of the cutters was It took an average of 23.01 sec. actual cutting OBSERVATION CARD. a: ee QUAPARTIR. OO MODINE on. .snssceecnscomnscsnnowetersesssetoneresesentansecscssesnransmmnesnennieheenionsensassnes z2eBereaec¢+crto.s C Fig. 4—New Observation Card. White, § x 8 in. time for the cutters to cut out one disk. The efficiency of the cutters was then 93.8 per cent. The speed of the cutters was increased to 11.78 r.p.m., and as a result it takes an average of 13.87 sec. actual cutting time to cut out one disk, and the efficiency of the cutters is advanced to 99.08 per cent. Thus the coefficient of slip between the cutters and metal was decreased 85.3 per cent., while the cutting time was decreased 39.8 per cent., and the standard output per hour increased from 96 to 160. As the only element of the operation affected is that of cutting time, during which time the operator was idle, the change does not work a hardship on the operator. All his elementary movements consume the same length of time as before, but his output has been greatly increased. The last consideration is that of the arrangement of machine tools about the shop. Some claim that in a shop doing miscellaneous repair work together with manufac- turing articles, such as herein described, the machines should be arranged to suit the manufacturing industry, while others claim that machines performing similar oper- ations should be grouped together. Probably the greatest factor to decide this is investigation along this line. “What percentage of all the work done is straight manufactur- ing?” and “Does the manufacturer of one article or one set of similar articles constitute the bulk of the work done by the shop?” If the manufacture of one article or set of similar articles is performed on a large enough scale to receive consideration in the arrangement of machine tools, unless there is great objection due to the manufacture of various other articles, the machine tools used in manufacturing the popular article should be so arrariged that the operations will proceed in a general direction through the shop. If there is doubt as to whether this condition holds in the manufacture of an article, it would be well to get a sketch of the shop showing the arrangement of machine tools. Draw on this sketch a line from the door where the raw material enters the shop to the location of its stowage. Draw a second line from the stowage to the machine on which the first operation is performed, and continue to draw straight lines from operation to operation until the exit door of the finished product is reached. This sketch may now show long lines intersecting and running the whole length or half the length of the shop. Each long line not only means an increased cost of transportation, THE IRON AGE 837 but it also means a congesting of work in the shop due to trucks having to stop along their passage in order to clear away materials for passage room. It would be well to get such sketches for various articles manufactured in the shops and thus use good judgment in the arrangement of the machine tools. Almost invariably the necessity of two conditions will be shown; first, all raw material should enter at one door; second, the stowage of the raw material should be near that door. After all operations have been studied and a standard set on each, the information gained from the observations should be compiled in a book form. The character of in- formation given in the book is indicated by the tabulation on page 835. A copy of the book is placed in the hands of the foreman and disposition made of other copies as OBSERVATION OE cases -cseseheoggsarisscccdesomasianes J9m8 OPERATION STANDARD OUTPUT PER HOUR ee nenennennnnnnnamenn anes A ctaddacedcceenenbsdecslignenianateninin th __ ASSISTANT z Q: e <: 3 es Fig. 5—Reverse of New Card seen fit. The observation cards are retained for refer- ence in case of doubt as to the correctness of the book. On the flyleaf of the book issued on paint drum manufac- ture is printed the following: Instructions for Manufacturing 10-Gallon Paint Drums Note; Each operation is given a number. This number should be written on the instruction card by the leading men —also the operation. Machine No. indicates the proper machine to use. Pay per diem indicates the maximum pay the operator should receive for the various operations, Standard output per hour is the number of operations the workman should perform per hour, adingmen will see that material is kept moving and not allow machines to become crowded. One or two helpers should be detailed to keep operators supplied with material and to remove same from operators, way when completed. The instrtictions contained herein will be followed in detail. Instruction cards, Fig. 2, are then issued to the men employed on the work. The card when issued has the | man’s name, check number, rate, pay, assistant’s name (if there is one), operation number, operation, standard output per hour, and instructions filled in by the leadingmen. The operator fills in under character of work performed, starting time, stopping time, and units performed. In event of a delay during his work between starting and stopping time, he will make an entry of the delay. Along the end of the card is a column of hours. Whenever a leading- man inspects the work of an operator, he punches the oper- ator’s card at the hour of such inspection. The punches of the leadingmen are different, so that the leadingman making the inspection is shown by the shape of the hole punched. At the close of each day the instruction cards are dropped in a box by the men as they pass out of the shop. These records are gathered and an efficiency is entered on each, the delays in each case being deducted from the total time before the efficiency is calculated. A slide rule is used to compute efficiencies. These efficiencies are then entered on a man’s efficiency record form on which the efficiency for each day during a month is kept. An average efficiency is determined every two weeks. In event a man’s efficiency is exceptionally high or exceptionally low, he is so notified on a form signe:l by the foreman, which emphasizes that a continuance of such records will result in advancement or reduction; as the case may be. These efficiency records are valuable when it is necessary to reduce the force or when co»- sidering raising or reducing a man’s pay. 838 A low efficiency is a signal to the foreman to investigate. A low efficiency on “Drilling around tops and bottoms” was investigated and it was found that the operator had only one twist drill. When that drill became dull it was taken out, and, because there were no facilities at that time in the shop for grinding drills, he took the drill over to a friend in a neighboring shop about 300 yd. away, and this friend, without authority, quit his work and ground the drill. The method of manufacturing an article is in a con- tinual state of evolution. Competition demands that no effort be spared in reducing cost. A standard set to-day may be out of date a month or a week hence, because of a new method, a-new machine, or a change in the design of the article. At this writing there are three changes being considered, viz.: 1—Omission of the center band; 2 —Change in the design of the collar and a change in the method of fastening it to the top; 3—Substitution of spot welding (by electricity) for riveting. As changes are made in the equipment for doing work in a shop so that the work will be done better and quicker, so are changes made in the equipment of the time study man. Fig. 3 shows the original type of time card. It was designed to meet the general need, but was especially adapted to the recording of observations in a machine shop. The use of the card soon proved it to be inadequate to general need. First it was possible to enter the time of only three complete operations, because columns 4 and 5 were used for “average” and “corrected average” respec- tively. In short observations no record is entered of the time required to set a machine, for, as a general rule in a manufacture of a type like the one here described, the operator does not set his machine. This is done by a higher-priced mechanic, and a separate time record is kept of this just as though it were a separate operation. The new observation card is shown in Fig. 4. After the standard output per hour has been established by a type C observation, the card is filed. The first division of the file is the shop; second division is the article; third division is the part and the fourth division is the operation number. A comparison of the two cards will show the numerous advantages of the new card over the old. In event the efficiency of a certain operation or the efficiency of an operation with materials in a different position other than the standard position is obtained, such information can be recorded under type A observations and the neces- sary remarks explaining this new method entered under “remarks.” In this way the card can be kept up to date. Should an operator find some more efficient method than the standard method, the high efficiency (probably over 100 per cent.) will be a signal again to set a standard using the new and more efficient method. In this way every operator performing that operation will be using the new and more efficient method. In the last ten months there have been approximately 20,000 drums of this type manufactured in the shop with which the writer is connected. The direct labor cost per drum for the first 1000 was $0.466, while the cost per drum of the last 1000 is about $0.31. At this rate there will be a saving of $3,750 per annum. This is the saving on only one article in a shop which pays its employees on a per diem basis and where the only incentive to increase the output is a desire for a high efficiency. It is believed that this cost can be still further reduced, even while paying on a per diem basis. In this study no reference has been made to the saving due to careful selecting of the sizes of sheets of galvanized iron nor to the economic use of material due to the arrangement of patterns on the sheet. Efforts are made to work no hardship on the operators. The standards are set by observing men working at a normal speed. The efficiencies of employees sometimes are in excess of 100 per cent. due to “speeding up.” When investigation shows that the high efficiency is due to speed- ing up, the standard is not changed. An efficiency of 60 per cent. on such operations as laying off work, punching bands or cutting out tops and bottoms, where the bulk of the time is consumed by manual and not machine work, is considered good. The reduction of direct labor cost is accomplished by adopting more efficient machines and more efficient methods. It is believed that the foregoing detailed explanation of the differential time study will be particularly valuable by way of suggestion to works managers. THE IRON April 4, 1912 AGE Combination Molding Machine A new jar ramming and squeezing and molding ma- chine has been designed by J. N. Battenfeld and is being placed on the market by the U. S. Molding Machine Com- pany, Cleveland, Ohio, which he has recently organized. The special feature of this machine is that, apart from the simplicity of its construction, there are no valves in the cylinder. The jarring cylinder at the bottom of the machine serves two purposes, since it acts as a cylinder for jarring and also for a guide for the large plunger for squeezing. The use of the small cylinder for jarring is said to result in a very marked reduction in the air con- sumption. Only two valves for controlling the supply of air are employed, these being the inlet valves for the jarring and the squeezing cylinders. There are two hose attachments, one for jarring and the other for ramming. A New Jar Ramming, Squeezing and Molding Machine Built by the U. S. Molding Machine Company, Cleveland, Ohio. In the accompanying engraving the machine is shown with the pattern plate and the pattern on the table. When the molder shovels sand into the flask the jarring cylinder is started, and by the time the flask is filled the ramming is practically completed and the flask is ready for the bottom board. At this time the bottom board is put on and the head is swung around and the mold is squeezed. If desired the ramming head can be set only % in. above the bottom board. The length of the squeezer stroke is 6 in. and that of the jarring stroke is regulated by the amount of air admitted. W. P. Snyder & Co., Shenango Furnace Company, Shenango Steamship Company and Shenango Steamship & Transportation Company announce the opening, March 15, of their Cleveland offices at 1612 Rockefeller Puilding. They are represented by Claude J. Peck. It cost the Government $1,931,208 more to build the battleship Florida at a Government yard than it cost to have the battleship Utah built by the New York Shipbuild- ing Company. The Utah complete cost the Government $7,768,118, and the Florida, including equipage, $8,527,325. April 4, 1912 A New Internal Grinding Machine A Brown & Sharpe Model Adapted for a Wide Range of Straight and Taper Work The No. 22 internal grinder which has been brought out by the Brown & Sharpe Mfg. Company, Providence, R. L, is particularly adapted for grinding holes in hardened steel gears, cutters, collets, bushings and similar work. It is designed to handle all varieties of work within its range that can be revolved, the swing being 18 in. Holes from 2¥, to 12 in. in diameter and 8 in. deep can be ground with the regular equipment, and with a special spindle holes less than 2% in. in diameter, but not exceeding 12 in. in depth, can be produced. Taper work can be done as easily and accurately as straight, and such work as the facing of recesses in automobile gears to obtain a bearing true with the hole can be done without removing the piece from the machine. The various parts are so designed and con- structed that great accuracy may be secured, and efficiency and convenience are conspicuous features. Front and rear views of the machine are given in Figs. 1 and 2 respectively. The work to be ground is fastened either in a chuck or on a face plate mounted upon the work spindle, which is driven from an overhead countershaft. The headstock carrying the work spindle is fastened to the table of the machine, which has an automatic reversing longitudinal movement similar to that of univer- sal and plain grinding machines. Bolted to the right end of the ma- chine bed is a heavy bridge of metal and upon this the grinding wheel stand and the slide are lo- cated. The grind- ing wheel spindle is driven by a web belt from an in- termediate shaft at the back of the wheel stand, this shaft in turn being belted to the over- head countershaft. Like the builder’s universal and plain grinders, the work speeds and the table feeds are completely separated. This arrangement, it is emphasized, constitutes an important factor in commercial grinding, since it is possible to obtain a correct table feed for any work speed. When it is desired to remove stock rapidly a slow speed and fast feed are available. Both work and table feeds are con- trolled by a single lever located on the front of the machine. The wheel spindle is of hardened tool steel and is inclosed in a heavy sleeve that extends practically its full length. The spindle and the sleeve together form an integral part which can be easily and quickly removed and another smaller one substituted. In the sleeve at the front end of the spindle adjacent to the wheel is a phosphor bronze box and at the pulley end an annular ball bearing is provided to take the pull of the belt and reduce friction. The different grinding wheels used are mounted upon a sleeve having a tapered hole which fits the end of the spindle. Interchangeable split pulleys on the intermediate shaft at the back of the wheelshaft provide for a wide range of independent spindle speeds. The regular equip- ment of the machine includes three pulleys, providing for speeds ranging from 3990 to 7300 r.p.m. To provide an even tension on the belt and take up the slack due to the use of pulleys of different diameters an idler pulley is employed. The wheel stand and the slide are both of heavy, rigid construction, an arrangement which enables them to resist the vibration that might produce inaccuracy in the work. The wheel stand swivels, permitting the spindle to be set at an angle when grinding the face of cutters or automobile gears. When a hole has been ground and it is desired to grind the face of the work before removing it from the machine this feature will also be found convenient. ‘The Fig. Two Views of the New No. 22 Internal Grinding Machine Built by the Brown & Sharpe Mfg. Company, Providence, ; 1—Front View THE IRON AGE 839 wheel stand slideways are long to insure correct alignment and are of ample proportions to resist vibration. A trans- verse movement of the slide is secured through a worm and worm wheel actuated by a hand wheel on the front of the machine. An automatic cross-feeding mechanism sim- ilar to that found on all the Brown & Sharpe universal and plain grinding machines can be furnished if desired. The headstock, which is of rigid design, is firmly fastened to the swivel table by bolts sliding in a T-slot. It swivels and can be set at any angle, the amount being indicated on the graduated dial on the base. The hollow spindle runs in bronze boxes having means to compensate for wear, and it can be locked in position when removing the face plate, chuck or an attachment. Over the pulley is shown the belt-tightening device, which consists of a small pulley mounted on a bracket that can be thrown forward or pushed back by manipulating a handle. When drawn forward tension is brought upon the belt, causing the spindle to revolve and drive the work, and when pushed back the spindle can be quickly brought to a stop or easily revolved by hand when centering a piece. A range of 12 work speeds varying from 40 to 340 r.p.m. is provided. The sliding table is of rigid construction with bearings scraped to fit the ways in the top of the bed, which are aligned with master straight edges. The table travel is automatic and entirely independent of the work and wheel speeds. It is controlled by adjustable dogs on the front of the table which operate against the reversing lever. These can be raised and the table run beyond the reversing point without dis- turbing their ad- justment, and this feature will be found convenient when withdrawing the work from the wheel. When the work is completed the trip dog near- est the wheel is raised and_ the table moves to the left until the power feed is au- tomatically disen- gaged. The table can then be moved farther away by manipulating the large hand wheel. After a new piece of work is put in, the table is moved up by the hand wheel until the work nearly reaches the wheel, at which time the power feed automatically re-engages. If desired, the machine can also be used for wet grind- ing, the soda water being pumped from a tank located in the bed and fed to the wheel through the hollow work spindle. This pump consists of a simple fan revolving in a case, and since it is immersed at all times, no packing is required. The tank is readily accessible for cleaning, and the hood or water guard shown on the machine covers the work. All the wearing surfaces are protected from grit, the ways of the bed are always covered and the oil holes are protected. All the screws, bolts and nuts requiring fre- quent adjustment are hardened. The hand wheels and levers for controlling the various movement of the machine are all placed at the front. In other details the construction of this machine differs but little from that of the company’s universal grinding machine. The swivel table, however, pivots on a stud located beneath the head- stock, making a long leverage from that point to the end and permitting very fine adjustments in setting to grind straight work and in bringing a taper to the exact angle after the headstock has been set as closely as possible by the graduations on its base. The equipment furnished with the machine is shown on the floor in Fig. 1, and an over- head countershaft which is not shown is also furnished. Six changes of table feed varying from 15 to 48 in. per minute are available for any work speed. Fig. 2—Rear View The furnace of the Pulaski Iron Company, Pulaski, Va., has been blown out for relining. It will be blown in again in about two weeks. Automatic Blast Furnace Charging Controller Eliminating Personal Equation in the Proper Manipulation of the Charging Bells and Rotary Distributor of the Burden BY DAVID BAKER The best devices for filling a blast furnace mechanically fail entirely if the operator does not maintain the proper sequence in the movement of the appliances for distribu- ting the stock on top of the furnace and in the operation of the bells. The following account of an actual experience with one of the best equipped and most modern stacks of East- ern Pennsylvania will illustrate. This furnace had been working irregularly for some time and the management 9 To bottom of Turnin Cylinder Fig. 3—Elevation of Control Valves Fig. 2—Side Elevation of Controller tion of the skip hoist and top. continued for some time, until one day a furnace manager BAKER MECHANICAL CONTROLLER FOR AUTOMATIC OPERATION OF BLAST FURNACE CHARGING APPARATUS was unable to locate the cause of the trouble, although they felt convinced it had some connection with the opera- This condition of affairs 840 from another state called at the plant and, in discussing furnace problems and practice, was told of the trouble this furnace was giving at that time. At noon the visiting manager refused luncheon, saying he would look about the plant until the superintendent returned. While walk- ing about the yard the visitor took a seat where he could observe the operation of the hoist and soon noticed that the skip operator lowered the small bell just before the skip dumped, instead of just after the load was discharged Fig. 4—Plan of the Controller Fig. 1—Front Elevation of the Controller from the skip. When the furnace manager returned he was told how the skip and top were being operated and he at once corrected the trouble by requiring the skip oper- ator to wait until the skip dumped before moving the lever April 4, 1912 to dump the bell. The instructions to this operator had been made correctly, but the man, in order to expedite matters, thought he was not doing anything wrong by low- ering the bell while the skip was hoisting. Within 24 hr. from the time the error in filling was corrected the fur- nace settled down to regular work. The story just related describes only one of the many mistakes that may be made willfully or ignorantly in the operation of the skip hoist. A common error is irregular lowering of the main bell. Occasionally a green operator will forget to lower the main bell until both hoppers are choked with stock, and he is made aware of his negligence by the rattling of stock falling on the cast house roof. Phere is another feature of the operation of the main bellathat has an important bearing on the distribution of the stock in the furnace, and that is the speed of opening thissbell when loaded with stock. If this bell is opened rapidly the stock has more impact against the wall, and there_is more sorting of the lumps and fines, the lumps rebounding farther than when the bell is opened slower, and hence more lumps are thrown away from the walls. When a mechanically filled furnace is equipped with a rotary distributor, some means is necessary to insure the proper sequence of operations of every part of the top, and where the movement of the distributor is depend- ent on the skip operator at the foot of the hoist some form of recorder is frequently installed that will give a record of the movement of the bells and distributor on a sealed chart that is changed every 24 hr. This would seem to furnish a perfect check on all the operations of the top mechanism, but that even this safeguard is not sufficient to secure uniform sequence of operation may be shown by the following experience at a large southern plant. At this plant a common form of rotary hopper dis- tributor was used, propelled by an electric motor, and of course it required more time to revolve this hopper when it was loaded with its ore charge than when running light. This fact the skip operator soon discovered, and found he could save time in filling the furnace by opening the small bell first and revolving the hopper afterwards empty, thus getting the records on the recording gauges all right and saving time in charging the furnace. His whole idea was to fill the furnace quickly and have more frequent periods for resting. This trick of the operator was not discovered until it had been in practice some months and not until disastrous effects had been produced in all of the furnaces. These two examples from actual practice only illustrate what happens frequently enough to cause severe losses in the operation of blast furnaces and great anxiety to the management. In order to reduce the opportunities of mistakes in fill- ing the blast furnace mechanically and to rob the operator of the temptation to deceive, the writer was convinced some time ago that these operations should be performed automatically, and, with that in view, designed the mechan- ical operator or controller to be described. With this controller the furnace manager fixes the cycle of opera- tions for the top mechanism, and the skip operator has simply to start the skip, attend to the sounding rod or stock indicator and the oiling; the movement of the bells and distributor are performed automatically. This con- troller was designed for and is used with furnaces equipped with single skip hoists and the Paker-Neumann rotary dis- tributors, but it is applicable to any other top distributor not operated by the movement of the skip. The controller is attached to the hoist engine. For this purpose the Otis Elevator Company equips its steam or electric hoists with its old form of automatic stop, where the scréw feeds a nut with a lug back and forward, de- pendent on which way the hoist is running. This nut and lug come in contact with a corresponding lug on a lever arm when the skip reaches the bottom. The Otis Company usually attaches this mechanism to the outside of the drum shaft. This lever arm just described is attached to the controller by means of a connecting rod, with ball and socket connection at each end. In Fig. 1 is shown a front elevation of the controller. In this a is the connection to the automatic attachment on the hoist engine. The drawing shows the position of the lever 24 attached to the automatic on engine when the skip is at the bottom of the hoist. Lever 24 operates, through pawl 37, the cam plate 22; the cam plate 22 operates lever THE IRON AGE 841 20, which is attached to lever 26, and operates lever 53, connected to valve 50, shown in Fig. 2, a side elevation of the controller. Lever 20 is controlled in its downward movement by the oil controlling cylinder 5. The cam plate 22 is loose on its shaft, but the lever 24 is keyed to the same shaft, which is connected to the three- way valve shown im the elevation of the valves, Fig. 3, and in the plan of controller, Fig. 4. The counterweight 34 is heavy enough to return lever 24 to its starting point as soon as the hoist engine has raised the skip out of the pit, and the automatic on the engine has released the connecting rod a. In the same way the counterweight attached to the lever 26 returns that lever and valve attached to starting point as fast as the oil in the cylinder 5 can pass through the by-pass ‘of that cylinder. Operation of Controller e Just as the skip car is coming to rest at the bottom of the skip pit the automatic attachment on the hoist engine pulls lever 24 in position, shown in Fig. 1, and the cam plate with it. This cam plate is laid out with as many notches between the cams for pawl 37 as there are to be skips of coke to the charge. Ordimarily the coke is hoisted in four skips and the ore and stone in four skips. It is customary to put the stone first in each ore skip, thus giving good distribution and keeping the skip clean in dumping. In Fig. 1 the cam plate is divided into four notches, so that after the four skips of coke are hoisted the cam raises lever 20 as the empty skip car passes into the pit to receive the ore and stone. Retracing our steps in this description a little, each time a skip of coke starts out of the pit, lever arm 24 is pulled back by’ the counterweight 34 to its starting point, thus operating the three-way valve at the right of Fig. 3 and opening the gas seal bell and lowering the distributor in position. Every time the skip returns to the pit empty arm 24 is pulled into the position shown in Fig. 1 and the cam plate is rotated one notch. This is done four times, giving a complete ring of coke on the bell when the fourth skip is dumped. At this point, when the coke is all on the main bell and the skip car has returned empty to the starting point, the three-way valve is thrown, the gas seal bell is closed, and just as the gas seal bell closes after four skips of coke have been dumped on the main bell the cam plate throws up the lever 20 and operates the four-way valve controlling the main bell to give a quick opening of the bell, but the closing is made slowly on account of the retarding oil cylinder 5. Underneath the cam plate will be noticed a brake oper- ated by a counterweight. This is to prevent the cam plate from working back when it is released by lever 24 and pawl 37 at the return of the empty skip to the pit. It will be noted that every time the skip returns to the pit steam is turned on the small bell cylinder, so that the top is sealed, and as the distributing plate of the Baker- Neumann top is attached to this gas seal bell it is rotated automatically as this bell closes, thus setting the distributor at the same time in the right position for receiving the stock for the next skip car that dumps. The cycle of bell and distributor operations is main- tained at all times the same, without regard to the honesty of the skip operator. The controller is now in use at sev- eral furnace plants and working very satisfactorily. It is protected by letters patent. The General Electric Company, Schenectady, N. Y., reports interesting sales of centrifugal air compressors. A 100 hp. set has been sold to the LaClede Gas Light Company, St. Louis, Mo., for blowing a water gas gener- ator. Other sets, varying from 10 to 50 hp. were sold for use in connection with oil-burning furnaces as fol- lows: Two to the Michigan Smelting & Refining Com- pany, Detroit, Mich. (repeat order) ; one to Warren Tool & Forge Company, Warren, Pa.; one to Vermont Car Company, Mt. Vernon, Ill. (repeat order); one to Cleve- land Metal Products Company, Cleveland, Ohio. For use with pneumatic ash conveyor systems, one set each to S. D. Warren Company, Cumberland Mills, N. C.; Gov- ernment Railways, New South Wales; Peoples Power Company, Moline, III. Turbo Compressors in Practical Service Commercial Promise of Turbo Blowers and Com- pressors—Efficiency of Different Installations and Exhaust Steam Turbine Drives Including the More than a decade ago Prof. A. Rateau, of Paris, first attracted the attention of the mechanical and metal- lurgical world with his practical experiments in the design of turbo compressors and blowers, and since that time both he and ethers have done much original work in this field. But not until within the past few years have their efforts been crowned with commercial success. At the present time there are in satisfactory service abroad (and to a limited extent in this country) several hundred such units, ranging from small blowers used for cupolas and shop compressors of moderate capacity to heavy blowers furnishing the blast for ore furnaces and compressors of the largest sizes for mine service. With it all very little information of a definite character has thus far been pub- lished on the subject. The advantages claimed for turbo compressors and blowers which have thus far been very well substantiated in service, may be summarized as follows: (1) Greatly decreased weight and smaller floor space occupied, as compared with piston machines. (2) Simpler and less expensive foundations, owing to the ab- sence of all shocks or knocking. (3) There are no ‘alves, reciprocating parts or rubbing sur- faces, and no opportumity of friction except at the bearings. The impellers revolve freely within the cylinder and the unit, as a whole, is not subject to the wear or deterioration of the piston type. (4) Very low consumption of oil, etc.; lubrication is required only at the bearings. (5) Attention reduced to a veritable minimum. (6) Flexibility of operation to suit the varying requirements of a pneumatic system or of blowing service. (7) Ready adaptability to automatic control in respect to con- stant volume or constant pressure of the delivered air with vary- ing pressure and varying volume, respectively. (8) Delivery of air to pneumatic service or blowing system in a steady stream of constant or varying volume, as required, in- stead of in puffs as with a pistom compressor. (9) Efficiency maintained within a few per cent. of its high full- load value over a wide ratge of output. (10) No drop in efficiency with lengthening service, and, if a turbine is used as the prime mover, no increase in steam consump- tion to still further lower the efficiency of the complete unit. (11) Lower cost of units for the larger capacities, as compared with reciprocating and less expensive buildings. (12) Can be driven by motors or by high pressure or exhaust steam turbine, in the last named instance affording an excellent means of utilizing exhaust from hoisting or rolling mill engines usually run non-condensing, as well as from other reciprocating units conveniently located. (13) Are adapted to all classes of service, as in mining work for the operation of pneumatic drills, cutters, etc., in blowing heat- ing furnaces and cupolas, blast furnaces and converters, or in shop service for supplying air to pneumatic tools. Efficiency of the Turbo Compressor Concerning the efficiency of turbo as compared with piston compressors there has been a great deal of contro- versy abroad. It is claimed for the former that efficiencies of 65 to 69 per cent. have been obtained, and ‘one test case mentioned in this article gave 67% to 68 per cent. but all of the figures thus far published appear to have been for very large units, both steam turbine and motor-driven, and there are very little reliable data. On the other hand the figures for piston compressors are almost equally scarce and, where the latter are driven by reciprocating engines, the engine losses have seldom been separated from the compressor loses. It would seem as though there were a good opportunity, at the present stage of development, to make accurate comparisons between piston compressors, engine driven, and turbo-compressors, steam turbine driven, by computations made in pounds of steam per iso- thermic air horse-power or per pound of free air delivered, and the efficiency calculated as thermo-dynamic efficiency or ratio of power needed to compress the air isothermic- ally to that theoretically obtainable from the steam in ex- panding from the initial to the final pressures. It is only in this way that a parallel can be drawn between the two types of compressor units and will also enable them to be compared, in turn, with electric motor driven compressors; for the figures can then be referred directly to power plant costs and the total investment in equipment. The method is advocated by V. Oswald Davis, and some inter- esting facts concerning means of arriving at compressor ®” efficiencies were given by him, in an article written for the A.E.G. Journal for November, published by the Allge- meine Elektricitats Gesellschaft, Berlin. Conceding that, in point of efficiency, the piston com- pressor has the advantage, precisely as the piston or plunger pump outranks the centrifugal, the numerous other advantages of turbo-compressors, as enumerated above, show that there are still strong arguments to be adduced for their use. Question of ‘Motor Drive for Turbo Compressors When it comes to the question of motor drive there are also differences of opinion, according to varying ex- perience. It appears, however, to have been pretty well demonstrated that, except for low pressures and small capacities or high pressures and large capacities, with fa- voring power conditions, electric drive is less efficient for either compressors or blowers than steam operation by means of a turbine direct coupled. The resultant economy is not only higher, but the regulation of the suction and discharge is much simpler and can be effected over a wider range. There is also the point in favor .of this combination of turbine and turbo-compressor that the effi- ciency of each is best developed under similar conditions. With motor drive the disadvantages are chiefly due to the fact that the unit ordinarily runs at constant speed, with regulation of the air supply by unloading or throttling. Such methods are never entirely effective. Considerably more favorable are the conditions with direct steam drive which, with a suitable governing device, enables a constant pressure to be maintained in the supply mains, within wid