THE GAS-ENGINE HANDBOOK, A MANUAL OF USEFUL INFORMATION FOR THE DESIGNER AND ENGINEER. BY E. W. ROBERTS, M. E., Author of The I. C. S. Textbooks on Gas, Gasoline and Oil Engines. FIRS T EDI TION. FI RS T THO US AND. CINCINNATI: THE GAS ENGINE PUBLISHING Co. GOODALL BUILDING. NOTEX It is the earnest desire of the publishers and of the author, that this work shall be a standard of reference for those interested in gas enginery. They will, therefore, be pleased to have their atten- tion called to any errors of commission or omission in this book. THE PUBLISHERS. COPYRIGHT, I9OO, BY THE GAS ENGINE PUBLISHING CO PREFACE. It was during the preparation of a series of textbooks on the gas engine for the International Correspondence Schools, that the author was most forcibly impressed with the dearth of matter upon American practice in this motive power. It is a recognized fact that designers on the other side of the Atlantic do not follow methods that meet w r ith the approval of engineers in the United States, yet the only truly valu- able works on gas-engine design that have made their appearance in the Bnglish lan- guage are by English authors. Unhappily, the average gas-engine manufacturer in this country, guards any information he may possess with the jealousy that is scarcely to be explained on ordinary grounds. While there are a number of good works on steam-engine design, the gas engine has been surprisingly neglected. Although the author does not, in the present work, hope to cover the entire subject of gas-engine design he has endeavored to 435986157 place upon its pages a sufficient number of rules and formulas, to enable any intelli- gent draftsman to design a gas engine without difficulty. Many of the formulas are new and have not appeared in any works by other authors. They have been derived for the most part from tables of dimensions of some of the most successful American-built engines, and have been used by the author in making up his own designs. It has been the endeavor, to place within the smallest possible compass a number of useful rules and hints, that may be of value not only to the designer, but also to the engineer who has the care of a gas engine. The chapter on testing has been given more attention than might perhaps have . been thought necessary in a book of this size, but it also covers many points regard- ing the calculation of horsepowers and other items purposely omitted from the other portions of the book. This portion of the work was founded upon the methods employed at the Cornell University. The author has drawn, in a few instances, from other works upon the same subject. In preparing the matter on design, he has received many useful hints from the works of Mr. Frederick Grover and Mr. William Norris, the two English writers already referred to. A part of the data in the table of heat values is from a similar table in the work of Mr. Gardner D. Hiscox. The mechanical tables are from various pocket- books and from the works already referred to. Much that the writer would have desired to include in the present work, it has been necessary to omit for lack of space. He has, therefore, given only such matter as he has judged most useful, and which he has found most frequently the subject of in- quiries in the question and answer columns of the technical papers. B. W. ROBERTS. Cincinnati, March, 1900. CONTENTS. CHAPTER I. PAGE Introductory : The principles of opera- tion of the cycles at present in use. . . I CHAPTER II. Comparison of the Three Cycles : The advantages and disadvantages of each. 13 CHAPTER III. Gas-Kngine Fuels : The principal fuels in use, with a table of heat values. ... 17 CHAPTER IV. Starting and Stopping: The order of procedure in each case 22 CHAPTER V. Care of an Kngine : How to keep a gas engine in good running order 27 CHAPTER VI. Gas-Engine Troubles: Where to look for the cause of trouble, and the rem- edies which apply 35 CHAPTKK VII. TMGE Gasoline Engines : How they differ from a gas engine. Special attachments re- quired for the use of gasoline 41 CHAPTER VIII. Handling a Gasoline Engine : Special rules for the care of a gasoline engine 49 CHAPTER IX. Igniters: Classification of the various devices and examples of the principal types 54 CHAPTER X. Valve Mechanisms : An explanation of the various systems in general use. ... 73 CHAPTER XI. Governors : A classification of the gov- ernor methods. The limitations of each method. Examples of the prin- cipal mechanisms 87 CHAPTER XII. Starters : Devices for starting large en- gines and how they are used 97 CHAPTER XIII. Engines for Automobiles : The chief requisites for engines of this class. The speeds at which they should run. How to avoid vibration. Regulation . 103 CHAPTER XIV. PAGE Gas-Engiiie Diagrams : How to lay out an ideal diagram. How to read a di- agram. Bxamples of good and bad diagrams 108 CHAPTER XV. Gas-Engine Dimensions : How to com- pute the cylinder diameter and the stroke of a gas engine of any size. How to find the speed at which the engine should run 121 CHAPTER XVI. The Cylinder : Rules and formulas for the design of a gas-engine cylinder. . . 129 CHAPTER XVII. Valves and Valve-Boxes : Examples of valve arrangement. Proportions of the various parts of a gas-engine valve. Formulas for valve- box design 136 CHAPTER XVIII. The Piston, the Connecting-Rod and the Crankshaft : Examples of each, and formulas for the proportioning of the various parts 146 CHAPTER XIX. The Engine Frame : An example of a frame for a horizontal engine.. 157 CHAPTKR XX. PAGE Flywheels : How to compute the weight of a flywheel for any gas engine and for any service 160 CHAPTER XXI. Balance-weights : Computations of the weight of a gas-engine counterbalance. Two methods that are in use 167 CHAPTER XXII. Foundations : Materials of which a foun- dation should be built. Formula for computing the weight of an engine foundation 171 CHAPTER XXIII. Miscellaneous Formulas : Formulas for the diameter of the camshaft, the vol- ume of the muffler, the horsepower of a marine engine, the diameter of a screw propeller 177 CHAPTER XXIV. Testing: A full description of all nec- essary operations for a gas-engine test for any purpose, with rules for work- ing up the data obtained 180 CHAPTER XXV. Selection : General rules for the selec- tion of a gas engine 203 TABLES. PAG* Heat values of various gases. Capacity of cylindrical vessels. Dimensions of gas and water pipe. . . 208 Areas and circumferences of circles. . . .209 Index 2I 5 CHAPTER I. INTRODUCTORY. The Gas Engine, properly so called, is a type of the internal-combustion engine. Its first conception seems to have been by one Abbe Hautefueille, who proposed a mo- tor driven by a gunpowder as early as 1678. From that time until 1862 many proposi- tions were made and some partially success- ful engines were built by various experi- menters. In the latter year, there were already in existence the gas engines of Ivenoir and Hugon, but very wasteful of fuel. In 1862 M. Beau de Rochas, a French en- gineer, made the following propositions, which he embodied in a patent of that date. They were as follows: I. The largest cylinder capacity with the smallest circumferential surface. II. Maximum speed of piston. III. Greatest possible expansion. IV. Highest pressure at the beginning of the stroke. These, he averred, should be the aim of 'every gas-engine designer to carry out in his engine. With the exception of propo- sition II, every engine, since the successful .advent of the Otto engine, has been de- signed with these principles in view. Ow- ing to certain practical limitations, high piston speeds have not been found advisa- ble. It was M. Beau de Rochas who first pat- ented what is popularly known as the " Otto Cycle." This cycle requires four strokes of the piston for its completion. The first stroke draws in the charge of gas -and air, the second stroke compresses the charge, on the third stroke ignition takes place at the dead point, followed by explo- sion of the gaseous mixture and the expan- sion of the products of combustion, and on the fourth stroke the burnt gases are ex- pelled from the cylinder. This series of operations is that employed by nearly all gas engines in use at the present time. This cycle was not put into a practical use in a working engine until, in the year 1876 Dr. Abel Otto brought out the first " Otto Cycle " engine, producing a motor which so far surpassed in economy all engines that superseded it as to drive its competitors from the field. This cycle of operation has, because of its being first put into practical shape by Dr. Otto, been improperly termed the " Otto Cycle," although the credit of its first inception belongs, without question, to M. Beau de Rochas. It is also known as a four-cycle engine, although this term, to be strictly correct, should be four-part cycle, because it consists of the four distinct oper- ations just described. Modern writers have been placing the credit where it belongs, and this series of operations is now generally termed the Beau de Rochas cycle. The au- thor, however, prefers the same descriptive term, " four-cycle," and as such it will be re- ferred to throughout this work. The practical operation of the four-cycle engine can be most clearly understood by reference to Fig. I, in which a diagramatic form of the engine is shown with the indi- cator diagram given by the engine placed directly above the cylinder. Since it re- quires two revolutions of the crankshaft K to complete the cycle, the principal points of the cycle are shown on the two circles drawn about the crankshaft as a center. Similar points upon the diagram, the piston travel and the rotation of the crankpin P, are indicated by the same letters. The let- ters indicating points on the piston travel are primed, and those indicating similar points on the crank circle have a double prime. Thus, the point of ignition is indi- cated on the diagram by B x on the line of piston travel, and by B x/ on the crank circle. Taking the series of operations in their regular order, they are as follows : Suction of air from a to A on the outward stroke, compression from A to C on the in- ward stroke, ignition taking place at B, just before the end of the stroke is reached, thus giving a " lead " to the ignition which should just be sufficient to bring the point of maximum pressure D, at the beginning of the next outward stroke ; this lead being necessary because of a lag in the process of ignition, which takes the time required by the piston going from B to C to fully in- flame the charge of gas and air. From D to F the piston is passing 011 a second out- ward stroke, the exhaust-valve X opening at K, just in time to allow the expansion line D F to reach atmospheric pressure at the end of the second outward, or expansion, stroke. The return stroke takes place from F to a, the piston making a second return stroke, driving the products of combustion through the exhaust-pipe by way of the valve X, . completing the cycle. A modification of the four-cycle engine was brought out by its inventor, Mr. Du- gald Clerk, in the year 1880. In this engine, using an auxiliary pump, a power stroke was obtained for each revolution of the crankshaft. The charge was admitted to the engine cylinder under pressure and just at the end of the expansion stroke, driving the exhaust gases out through ports in the sides of the cylinder, which were uncovered by the piston just at the end of the stroke. This arrangement eliminated both the suc- tion and the exhaust strokes required in the regular Beaii de Rochas cycle. Clerk's engine was the pioneer of what is now known as the two-cycle type ; that is, an en- gine which completes its cycle in two revo- lutions of the crankshaft. Following the introduction of the Clerk engine came a class of engines which, in- stead of having an auxiliary cylinder to compress the charge, used for that purpose an enclosed crank-chamber. The inward stroke of the piston draws into the crank- chamber the charge of gas and air through a check-valve, and on the outward stroke the charge is compressed. Just before the piston reaches the end of its stroke its end passes an exhaust-port in the side of the cylinder, and in some engines a port lead- ing to the crank-chamber is passed imme- diately afterward. In another type, communication was made by means of a poppet valve. The pioneers of this type were the Nash and the Day engines. The former used a mechanically opened poppet valve, and the latter an ad- mission port in the cylinder wall. Quite a number of engines of the Day type are in successful operation at the present time. The principle of the operation of the Day type of the two-cycle engine is shown in Fig. 2, in a similar manner to that of the four-cycle engine shown in Fig. I ; the same method of notation as regards letters being used in this figure, the only difference be- ing that where the two crank circles are used, the inner one indicates the series of operations taking place in the crank-cham- ber, and the outer indicates the series of operations taking place in the cylinder. The piston makes an inward stroke from b to a, causing a partial vacuum in the crank- chamber and drawing the explosive mix- ture through the valve P. On the following outward stroke the mixture in the crank- chamber is compressed to a pressure of about five pounds to the square inch. As the piston uncovers the inlet port S, the pressure within the crank-chamber drives a portion of its contents into the cylinder C, and on the next inward stroke of the piston this charge is compressed, as shown on the diagram from A to C, ignition taking place as in the four-cycle at B, pressure rising to D, the piston making outward stroke, exhaust- ing at E, and the gases reaching atmospheric pressure again at A, the exhaust-port being opened by being. uncovered by the piston. In the meantime a charge is being drawn into the crank-chamber as before, has been compressed, and the inlet-port S being un- covered immediately after, the contents of the crank-chamber rushing through S is deflected by means of the plate R on the piston to the top of the cylinder, effectu- ally driving the products of combustion through the port M. Returning to the se- ries of operations in the crank-chamber, the port S opens just as the piston reaches the point E, the pressure immediately drop- ping and reaching the pressure of the at- mosphere at about the time port S is closed by the return of the piston. It is unneces- sary to repeat the explanation of the points on the diagram taken from the cylinder C, as they are in all respects similar to the four-cycle engine illustrated in the Fig. i. The Clerk engine is no longer built. Nearly all two-cycle engines being manufac- tured at the present time are built after the Day pattern. In the latter part of 1897, a new cycle was brought to the attention of the public. This cycle comprises the se- ries of operations which were the inven- tion of Herr Rudolph Diesel, a German scientist. The Diesel cycle is, like the invention of Beau de Rochas r a four-part cycle, requiring two entire revolutions of the crankshaft to complete the series of opera- tions. The first, or outward, stroke draws into the cylinder a charge of pure air. On the following return stroke the charge is compressed into a space at the end of the cylinder, equal to about 7 percent of the entire cylinder capacity, the pressure at the end of the stroke being approximately 550 pounds per square inch. This high com- pression produces a temperature of the air equal to that produced by the combustion of the fuel, and, in consequence, fuel ad- mitted at the end of this stroke is sure to be ignited as it enters the cylinder. The fuel, whether it be gas, oil or other com- bustible material, is forced into the cylin- der at a pressure higher than that produced by compression, and just as the piston is about to start on its outward stroke. The fuel, burning as it enters, keeps the temper- ature of the cylinder contents up to that produced by compression. After the pis- ton has reached a point of the stroke representing 10 percent of the whole, the fuel is cut off and the products of combus- tion expand. The ensuing return stroke drives the exhaust gases from the cylinder. It is readily seen that, owing to the small compression space, the exhaustion is very much nearer being complete than in those engines using the cycle of Beau de Rochas, and that but a very small quantity of the products of combustion remain to mix with the fresh charge This series of operations may be better understood by reference to the Fig. 3. The same manner of representation is employed as in Fig. i. Suction of a charge of pure air takes places from a to A on the outward stroke, the following inward stroke compressing this charge until the point B is reached, when the fuel is admitted while the piston is passing from B to C and is cut off. The temperature of the gases is con- stant during admission of the fuel, the line from B to C being one of equal tempera- ture known as an isothermal. From C to E, expansion proceeds in a manner similar to that in the original four-cycle engine, the exhaust-valve opening at D and the pres- sure inside the cylinder falling to atmos- pheric at B. The range of temperature from C to D is much greater than that from D to F, in Fig. I, and represents a much higher heat efficiency than that shown on the diagram of the four-cycle engine. CHAPTER II. COMPARISON OF THE THREE CYCLES. The four-cycle engine is that most gener- ally manufactured, and it has the advantage over the two-cycle of being more readily controlled. With a properly designed four- cycle engine, the behavior of the gases within the cylinder is known beforehand. The idle stroke gives the cylinder a chance to cool by radiation, no pumps nor enclosed crank-chambers are necessary, and any waste of fuel may be easily remedied. On the other hand, the four-cycle engine must be built in large sizes when compared to the power de- manded of them, the many idle strokes ne- cessitate extremely heavy flywheels and make close regulation most difficult to ob- tain. Furthermore, the operation of the valves occurs but once during two revolu- tions of the crankshaft and necessitates some form of reducing motion between the crankshaft and the camshaft which operates the valves. The two-cycle engine is, as a rule, more wasteful of fuel than the four-cycle engine, and such wastes are more difficult to reme- dy in this type In many of these engines much trouble is experienced with prema- ture explosions, usually called " back- firing," and in the Day type of two-cycle engine the charge quite frequently explodes in the crank-chamber. The bulk of these troubles is fortunately confined to engines built by designers who have an imperfect knowledge of the type, and who fail to de- sign their engines with the proper propor- tions. Time should always be given the exhaust gases to fall below the pressure of the crank-chamber before the inlet-port to the cylinder is uncovered. It is also important, when running a two-cycle engine, to re- member that much more heat is being given off through the cylinder walls than in a four-cycle engine of the same power dur- ing the same period, hence the supply of jacket water must be greater in consequence. A two-cycle engine will shut down more quickly from lack of water than will a four- cycle. Also, because of the rapid succession of explosions in the cylinder of a two-cycle engine, greater pains should be taken to avoid projections into the cylinder which are not so placed that they may be kept at a comparatively low temperature. M These engines are being better under- stood as the makers gather knowledge from experience, and the above objections have, in several cases, been entirely overcome. So many and manifest are the advantages of more frequent impulses and less weight, both in the engine itself and in the fly- wheels, that the adoption of the two-cycle type is being seriously considered by a number of manufacturers. In fact, the two- cycle engine, for large power-units where the fuel employed is blast-furnace gas, has been in use for some time and found to give satisfaction. Instead of having an enclosed crank-chamber, these engines are supplied with a pump which first drives a volume of pure air through the cylinder, clearing it entirely of the products of combustion re- maining from the previous charge. The fresh mixture follows the charge of pure air, and none of the fuel is wasted, while the clean cylinder increases both the efficiency and the capacity of the engine. The Diesel motor is not at present upon the market in sufficient numbers, nor for a length of time, that would make it feasible to draw conclusions of any kind with regard to the future of the engine. The engine is a grand conception theoretically, and the heat efficiency is much higher than that of any heat engine in use. The scheme of regulation is similar to that which has proven so successful in high-speed steam engines, and it has been shown by actual tests that, in practice, the engine surpasses any other in economy of fuel, and, besides, it is capable of being run with such inex- pensive fuels as the low-grade petroleums classed as fuel oils and selling in quantities for from two to three cents per gallon. The question therefore remains: will the great economy of the motor jtistify the added ex- pense for first cost of the engine? Again, there is the question of repairs and the durability of the engine under the high pressures which it is necessary to sustain in the cylinder. All this remains co be seen. 16 CHAPTER III. GAS-ENGINE FUELS. For driving a gas engine, the fuels avail- able are all of those which are in use for other purposes. Many of these fuels need to be transformed into gas before they are available for the gas engine, but many of the liquid and all of the gaseous fuels may be employed directly in the cylinder with- out the necessity of an intermediate proc- ess. There is no fuel not in a gaseous state already that may not be transformed into gas by one of the many processes known to science. The garbage gathered from the streets of our large cities, the waste fat from the slaughter-house, the fat abstracted from wool, as well as all of the animal, vegetable and mineral oils, have been at one time or another transformed into gas which could be used to drive a gas engine. The amount of power derived from a cer- tain quantity of fuel is always greater by 3 large amount when the fuel is first made into gas and then used to drive a gas en- gine, than when the fuel is consumed in the firebox of a steam boiler and the power de- rived from a steam engine. Bven with coal as a fuel, if the coal be made into gas and the gas used in a gas engine, it is found that where the very best steam engines give but a horsepower for one hour on i*^ Ibs. of coal, the same amount of fuel has ^iven 1.8 horsepower for one hour. The gas-producer which is required for the use of coal in connection with the gas engine, is a much simpler device than the steam boiler, and requires less attention than the latter. There is very little danger of explosion with a gas-producer and gas- engine power plant, and there is usually a supply of gas on hand in the gas-reservoir sufficient to permit the engines to be started at a moment's notice without the necessity of waiting, as when getting up steam for a steam engine. Fuels which are most difficult to burn under a boiler, may be turned into gas in a gas-producer. Even gases which are diffi- cult to ignite may be used in a gas engine. For instance, the gas produced during the process of extracting iron from ore in the blast furnace is at times so poor in quality that it may not be used under a boiler, while 18 when compressed in a gas-engine cylinder it is found to ignite readily. The value of fuel for use in a gas engine, is determined in a great measure by the number of heat units produced when a quantity of the fuel is burned. There are, however, several other considerations, such as the pressure derived from the fuel and the percent that may be burned in the gas- engine cylinder. In the process of making gas from fuels a percentage of the heat is employed in liberating the gas, and this, of course, can not be given account of in the gas engine. The heat values of the various fuels are given in Table i. It will be seen that the highest heat value is found in natural gas. This is due to the large quantity of marsh gas contained in natural gas. The value of natural gas as a fuel varies, even when the gas is taken from wells very close to one another. It may be said, however, that the natural gas obtained from the wells in west- ern Pennsylvania exceeds in heat value that of the gas in the Ohio fields from Io% to Gasoline, in spite of the low heat value per cubic foot of vapor shown in the table, will give, for the same size engine, a power equal to that derived from the best natural gas, and the usual custom is to credit gaso- line with giving 10% more power than an average quality of natural gas. It is not possible to compute the horsepower to be derived from a certain engine by compari- son of the fuel values, since the gases of low heat value require a smaller quantity of air for their complete combustion, and hence a larger quantity of gas may be taken into the cylinder of the engine at each .stroke. HEAT VALUES OF FUELS. FUEL. B. T. U. per Ib. B. T. U. per cu. ft. Hydrogen (n 32 F. . . . Carbon . ... 62,030 14 500 343 Carbon monoxide (CO). Penn. heavy crude oil. . . Caucasian heavy crude oil, Caucasian light crude oil, Petroleum refuse . . . \nthracite gas 4,396 20,736 20,138 22,027 19,832 2 2A.8 539 Bituminous 2fas 1 d8/l 28-candlepower ilium, gas, 19- J5- New York city water-gas*, London coal gas. . . . 1 8 448 950 800 620 710.5 Ave 668 Gasoline and its vapor . . Ethylene C 2 H 4 Marsh gas (Methane) C H 4 Nat. gas, Leechburg, Pa., Nat gas, Pittsburg, Pa . . 11,000 21,430 23,594 690 i,677 1,051 1,051 802 Acetylene C 2 H 2 21,492 868 iSs I ^O NOTE. The values shown in the above table are given on what is deemed good authority, but they will not be found to agree with all similar tables. * Carbureted gas at 60 F. and at 30" water pressure. 21 CHAPTER IV. /! STARTING AND STOPPING. Starting a gas engine is a simple oper- ation if a few easily-remembered rules are borne in mind. These rules are briefly as follows : A gas engine will not start itself, like a steam engine, and must receive sufficient turning power from an outside source to enable it to take up its cycle of operations. The mixture of fuel and air must not be too rich in fuel nor too poor in fuel, for in either case an explosive impulse can not be obtained. More frequently, trouble oc- curs from the mixture being too rich in fuel. On large engines the compression should be relieved and the ignition given a nega- tive lead, especially when starting by hand. Never attempt to start a gas engine when it is connected to a load, unless it has a powerful starter. The ignition apparatus must be in good working order, and this must be seen to before attempting to start the engine. To start a gas engine, it is always a good idea to have a regular order of procedure and to stick to this order so as to avoid the chance of confusion and of omitting some- thing. A good way to proceed is as follows : First, oil the engine thoroughly, filling every oil-cup, no matter if one or more is already nearly full. See that the ignition appliance is in good order. If it is a tube igniter, bring the tube to the proper temperature (usually a cherry red) and see that the burner is at the right height to keep the tube at the required heat. If an electric igniter is used, unfasten the wire from the insulated electrode and brush it on some part of the engine to see that it gives a good flash. If there is a sight plug, it would be well to try the igniter while observing its behav- ior through the hole, or it can be tried by first pressing the points together by means of the igniter mechanism and seeing if there is a circuit by drawing the wire al- ready removed over the end of the insulated electrode, then letting go of the mechanism and determining in the same manner as before if the circuit is broken. If these conditions are fulfilled, and when the engine is turned over it is found to 23 close the circuit on the igniter-points, the igniter is in working order. The igniter should be examined periodic- ally and the points cleaned of any accumu- lation of soot or other foreign matter which may occur. The frequency of this exami- nation depends entirely upon the fuel used, and to some extent upon the class of cylin- der oil employed. After determining that the igniter is in working order, such oil-cups as can not be conveniently reached when the engine is running should be turned on. Next set the starting-cam to start, or open the relief-cock, as the case may be, and if the engine is fitted with a changeable ig- niter, set this also to the starting position, or so that it will fire after the crank has passed the center. If the engine is to be started by hand, set the gas-valve open to about one third the opening used when running; usually at a point on the gas-valve dial marked " Start." Then turn the engine over in the running direction until it takes up the cycle, j. e., until an explosion is obtained. Next take hold of the gas-valve handle immediately, and open it gradually to the running position as the engine increases in speed. Don't be too precipitate about open- ing the gas-valve, as the engine may get 24 too rich a mixture and, failing to ignite, slow down and stop. If it shows signs of stopping, close the valve a little at a time until it starts receiving impulses again. While opening the gas-valve with one hand, the other may be occupied in closing the relief-cock, or in throwing the starting- levers to the running position. When the engine has reached its full speed, the load may be thrown on, and then the remaining oil-cups must be opened and the water turned on to the water-jacket. After the engine has been running a short time, and if the load is very nearly a constant one, the water should be turned on or off until the exit water is of a tem- perature that can just be borne comfortably by the hand. If the engine is fitted with a starter, the gas-valve should not be turned on until the engine has made one or two revolutions. The methods of using the starter vary with the several types, and their use will be explained in a chapter on that subject. To simply stop a gas engine, it is neces- sary to do no more than turn off the fuel. The order for turning off the oil-cups, etc., should be about the reverse of starting. First turn off the gas, and if it is desired to save time in stopping, brake the flywheel with a plank. Then turn off the jacket water and turn off the gas from the burner, or throw the igniter-switch. If there is 110 switch, disconnect one of the wires and hang the end of it up so that a short circuit may not occur. Turn off the oil-cups and drain the water-jacket. This latter proce- dure is not absolutely necessary except in cold weather, but it is not a bad way to prevent the accumulation of sediment in the jacket, and it takes but little time. The habit of draining the cylinder may save the engine at some time when you " didn't know it was going to freeze." 26 CHAPTER V. CARE OF AN ENGINE. The proper care of a gas engine should be the pride of every engineer who runs one. In every shop or factory where a gas engine is in use, it should be left in the care of one man and that man should be respon- sible for the condition of the engine at all times. " Everybody's dog is nobody's dog," and so it is with an engine which is left in the care of any man who happens to be about at the time it needs attention. Above all things keep the engine clean, well oiled and with a plentiful supply of water in the tank, when a tank is used, or with sufficient flow of water when connected to a pressure system. If the engine is a new one when first re- ceived, it is well to determine the exact point of the stroke where the ignition takes place. This can be done very easily, with an electric igniter, by turning the engine over until the igniter snaps for breaking the circuit, and marking the piston so as to 27 have a point of reference for future settings of the igniter. All igniter mechanisms wear more or less, and the mark on the pis- ton will avoid the necessity of taking an in- dicator diagram to set the igniter. If the piston is so situated as to make it inconvenient to make a mark upon it, the mark may be placed upon the flywheel. To mark the flywheel for this purpose, proceed in one of the following ways : Turn the en- gine over slowly until the igniter just snaps. If the engine has been turned too far it will be necessary to make a complete revolution again in order to be certain that all lost motion is taken up. After the flywheel is in the right position, drop a .plumb-line past the center of the crankshaft and mark on the rim of the flywheel opposite where the line passes it, or level up a straight-edge so that it passes the center of the crankshaft and mark the flywheel as with the plumb- line. It will be more convenient even than the above methods, if the flywheel passes close to some part of the engine-frame or bed, to mark both the wheel and some point on a stationary part of the engine. A good grade of machine oil should be used on the engine bearings, but in the cylinder there should always be used an oil that is made expressly for this purpose. Never use the heavy cylinder oils that are 28 sold for steam-engine cylinders, or you will soon find that the passages are becoming choked with carbon. The proper oil for use in a gas-engine cylinder is a thin lubricant which will not carbonize under the high temperatures present in the gas-engine cylinder. Don't think that because steam- engine cylinder oil costs four times as much as gas-engine cylinder oil, it is four times as good. The direct opposite is true. For an engine with an inclosed crank-case an- other special oil is required and is known to the trade as " crank-case oil." Ordinary oils will churn into lumps in the presence of water and soon become practically use- less as a lubricant. It is not a bad practice to make an occa- sional examination of the exhaust-passages in order to determine whether they have an accumulation of carbon in the form of soot. With some gases this deposit does not amount to a great deal, while with others the deposit is such as to in time cause sufficient back pressure to materially de- crease the power of the engine. Should f the water used for cooling purposes contain substances that are likely to cause a deposit of sediment in the water-jacket, the jacket should be cleaned occasionally by means of a long iron rod with a hook on the end, similar to a poker. 29 Examine all valves occasionally, paying particular attention to the exhaust-valve to see if it is cutting. If they show the least sign of leaking they should be at- tended to at once and ground to a good fit to the seat with flour emery. Never neg- lect a leaky valve, for, once started, the leak will increase rapidly in size. Examine the springs frequently to see that they pull the valves securely to their seats. This can be determined by taking hold of the valve- stem and pulling it back from its seat. The resistance to the pull should be a stiff one, even at the start, for all valves excepting a suction inlet. In the case of an inlet-valve operated by the suction of the piston, it should respond promptly to the vacuum in the cylinder, a very small movement of crank past the center causing it to open. It should be carefully adjusted so that it will seat firmly and yet respond promptly to the vacuum. The gas-valve should also receive its share of attention. With engines using electric ignition, the care of the igniter mechanism is one of the most important points. Never neglect the igniter. Much of the prejudice against the electric igniter is the result of improper care of the ignition apparatus. In case a primary cell is used for the battery, a full set of renewals should be kept on hand at 3 all times and the battery should be renewed before it is too weak. In case an ignition-tube is employed, it pays at all times to get the best that can be obtained. Nickel alloy is much the best material to employ for this purpose, as a tube made from the alloy will last, with ordinary care, from six to eighteen months, according to the work the engine has to perform. Wrought-iron tubes have to be replaced anywhere from every two days or so to every two weeks. The shut-down nec- essary while replacing a tube is annoying, to say the least, and it may cause expensive delays. A new tube should be kept in stock at all times, and when tubes that last but a short time are used, it would be well to keep several on hand. Do not keep the tube too hot at any time, as a high temperature reduces its strength and makes it wear out so much sooner than it otherwise would. The proper tempera- ture to keep the tube is at a bright cherry, not a white heat. The temperature of the tube for ignition, will vary somewhat with gas of different qualities and the engineer should determine for himself the lowest temperature at which the tube will work successfully. If at any time the engineer has forgotten to turn on the jacket water and the engine begins to throw out volumes of smoke through the open end of the cylinder, turn on the jacket water, but do so with great caution, as, if the cylinder walls are cooled too rapidly, the cylinder may contract be- fore the piston has time to do so, and it may bind the piston so as to materially damage the engine. The cylinder oil-cup ^should also be opened wide at the same time, to give the cylinder an excess of oil and ward off the possibility of its binding. Feel of the bearings occasionally, and keep watch of the sight feed of the oil-cups, so that no injury may be done to the engine by cutting of the journals. Don't attempt to cool an overheated bearing with ice un- less you thoroughly understand what you are about. You may get an effect the oppo- site of that desired, and find that the sudden cooling of the outside of the bearing has caused it to grip the shaft like a vise. If the ice can be applied to the shaft itself so as to cause the shaft to contract before the surrounding metal, all trouble may be averted and the desired effect obtained. Examine the governor occasionally to see that it is working freely and is not clogged by dirt or any foreign matter. The regularity of motion is dependent as much upon the sensitiveness of the governing device as upon the flywheel. If the governor is a hit- 32 aiid-niiss, see that the gas-valve is open just far enough to cause an explosion to follow immediately upon the action of the gov- ernor. Nearly all governors of this type use a pick-blade, and if the gas is not turned on sufficiently, the blade will strike or let go according to its method of operation some time before the engine will receive an impulse, and the consequence is a slow- ing down of the engine until the mixture becomes rich enough to produce an ex- plosion. If there is a pet cock on the engine having communication with the compression space, the proper state of the mixture may be de- termined by opening the pet cock just at the time of ignition and observing the color of the flame. If the flame is an extremely pale blue, the mixture has too little gas and the gas-valve should be opened a little wider. If, on the other hand, the flame is tinged with red, orange or yellow the mixture is too rich. The proper color of the flame is a deep blue approaching a violet. Do not depend for this test upon a pet cock situ- ated close to the gas-valve opening, as the mixture at this point of the compressior space is usually much richer in gas than at the center. At intervals of six months, if the engine is in constant use, remove the piston anc clean the piston and the inside of the cylin- der with kerosene. Take out the piston - rings and give them and the grooves in which they lie a thorough cleaning as well. The valve-stems should be given an oc- casional spray of kerosene with a squirt can. Never use heavy oil for this purpose, as it will cause the stem to clog and stick. This is especially true of the exhaust- valve stem. 34 CHAPTER VI. GAS-ENGINE TROUBLES. While it is not possible to anticipate ev ery trouble that may occur in the running of a gas engine, a general outline of such troubles as are of frequent occurrence, and the remedies that apply in each case, will give a clew to the solution of those prob- lems not considered in this chapter. Failure to Start. See first that the fuel- valve is not open too wide, nor that it has not been open a small amount for a length of time that would allow an excess of gas or gasoline to leak into the cylinder. Then examine the igniter. If a tube, see that it is not too cold. If an electric igniter, look it over thoroughly as directed in the chapter on Starting. See if the gas supply is inter- rupted in any way. It may not have been turned on. In cold weather, a gasoline en- gine may fail to start because the gasoline does not vaporize. In this case, warm the air-inlet pipe, or fill the water-jacket with 35 hot water. In very severe weather it is a good plan to do both. Engine Starts, but with W'eak Explo- sion. The igniter may be set too late, and it should be adjusted to ignite just before the engine reaches the rear dead center on the compression stroke. The fuel-valve may not be open wide enough. The engine may lose compression from various causes. One or both of the valves may leak or may not seat itself with sufficient force. The starting lever may be still in the starting position. The relief-cock may be open. The piston may leak, but this is not often sufficient to materially affect the power of the engine. See that the valve springs are not too weak. Explosions in the Exhaust Passages. The exhaust-valve may leak or may not seat property. The ignition-tube is too cold, or the electric spark is weak. The fuel -valve may be open too wide, so that, occasionally, the mixture in the cylinder is too rich to take fire. The primary cause of this trouble, is always the occurrence of unburned fuel in the exhaust passages. Jingine Slows J)oicn and Finally Stops. This may be due to overheating of the piston or the cylinder, because of an insuffi- cient supply of water, or insufficient oil. One or more of the bearings may be overheated. 36 notably the crankpin or the crankshaft bearings. The engine may be overloaded. See also the comment under Weak Explo- sion above. For the treatment of an over- heated cylinder or bearing, read the chapter on the care of a gas engine. See paragraph on " Back Firing." Explosions Cease. See paragraph on Failure to Start. Premature Explosions, " Back Firin^ This trouble is of most frequent occurrence with fuel of a low ignition temperature, such as gases rich in hydrogen, and gaso- line. If the ignition apparatus is properly adjusted, the source of the trouble may be traced to an overheated cylinder and too high compression, or to highly heated pro- jections within the compression space. The latter cause of this annoying trouble has frequently been a puzzle for some of the best gas-engine men to find. A thin projection of metal within the cylinder may be so situated that it becomes heated to a comparatively high temperature and acts in the same manner as an ignition-tube. Again, there may be a projection within the cylinder upon which carbon will deposit in the shape of a cone. This cone of carbon will become incandescent, or nearly so, and cause premature ignition, even as early as on the suction stroke. Projections upon the piston head such as the heads of fol- lower-bolts, nuts, etc., quite frequently make trouble in this way. In two-cycle engines of the Day type, explosions will sometimes occur in the crank-chamber because of an insufficient fuel supply. Flame Blown Out. If in engines of the Otto slide valve type, the flame is frequently blown out and there is no draft to which this trouble may be traced, it is a sign of a leaky slide valve, and that either the springs need tightening or the valve itself needs facing. Should the flame on a tube- igniter be blown out, there is a leak either in the tube itself or in some surrounding part of the engine. Spark Gradually Weakens. A spark that is of the proper strength when the engine is started, may gradually get w r eaker after the engine is running, until it is finally too weak to ignite the charge. If the spark is furnished by a battery, this is a sign that the cells need recharging or that the cell is ' not adapted to the work. Battery cells that are made for open circuit only, as those of the sal-ammoniac type, are un suited for gas-engine ignition, because they polarize rapidly and there is not sufficient time for them to recover between the sparks. Occa- sionally a magneto will grow weaker after a few hours use, showing that the magnets are not strong enough to stand the work required of them. The remedy is to get a better magneto. Engine Pounds. -Look the engine over carefully to determine if there are any loose bearings. Lost motion is the prevalent cause of noise in any machine or mechan- ism. If the bearings are " snug," note if the igniter has too much lead, or if prema- ture explosions occur from any other cause. Engine does not Develop Full Power. Note what is said in the paragraph on Weak Explosion; then, if the trouble be not found, see if the exhaust passages are obstructed in any way. See also Engine Slows Down. Smoke. A black smoke may sometimes be observed issuing from the open end of the piston. In this case the piston is leak- ing. The remedy will suggest itself, upon taking out the piston and examining its condition. If the cylinder is badly out of round it should be rebored. The packing rings may need renewing. See if they are too small to expand to a size slightly greater than the bore of the cylinder. Smoke from the open end of the cylinder may also come from overheating. Smoke issuing from the exhaust-pipe, is due to an excess of fuel in the mixture. Leaks. To stop a leak at any point about 19 the engine, first try tightening up the bolts or nuts that hold the parts. If this does not stop the trouble, and the joint is packed, renew the packing. Should the joint be a ground joint, it should be re- ground with flour emery and oil, and the joint wiped perfectly clean after the opera- tion. The best remedy for a leaky valve- stem is to ream out the bearing and put in a bushing or a larger stem, being careful to see that the stem is in line when the job is complete, and that the bearing centers with the valve seat. 40 CHAPTER VII. GASOLINE ENGINES. In general details and appearance, there is little to distinguish between the gas engine and the gasoline engine. The only point of difference being, that a gasoline engine has a special attachment for the purpose of supplying fuel to the engine cylinder, either in the form of vapor or a finely .divided spray. In engines where the compression within the cylinder is carried to the practical limit, it is found that the limit is a somewhat lower pressure for gasoline than for gas of the average quality employed for a gas-engine fuel. A gasoline engine will develop more power than can be obtained from the same engine using a good quality of natural gas. A mixture of gasoline and air will become entirely ignited in much less time than a mixture composed of air and gas. This peculiarity of gasoline produces a much more powerful blow at the beginning of the power stroke, and usually causes the indicator to show a much higher pressure at this point than is prob- ably present in the engine cylinder, owing to the inertia of the pencil mechanism being too great to allow it to stop when the maximum pressure is reached. A study of gasoline-engine indicator diagrams will show this effect, as illustrated in Chapter XIV. Devices for supplying gasoline to the engine, may be divided into three classes; carbureters, vaporizers and jets. A carbureter is a device for transforming liquid fuel into a vapor by passing air either over or through a body of the liquid, and carrying off a portion of the liquid in the form of vapor with the air. Carbureters usually operate at ordinary temperatures, but for fuels that have a low specific gravity the air or the fuel and sometimes both, are heated. This mixture of gas and air is usually too rich in fuel to be explosive, and a further addition of air in the engine cylinder is required before it is suited to the w r ork. A vaporizer is an appliance for transform- ing into vapor, just the quantity of gasoline that is required for one impulse of the engine and no more, and it differs from the carbureter in not having a supply of vapor constantly on hand. Either the proper 42 *H* Fig. 4. quantity of fuel is caused to flow directly into the path of the entering air, or the air is passed over a pipe connecting with a small gasoline reservoir and a current of the fuel is induced into the path of the entering air. Jets are what the name implies, a jet of liquid usually controlled by a small pump. The pump throws a jet of the liquid into the air pipe so that it strikes the side of the pipe and breaks into a spray, or, as in cer- tain classes of kerosene engines, into a compartment of the compression space and against the side. Jets are sometimes classed as vaporizers, but placing them in a class by themselves makes them much more con- venient to refer to. Carbureters may be divided into two classes, surface carbureters and Jittering carbureters. In Fig. 4 is shown an example of a surface carbureter. The carbureter is constructed in the form of a spiral in order that the air passage through it may be a long one. The bottom of the carbureter is covered with gasoline to the height ,rjy, and the wicking ww absorbs the liquid so that a large sur- face of fuel is exposed to the air as it passes through. According to Mr. Gardner His- cox, the height of the gasoline should be not over 3 inches and the total height of 44 p Fig. 5- 4.S the carbureter not over 8 inches. The air enters the spiral through the clack-valve v and passes to the engine through the pipe e. A filtering carbureter is shown in Fig. 5. The air enters the carbureter through the holes h and passes downward through the pipe p to the gasoline, whence it bubbles up carrying with it particles of vapor. A float F carries the pipe p in order that the lower end may be constantly at the same distance below the surface of the liquid. In passing upward, the carbureted air goes through the wire gauze g so any drops of the fuel that may be held in suspension will be caught and left behind. The mixture passes to the engine through the pipe e. A good example of a vaporizer is shown in Fig. 6. Gasoline enters the vaporizer through the needle valve n and air through an opening leading to the space A. The double-seated valve A is lifted at each in- duction stroke of the engine, the larger seat opening a passage for the mixture while the smaller seat on lifting opens the passage for the gasoline. As the air is warmed previously to coming in contact with the fuel, it vaporizes readily, and the proportions of gasoline vapor and air may be regulated by the needle valve. An example of the jet is shown and described in Chapter IX, see Fig. 9. The 46 Fig. 6. 47 method explained in that chapter is one in use in an oil engine and a similar feed is used with a gasoline engine, with the excep- tion that the jet is thrown into the air-inlet pipe, usually against the side. 157 CHAPTER VIII. HANDLING A GASOLINE ENGINE. For starting, stopping and the care of a gasoline engine, the same general rules apply as for a gas engine. In starting a gasoline engine, especially when the engine is cold after standing idle for some time, it is a good plan to put a quantity of gasoline in the cylinder and allow it to remain there for about a minute before starting the en- gine. With engines employing a pump to raise the gasoline from a tank or reservoir, it is necessary to operate the pump by hand for a few strokes in order to get a supply of fuel in the reservoir. The fuel supply to the engine is usually regulated by means of a needle valve, which should be carefully cleaned at regular interval-s. In engines using a jet feed, the supply is regulated by adjusting the stroke of the pump, or by regulating the opening in a by-pass, so that a portion of the fuel is pumped through the by-pass and returns to the source of supply. 49 HTCov. With engines using a carbureter, it may be found necessary, in extremely cold weather, to arrange some means of supply- ing heat to it, because the transformation of the fuel into vapor produces a refriger- ating effect which will chill the liquid to such an extent that it will not vaporize. Two methods of heating the carbureter may be employed ; one is to pass the exhaust- pipe through the liquid and the other is to warm the fuel by means of the outgoing jacket water. The latter method is un- doubtedly the most satisfactory as it avoids excessive heating of the gasoline. In kero- sene-oil engines which operate on the car- bureter or vaporizer principle, it is found .absolutely necessary to heat the fuel before it enters the cylinder. In some forms of jet supply, it is not necessary to heat the kerosene before it enters the cylinder, as it is either injected against a highly heated surface or into a body of air that has been brought to a high temperature by compres- sion, as in the Diesel motor. The time of ignition in the gasoline engine may be made a little later in the cycle, and thus avoid the hard blow pro- duced at the time of the explosion, without a noticeable loss in power. In starting a gasoline engine, it is often necessary to iirst keep the regulating valve entirely 50 closed and to open it a little at a time immediately after the engine receives its first impulse. In general, a carbureter will produce a more perfect mixture than the jet or the vaporizer, thus insuring complete combustion and a consequent absence of smoke and odor at the exhaust. But this advantage is, in many engines, offset by the many disadvantages of this method. The carbureter usually takes but the lighter portions of the fuel, leaving in the bottom of the tank a residue which is of little or no value, as it is of too low a specific gravity to be used as a fuel in the engine. On the other hand, the vaporizer or the jet insures the using of all the fuel irrespective of specific gravity, if the aver- age of the mixture is equal to that required by the engine. In the occasion of the engine being located near a refinery, it is probable that the residue resulting from the use of a carbureter may be sold at a price that will overcome that objection to this method. In general, however, the vaporizer or jet will be found the least troublesome, although some very successful engines are employing carbureters. At no time should there be any foreign matter permitted to enter the gasoline- supply pipes or the valves. Gasoline tanks should be filled through a fine-wire strainer or a piece of closely woven muslin. The opening of the supply-pipe into the tank should be covered with a strainer and every precaution taken to prevent trouble from this source. Never put the gasoline in anything but a perfectly clean receptacle. Old paint or varnish cans or barrels are especially to be avoided. At regular inter- vals, drain the gasoline tank and clean it thoroughly. Where the tank is under ground or in a similar position, in which it would be impractical to clean it, extra precaution should be taken to avoid the entrance of foreign matter. Much of the trouble experienced with gasoline engines occurs from neglect of this precaution. Large quantities of gasoline should be stored at some distance from buildings, and the tank should be protected from the direct rays of the sun by means of a shed or a covering of earth. If the tank is made of sheet iron and fitted with a safety valve that will allow the escape of the vapor, so that the pressure induced by overheating may be relieved before endangering the tank, much of the loss by evaporation will be avoided. Many gasoline-tank explosions are due to excessive pressure caused by the overheating of a tightly closed receptacle, and not from the application of a flame. Usually, the evaporation of this light hydrocarbon is so rapid that all air is driven from the vacant space above the liquid and a flame applied to an outlet of the tank would not cause an explosion, as the mixture would be too rich. If, by any mishap, a tank of gasoline takes fire at a small outlet, run to the tank and not away from it, and either blow or pat the flame out. Never put water on burning gasoline or oil, for the oil will float on top of the water and the flames spread so much the more rapidly. Throw fine earth, sand or flour on top of the burning liquid. Flour is best, because it will float and less will be needed. The best fire-extinguisher for a fire of this sort in a room that may be closed is ammonia. Several gallons of ammonia, thrown in the room with such force as to break the bottles which contain it, will soon smother the strongest fire if the room be kept closed. Very often, simply striking the opening from which the flame is issuing, with the palm of the hand will put out the fire. CHAPTER IX. IGNITERS. A good igniter is one of the most impor- tant parts of a gas engine. It is required of an igniting device: that it shall fire every charge without fail, that it shall ignite the charge at the proper point of the cycle, and that it shall require a minimum amount of attention both for cleaning and renewal of parts. There are four distinct ways in which the charge may be ignited I. Ignition by means of a naked flame. II. Contact with a surface which is at high temperature. III. The flame of a small electric arc. IV. Raising the temperature of the cylin- der contents by high compression. Flame ignition has its best known example in the Otto slide-valve engine. The general principle involved is, however, much better illustrated in Barnett's igniting cock, Fig. 7. In this method of ignition two gas jets are necessary. One, the flame f which is 54 Fig. 7. 55 employed for ignition of the charge, and the other F for relighting^" when blown out by the explosion within the cylinder. The plug valve A is shown in the proper posi- tion for relighting the flame f. At the right moment, the valve makes a quarter turn so that the opening in the plug is opposite the opening into the cylinder indicated by the dotted lines. The resulting explosion extinguishes the flame f, and the valve .re- turns to the position shown in the figure, the gas rushes out through the opening and ignites at the flame F. Contact with a highly heated surface finds its best example in the hot tube. In several makes of kerosene oil engines, a portion of the cylinder is divided from the remainder by a narrow passage and left unjacketed, so that it reaches a tempera- ture sufficiently high to ignite the charge. Another plan, due to Mr. Dugald Clerk, is to drive the charge through a grate built of thin strips of platinum. The grate is brought to the proper temperature by means of an exterior flame before the engine is started and, thereafter, the grate receives sufficient heat from the burning gases to ignite the following charge. This method is useless in a hit-and-miss type of engine, because a few idle strokes will allow the grate to cool to a temperature below 56 57 that necessary for ignition. Thin rods and even small bolts have been employed in much the same manner as the platinum grate. The earlier tube igniters used what is known as a timing valve. This valve opened communication with the tube, so as to time the ignition at the proper point of the cycle. It is now the common practice to time the firing point automatically by the compression of the charge. When the engine exhausts, the pressure within the tube falls to that of the atmosphere, and it is filled with the products of combustion. When the engine compresses a fresh charge, a portion of the mixture is driven into the ignition tube, forcing the products of com- bustion ahead of it, and when the pressure within the cylinder has reached the right amount, the fresh mixture is brought into contact with the heated portion of the tube and ignites. A tube of this kind with adjustable burner, is shown in Fig. 8. The tube is shown at t, and it should be of either nickel alloy or porcelain. The gas is driven by the compression of the piston into the tube through the port p. The tube is heated by the Bunsen burner b, the flame entirely surrounds the tube and is confined by means of the chimney c. The chimney is lined with a tube of asbestos, the asbestos being an important addition, as it prevents the heat from being carried off through the walls of the chimney. Minute adjustments of the flame may be obtained by swinging the burner b about the set-screws s, while for larger movements the chimney and the burner may be changed by loosening the screw y and sliding the chimney along the rod r. A pet-cock is provided at g for blowing the soot from the tube. In general the flame and the chimney are not made adjustable, their proper position having been determined by experiments at the factory, and the chimney and burner being fastened in the position found most favora- ble for the operation of the engine. Ignition by contact with a hot surface in the combustion chamber, is illustrated in Fig. 9. This set of diagrams shows the series of operations which take place in the Hornsby-Akroyd oil engine. Before start- ing the engine the chamber c is brought to a temperature very nearly that of a red heat and this temperature is afterwards main- tained by the combustion within the cylin- der. During the suction stroke of the engine, a jet of oil is forced into r by means of a pump and, striking the hot surface of the chamber, it is transformed into vapor. The cylinder of the engine when the piston is at the end of the suction stroke contains 59 pure air, while the chamber c is filled with oil vapor and products of combustion left from the last cycle. In the figure, the oil is represented by small circles and the air by crosses, while the products of combus- tion are shown by small squares. In dia- gram (B) the piston has compressed the air, driving it into c and, as soon as the con- tents of the chamber is of an explosive nature, it takes fire from the heated surface of c. In diagram (D) the piston has started on a forward stroke, expanding the products of combustion. The electric igniter is slowly but surely taking the place of all others, because only by its use can the ignition of the charge be timed to a certainty. The form of electric igniter which is in most general use, op- erates upon the following principle: An electric circuit from a battery or other source of electrical energy is closed, by means of contact points within the com- pression space, through an inductive resist- ance in the form of a spark coil. Upon breaking the circuit, the inertia produced by the induction raises the pressure of the circuit and causes a hot spark to arc across the terminals. This method is known as the make-aiid-break, and it may be pro- duced either by forcing the two contacts to- gether and then throwing them suddenly 61 00- Fig. 10. apart by means of a spring, or by wiping one contact on the other in a manner that is known as the wipe break. This latter method produces a very hot spark, but pro- vision must be made for adjustment as the points wear out quite rapidly. In either method, good judgment must be used in adjusting the current supply to the requirements of the ignition device. Too little pressure will produce but unsat- isfactory ignition, while, should the pres- sure be such that the current will be large, the contact points will wear out in a very short time. Where the battery employed is of low resistance, as in the case of a storage cell, the pressure at the battery terminals should be much less than with cells having a high internal resistance. Two methods are available for reducing the rapid destruction of the points by too large a flow of current, when it is inadvisable to reduce the number of cells. The destruct- ive action of the spark may be annulled by placing a condenser in parallel with the break as shown in Fig. 10, or by putting a noninductive resistance in series with the circuit as shown in. Fig. loa. To make a noninductive resistance, wind the wire in a coil, about a core that contains 110 iron, and begin to wind at the center of the length of wire so that, when finished, one half the 63 6 4 current will flow in one direction around the coil, and the remaining half in the opposite direction. If the resistance coil is not made iioninductive, it will be quite sure to destroy the sparking power of the spark-coil. An example of the first style of make- and-break is shown in Fig. n. This figure is diagrammatic only and is not intended to represent any special make of igniter. From a source of energy B, one wire is grounded on the frame of the engine. The other side of the circuit is attached to the insulated electrode e, and it contains the spark coil c connected in series. The cam C, rotated in the direction of the arrow, depresses the spring s which carries with it the electrode p until the point of p meets e. Further rotation of the cam merely deflects the spring, increasing the pressure between the contacts. Still further rotation of the cam allows the end of the spring to slip off the lip and it flies back, carrying with it the plunger />, making a quick break between the contacts and producing the spark. In practice, much trouble is experienced with flat springs as they are more liable to break than are helical springs. Note that by helical spring is meant a spring that is wound in the form of a screw thread as shown at A, Fig. 12. This form of soring is 65 bfl s 66 :>ften but erroneously called a spiral spring. The true spiral spring is shown at B in the figure, and is of the kind so much used in clocks ; it is not so reliable as the helical spring. An example of a wipe break is shown in Fig. 13. The electrodes X and Y are so placed that they lie directly in the path of the incoming gases in order to keep the electrodes cool. The electrode Fis rotated by the crank E, and rubs against the spring electrode. The rubbing has the advantage that it keeps the contact surfaces always clean. The electrode Fis made adjustable, so that it can be pushed further into the valve box as it wears off. A very hot spark is produced by this form of break and,, although it requires more frequent adjust- ment and renewal of parts than some other forms, there is no platinum used, so that it is quite a favorite with some designers. It should be borne in mind by the designer that, in both these forms of make-and- break igniters, contact must be made with pressure and the break must be a quick one. Another method of electric ignition that is finding great favor, especially among automobile builders, is the jump-spark. This spark is so called because it will arc or "jump" across an air gap without pre- 67 68 viously bringing the terminals into contact. It is produced by means of an induction coil, or a RuhmkorfF coil as it is sometimes called. In this coil there are two windings, one of comparatively coarse wire wound on a core of iron wire, and a coil of much finer wire w T ound on the outside of the coarse- wire coil. The current from the source of energy is allowed to flow through the coarse or primary winding, and any varia- tion in the strength of the primary circuit will induce a current in the fine wire form- ing the secondary circuit. In a Ruhmkorff coil the secondary current is usually pro- duced by rapidly opening and closing the primary circuit, either by means of a magnetic vibrator or by a toothed wheel. This method produces a series of sparks across any gap in the secondary circuit that is not too great for the capacity of the coil. Sparks are produced both on closing the primary circuit and when it is broken, but the spark at the break is much the more powerful of the two. The induction coil is used in two ways. In one the vibrator is employed and, upon closing a switch on the cam shaft, a series of sparks is sent across the spark gap in the combustion space. In the other, 110 vibrator is used, and two sparks are produced, one on the opening and one on the closing of the 69 b/J S switch. The second spark is that which is depended upon for igniting the charge, as the make spark is not strong enough. The arrangement of the circuits for pro- ducing the jump-spark is shown in Fig. 14. The primary circuit passes through the switch S on the cam shaft to the primary of the coil, through the primary and thence back to the source of energy B. The sec- ondary circuit is from the terminal a of the secondary to the terminal x of the ignition plug and from the terminal y of the ignition plug to the terminal b of the secondary. For the secondary circuit, the secondary winding of the coil becomes the source of energy. If there is a vibrator in the circuit, the time of ignition will be immediately after the switch closes the circuit. If there is no vibrator in the circuit, the time of ignition will be just as the switch breaks the circuit. To get a sure ignition, it has been found that the points should be set a distance apart equal to about one quarter the maximum sparking distance of the coil. A 3^-inch coil with a J^-inch gap has been found to give very good results. There are coils on the market designed expressly for gas engines using the jump-spark, and the author strongly advises the reader to buy a coil rather than attempt to make it himself; for the reason that considerable skill and experience is requisite for the design and construction of a coil which will stand the hard service required of it when used for gas-engine ignition. Raising the temperature of the cylinder contents is a method that has proven quite successful in the Diesel motor, and for a description of this method the reader is referred to Chapter I. 72 CHAPTER X. VAI^VK MECHANISMS. The usual form of valve used in a gas engine is that known as the mushroom type and is shown in Fig. 29, Chapter XVII, where the general proportions of the valve are treated. The reason that this type of valve is best for the gas engine, is because there are no rubbing surfaces and there is very little wear. The high temperature within the cylinder of the gas engine, makes it difficult to keep the valves and their seats at a pressure at which they will wear for any reasonable length of time. Hot gases of themselves, when passing at a high rate of speed through a small opening, score the metal surrounding the opening, and, for this reason, if a hole is once started it enlarges rapidly and soon the engine is not working at its full power. In nearly all gas engines of the four-cycle pe, the valves are operated by means of ams on a shaft called the lay- or camshaft, 73 which makes one revolution to two of the crankshaft. Occasionally the use of the camshaft is avoided by employing as pecial mechanism in connection with an eccentric, in which the eccentric rod is made to open the valve at every other stroke, the interme- diate strokes being idle ones. In an engine manufactured in England, a rotating valve is employed which is on a shaft making one revolution to four revolutions of crank- shaft. The valve ports are so arranged that they are open to the cylinder twice in each revolution of the valve-shaft. In many small engines, only the exhaust valve is opened by mechanical means, the inlet valve being operated by the suction of the engine. This of necessity causes some wire-drawing during the suction stroke, as there must be a partial vacuum in the cylinder before the valve will operate. On large engines, wherein a small propor- tionate loss of power becomes of impor- tance, it is not advisable to employ the suction valve, and the inlet valve should be operated in the same manner as the exhaust valve. In fact, some gas-engine designers claim that a suction valve will give trouble because, if it sticks at all, the valve opening is sure to be reduced, while with a mechanically operated valve, a little extra friction is of no consequence. There 74 75 are, however, some very good engines and a few large ones which have been operating for several years- with suction valves. The only advantage the author can see in such a valve is simplicity and less first cost, be- cause of the absence of the valve mech- anism. In two-cycle engines of the Day type, the only valve required for either the exhaust or the inlet, is a check-valve opening into the crank-chamber. For a description of this engine and the valve arrangement the reader is referred to Chapter I. The check- valve will cause a slight wire-drawing in the crank-chamber, but, with a properly proportioned engine, there is no vacuum formed in the cylinder at any portion of the stroke. It is customary, in the design of a valve mechanism, to transmit motion from the cam to the valve-stem by means of a lever in order to avoid cramping of the valve- stem in its bearings. If, however, the valve-stem be given a bearing near the cam with a long surface and the stem be en- larged if necessary in order to stiffen it, the intermediate lever may be omitted. A valve- operating device with a lever is shown in Fig. 15. The lever L is pivoted at x and carries the roller r which bears against the cam C. The center of gravity of the lever 76 L is placed far enough to the right of x that the lever will incline toward the cam after it has been thrust by the valve -stem s as far as allowed by the head of the valve. The hardened steel contact piece a is put in the end of the valve-stem and a hard- ened plate b is dovetailed into the lever. In Fig. 16 is shown a valve-stem without the lever. The stem is screwed or keyed into the sliding block B, which is forked on the end so as to provide a bearing for the shaft of roller r. If the camshaft rotates constantly in one direction, the center line of the camshaft should be set out of line with the roller shaft as shown in the figure. The direction in which the shaft c should be out of line is determined by the direc- tion of motion of the camshaft. The line w v should be on the side of x y opposite the projection on the cam, when the pro- jection is approaching the roller. In laying out a cam, the simple method shown in Fig. 17 will give an outline that is suitable for nearly every condition. The cam is made up of two parts, a portion C concentric to the shaft, and the eccentric portion P. The part C is usually turned a little smaller than the circle shown by the dotted line, which is a circle that would be described by the roller when the valve is down upon its seat and all lost motion 73 79 is taken up between the stem and any mechanism which carries the roller. In laying out the eccentric portion, it should be remembered that the valve does not begin to open until the point/, where the line m n meets the dotted circle, is in con- tact with the roller. That portion of the revolution of the cam during which the valve is to remain open, should, therefore, be laid off on the dotted circle and not, as the author has seen it done, on the concen- tric outline of the cam. The arc P is usually so laid out for the exhaust-valve, that it will open the valve when the piston has completed about .9 of the working stroke and close it just as the exhaust stroke is completed. Hence, for the exhaust cam the arc P should be i.i times a quarter cir- cle or 99, say 100 in round numbers. After laying off 100 on the dotted circle, draw tangents through p and q to the circular outline of the cam, meeting it in the points m and/. Lay off the radial distance / equal to the movement of the roller when it is operating the valve, and describe the arc n k slightly rounding the corners at n and k. For the inlet cam the arc P is usually slightly under 90 in order that it may not open the inlet valve before the piston starts on the suction stroke nor hold it open after the compression stroke has begun. The 80 bo S designer should exercise his judgment to some extent in this matter, but he will find 85 to be very close to the proper length of the arc. For the gas valve, a great many designers use the same cam as for the air valve. In any case, the time that the gas valve should remain open is practically the same as for the air valve, some engines be- ing governed within limits by varying the time during which the gas valve remains open. An example of a valve operating mechan- ism in which there is no camshaft, is shown in Fig. 18. In the figure, ^ is the exhaust- valve stem. The slide W^ reciprocated by the eccentric E on the crankshaft C, car- ries a toothed wheel X. To the wheel A' is pinned the ratchet wheel Y which has just twice the number of teeth that are on A". At each stroke of the slide W, made in the direction of the crankshaft, the pawl p rotates the wheels X and Y so that a tooth T and a notch jVare presented alternately to the stem s. Hence the valve is opened but once during two revolutions of the engine. It is claimed by those manufactur- ers who employ this and similar devices, that the valve opens much more quickly than with the cam, and a better perform- ance of the engine is obtained in conse- quence. There are several varieties of 82 valve-motion dependent upon an eccentric and a device similar to that shown in Fig. 18, nearly all of which are the same in prin- ciple as the device shown. A number of devices for avoiding the two-to-one reduc- tion have been patented, a few of which are in use, but the majority of the four-cycle engines are using the camshaft. For transmitting motion from the crank- shaft to the camshaft, there are three kinds of gears employed the ordinary spur gear, the bevel gear and the skew gear. The spur gear is that which has its teeth on the periphery of a disk, and is used for trans- mitting motion between shafts which are parallel. It is the most familiar of the three. The bevel gear has its teeth on the surface of a cone and is used for transmit- ting motion between shafts at an angle but which lie in the same plane. The skew gear is one in which the teeth are in the form of a screw thread or helix a screw with a multiple thread and is used for transmitting motion between shafts which are at an angle and which do not lie in the same plane. Examples of these three varieties of gear are shown in Fig. 19. At (A) is shown a pair of spur gears for transmitting motion from the crankshaft s to the camshaft S f . and at the same time reducing the velocity of 83 S 4 5 to half that of s. The size of pinion p is made one-half that of the gear <7, in order to make the proper reduction in speed. The two shafts are parallel. At (B) is shown a pair of bevel gears for changing the direction of motion and at the same time making the two-to-one reduction. The shafts lie in the same plane, that is to say, straight lines, as x y, p q, etc., may be so drawn as to pass through the centers of the two shafts. At ( G) is shown a pair of skew gears in which the speed reduction may be made without making the slower rotating gear the larger. In fact, the gear on the camshaft may be, and usually is made smaller than that on the crankshaft. An- other advantage is that the two shafts do not lie in the same plane. Motion may be transmitted with these gears without the noise that is more or less evident when the spur or the bevel gear is used. Engines, which are fitted with spur gears, usually have the camshaft near the crank- shaft, and the valve-stems project from the valve boxes in a direction parallel to the axis of the cylinder. This arrangement necessitates either long stems on the valves, or a long rod from the camshaft to the end of the valve-stem. In some engines, this feature is avoided by the use of a train of spur gears which brings the camshaft close to the valve boxes. In a carefully designed engine, there is 110 objection to the use of spur gears, and their use often saves ex- pense in construction. Very few engines employ the bevel gear, as it is a somewhat difficult matter to so arrange them on the engine that they will not take up an undesirable amount of room. The skew gear obviates this latter difficulty to a great extent, and it allows the cam- shaft to be placed in a position from which it is more convenient to operate the valves. Either gear may be used to operate a cam- shaft the axis of which is parallel to the axis of the cylinder,. and by the use of the second pair of gears the shaft may be given a second quarter turn and run across the back of the cylinder head to operate valves that are placed in the head. Where there are two or more cylinders, it will add to the economy of construction, if a single cam- shaft is employed to carry all the cams. This shaft should then be placed so that the axis is at a right angle to the axes of the cylinders. 86 CHAPTER XI. GOVERNORS. All gas engines are governed by eitlier cutting out or cutting in impulses to the piston, or by reducing or increasing the force applied the piston according to the requirements. Regulating the speed of the engine by varying the number of impulses given to the piston is called the hit-and-miss system. There is no term at present in use that will cover all cases of governing by varying the strength of the impulse, and the author suggests the term variable impulse. There are three methods of gov- erning under each of the above two systems of regulation, as described below. HIT-AND-MISS. I. Holding the gas valve closed during one or more revolutions of the engine when running below full power. During the idle strokes the engine is compressing and expanding a charge of pure air. 87 II. Stopping the action of the exhaust valve, holding it either open or closed dur- ing the idle strokes. When the valve is held open there is not sufficient suction to open the inlet valve, this method usually being employed on engines having a suc- tion valve. In the latter case the engine retains the products of combustion, the pressure within the cylinder being at all times too great to allow a fresh charge to enter. III. Cutting off the current from the igniter, employed, of course, where an electric igniter is used. The governor is, in this case, attached to a switch, which is opened whenever the speed passes the limit. In this case the charge is alternately compressed and expanded until the switch is closed and ignition takes place. A fourth method might be added to this class, i. e., the stopping of the camshaft. With this method all operations are halted, usually with the exhaust valve held open. VARIABLE IMPULSE. IV. Partial stoppage of the gas supply. The mixture of fuel and air retains its explosive properties within certain limits of the proportion between the two. If the mixture is poor in gas the impulse is cor- respondingly weakened, and up to the point 88 whereat the charge explodes with the greatest force, the strength of the impulse may be increased. There is also a lower limit, beyond which the mixture becomes so poor in gas that it will not explode. After passing this limit it is necessary to resort to the hit-and-miss method I, and cut out the gas supply entirely. ' V. Throttling the charge. This method reduces the strength of the impulse by reducing the quantity of the charge that enters the cylinder, the proportions of fuel and air remaining the same at all times. VI. Varying the point of ignition. The greatest ' mean effective pressure in a gas engine is obtained when the spark has sufficient lead to bring the maximum pres- sure at the beginning of the stroke. If the lead is reduced, or made negative (ignition delayed until the crank has passed the center), the strength of the impulse may be varied. Hit-and-miss regulation is that which gives the greatest economy of fuel con- sumption. There is little choice between the three methods given above, as all have advantages and disadvantages that make them about equally good. Of the variable impulse methods, No. IV is that which gives the best fuel economy, the compression is the same at every cycle. The objection to this plan is, the return to the hit-and-miss system as soon as the speed falls to the limit below which the charge would be too poor in gas to explode. Method V has been used quite successfully for engines driving dynamos, but it does not give the economy that is derived from an engine employing one of the first four methods. The method de- scribed last, is perhaps the most wasteful of fuel after the lead is late enough to form a perceptible reduction in the area of the indicator diagram. The result obtained is shown very plainly in the chapter on indi- cator diagrams. This method of speed regulation is of little service where econ- omy is desired, but where economy is a secondary consideration, as in an automo- bile engine, it is a very simple method of governing. Governors of three different kinds are employed on gas engines. Two of these are of the centrifugal type, similar to the governors in use on the steam engine, the other is a governor made possible by the hit-and-miss system of regulation and is known as the pendulum or inertia governor. The centrifugal governors are the conical pendulum and the shaft governor. Hither of these two may be employed for any method of regulation while the pendulum 90 governor may he used only for the hit-and- miss. An example of the conical pendulum gov- ernor is shown in Fig. 20 as arranged to operate a throttling device. The two balls bb are rotated from the camshaft, or other convenient revolving portion of the engine, through the bevel gears pq. As the speed of rotation increases, the balls move in the direction of the arrows, compressing the spring s and depressing the stem S of the valve Fby means of the collars. As the valve moves downward, the holes a and g are gradually covered by the wall of the passage, contracting the opening from the gas passage G and the air passage A into the chamber X, and preventing the mixture from flowing in as great a quantity as before, into the inlet port /. An example of the shaft governor is shown in Fig. 21, in connection with a de- vice for opening the gas valve only when the engine is below speed. The balls bb are on the end of the short arms // pivoted at pp. As the crankshaft C revolves, the balls tend to swing in the direction indicated by the arrows, but they are withheld by the springs s. As the balls swing about the pivots />/>, the opposite ends of the arms // move toward X and carry with them the collar c. A long lever m with a fulcrum pivot at P is 92 93 carried with the collar and the long end pushes the point out of line with the stem v of the gas valve. The point x is attached to the air valve mechanism so that it rocks backwards and forwards at every movement of the air valve. When the speed is too high the governor causes the point x to miss the valve-stem and no gas enters the cylinder. In Fig. 22 is illustrated one form of the pendulum or inertia governor arranged to hold the exhaust valve open so long as the engine is above speed. The valve stem is operated from the camshaft by means of the slide M. On the slide is carried the pendulum P which swings about the pivot J>. If, upon the return of the slide, the speed is higher than that for which the governor is set, the pendulum lags behind and the end of the long arm a of the bell- crank to which the pendulum is attached, strikes the pin c and throws it in the path of the block B, holding the valve open. Upon the following stroke of the slide, the valve-stem >S receives a thrust sufficiently long to allow the pin c to drop out of the way of B unless the arm a is still in the way. If the arm allows the pin to drop, the valve closes and the engine takes up its cycle once more. The speed of the engine may be varied by a simple adjustment of 94 b s 95 the pendulum ball P. If it is desired to increase the speed of the engine, the set screw n is loosened and the ball P is raised. L/owering the pendulum makes the gov- erner act at a lesser speed. This form of governor is also much used to control the gas valve. The long arm a of the bell-crank is pointed and it is used to give the necessary thrust to open the valve. In case the speed rises above a cer- tain amount the arm a swings aside and misses the gas valve stem. 96 CHAPTER XII. STARTERS. Engines larger than eight or ten horse- power are difficult to start by turning the flywheel by hand, and, in order to lessen the amount of labor involved, larger en- gines than the above are usually supplied with some device by means of which an impulse may be given to the piston when the engine is at rest, that will be strong enough to propel the engine for three or more revolutions and until it can take up its cycle and receive an impulse in the regular way. The methods in use for this purpose are so numerous that but a few examples may be illustrated in a volume of this nature. The first method to suggest itself to an engineer would obviously be to apply pres- sure to the engine for one or more strokes- from an exterior source, as from a steam boiler or from a tank of compressed air. In fact, the compressed air starter is a most 97 convenient form. It is very simple and -consists of a small air pump and a storage lank for the air. A pipe connects the tank to the cylinder of the engine, and this pipe is supplied with a two-way plug-valve. The lank is filled with air, at a pressure of forty pounds or over^when the engine is running, the pump being operated by means of a belt from the line shaft or other convenient point. When it is desired to start the en- gine the flywheel is turned until the cam- shaft is in the proper position for the exhaust stroke, and until the crank-pin is just above the back-center. The air-valve is then opened wide, and closed again as soon as the piston nears the end of its stroke. If the impulse is not strong enough to give the engine a sufficient start it should be repeated at the next revolution l)ut one. Two impulses with the air are usually more than enough for the purpose. After the engine is under way, the fuel should be turned 011 as directed in the chapter on Starting. If the engine has more than one cylinder, one cylinder may be used to start while the remaining cylin- ders are taking up the cycle. Occasionally, the engine is fitted with a special valve mechanism, for the purpose of starting with compressed air, so arranged that an impulse is given to the piston by the air at every 9 s Fig. 23. 99 revolution until the speed is sufficient to lake up the cycle. A class of starters that is very much in use on engines of fifty horsepower and l>elow, is simply a means of igniting a charge that has been drawn into the engine by turning it over by hand. For instance, a certain manufacturer equipped his engines with what is called a match igniter. This igniter is shown in section in Fig. 23. The igniter consists of the casting B which is screwed into an opening into the valve-box. It contains a plunger consisting of a tapered and roughened head A, a stem d and a cap C. The plunger is held at the top of its stroke by means of the spring S. A small pet cock s is screwed into the side of B, through which is inserted a parlor-match m, the plug h is then turned so as to break the match-stick and hold it in place. When the engine is under way the cock s is opened and the burnt match expelled by the next explosion within the engine. To start the engine, it is turned over until the igniter snaps, and a charge of air is drawn in by turning the flywheel. At the same time, a small quantity of gasoline the amount depending upon the size of the engine is drawn in with the air. The en- gine is now turned over until the piston is at nearly the end of the stroke ; the engine is then reversed compressing fh^ ciii.r^e by a quick backward turn of the flywheel, the cap C of the igniter is struck with the hand lighting the match m and firing the charge. The explosion is usually of suffi- cient force to turn the engine until it can take up its cycle. In some engines, the use of the match igniter is avoided by means of a device which operates the electric-igniter at the proper moment. When the jump-spark is employed, a switch which will close the primary circuit, is all that is necessary. In j other engines the flywheel is turned until the camshaft is at the proper position for the beginning of the suction stroke, and the crank-pin is on the inner center. A charge of fuel and air is then pumped into the cylinder by hand. The pressure within the cylinder can be determined by the force necessary to work the pump or by setting the engine a little past the center and hold- ing the flywheel, noting when it begins to pull. When the required pressure is ob- tained, the flywheel should be turned until the pin is about 10 past the center and the charge fired by means of a match igniter or equivalent device. In large engines, where the horsepower is 100 h. p. or above, a small gas engine, which may be easily started bv hand, is . employed to' Inrn t#e large engine over until it takes up its cycle, and the starting engine is then uncoupled automatically. Sometimes the valve mechanism of the en- gine is so arranged that, after the engine is uncoupled from the line shaft and before it is stopped, it is employed to charge a pres- sure tank by permitting a portion of the expanding charge to pass into the tank. This is accomplished by opening communi- cation with the tank after the piston has made a portion of its expansion stroke, the first portion of the stroke being used to impart motion to the engine. CHAPTER XIII. ENGINES FOR AUTOMOBILES. The requirements of an automobile en- gine are lightness, compactness, minimum vibration, wide range of speed under con- trol of the operator, and simplicity. Econ- omy of fuel consumption is a secondary consideration. The fewer the cams, levers, etc., on an automobile engine, the less liable will it be to get out of order. It was for- merly supposed by the gas-engine builder that a gas engine must run at a constant speed, and all changes of speed made by means of gears. Recent experiments have exploded this theory and successful engines are now on the market which will run at speeds varying from 100 to 1000 r. p. m. for the same engine. At least one manufacturer claims that his engine may be run at speeds ranging from 100 to 2000 r. p. m. Two- methods are employed for governing the speed of the engine. One is by throttling the charge, and the other by changing the 103- lead of the spark. Hither of these methods gives an impulse at each cycle. Bngines operating on the hit-and-miss principle, are useless for motor vehicles. For single-seated carriages, an engine capable of giving at least four horsepower at between 500 and 600 r. p. m. should be employed. If the carriage is likely to be called upon for heavy hill-climbing or rac- ing speeds the horsepower should be in- creased to six. For motor bicycles one liorsepower is usually sufficient, while for tricycles two horsepower is that generally employed. Single-cylinder engines will not run in a horizontal position without caus- ing an annoying amount of vibration to be transmitted to the vehicle. If it is desired to use a single-cylinder engine it should be placed with the cylinder vertical and the engine very carefully balanced. Two-cylin- der engines with the cylinders placed end to end, each connecting-rod being attached to a separate crank-pin and the crank-pins placed 180 apart, will, if all parts are care- fully made of the same weight, run with , very little vibration. The reciprocating and rotating parts of the two engines balance each other. Two cylinders with their axes parallel and the open ends in the same direction, work very well when the cranks are set at 180, but still better results in 104 the matter of balancing are obtained by the use of three or four cylinders with the cranks equally spaced about the crank- circle. Among the automobile builders, the jump- spark is finding favor, as it may be better controlled at the high rotative speeds at which these engines run. The change of lead for the spark may be obtained by rotat- ing the cam or switch, as the case may be, around the camshaft. Governors for auto- mobile engines are very well in theory, to prevent the engine from speeding up when an obstruction throws one of the wheels off the ground, but there is no governor w r ithin the knowledge of the writer that will act within a time sufficiently short for the pur- pose. The speed of the engine should always be at the control of the operator. If the engine speed is under control, there will be but two changes of speed required, one for levels and easy grades and the other for hill-climbing. The reversing gear may be connected to the hill-climber as a high speed is undesirable for going backward. For cooling the cylinders, both a closed circulating water system is employed, and also ribs radiating from the cylinder-walls. The latter may not be used successfully on cylinders over 3^ to 4 inches diameter. For an engine employing a water-jacket 105 there should be some method of cooling the water, otherwise the amount of water which it is necessary to carry becomes excessive. In France, where the gasoline automobile is in extensive use, a coil of pipe with collars to increase the radiating surface of the pipe is considerably employed. These collars or disks are stamped from sheet metal and punched with a star-shaped hole in the center and then forced over the tube, while the tube is straight. Afterward the tube is coiled on a comparatively small radius. This coil is made a part of the circulating system and answers its purpose very effectively. The author would suggest that if means were taken to guide air through these coils by means of a curved shield under the carriage it would increase the cooling efficiency of the coil. The gasoline and the water-tanks should each have some means by which the quan- tity of the liquid contained in them could be determined at any time without it being necessary for the operator to leave his seat. All parts of the mechanism should be com- pletely enclosed to protect them from dust. For this reason an enclosed crank-chamber is a desirable feature. Lubrication of every part should be strictly automatic, so that when the carriage has been given the proper attention at the outset of a trip it 106 will need no further attention for a reason- able time at least. The fuel feed should be so regulated that 110 odor of consequence is noticeable at the exhaust opening. The fuel should be under control in order that it may shut off when going down grade, or when stopping the engine. The exhaust- muffler should be a good one, as excessive noise is an objectionable feature. It is nec- essary to sacrifice a little power to get good results with the muffler, but, as already noted, the fuel economy is a secondary consideration. 107 CHAPTER XIV. GAS-KNGINE DIAGRAMS. It is quite a general custom, among the manufacturers of gas engines, to base the rated horsepower of an engine upon its per- formance with natural gas of an average quality. If, therefore, a diagram may be laid out which will show the conditions within the engine cylinder during the cycle, when working under average conditions, it will prove of valuable assistance to the designer. In order that the reader may know what to expect, when about to con- sider a departure from the practice of the present builders w r ith reference to compres- sion, and also that he may have an ideal dia- gram with which to compare those he may obtain from the engine after it is built, the author will show just how such a diagram may be built up by the aid of formulas. An ideal diagram for an engine working with natural gas as a fuel, is shown in Fig. 24. Before laying out the diagram it is 108 necessary to decide upon, either the com- pression pressure desired, or upon the ratio of the volume of the compression space to that displaced by the piston during the stroke. For the diagram in the figure, the ratio has been taken as 30 percent. The total volume of the cylinder, when the pis- ton is at the end of its outward stroke, is therefore i -f- .3 = 1.3 of the piston displace- ment. As it is much more convenient to consider the total cylinder volume as unity for the purpose of making the calculations, it is necessary to find what proportion of the total cylinder volume is included in the compression space, or to divide .3 by 1.3, giving .2308 as this ratio. In calculating the pressures at various points upon the curves, the following formulas should be used, because they represent the average behavior of the gases shown by actual indicator diagrams taken from engines in operation. For the compression curve PV-*=K (i) Wherein P= the pressure above a vacu- um or absolute pressure ; V= the volume of the gases at the time they are at the pressure P; K = a number or constant de- pending upon the conditions, but which is the same for all parts of the same curve. When the volume of the cylinder is con- sidered as unity, A" becomes the absolute pressure of the atmosphere, usually taken as 14.7. When natural gas is employed at the proportion to air which gives the best effect, the pressure of the gases after explo- sion is four times the pressure after com- pression, both these pressures being those above the atmosphere. The formula for the expansion curve is />F'-35= C (2) Wherein P and V have the same signifi- cance as in formula (i) and C is a constant depending upon the maximum pressure. These equations may also be written as follows : Py*-3=PV*.3 t />^i. 35=J pK.3S > a form which shows the exact relation be- tween any two points on the curve. In order that the reader may better see how to apply the above formulas, the com- putation of the diagram in Fig. 24 will be given. Taking first the computation of the pressure at the end of the compression stroke, and applying formula (i) />K'.3 = ^=i 4 . 7 , and P=m = In order to find the denominator of this fraction, it is necessary to multiply the true logarithm of the number by 1.3. The log- arithm of .2308 as found in a table of log- arithms (the tabular logarithm) is 1.363236, and since the logarithm has a negative characteristic (number to left of decimal point), the true logarithm must be found by adding the mantissa (number to right of decimal point) to the characteristic as follows : i. oooooo 363236 .636764 1910292 636764 ^.8277932 1.172207 (subtracting from i and add- ing T to get the tabular log) . Log 14.7=1.167317, and subtracting the log of (.23O8) 1 - 3 the resulting log is that of the compression pressure at the end of the stroke. 167317 1.172207 1.995110 = log of 98.88. 98.88 Ib. is the compression pressure in Ib. per sq. in. above a vacuum. For practical purposes, three figures will be sufficient and 98.9 may be taken as the pressure. The pressure above the atmosphere would be 98.9 14.7 --= 84.2 Ib. per sq. in. To get the maximum pressure, this amount should be multiplied by 4, giving 84.2 X 4^=33 6 - 8 lb -> and 33 6 - 8 -f 14.7 351.5 lb. absolute pressure. In order that the proper shape of the curve may be very closely approximated, it is best to compute the pressures at three points between the ends. The points for which the calculations in the figure were made are those when the volumes are .35, .5 and .75, and in case of any uncertainty a point should be taken midway between .35 and the higher end of the curve. The pressures corresponding to the points given, are 57.6, 36.2 and 21.4 respectively. The pressures for the expansion curve are found in the same manner as for the compression curve, but by means of formula (2). Before this formula is applied, how- ever, it is necessary to find the value of the constant C, which is the pressure at the end of the stroke when the volume is equal to i. Hence PI' 1 ** C becomes 351.5 X (.2 3 o8) I -35 = C. Log (,23o8) I -35 == 1.140368, Log 351.5 = 2.545925. Adding these logs gives, for the sum, the log of the constant C, thus : 1.140368 2.545925 1.686293 = lo 4 8 -5 6 - 4 8 -5 6 lb s. is the terminal pressure of the gases in the cylin- der, should release take place at the end of the stroke. Hence the equation of the expansion curve is PV* *& = 48.56. The intermediate pressures for the rer mainder of the expansion curve are found by subtracting the logs of .35, .5 and .75 raised to the 1.35 power, from the log of 48.56. They are 200.4, I2 3-8 an d 71.6 respec- tively. After these points are located, the> diagram is constructed by drawing the curves through them as shown in the fig- ure. The length of the diagram is 3^'' from the line D C to , and the scale of the spring represented is 160 Ib. Hence unit volume is represented by a length of 3% // , and a height of one inch on the diagram represents a pressure of 160 Ibs. per sq. in. In order to make the diagram complete, it is necessary to add the two small curves .at i and r e. The curve at i is that due to the lead of the ignition which causes an in- crease of the rate at which the pressure is rising. It can be drawn but approximately, as it varies in size for any change in the piston speed. Its effect upon the area of the diagram is too small to be of conse- quence, and the only value it has, is to serve as a memorandum when using the diagram for the purpose of comparison with those actually taken from the engine. The curve r c is that produced by the release taking place before the end of the stroke is reached. 114 The point of release r should be such as to bring the point e where the expansion line meets the atmospheric line A B at the end of the stroke. To draw this curve as it is shown in the figure, erect the vertical line j? r at a distance from Xm equal to .9 time the length of the diagram, which, in the figure, is .9 X -77 -693 time the unit length. Through r and c describe a circular arc, which shall be tangent to the expansion line at r. Joining the points i and m with a straight line completes the diagram. The diagram having been drawn, the de- signer may find the M. E. P. by means of planimeter as explained in the chapter on Testing. From the M. E. P. obtained from this diagram, he may calculate the horse- power of an engine of any size which it is proposed to build, or he may determine the dimensions of the engine. The design of the valve motions may, to a certain extent, be founded upon the diagram, as it shows the proper time for opening and closing the valves. It should be the aim of the designer to build an engine that will give as nearly as possible, such a diagram as that shown in the figure, for only with a diagram of this shape are the best results obtained. A few diagrams will now be shown which have been made by an engine when not working under the conditions necessary for producing a diagram similar to that in Fig. 24. In Fig. 25 are shown diagrams taken from various engines while in actual opera- tion and they illustrate just what occurs under several conditions. At (A) is shown a diagram taken from an engine which is operating with everything in good working order with the exception of an exhaust re- lease which is a trifle late, indicated by the expansion line not reaching the atmpspheric line X Y until the piston has returned to d. The point of release is shown by the sudden change in the curvature of the ex- pansion line at , the point of ignition by the sudden change of curvature in the com- pression curve at i. On either side of the atmospheric line is a curve which is exaggerated in the draw- ing in order that it may be distinct. The upper curve is that produced by pressure within the engine cylinder on the exhaust stroke, and is due to an unnecessarily con- tracted exhaust passage, or to obstructions that have accumulated in the exhaust-pipe. The lower curve is caused by the pressure within the cylinder falling below that of the atmosphere, forming a partial vacuum be- cause of an obstructed inlet passage. Ordi- narily, these curves are too small to be of any moment, but in case they are a quite distinct departure from the atmospheric 117 line, they show that an unnecessary amount of work is being done by the piston in forcing the gases through the passage, and steps should be taken to remedy the defect. A condition of this kind is spoken of as wire-drawing. Diagram (B) is that produced when the ignition of the charge is late. In this case ignition takes place at i after the crank has passed the center. The loss in area of the diagram is quite noticeable when comparing (B) with (A). The reader should learn to distinguish between a diagram produced by late igni- tion and that produced by a weak or a throttled mixture. In (B) the curve indi- cating rise of pressure immediately after ignition, is concave, while in (D) and (E), which are diagrams produced by a w r eak mixture, this curve is either a straight line or is convex. (G) is a diagram in which the ignition is very tardy, and the area lost be- cause of late ignition is shown by the dotted lines. Premature ignition is shown by (C). , In this diagram the ignition is at / and the loss of area is illustrated by the proper ' diagram being shown in dotted lines.* Dia- gram (F) is an example of the effect of late release together .with obstructed exhaust passages. When an engine is operating on fuel which ignites at a low temperature, as is 118 the case with gasoline and acetylene, the propagation of the flame throughout the mixture is very rapid, causing an explosion which is almost instantaneous. An example of the sort of diagram usually obtained with these fuels is shown in Fig. 26. The sudden blow given to the piston of the indicator, sets the spring in vibration and produces the wavy line which appears at the point immediately following maximum pressure. This effect has been attributed to various causes, such as a rapid succession of explosions, and these in turn have been said to be caused by an uneven distribution of the fuel within the air (a stratified charge). The author believes that very few of these curves are of such a nature as to justify such a reasoning. They are all of the nature of a sine curve, the waves grad- vially lessening in height as the piston proceeds on its stroke. In using these fuels, it is best to make the lead of the igniter somewhat less than when using natural gas or manufactured gas. By lead is meant the time or the distance before the end of the stroke, that ignition takes place. CHAPTER XV. GAS-ENGINp; DIMENSIONS. The power of any engine is dependent upon four factors, the average or mean effective pressure upon the piston during the stroke, the area of the piston upon which this pressure is exerted, the length of the stroke, and the number of times per minute the pressure is exerted. The power of an engine computed from the above fac- tors is known as the indicated horsepower (I. H. P.). The power delivered at the pul- ley or obtained from the engine, and which is available for power purposes, is called the delivered horsepower (D. H.P.) or the brake horsepower (B. H. P.) and by some writers the effective horsepower (E. H. P.). The use of the latter abbreviation is, however, likely to cause confusion as the same abbre- viation is employed for electrical horse- power. The tef\n brake horsepower is de- rived from the manner of testing the output of the engine, which is usually with some form of brake. The D. H. P. is the I. H. P. minus the power absorbed in the friction of the engine, the latter being called the friction load. The ratio of the D. H. P. to the I. H. P. is called the mechanical effi- ciency of the engine, sometimes abbreviated to M. E., and M. B. = D ' H ' ^' , this ratio being usually expressed in percent. Thus, if the I. H. P. of an engine is 10 H. P. and the D. H. P. is 8 H. P., the M. E. = T 8 o = 80 percent, and the friction load = 10 8 = 2 horsepower. The mean effective pressure (M. E. P.) in a gas-engine cylinder varies with the fuel, the pressure after compression, the propor- tion of the mixture and the time of igni- tion, and in a minor way on several other conditions which are not of sufficient im- portance to consider at present. If the mixture contains the most effective pro- portion of gas and air, and the ignition is properly timed, as it should be, it leaves the matter of compression and quality of fuel as the two important points to be con- sidered, with reference to the M. E. P. The highest mean effective pressures are those obtained with natural gas and gasoline, then come the illuminating gases in the order of their light-giving values water- gas and semi-water-gas the latter being usually known as producer-gas. For the compression pressure usually employed, the M. E. P.s range between 45 Ib. per sq. in. and 80 Ib. per sq. in., with the average not far from 65 Ib. per sq. in. for an average quality of natural gas. For gasoline the M. H. P. may safely be taken at 70 Ib. per sq. in. The author has found the following formulas are very well borne out in practice for the best performance of the average gas engine : Let D = the diameter of the cylinder in inches ; L = the length of the stroke in inches ; R = the number of revolutions per min- ute ; Then for a four-cycle engine 19,000 For a two-cycle engine ' IO,OOO For engines using gasoline these denom- inators should be reduced to 18,000 and 9,500 respectively. These formulas may be used for determining the dimensions of any engine that is being designed, if the performance of any engine of the same kind and on the same quality of fuel is already known. The denominator of the fraction then becomes an unknown quan- tity and it may be found as follows : Calling the unknown denominator A' X = X -^ X ^ / \ (D. H. P.) Suppose an engine which is a good ex- ample of a series already being built has a cylinder 2O // diameter and a stroke of 3O /X , i. e., a 20 XX X 30 XX engine, and that it runs at 120 r. p. m. giving 70 D. H. P., the value of Xis v= (20)2 X 30 XI20 _ ^- 20,577 -1 call X 20,600. The formula fer these engines is then nz v 7 v I? D. H. P. = A ^ A ^ 20,600 To find the diameter of the cylinder for any engine to give a required horsepower at a given speed, the ratio of the stroke to the diameter should first be determined. For a four-cycle engine, prominent author- ities agree that the best proportion is to make the stroke ij^ times the cylinder diameter. For a two-cycle engine, practice varies between making the diameter of the cylinder and the stroke the same, to a stroke equal to i^ times the diameter. It is also necessary to decide upon the speed at which the engine shall run. Gas engines built in the United States, are seldom run at a higher piston speed for horizontal en- gines than 600 feet per minute and for vertical engines 700 feet per minute. The following formulas representing average practice among manufacturers will be found valuable in making the first approximate calculation : Let H = the D. H. P. of the engine ; Let R = the rev. per min. ; Then for a four-cycle engine A* =-^ ' " " two-cycle " R = -44. W In order to solve the above two equations it is necessary to use logarithms. Suppose it is 'desired to find the speed of a fifteen- horsepower four-cycle engine. Take a table of logarithms and find first the logarithm of 15, which is 1.176091 ; multiplying by .21 the result is .24697911, which is the log of 15 to the .21 power. The log of 380 is 2.579784; subtracting the log of (i5)- 21 from this 2.579784 .246979 2.33 2 705 = log 215-1 + And the proper speed for this engine would be 215 r. p. m. or thereabout. For a four-cycle engine, formula (6) gives a piston speed of 600 feet per minute at 32.5 horse- 125 power, and for a two-cycle engine, formula (7) gives a piston speed of 600 feet per min- ute at 71 horsepower. Beyond these powers, the formulas (6) and (7) should not be used but the computations made from the piston speed. At 600 feet piston speed the r. p. m. is found as follows : (8) JL, When the piston speed of the engine is 600 feet per minute, much simpler formu- las than (3) and (4) may be used to find the diameter of the cylinder. Using the same value of X as before, D = 2.3 ^/ H for a 4-cycle engine ; A' = 19,000. (9) D = i.67/y/ H for a 2-cycle engine ; X = 10,000. ( 10) These formulas should not be used ex- cept for a piston of speed of 600 feet per minute. The stroke of the engine can afterwards be determined by its ratio to the diameter, and from that the r. p.m. may be found by means of formula (8). It should be remembered that these equations are applicable as they stand, only for engines with one cylinder. When the engine is to be built with two or more cylinders, the horsepower must be divided by the number of cylinders. Thus, if a three-cylinder en- 126 glne of 45 horsepower is to be designed, the calculations for cylinder diameter, r. p. m., etc., should be made for a i5-horsepower engine. Should the reader desire to calculate the power of a gas engine from a known M. B. P., the following formulas will assist him. When the piston speed is 600 feet per min- ute and the stroke of the piston is i^ times its diameter p Wherein P = the M. B. P. When the piston speed is 700 feet per minute D= 1.55 V 1 - H - p - ( I2 ) ~P Both of these formulas are for the 4-cycle engine. If it is desired to find the D. H. P., an M. B. of 80% may be safely employed for engines under 25 horsepower. For larger engines having two or more cylin- ders 85% may be used. The reason that a higher efficiency may be obtained from multiple cylinder engines, is that the fly- wheels are much lighter, and the load is better distributed throughout a revolution, reducing the friction of the engine. It is fast becoming the custom to build engines larger than 50 D. H. P. with two or more 127 cylinders. Although four-cylinder engines have been built and are in use, it is the opinion of many gas-engine experts that no material advantage is gained by increas- ing the number of cylinders beyond three. The cylinder diameter of a gas engine may be made the basis of nearly all other dimensions of the engine and not make them vary but a few percent from what would be given by the best of formulas. Although every dimension of the engine should rightly be discussed in the present chapter, the author will give each part a chapter by itself in order to make them easier of reference. 128 CHAPTER XVI. THE CYLINDER. A good example of a gas-engine cylinder is shown in Fig. 27. The dimensions of the various parts have been based upon average practice of gas-engine designers in this country, and, although he will not be able to take up any gas-engine design he chances to come across and find the di- mensions exactly fitting these formulas, he will, by the use of the following equations, be able to produce a design which will be in good proportion. For the thickness of the cylinder wall it is necessary to allow for reboring to a cer- tain extent, and also to have sufficient metal for stiffness, and to make a good casting when the cylinder is a small one. The thickness of the wall is therefore made quite a little heavier than is necessary for mere resistance to pressure. The following formula is that which represents average practice. I2Q /=.o9/>; (13) Wherein t = the thickness of the cylinder wall ; D the diameter of the cylinder. The depth of the water-jacket is usually as given by the following formula : y=..i/V (M) Wherein/ = the depth of the water-jacket. This depth is measured on a radius of the cylinder across the water-jacket. No regular proportion appears to be fol- lowed at all closely for the thickness of the outer wall of the water-jacket. The author has used for engines of his own design a thickness which is half that of the cylinder wall, or .045 D. For very small engines these formulas give dimensions which produce walls too thin for the foundryman. In order to get good castings it is well to limit the thick- ness of the cylinder wall to T 5 g- inch, the depth of the water-jacket to y% inch, and the thickness of the jacket wall to % inch. In order to find the diameter of the cyl- inder head studs the maximum pressure within the cylinder must be known as well as the number of studs. It is a good plan to limit the distance between the studs to six inches, unless there is a feature in the design of the cylinder or the head that will prevent such a space being used. An even number of studs is almost invariably em- ployed. The following formula should be used for finding the diameter of the stud : L,et A = the diameter of the cylinder ; z = the diameter of the stud at the root of the thread ; S 1 the safe stress in Ib. per sq. in.; p = the maximum pressure in the cylinder ; n = the number of studs ; Then </= or d = D\JL_. (15) sn sn The flange by which the cylinder is at- tached to the frame, may be either square or circular. It is usually square on vertical engines and circular when the cylinder is horizontal. The sizes of the bolts which are used to fasten the cylinder to the frame may be found by formula (15) making z the diameter of the bolt at the root of the thread and n the number of bolts. The thickness of the cylinder wall between the end of the water-jacket and the flange should be .125 D. The thickness of the flange should be equal to the diameter of the bolt -f- %". On engines smaller than six horsepower, it is customary to make the cylinder and the frame in one casting. The water-jacket should extend, beyond the end of the piston nearest the cylinder head, tc 132 H : ^^^^^^N^\\^^^^\v^ ^\^\\\x s \\x\i i? i i a JSJSSSE^SXSSSS^B g gy ^^^ix^^vj 00 c* 133 a distance equal to about 10% of the stroke. The size of the water inlet and outlet pipes should be .15 times the diameter of the cyl- inder. Two methods are in use for making the core for the water-jacket space. In one the core-box consists of two cylinders of wrought iron, placed one within the other and of the proper dimensions to make the space between them equal to the size of the core. The core is then rammed up between the two cylinders and put into the core- oven in the core-box. With this method of making the cores there is no necessity of making draft in the core-box for the metal about the studs. Another method of mak- ing the core is to make one half of the core at a time, using the same core-box for each half, the core being lifted out sideways. With this form of core-box it is necessary to arrange the metal as shown by the dot- ted lines in the figure in order to make a draft on the parts y y. The cylinder head, when it contains no valves or other mechanism, is made as shown in Fig. 28. The formulas for the thicknesses of the walls and the depth of the water-jacket are the same as for the cylinder. In fact, it may be said to be an extension of the cylinder. In some engines the end of the cylinder is closed and is made in the form of a sphere. In others the cylinder is closed but a head is used to cover the end of the cylinder in such a manner that it forms the outer wall of the water-jacket at that point. In many en- gines, the valves are placed in the head while in others, the head contains but one valve or the igniter. 135 CHAPTER XVII. VALVES AND VALVE-BOXES. The methods employed for handling the charge and the exhaust was discussed in Chapter X. The proportions of the valve passages, the valves and the valve-boxes will be discussed in the present chapter. The proportions of the passages for both the entering charge and the exhaust should be founded upon the speed at which the gases will be compelled to pass through them. Hence the areas of these passages depend upon both the area of the cylinder and the speed of the piston. Too frequently these areas are made smaller than they should be, with a consequent wire-drawing and loss of power. A careful discussion oi this matter with prominent designers, both in this country and in Europe, shows the concensus of expert opinion to be, that the speed of the gases should be limited to loc feet per second in the inlet, and to 85 fee' 136 per second in the exhaust passages. The 85 feet per second allowance for the exhaust, is made on the assumption that the prod- ucts of combustion are driven from the cyl- inder when at atmospheric pressure, this assumption being made for convenience only. As a matter of fact the pressure of the gases at the moment of release range between 30 and 40 lb., and the lower limit of speed for the gases in the exhaust pas- sages is adopted for this very reason. As the piston speed of gas engines varies con- siderably, even for engines of the same power when built by different manufactur- ers, it should always be taken into consid- eration when proportioning the passages. It should also be remembered that to make the areas of the inlet and the exhaust pipes according to the above, and to then choke the column of gas by a small valve opening or cylinder port, is not very good practice. The formulas given below will give the de- signer a short method of determining the sizes of the passages : Let .9 = the speed of the piston in feet per minute ; A. = the area of the cylinder ; a = the area of the inlet passages ; a 1 = the area of the exhaust passages; D = the diameter of the cylinder; d = the diameter of the inlet passage ; 137 -p- - /i d l = the diameter of the exhaust pas- sage ; L the length of the stroke in inches ; R = the r. p. m. of the crankshaft; then (16) ora = ARL (i6a) for - 6,000 36,000 the inlet. l = -(17) or a 1 = ARL (lya) for/ 5,100 30,600 the exhaust. for the inlet (i6b). d l = . 00572/2 ^ /?/, for the exhaust (lyb) When the piston speed of the engine is at the usual limit for horizontal engines these formulas may be still further simplified to : a .1 A (i6c); a 1 = .12 A (iyc) ; and for passages of circular cross-section to dl = .35 D (i 7 d). The designer should use the formula best suited to his work and he will find that, in many cases, he may calculate the dimen- sions by one formula and use another as a check. The proportions of the valve are shown in Fig. 29. The author does not insist that these proportions should be followed in every case but they form a very convenient 139 140 basis for the design of a valve, especially in those cases where it is hard for the designer to decide what to do, as the ratios suggested below will give a valve of good proportion. The dimensions are based upon the diam- eter of the opening as a unit. They are : V= 1.14 d; b = .14 d\ c--= .07 tf; s -= .2 d, for short stems ; s -= .23 d, for long stems ; r = .25 d, the minimum lift of the valve, to which should be added J^ inch to allow for wear. At B is shown the usual method of mak- ing an exhaust valve, the head of the valve being riveted to the stem. This is neces- sary because the head of the exhaust-valve should be made of cast iron, as wrought iron or steel will not stand the abrasive action of the hot gases. Three methods of arranging the valves 011 the engine are shown in Figs. 30, 31 and 32. Fig. 30 is a cylinder head arranged for both the inlet and the exhaust-valve, and the proportions of the various dimensions to the cylinder diameter are marked upon the different parts. It is always a good practice to water-jacket the exhaust-valve, so that the heat may be carried away from both the bearing for the stem and the valve-seat. 141 142 In the 11 the figure, the inlet-valve is also shown water-jacketed. It is not necessary to jacket the inlet-valve, as the incoming air suffices to keep it cool. The jacket is continued about the valve in this case, in order to save iron and to retain a uniformity of construc- tion. In Fig. 31 is shown a valve-box designed to be attached to the side of the cylinder and to so place both valves that their stems are parallel and point in the same direction. This box is a style used on both horizontal and vertical engines and is employed where the camshaft is at a right angle to the axis of the cylinder and between the valve-box and the crankshaft. In order that the valves may be removed from the box with- out taking them apart, openings are made in the box at Jt and S and the holes are closed with a cap A. The cap is fitted to the box with a ground joint and is held in place by the clamp B. The proportions of the various parts of the box are shown by formulas on the figure. But one view of the box is shown, a cross-section through the axes of the valves. It is bolted to the cylinder by means of studs passing through the holes j, the studs serving also to hold the cover of the box. The inlet gases enter through a cored passage in the cylinder and the opening /, from whence they pass 143 144 through the inlet-valve to the cylinder-port P. From the port P the products of com- bustion pass through the exhaust-valve opening to E and through to a cored ex- haust passage in the cylinder. Fig. 32 is an illustration of a valve-box so arranged us to place both valves in a box of the smallest possible compass. .The ex- 'haust-valve is operated by means of a cam, but the inlet-valve is intended to be ope- rated by the suction of the engine piston. The box is shown in section through the plan. In order to permit of removal of the valves, the seat for the inlet-valve and the bearing for its stem are placed in a separate casting, which is bolted to the box as shown. The opening from the valve- box into the cylinder is shown at B and from the jacket surrounding the exhaust- valve to the jacket of the cylinder at A. The exhaust-valve opening is shown at E and the inlet-valve opening at /. The ex- haust-valve is bolted to the box by means of a flange, while the inlet is screwed into the side of the cover at C. It would be well to put a union into the inlet pipe near the box, in order that the cover may be taken off without difficulty. The dimensions of the the various parts of the box are shown b} equations on the figure, as in Figs. 30 and 31, u.s CHAPTER XVIII. THE PISTON, THE COXXECTING-ROD AND THE CRANKSHAFT. With few exceptions, all gas engines are fitted with trunk-pistons. The principal reason for the adoption of the trunk-piston is, that if the engine were made double-act- ing, as are most of the steam engines, the hot gases would surround the piston-rod heating it to a temperature that would cause it to cut in the stuffing-box. The trunk piston taking an impulse on the end where there would be no rod were the engine double-acting, gives no trouble from this source. As the other end of the cylinder is not used, there is no necessity for closing it, and by making the piston long enough to act as a guide, there is no need for either a piston-rod or a cross-head. A few engines are manufactured which are double-acting, the trouble from an overheated piston-rod being avoided by providing means for cool- ing the rod. In one of these engines the water-jacket is extended so that it surrounds 146 ie stuffing-box, while in another, the rod is made hollow and water is circulated through it. The present discussion will be confined entireh- to the trunk-piston. For a treat- ment of the piston for the double-acting engine, the reader is referred to works on steam-engine design. An example of the trunk-piston, as usually employed with the gas engine, is shown in Fig. 33. The pro- portions are, with the exception of the pin, based upon the cylinder diameter as a unit. All the necessary equations are placed upon the drawing. For small pistons of S" diameter and under, the ribs R may be omitted, and the piston designed as shown by the dotted line. In some engines there is a supplementary piston-ring added, as shown by the dotted lines at Z. Up to and including pistons \Q" diameter three rings will be found sufficient. For sizes above S", four rings should be used, one of which may be placed as shown at Z. In pistons above \2" diameter four rings should be placed between the pin and the inner end. and a fifth ring added at Z if it is desired to place a ring at this point. The extra ring is of somewhat doubtful utility, so far as the packing ring Z is concerned. It is con- tended by some that it relieves the piston of pressure on the cylinder, but as the piston U7 N i N a 148 is not in any way supported by the rings, it is difficult to see what foundation there is for such an argument. In using the equa- tions given in Fig. 33, the designer should be careful to observe the limits, on the smaller sizes, that it is necessary to consider for securing good castings. The piston wall should not be less than T \ x/ , and even then the pattern should allow considerable stock to be turned out in the lathe. The difference in the thickness of the rings at the smaller and the larger ends may be less or more than T ^ x/ according to the method of turning them up. A ring which would be of equal strength throughout would taper off to nothing at the smaller end. In the shop when turning piston-rings for a gas engine, some method should be followed whereby the outside of the ring is turned to a perfect circle, the diameter of which shall be exactly the same as the bore of the cylin- der. In order to do this it is necessary to first cut the rings a"nd then to clamp them in a jig which will hold them sprung very nearly together. The ring is then turned to the diameter of the cylinder while held in the jig. This is the secret of securing a tight piston, and one that will successfully withstand the high pressures due to the increase in compression, which is now so extensively coming into use. 149 "^iiftiz The diameter p of the piston-pin should be determined by the bearing surface re- quired ; the length of the pin should be r wherever possible, the same as that of the crank-pin. The average pressure which is allowable upon the piston-pin per sq. in. of projected area is 7.50 Ibs. In order to find the diameter of the pin, when the length is known, the designer may make use of the following equation. f= ( l8 ) 750 Xe Wherein p = the diameter of the piston pin ; A = the area of the cylinder : P = the M. B. P. ; e = the length of the piston pin. The connecting-rod of a gas engine is made in various styles, but what is known as the marine type, and illustrated in Fig. 34, is that in most general use. This rod is of circular cross-section and is easily ma- chined, as it is nearly all of it turned in the lathe. It is made slightly tapering, the larger end being that nearest the crank-pin. The taper given is dependent to some ex- tent upon the length of the rod, the crank end being from ^^ to ^" larger than the piston end. Thus, for a 24" rod the differ- ence would be ^''^ and on a 60" rod a dif- ference of Y%" , or a difference of }/%" for ?ach foot between centers. The mean di- ameter of the rod, /. ^., the half sum of the diameters of the two ends may be found by means of the formula given below. r = .035 \/w; (19) Wherein r = the mean diameter of the rod ; D = the diameter of the cylinder ; / = the length of the rod be- tween centers in inches ; m = the maximum pressure within the cylinder in Ib. per sq. in. For the convenience of the designer the author has deduced the following formulas for the mean diameter for engines in which the stroke is i% times the diameter of the cylinder. When / = 2, L being the length of the stroke, r== .06 D i/ m (19 a) When / 2.5 L, r = .068 1) \ /- ^ (19 b) l=$L, r === .074 D j 1 /"^" (19 c) These formulas may be still further sim- plified by extracting the fourth root of m for a range of pressures within ordinary use. Thus, when / = 2/, and m is 240 Ib. per sq. in. r .236 D. Call the coefficient of D in this equatioi: /'"and the formula becomes r =* FD. (19 d] The values of F for various proportions of / and L and for different values of m an 152 given in the following table. The designer may take the values nearest to those he expects to use, and from these find the di-. mension of r. When 7=2/. When 7=2.5^ When 7=3/, ;;/ /v . F=^ F = 240 .236 .268 .291 280 .245 .277 .303 320 .253 .288 .313 360 .261 .296 .322 400 .268 .304 .331 The proportions of the various dimen- sions of the brasses, etc., to sizes of the pins are shown on the figure. The crankshaft of a gas engine has to withstand not only the strain due to the transmission of the entire power of the engine, but this strain, at the time it conies upon the shaft, is equal to four times the average power transmitted. This is one reason why the crankshaft of a gas engine must be so much larger than that of a steam engine of the same power. To find the di- ameter of the crankshaft, the diameter of the cylinder and the maximum pressure within the cylinder, is taken as the basis of computation. Let 6"= the diameter of the crankshaft ; D and m as above ; Then for steel ,-.059 D f / ~f^ (20) For wrought iron ,V .064 D i^"^T (20 a) 153 These formulas are lor the average case of L = i% D. When there is much of a departure from this ratio the following for- mula should be used : For wrought iron, 5= .056 ^~^ZJ9 a ~ ( 20 *) For steel, S= .052 f m L D * (20 c) These formulas will give somewhat larger crankshafts than are in use on many gas engines, but it is a matter worthy of note that too many crankshafts are made too small. The diameter of the crankpin is usu- ally made i^ times that of the crankshaft. Practice varies in this respect from making the pin the same diameter as the shaft to 1.25 times the shaft. It is a good plan to select the larger rather than the smaller ratio. Mr. Frederick Grover, an English writer, recommends the ratio 1.2 as the lowest limit. It should be remembered that the larger the diameter the shorter the pin. The projected area of the crankpin should be such that the average pressure does not exceed 400 Ib. to the sq. in. From this the following formula is derived : /= dJ? (j 400^7 Wherein q = the diameter of the pin ; fr= the length of the pin ; A and P as in formula (18). I o- -b + 155 The length of the crankshaft bearing is usually from 2S to 2.5,5", the smaller ratio 2S being that in most general use. Fig. 35 shows a good example of a crankshaft with those proportions not already given, marked upon the figure. 156 CHAPTER XIX. THE ENGINE FRAME. As with many other parts of the gas en- gine, the frame is built in so many different styles that to give an example of every one would take up more room than is at the disposal of the author in this work. A good example of the style of frame in use on a great many horizontal gas engines is shown in Fig. 36, w r ith the proportions of the various parts shown by their relation to the cylinder diameter of the engine. The angle of the parting line of the bearing brasses is taken at 45, in order that the thrust of the piston may not come directly against the stud, but against the frame. The most rational angle for the parting line would be at 90 to the resultant of the average total pressure of the piston and that of the flywheels. This would, however, make the angle a very steep one, and it makes a much neater-looking design to make the angle as shown, while keeping 157 If 158 the pressure of the thrust from the piston, on the frame. The cylinders of gas engines are quite frequently cast with projections on the side, by which they are bolted to the frame, as shown in cross-section at Z. It is claimed that the overhanging piston will not get out of shape when the tem- perature rises while the engine is at work. The author must confess that he has been unable to detect any difference in the effect upon the cylinders with an engine built by either method. In some engines the frame is set directly on the foundation, while with others a sub-base is furnished, which is practically a continuation of the frame. It is usual to make the lower part of the frame hollow when there is no base, and to draw the air supply from this space in order to v make the suction a quiet one. When there is a base, the air is drawn from the hollow portion of the base. 159 CHAPTER XX. FLYWHEELS. The great proportion of idle strokes to those in which work is actually being done upon the piston makes it imperative that the gas engine be supplied with a very heavy flywheel in order to regulate the speed within reasonable limits. For dif- ferent requirements in speed regulation there will be a difference in the weight of the flywheel. Steadiness of speed is usually spoken of as percent variation, for the rea- son that there is no such thing as absolute uniformity of speed, but a variation between limits that is determined by the efficiency of the governor mechanism and the regu- lating power of the flywheel. Thus, a gas engine that is operating with a speed vari- ation of 2% and at an average speed of 200 r. p. m., has a difference between the high- est and the lowest speed of the engine occurring between one impulse stroke and 1 60 ;he next, of 200 X- 2 4 revolutions; and since the average speed of the engine should be 200 r. p. m., the speed should vary be- tween 202 r. p. m. and 198 r. p. m. It is as well to note, at this point in the discussion, a mistake that is occasionally made. This is, to consider the steadiness of speed as the percent of difference of speed between no load and full load. To illustrate : If the engine ran at 220 r. p. m. at no load, and at 212 r. p. m. at full load, then the speed va- riation would be considered as 220 212=8 r. p. m., and the engine said to regulate within 8-^-216=3.7%" nearly. A little reflec- tion will show that the steadiness of motion is not to be determined by the difference between the speeds at no load and at full load, this matter being one that is depend- ent in a great measure upon the regulating power of the governing mechanism ; or, in other words, upon the efficiency of the gov- ernor. The steadiness of speed between one impulse and the next is dependent entirely upon the power for storing energy that is contained not only in the flywheel of the engine but in the moving parts of the machinery it is driving. Thus, the ar- mature of a dynamo adds somewhat to the power of storing energy, as do the pulleys, etc., of any pieces of machinery being driven. In the operation of electric gener- 161 ators for incandescent lighting, the regu- lating power of the engine flywheel is quite frequently augmented by the use of an ad- ditional flywheel on the armature shaft, oil a jack-shaft between the engine and the dynamo, or on both the dynamo shaft and the jack-shaft. It is obvious that the rational method of calculating the energy storing power re- quired in the gas-engine flywheel is to make a graphical diagram of the operations that take place within the engine cylinder and to transform them into a diagram show- ing the resultant effect of these forces upon the crank-pin. There are two ways in which this may be done. One of these methods considers only the effect of the pressures within the cylinder, while the other and most accurate method brings in the effect of the reciprocating parts of the engine, /. ^., the piston, the connecting-rod and the cross-head when it is employed. Both of these methods require tedious calculations, and the reader is referred to works of greater scope than the present one for description and explanation of these methods. For the case of a gas engine operating under aver- age conditions the following formulas will be found to give good results in practice. They give values that are within a very few percent of those obtained from the actual 162 diagram of the engine, and show good re- sults in practice : Let W = the weight of the flywheel rim in pounds ; f= the diameter of the flywheel at the center of gravity of the rim in inches ; N= the r. p. m. of the crankshaft; E= the coefficient of unsteadiness per- missible ; I. H. P. X 111,600,000,000 Then W= f* N *R ( 22 > when the engine has an impulse at every fourth stroke. If the I. H. P. is taken as the power of the engine when operating at less than the maximum number of explosions, the value of the figure in the numerator must be increased in proportion. Thus, should the calculations be made for an I. H. P., taken when the engine is getting an impulse only every eighth stroke, the weight found by the above formula must be multi- plied by 2. The value of the coefficient E for the various classes of work is given by competent authorities to be: For pumping water and all ordinary duties 05 ; For driving machine tools 03 ; For driving textile machinery.. .025; For driving, dynamos 02 ; For driving spinning machinery .01. 163 164. If the flywheel on the engine is to be assisted by flywheels on a jack-shaft, or on the armature shaft of a dynamo or other machinery it may be driving, the power storing effect should be divided among the various flywheels. This is done by first deciding what power is stored by the fly- wheels on the engine at the coefficient of steadiness employed, and then finding by means of formula (22) what weight of fly- wheel is required on the jack-shaft to store the remainder, using for f and ^Vthe diam- eter to the center of gravity of the jack- shaft flywheel and the r. p. m. of the jack- shaft. The general proportions of a gas-engine flywheel are shown in Fig. 37. Taking the diameter of the crankshaft as a basis of the dimensions of the hub and the spokes, the proportions are as follows : h = ic. i = y d = .8 to I.2C. b = .\d to .$d about. This dimension should, however, be calculated from the following formula : nNd* Wherein B. H. P. = the maximum brake horsepower of the engine ; ! (23) 165 n = the number of spokes ; N "= the r. p. in. of the crank- shaft. It appears to be the universal practice to use six spokes for a gas-engine flywheel. The length of the hub should be at least 1.5^, and is usually from 1.75 to 2.5 times the diameter of the crankshaft. The di- mensions of the rim must be made to suit the requirements of the engine. The width of the rim will depend upon whether the engine will use flywheel as a pulley or not, and upon the weight required. It should be remembered that the wider the rim the less will be its weight for a given amount of energy storage. The outside diameter is usually from four to five times the stroke of the engine. It is customary to limit the speed of the rim to 6,000 feet per minute, or a maximum diameter of D = -3 N wherein D the outside diameter of the wheel and N ^= the r. p. in. * Kent's M. K- 1'ocketbook. 166 CHAPTER XXI. BALANCB-WKIGHTS. The proper method of balancing an en- gine, and also the correct formula for cal- culating the weight of the balance-weight, is a subject of much discussion among en- gineers. The writer finds it impossible to attempt to give what would be called aver- age practice in this regard, so wide is the difference of opinion. From the conditions under which a gas engine operates, espe- cially engines with a single cylinder, it is not possible to balance an engine perfectly by means of a rotating weight. If the engine be balanced for horizontal vibratory effect it is found that it is too heavily counterbal- anced for vertical movement. It is therefore the custom to attempt a medium between the two. Some builders balance what they consider to be the rotating part of the en- gine, and make no allowance for the recip- rocating parts ; others balance the rotating weights, and add to this one half the recip- 167 rocating weight. By the first method only the crankpin, the crankarms and that por- tion of the connecting-rod which is consid- ered as rotating, is balanced. This gives rise to two sets of formulas, which, strange to say, give apparently good results in each case. The question now remains, what proportion of the connecting-rod should be considered as having a rotating effect ? This may best be answered by weighing the connecting-rod in the following man- ner : Support the piston end of the rod 011 a trestle or other convenient support, and let it rest at a point opposite the center of the bearing on a knife edge ; support the crank end of the rod on a platform or other scales, also by a knife edge, and see that the center line of the rod is horizontal. The weight of the crank end as shown on the scale is that which should be taken for giv- ing rotating effect. If, however, five eighths of the rod be taken as rotating, the result will not be far from wrong. Taking the two cases cited above, the following formulas apply : Let B = the weight of the balance-weight; M '== the weight of the crankpin -{- the rotating portion of connect- ing-rod ; K '= weight of reciprocating parts, including the remaining portion 168 of the connecting-rod, the pis- ton and the piston-rod and cross- head in engines that employ them ; J = weight of both crankarms ; m = crank-radius = one half the stroke of the engine ; j = the distance to the center of gravity of the crankarm from the center of the crankshaft ; q = distance to center of gravity of balance-weight from the center of crankshaft ; Then B Mm ' ? for balancing the ro- q tating effect alone ; (24) (M + *)m + Jj , , . And B = :* - for balancing the rotating weight and one half the reciprocating weight. (24 a] The proper place for the balance-weight is considered by a great many designers to be undoubtedly on the crankarms. There are many builders who place a counter- weight on the flywheel near the rim, and some who core out the rim on the side of the flywheel nearest the crankpin. There are two advantages for the latter method; one is that the counterweight may be much lighter, and the other that it is cheaper to make. On the other hand, it is necessary 169 to make some allowance in an increase of the size of the crankshaft over what would otherwise be necessary, in order that it will withstand the wrenching effect of the coun- terbalance when placed in the flywheel. 170 CHAPTER XXII. FOUNDATIONS. Without a good foundation, an engine may be expected to give more or less trouble by vibration, and in time work itself loose from such a foundation as has been pro- vided for it. No engine should be bolted directly to a floor for anything other than a temporary job ; and even when the engine is to be placed upon an upper floor a foun- dation should be built in a hanging frame below the floor. Foundations are usually built of either concrete, stone or brick. On top of the concrete is set a capstone, the best material for which is granite, limestone, bluestone, or any stone of a close-grained structure may be used for the purpose, but the ordinary sandstones will not answer. Concrete should be made of good sand and a good quality of cement and broken stone. The stone should pass through a two-inch ring but not go through a one- .nch ring. Gravel, broken brick and ciii- iers have been used, the latter making a tnuch better concrete than might be sup- posed. The following proportions are those best for concrete : Cement I part. Sand 2 parts. Gravel or stone 5 parts. If the cement is a good quality of Port- land, three parts of sand may be used. The concrete should be laid in a crib in layers not over six inches deep, and each layer should be thoroughly tamped before the next is put in. A very good foundation may be made cheaply and quickly in the following manner: Lay a brick wall in cement mortar, making the outside bound- ary that of the finished foundation; then fill in the center with layers of brick or bats, laid flat as in a wall, and loosely. The intermediate spaces should then be filled in with a mixture of one part cement to one part sand. All foundations should be al- lowed three days, at the least, to set before the engine is put in place, and a week is a much better allowance, if there is time to wait that long. It is always best to start the foundation from solid rock or hard pan wherever possible, and it should, at the very least, be started from the ground, i Foundations hung from an upper floor or | 172 built upon it, should be placed as close to a wall as practicable. When building the foundation, it is necessary to insert in the concrete, gas pipe of an internal diameter equal to twice the outside diameter of the foundation bolts. The nut or threaded plate into which the foundation bolt screws is placed at the bottom of the tube, or the foundation bolt is set in with the tube. After the engine is in place the tubes should be filled with cement. Capstones are not necessary with the smaller engines, and it is often a good plan to lay wooden beams on top of the foundations, and to then put the engine on top of them, so that when the engine frame is bolted down it beds itself into the timber. The timber cap often stops an annoying vibration when it can be overcome in no other way. In order to determine the proper dimen- sions of an engine foundation for any en- gine the following formula will be found useful : L,et F = the weight of the foundation ; E = the weight of the engine ; R = the r. p. m. ; Then F = .21 E 1 /~^~ (25) The weight of brick per cu. ft. is 112 Ib, in the wall or foundation. The average weight of concrete is 137 Ib. per cu. ft. It is not customary to make any difference in 173 174 the foundation design for the various ma- terials, and the author has found that the practice of many designers is to base their dimensions on the weight of concrete. As a concrete foundation is % heavier than a brick foundation of the same size, this is evidently an error. The best way would undoubtedly be to make a separate drawing or list of dimensions for brick, or to at least calculate the dimensions of the foun- dation on the basis of 125 Ib. to the cu. ft., an average of the two weights. A very good style of foundation is illus- trated in Fig. 38. It is easily constructed of any material by a fairly good workman. The inclination of wall from the top to the bottom is usually known as the batter, and it is customary to make the batter from 3" to 4 X/ to the foot in height. Instead of building the foundation in tiers, as shown, some designers prefer to make the sides with a gradual slope, as indicated by the dotted lines. The number of foundation bolts varies from four to eight, according to the size of the engine. No less than four bolts are used, and to determine roughly the necessary number of bolts, divide the horsepower of the engine by 6, always using an even number. The sizes of the bolts to use, may be determined approximately as follows : Let //= the horsepower of the engine ; y = the area of one bolt ; k = the number of bolts ; (26) _ k It should be observed that this formula is entirely an empirical one, and that the designer should consider the strain upon the bolt that is farthest from a line directly under the crankshaft, in cases wherein the shaft is at an unusual distance from the top of the foundation. As a rule it will be found that the above formula gives bolts of ample size for engines of the usual design. In case it is necessary to go very deep to find a solid bottom on which to set the foundation, piles may be driven over an area slightly larger than that taken up by the base of the regular foundation, or a crib of timbers laid in. A sub-foundation of timber or piles should be laid, if pos- sible, at a sufficient depth to keep the tim- ber covered with water or very damp earth. CHAPTER XXIII. MISCELLANEOUS FORMULAS. The following formulas may prove them- selves useful to those who design or handle gas engines : For the diameter of the camshaft : Let c = the diameter of the camshaft ; D = the diameter of the cylinder in inches ; Then c = .057 D -|- .625. For the volume of the muffler ; Let M = the volume of the muffler in cu. in. ; L = the length of the stroke in inches ; Then^/= 3.5 />. For a closed circulating system of cooling the cylinder the capacity of the tanks is made from 20 to 50 gallons per I. H. P. Thirty gallons is usually considered an ample allowance. The allowance for evap- oration is o.i gallon per D. H. P. per hour. For water flowing through the jacket Mr. 177 Frederick Grover recommends 4^ gallons per I. H. P. per hour. The proper size of marine engine to use for launches up to 50 ft. long may be roughly estimated by means of the following equa- tion : Let JB= the length of the boat on the load waterline in feet ; H = the horsepower of the engine ; Then H=* f 9. For the smaller craft, to which so many gasoline engines are applied, the usual rules for the diameter and pitch of a screw pro- peller do not seem to be applicable. The author has derived the following empirical formula for pleasure boats which use en- gines of from i to 12 horsepower. Let d = the diameter of the propeller in inches ; H = the horsepower of the engine ; R = the r. p. in. ; Then for a 3-bladed propeller with pitch and diameter equal, rf= V22-S2SL2+ 164 R For a 2-bladed propeller the diameter should be about 8% greater than for the 3-bladed propeller. It may be a new propo- sition to many to make the pitch and the diameter equal, as it is the custom of a great many boat builders to make the pitch 1.3 d. One of the most successful launch builders in this country uses the ratio pitch = diameter, and speeds his engines up to correspond. 179 CHAPTER XXIV. TESTING. Hvery manufacturer of gas engines has a department known as the testing floor, equipped more or less fully with apparatus for determining, before the engine leaves the factory, if it is working satisfactorily. The primary objects of the test are, first, to find out if the governor is so set as to give the engine its proper speed; second, by means of the indicator, to discover if the igniter is properly timed, whether or not the valves open at the proper points in the cycle, and to see if the compression is of the right amount. The gas-engine indica- tor diagram is of value for the determination of other points in the working of the engine, as has been discussed in Chapter XIV. The third object of the test is to determine if the engine is giving the required amount of horsepower at the pulley. This is known as the delivered or brake horsepower, de- noted bv either the letters D. H. P. or B. H. P. 180 It is determined by means of a form of ab- sorption dynamometer known as the prony brake. The fourth object of the test is the determination of the fuel consumption of the engine per I. H. P. per hour. The necessary apparatus for a gas-engine test, as it is usually carried out on the test- ing floor, is as follows : A good steam-engine indicator, with a piston having an area of X square inch, and, if possible, a pencil movement of extra strength, is the princi- pal requisite for determining the I. H. P. There are indicators now to be had that are designed expressly for the gas engine and with the above features. There should also be some good type of reducing motion in order that the movements of the stroke of the piston may be reduced to a stroke of between 2^ x/ and 3^", which is the usual length of the indicator diagram. There should be, if it is possible to obtain it, a test meter reading in single cubic feet in order that the fuel consumption, when the gas engine is used, can be measured with some degree of accuracy. There should also be a prony brake large enough to be attached to a pulley of considerable diameter. The brake should have some means by which its grip upon the wheel may be quickly al- tered, and there should also be a platform fe, and the scale should be accurately 181 tested beforehand with a standard weight to determine if its readings are correct. The experimenter should be supplied with a good form of revolution or speed counter, those being the best which have a soft rub- ber tip for placing in the center countersink at the end of the crankshaft. For a more refined test than is usual on the testing floor of a factory there is needed a means of measuring the jacket water, thermometers for taking the temperature of the water before passing into the jacket and as it is leaving, a pyrometer for determin- ing the temperature of the exhaust gases, a pressure gauge or a manometer for de- termining the -pressure of the gas, a ther- mometer for determining the temperature of the same, a thermometer to be hung on the wall near the engine to determine the tem- perature of the room, and a barometer for measuring the pressure of the atmosphere at the time the test is made. This appa- ratus is necessary only if making the test for discovery of faults in design, and when it is desired to obtain data upon which to make improvements in the engine. It would, however, be a good plan to always take the pressure of the gas, particularly when natural gas is used and the pressure is more than a few ounces per square inch. Otherwise the experimenter mav be led into 182 grave errors by striving to calculate the fuel consumption of his engine. No gas engine should ever leave the factory without hav- ing undergone a test for speed and for de- livered horsepower, nor without several indicator diagrams having been taken from the engine. A gas engine rigged up for the purpose of making a thorough test is shown in Fig. 39. The prony brake is shown applied to the flywheel, and the engine is running over, as shown by the arrow. The brake consists of the strap a with the blocks b bearing upon the periphery of the wheel, and the brake-arm made of the two boards ff. Two cast-iron or wrought-iron angles cc are placed where the band is parted, and through them is put the bolt d> used to tighten the brake by means of the threaded crank e. A platform scale is placed at /, and the thrust of the brake is applied to the scale by the arm f acting through the knife edge g, the iron block h and the wooden stand j. The distance L from the center of the crankshaft to the point where the knife edge rests upon the block h is known as the length of the lever arm. The indicator is shown at z, the cord from the indicator passing to the smaller drum of a reducing wheel r, and a cord from the larger wheel being attached to the piston. i 184 as shown in the small sketch at the upper right-hand corner of the figure. To attach the cord to the piston a piece of ^ x/ round iron is bent into the shape illustrated and flattened at the end opposite the eye m. Through the flattened portion two holes are drilled, and the rod is fastened to the inside of the piston by means of machine screws or capbolts. Water enters the water-jacket through the pipe ze/, controlled by the valve z>, and it leaves the jacket through the pipe w. Thermometers are placed at t and t to de- termine the temperature of the water as it enters and as it leaves the jacket. These thermometers do not set directly in the water but are in small cups filled with oil. In leaving the jacket the water is caught in the box B and either weighed or meas- ured as is most convenient. In case it is desired to weigh the water, two receptacles should be provided and one emptied while the other is being filled. Catching the weight "on the fly" is not an accurate meth- od and should be avoided. When using two receptacles, the stream of water should be changed just at the time the signal is given for taking the reading. When it is desired to measure the water by volume, the receptacle may be made in the form of a box of the following dimen- 185 sions, in order to make the measurement easier to read. If the dimensions of the tank are 37^ x 37^- inches, and the meas- uring stick s is marked off in half-inch divisions, each j^-inch in depth will indi- cate 25 Ib. of water. If a smaller tank is desired, it maybe made 26^ x 26$ and each inch in depth will indicate approximately 25 Ib. of water. These measurements are made for water at a temperature of 150 F., which is about the average temperature of the water leaving the water-jacket. For very accurate determinations the water should, of course, be weighed. Weighing or meas- uring the water is necessary when making a thorough test, in order to find the amount of heat carried off in the water. A meter is shown on the wall at M with the gas bag at G. On the wall near the gas bag, hangs the thermometer T for determin- ing the temperature of the atmosphere and a barometer X for determining the atmos- pheric pressure. The temperature of the exhaust gases, as they leave the engine, is measured by means of the pyrometer p. The temperature of the gas is taken just before it enters the meter, by the thermom- eter t" ', and its pressure by means of the manometer m' . It is also advisable to use a meter for the measurement of the air when it is necessary to compute the ratios of the 1 86 fuel to the air under the various conditions. It has been the custom in many testing rooms to compute the volume of the air from the difference between the quantity of gas entering the cylinder at each working stroke and the volume of the piston dis- placement. Such a method is an erronous one, as there is usually more or less wire- drawing, and a certain quantity of the prod- ucts of combustion remains from the pre- vious charge, hence the air should be meas- ured by a meter. In order that the reader may thoroughly understand the method of making a com- plete test and of working up the data, it will be explained in its entirety. In ordinary testing he may select such portions of the complete test as will suit his purpose, and, before starting the test, a log should be made out as shown on page 188. This log shows at the head of each column just what readings should be taken. The last three columns are for convenience when working up the data after the test, while the balance of the log is for data taken while the test is in progress. In making the test, particularly when a large number of readings are to be taken, the engineer in charge should have a suffi- cient number of assistants, in order that the readings may be taken immediately after 187 S I 'd 'H 'I I81-BAY JOS8 -qoni 'a njoAai .lad spuuod-^oo^ spnnod spnnod u aad 'Aa>[ 1 88 the signal. One man should be posted at the brake to keep the pressure upon the scale as nearly as possible at the same point during one run. Another man should have charge of the indicator, and a third should read temperatures, measure the water, and take the readings of the two meters and the barometer. This gives the third man the greatest part of the work, and a fourth man could therefore be used to advantage. The man in charge of the indicator may also take the speed of the engine and count the number of explosions per minute. For taking the speed of the engine, a continuous counter will be found more convenient than an ordinary speed counter, as the speed can be taken in less time. When there are but two observers, readings should be taken at intervals of ten minutes, and the man in charge of the indicator should take all read- ings, as the brake requires constant atten- tion. The test should be divided into runs, each run being made at the same horse- power throughout. One run should be made at the full power of the engine, another at the rated horsepower, and a third at 110 load with the brake off. The number of runs, at horsepowers other than the above, will depend upon the size of the engine and the time at the disposal of the 189 experimenter. The author would suggest making a run at quarter load, half load, and three-quarter load as well as at the loads already noted. At least ten readings should be taken during each run, and fifteen or twenty readings would be much better, if accurate results are desired. The test should, if possible, be in charge of a competent engineer. He should be provided with a whistle which should be blown as a preparatory signal thirty seconds before the time of taking the reading. Two blasts of the whistle should be the prepara- tory signal and one blast should be the signal for taking the readings. Promptness should be observed, each man taking his post at the preparatory signal. Under no circumstances should an outsider be allowed to interfere, and, above all, no one but an observer engaged upon the test should be permitted to take a reading. Always enter the data upon the regular log. Keeping notes upon loose slips of paper leads to confusion, and it should be avoided. At some convenient time, either before or after the test, the diameter of the piston, the length of the stroke, and the clearance should be carefully measured. In order to measure the clearance, the crank should be placed exactly upon its inner dead center and the valves firmly seated. Carefully 190 ^ weigh a quantity of water and fill the com- pression space so that it is just full and no more, being careful to spill none of the water. Then weigh the remaining water and the difference between the two weights will be the quantity necessary to fill the compression space. Water at 39.1 F. weighs 62.5 Ib. per cu. ft. and dividing the weight of the water taken to fill the compression space, by 62.5 gives the volume in cu. ft. If the water is much warmer than 39.1, the weight may be found by the following formula : 62 ' 5 X 2 = Wt.percu. ft. , / + 461, 500 f3i * 500 t -\- 461 Wherein t = the temperature of the water. This formula gives a close approximation to the correct weight. For the usual hy- drant temperatures, dividing by 62.5 gives results sufficiently close for the purpose. In making up a report of the test, the form shown on page 192 will be found con- venient. The first line should be filled in with the name of the engine and the manufacturer, the second line with the name of the party by whom the test is made or the engineering firm of which he is the representative, and the third line by the locality at which the test is made, to- gether with the date. 191 REPORT OK TEST. (las Engine Test made by At... ..19.. DIMENSIONS OF ENGINE. Diameter of piston In. Area of piston Sq. in. Length of stroke Ft. Piston displacement Cu. ft. Clearance Cu. ft. Clearance. . . Per cent. DATA. Duration trial. . Hrs. Gas per hour Cu. ft. Air per hour .Cu. ft. I Ratio, gas to air Jacket-water per hour. T/b.j Jacket-water temperature, inlet I Jacket- water temperature, outlet i Jacket-water temperature, range. . F. c j Revolutions per minute . . . .Average. Revolutions per hour Explosions per minute Average Explosions per hour Temperature exhaust F. c Temperature room F. c 192 Length of lever arm Ft, Brake load, average Lb, Gas Weight of cubic foot Lb, Air Weight of cubic foot Lb Mixture Weight of cubic foot. . .Lb Specific heat, gas Specific heat, air Specific heat, mixture Heat value cu. ft. gas B. T. U RESULTS. Work Ft. Ib. per ruin Average, W T ork Ft. Ib. per hour Average D. H. P Average Indicated M. B. P Average, Indicated H. P Average Gas per I. H. P Cu. ft Gas per D. H. P Cu. ft Mech. Eff. D., H. P. -:- I. II. P Friction Loss I. H. P. D. H. P HEAT PER HOUR. Supplied by gas B. T. U Absorbed by jacket-water. . . .B. T. U Exhausted B. T. U Absorbed in work B. T. U Radiation B. T. U: Thermal efficiency Percent B. T. U. per I. H. P ' A great many of the items in the report need no explanation, and only those that are not likely to be understood by the reader will be explained. The percentage of clearance is found by dividing the vol- ume of the clearance by the piston dis- placement. The ratio of the gas to the air is the quantity of gas used per hour divided by the quantity of air. In engines which take air into the cylinder at each cycle whether gas enters or not, the ratio may be obtained only when the engine runs with- out missing an explosion. The range of temperature for the jacket water is the difference between the temperatures of the outlet and the inlet water. The weight of the gas may be obtained from the gas com- pany or it may be computed from the results of an analysis of the gas by multi- plying the weight of a cubic foot of each constituent by the percent contained in the gas and adding the results. The following table gives the weights per cu. ft. of those gases which occur most frequently and in the greatest quantities in gases used for gas engines, together with their specific heat at constant volume. 194 L,b. per Spec, Cu. Ft. Heat 0447 .470 Olefines 1174 -33 2 Hydrogen o559 2.406 Carbon monoxide 0783 .173 Nitrogen 0783 .173 Carbon dioxide 1060 .171 Oxygen 1060 .155 These weights are for the gases when at an atmospheric pressure of 14.7 Ib. per sq. in. and a temperature of 32 F. The spe- cific heat of a mixture may be found in the same manner as the weight, by multiplying the specific heat for each gas by the percent contained in the fuel and adding the results. The weight of air at 32 F. and at a pres- sure of 14.7 Ib. per sq. in. is .08082 Ib. The specific heat of air at constant volume is .1688. The weight of a cu. ft. of the mix- ture may be found by multiplying the weight of the gas as already found by the percent of gas in the mixture, and then multiplying the weight of air per cu. ft. by the percent in the mixture. The sum of these two results will be the weight of the mixture. The heat value of the gas should in every case be determined at a laboratory. When it is necessary to know the heat value a sample of the gas should be sent to a lab- oratory for examination, taking samples of 195 the gas at various times and mixing them to obtain an average. The heat values of various fuels will be found in Table I ; but gas is an uncertain quantity in this respect, and the table is not to be relied upon when accurate results are desired. Always, when making a gas-engine test, the volumes of the gases, both air and fuel, should be reduced to a standard tempera- ture and atmospheric pressure for the pur- pose of comparison. It has long been the custom of engineers to use as standard/. the temperature of freezing water and tlit average pressure of the atmosphere at the sea level, or 30" of mercury. To reduce the gas from the volume at which the tesc is made to its corresponding volume at 32" and at 30" of mercury in the barometer, the following formula should be used: Let 7^= the temperature at the time of the test ; /= the pressure in inches of mer- cury ; i 1 -- the volume at this pressure and temperature; / the volume at 30" of mercury and at 32 P. Then for air : 30 x (1 -1-461 ) This formula will also give a sufficiently 10 close approximation when used for gas. The pressure by barometric measure has been used in this formula, as more conven- ient than the ordinary method of comput- ing on the basis of pressure in Ib. per sq. in. In order to reduce the pressure of the gas, when measured in inches of water, to an equivalent pressure in inches of mercury, divide the pressure in inches of water by 13.62. If the pressure of the gas is meas- ured in Ib. per sq. in., multiply this pres- sure by 2.033 to reduce to inches of mercury. These ratios are for temperatures of 32 F., and will be found sufficiently accurate for ordinary temperatures without temperature corrections. The work in foot pounds per minute is computed from the indicator diagram. It is the continued product of the M. B. P., the area of the cylinder in sq. in., the length of the stroke in feet and the number of explosions per minute. The quantity of heat supplied by the gas per hour is the product of the heat value of the gas per cu. ft. by the number of cu. ft. used per hour. The heat absorbed by the jacket water is the product of the temperature range by the weight of the water supplied. The heat absorbed in work is the number of ft. Ib. per hour divided by the number 778, as 778 ft. Ib. is the equivalent of a B. T. U. (British 197 thermal unit). A British thermal unit is the quantity of heat necessary to raise one pound of water through a temperature of one degree Fahrenheit. The heat carried off by the exhaust gases is found by the formula given below : Let S = the specific heat of the mixture ; w = the weight of i cu. ft. of the mix- ture in pounds ; q = the quantity of the mixture in cu. ft. exhausted per hour ; T = the temperature of the room or the average temperature of the air and gas entering the engine ; T' = temperature of the exhaust as measured by the pyrometer ; U = the quantity of heat carried off by the exhaust in one hour. Then U Swq ( T T i. (33 ) The volume of the mixture passing through the exhaust, is the sum of the gas and the air used by the engine, and it in- cludes all air passing through the exhaust for any reason. The heat otherwise unaccounted for, and determined by subtracting the above three results from the heat supplied by the gas, is usually credited to loss by radiation. To be strictly accurate, the quantity of un- burned fuel passing through the exhaust should be determined, and the heat value 198 )f this waste fuel subtracted from that sup- plied by the fuel taken into the engine. The thermal efficiency is the quotient of ';he heat absorbed in work divided by the heat supplied to the engine in the fuel. It is usually written as a percentage, and it is the only basis upon which two engines should be compared for efficiency. The indicated horsepower of a gas engine is determined from the indicator diagram. The area of the diagram should first be found by means of planimeter.* The M. E. P. is determined by multiplying the mean ordinate by the scale of the indi- cator spring. In indicators using a special piston for indicating the gas engine, with an area of the spring of X sq. in., the scale of the spring must be multiplied by two for the scale of the diagram, unless it is ex- pressly marked for the smaller piston. To find the I. H. P., when the M. E. P. is known, the following formula should be used : *For the method of using the planimeter the reader is referred to works upon that subject, as space will not permit of its treatment in this vol- ume. The area of the diagram should then be divided by its length, and the result will be the mean ordinate. 199 / H . P. L (34) 33,OOO Wherein P = the M. E. P. ; / the length of the stroke in feet; a = the area of the piston in sq. in. ; n = the number of explosions per minute. The reader should observe the difference between the value of n in this formula and value of the same letter in the similar for- mula for steam engines. In the formula for the steam engine, n is the symbol for the number of revolutions per minute. To find the D. H. P. of the engine, the length of the brakearm L, Fig. 39 the r. p. m. and the pressure exerted upon the scale must be known. The formula for the D. H. P. is : D. H. P. = .0001904 pin (35). Wherein p = the pressure upon the scale (net) ; / = the length of the brakearm ; ;/ = the r. p. in. The net pressure upon the scale should be found by subtracting the pressure ex- erted upon the scale by the unbalanced portion of the brakearm and the weight of the block /, Fig. 39, from the total pressure exerted upon the scale. In order to find the effect of the unbalanced portion of the brakearm, proceed in the following man- ner : When the engine is not running, have a man stand upon the scale platform and balance his weight, i. e., weigh him. Now balance the scale when he has hold of the brakearm and is pulling upward against a slight friction on the brakewheel or pulley, and again when the man is pushing down on the arm. The brakearm should be grasped close by the knife edge and the scale balanced when the brake is in motion. Add together the weight on the scale when the brakearm is being pulled up and the weight when it is being pushed down, and subtract the weight of the man from half this sum. The result will be the pressure due to the unbalanced portion of the arm. Thus, suppose the man to weigh 160 Ib. and that when the arm is being pulled up the scale balances at 180 Ib., while when the arm is being pushed down, the scale balances at 170 Ib., the effect of the arm would be : '7-'- l8o _i6o = i 5 lb. 2 When making a brake test the man in charge of the brake should keep constant watch of the scale beam, with his hand always on the lever e, Fig. 39. Owing to the wide fluctuations in pressure upon the piston of a gas engine, especially in a hit- and-miss engine, this is no easy task, and a good man should be selected for this post. The scale beam should be kept floating at all times, otherwise all the computations for D. H. P. will have no good foundation. When making a brake test of any duration, or when the engine is a large one, running water should be kept on the band while the engine is going. On the factory testing floor, it is not cus- tomary to do any more than take the D. H. P., the I. H. P. and the fuel consumption. Sometimes a brake test only is made. It is always best, however, to take a sufficient number of indicator diagrams to determine if the ignition is properly timed and if there is any derangement of the valves. CHAPTER XXV. SELECTION. In selecting a gas engine beware, first of all, of the oily tongue of the salesman. Every gas-engine manufacturer makes the ' best gas engine on earth." Yet the mar- ket is overstocked with poor gas engines. To find out what gas engine is best adapted to your special line of work, consult several users of gas engines who are employing the engines for the same class of machinery which you intend to drive. The engine that will give satisfactory service pumping water may be a good engine for running an electric light plant, but it is doubtful if the regulation is sufficiently close for the purpose. On the other hand, the engine having heavy flywheels and a sensitive governing device, built expressly to meet the rigid requirements of an electric lighting service, is too costly a machine to be pur- chased for pumping water, because a cheaper engine will answer the purpose just as well, and the difference may be placed in the owner's pocket. When you go to consult your neighbor with reference to his engine be prepared with questions regarding the operation of the engine, and also make it a point to see the engine in operation and watch it at work. Then find out as nearly as possible how much the engine costs the owner for repairs, and how often and for how long the engine has been laid up for repairs. Deter- mine the time since the engine left the fac- tory, and whether it has been in constant service ever since that time. If the engine is in a filthy condition, or if it is running with its parts out of joint when it is apparent that they could just as well be adjusted to run properly, score a few points in favor of the engine and against the engineer. An engine that will do fairly good work under bad management has something to recom- mend it. If, however, the engine is clean and all possible adjustments made, and yet is running in a noisy, jerky way, score a point against it. If the engine is counterbalanced with the counterweights in the flywheel, stand in a position in line with the axis of the cylin- der and observe if the flywheel is running true at the side. If it sways back and forth it is a sign that the crankshaft is not strong 204 enough to withstand the strain produced by the counterweight. Determine if the bearings give much trouble from overheat- ing, especially when the engine is working under full load. Ask if the' igniter mechan- ism gives much trouble, and if frequent re- newals of its various parts are necessary. In fact, the necessity for the frequent re- newal of any part of the engine is a black mark against it. Find out how much attention, 011 an average, the engine requires per day, and just what the nature of this attention is. If the engine is required to run a dynamo as a considerable proportion of its load, watch the lights when the dynamo is run- ning. If the lights show no perceptible winking, when the light is observed indi- rectly, the engine will give satisfactory ser- vice for electric lighting purposes. A good way to test a light is to try to read fine print by it. If the light throbs to too great an extent, the throbbing will be quite distinct and the eyes will quickly tire. If the pres- sure 011 the lines is 100 volts, the swing of the volt-meter needle should not exceed two volts when swaying back and forth. On a 5o-volt circuit this swing should not exceed one volt, or 2% of the pressure, in any case. If the engine shakes considerably each 205 time it receives an impulse, the trouble is due to an insufficient foundation, but if this shaking continues when the engine is run- ning light, i. e., between explosions, the effect is due to improper balancing. In general, select an engine with the fewest number of parts, and with the parts so arranged that they maybe easily reached in case of accident or when repairs of any kind are necessary. Don't think, because the parts are encased, that the engine is a simple one, for the opposite may be the case. An engine with an enclosed mechan- ism may be a very good one for all that. The point that the writer wishes to impress upon the reader is, that it should not be necessary to pull the entire engine to pieces in order to make some insignificant repair or adjustment. 206 LENGTH OF A CYLINDRICAL TANK REQUIRED TO HOLD ONE GALLON. Diam. Length Diam Length Ft. In. per gal. in feet. Ft. In. per gal. in feet. 1 .1704 11 .00141 1 3 .108 11 8 .00131 1 6 .0756 11 (') .00129 1 9 .0556 11 9 .00123 2 .043 12 .00118 > 3 .0337 12 3 .00113 2 6 .0273 12 C> .00109 2 9 .0225 12 9 .00105 3 .0189 13 .00101 3 3 .0161 13 ;; .000970 3 6 .0139 13 6 .000934 3 9 .0121 13 9 .000901 4 .0106 14 .000868 4 3 .00943 14 3 .000838 4 6 .00841 14 6 .000809 4 9 .00755 14 9 .000782 5 .00681 1") .000756 5 3 .00617 15 3 .000732 5 6 .00564 15 6 .000708 5 9 .00515 15 9 .000686 6 .00472 16 .000665 6 3 .00435 li 3 .000645 6 . 6 .00403 l(i 6 .000(525 i; 9 .00374 16 9 .000606 7 .00347 17 .000589 7 3 .00323 17 3 .000572 7 6 .00303 17 6 .000556 7 9 .00283 17 9 .000540 8 .00266 IS .000525 s 3 .00250 18 .000511 8 6 .00235 18 6 .000497 8* 9 .00222 18 9 .000484 9 .00210 19 .000471 9 3 .00199 19 3 .000459 9 6 .00189 19 6 .0004 IS 9 9 .00179 19 9 .000437 10 .00170 20 .000426 10 3 .00162 20 3 .00041.". 10 6 .00154 20 6 .000405 10 9 .00147 20 9 .000395 g w JH ^ </} 3 I fi- 3 rf M Jl O IIU* -2* R.d r l-COXJ^^i-ii-li-Hr-^CCXXXXXXJXX II THr-KNC^OOL^r>CiO^^GCC-X 1C O OS O CO '-b OS CO 10 X X" 00 CO ?0 O X Tt "T O T I r-J CO 1C 30 "^ O CO 1 . CC CO I- Oi Ci CC I> O T-KMro^t^ocato^^ooo ^-- i - i-('MCOT}HLOCOO'M II 5 1C iC lO - rjl iO i I O 1^- r 5 00 O CC O C5 C t X ' T Nom Inter 208 CIRCUMFERENCES AND AREAS OF CIRCLES ADVANCING BY EIGHTHS. Diam. Circum. Area. Diam. Circum. Area. 1-64 .04909 .00019 1. 3.1416 .7854 1-32 .09818 .00077 1-16 3.3379 .XS66 3-64 .14726 .00173 y& 3.5343 .9940 1-16 .19635 .00307 3-16 3.7306 1.1075 3-32 .29452 .00690 M 3.9270 1.2272 /^ .39270 .01227 5-16 4.1233 1.3530 5-32 .49087 .01917 'A/ 4.3197 1.4849 3-16 .58905 .02761 7-16 4.5160 1.6230 7-32 .68722 .03758 1^ 4.7124 1.7671 9-16 4.9087 1.9175 34 .78540 .04909 5 /s 5.10'VL 2.0739 9-32 .88357 .06213 11-16 5.3014 2.2365 5-16 .98175 .07670 % 5.4978 2.4053 11-32 1.0799 .09281 13-16 5.6941 2.5802 % 1.1781 .11045 % 5.8905 2.7612 13-32 1.2763 .121)62 15-16 6.0868 2.Dls:-5 7-16 1.3744 .15033 15-32 1.4726 .17257 >. 6.2832 3.1416 1-16 6.4795 3.3410 X 1.5708 .19635 i ' 6.6759 3.5466 17-32 1.66DO .22166 3-16 6.8722 : 5. 75s: 5 9-16 1.7(571 .24850 I/ .0686 3i97Cl 19-32 1.865:5 .276S8 5-16 .264!) 4.20CO % 1.9635 .30680 :|/ .4613 4.4301 21-32 2.0617 .33824 7-16 .6576 1.66C1 11-16 2.1598 .37122 1 A 7.8540 4.90F7 23-32 2.2580 .40574 9-] 6 S.050. , 5.1572 5^ 8.2467 5.4110 M 2.3562 .44179 11-16 8.44:50 5.6727 25-32 2.4544 .479:57 /4 8.6394 5.9396 13-16 2.5525 .51849 13-16 8.8357 6.2T26 27-32 2.6507 .55914 i/ 9.0:521 ().4918 29-32 2.7489 2.8471 .60132 .64504 15-16 9.2284 6.7771 15-16 2.9452 .69029 3. 9.4248 7.06S6 31-32 3.04:54 .73708 1-16 9.6211 7.3662 209 CIRCUMFERENCES AND AREAS Continued. Diam. Circum Area. Diam. Circum. Area. ' 3-16 9.8175 10.014 7.6699 7.9798 13-16 18.064 18.261 25.967 26.535 IX" 10.210 8.2958 % 18.457 27.109 5-16 10.407 8.6179 15-16 18.653 27.688 '.y 10.603 8.9462 7-16 10.799 9.2806 6. 18.850 28.274 i/ 10.996 9.6211 i/ 19.242 29.465 9-16 11.192 9.9678 ~y 19.635 30.680 5/ 11.388 10.321 % 20.028 31.919 11-16 11.585 10.680 1/2 20.420 33.183 13-16 11.781 11.977 11.045 11.416 P 20.813 21.206 34.472 35.785 12.174 11.793 % 21.598 37.122 15-16 12.370 12.177 7. 21.991 38.485 4. 12.566 12.566 }'$ 22.384 39.871 1-16 12.763 12.962 % 22.776 41.282 YB 12.959 13.364 % 23.169 42.718 3-fe 13.155 13.772 Yi 23.562 44.179 % 13.352 14.186 / 23.955 45.664 5-16 13 548 14.607 3^ 24.347 47.173 '*% 13.744 15.033 /& 24.740 48.707 7^16 13.941 15.466 14.137 15.904 8. 25.133 50.265 9 7 16 14.334 16.349 i '\ 25.525 51.849 /^ 14.530 16.800 / 25.918 53.456 11-16 14.726 17.257 '% 26.311 55.088 14.923 17.728 % 26.704 56.745 13-16 15.119 18.190 % 27.096 58.426 /8 15.315 18.665 d/ 27.489 60.132 15-16 15.512 19.147 % 27.882 61.862 -1. 15.708 19.635 9. 28.274 63.617 "l-16 15.904 20.129 % 28.667 65.397 % 16.101 20.629 % 29.060 67.201 3-16 16.297 21.135 II 29.452 69.029 M 16.493 21.648 29.845 70.882 5-16 16.690 22.166 a 30.238 72.760 % 16.886 22.691 3^ 30.631 74.662 7-16 17.082 23.221 /8 31.023 76.589 i/ 17.279 23.758 9-16 17.475 24.301 10. 31.416 78.540 17.671 24.850 I/ 31.809 80.516 11-16 17.868 25.406 J;t 32.201 82.516 CIRCUMFERENCES AND AREAS Continued. Diam. t Circum. Area. Diam. Circum. Area. 10.% 32.594 84.541 15. K 47.909 182.65 32.987 86.590 '&/ 48.302 185.66 1 33.379 33.772 88.664 90.763 1 48.695 49.087 188.69 191.75 7/ 34.165 92.886 'A/ 49.480 194.83 H 49.873 197.93 11. 34.558 95.033 34.950 97.205 16. 50.265 201.06 l/ 35.343 99.402 i/ 50.658 204.22 % 35.736 101.62 *% 51.051 207.3U s 36.128 103.87 '%> 51.444 210.60 % 36.521 106.14 i^ 51.836 213.82 % 36.914 108.43 52.229 217.08 J X 8 37.306 110.75 3^ 52.622 220.35 ^8 53.014 223.65 12. 37.699 113.10 i^ 38.092 115.47 17. 53.407 226.98 /* 38.485 117.86 % 53.800 230.33 % 38.877 120.28 i/ 54.192 233.71 H 39.270 122.72 % 54.585 237.10 5/ 39.663 125.19 /-/ 54.978 240.53 '% 40.055 127.68 zi 55.371 243.98 % 40.448 130.19 :{/ i 55.763 247.45 j 56.156 250.95 13. 40.841 132.73 % 41.233 135.30 18. 56.549 254.47 M 41.626 137.89 i/ 56.941 258.02 % 42.019 140.50 /4 57.334 261.59 i^ 42.412 143.14 : Vs 57.727 265.18 42.804 145.80 1? 58.119 268.80 43.197 148.49 5/'' 58.512 272.45 /b 43.590 151.20 % 58.905 276.12 % 59.298 279.81 14. 43.982 153.94 ^8 44.375 156.70 19. 59.690 283.53 % 44.768 159.48 i^ 60.08;; 287.27 % 45.160 162.30 /4 60.476 291.04 I/ 45.553 165.13 3/ 60.868 294.83 45.946 167.99 i| 61.261 298. 65 K 46.338 170.87 "> s 61.654 302.49 % 46.731 173.78 :? ( 62.046 306.35 J? 62.439 310.24 15. 47.124 176.71 K 47.517 179.67 20. 62.832 314. 1 CIRCUMFKRKNCKS AND ARK AS Continued. IMam. Circum. Area. Diam. Circum. Area. CO. % 63 22-) 31* 10 2. 78.540 490.87 X 63.617 322.06 Ye 78.933 495.79 % (U.010 326.05 y* 79.325 500.74 Y* 04.408 330.06 % 79.718 505.71 7* 61.795 334.10 Y 80.111 510.71 i\ 65.188 338.16 % 80.503 515.72 65.581 342.25 / 80.896 520.77 /Q 81.289 525.84 > | C5.973 346.36 A^ C.'i :)66 350.50 26. 81.681 P30.93 y <;C).759 354.66 YB 82.074 536.05 / 67.152 358.84 H 82.467 541.19 i 67.544 (i7.9:57 363.05 367.28 % 82.860 83.252 546.35 551.55 :$ 4 68.330 371.54 5 /8 83.645 556.76 % 68.722 375.83 % 84.038 562.00 Y*. 84.430 567.27 22. (>9.115 380.13 N 69.508 3S4.46 27. 84.823 572.56 g 69.900 388.82 ', 85.216 577.87 P 70.293 393.20 ', 85.608 583.21 70.686 397.61 :: , 86.001 588.57 % 71.079 402.04 ^2 86.394 593.96 A 71.471 406.49 H 86.786 599.37 Ys 71.864 410.97 % 87.179 604., il % 87.572 610: ?7 23. 72.257 415.48 H 72.649 420.00 28. 87.965 t>15.7 -> 8 73.042 424.56 % 88.357 621.26 % 7:5.435 429.13 ', 88.750 626.80 s 73.827 4:5: 5.74 % 89.143 632.36 's 74.220 438.36 '- 89.535 637.94 % 74.613 443.01 :>^ 89.928 C.43.55 75.006 447.69 %! 90.321 649.18 Ys t)0.7i:j 654.84 24. 7.").39S 452.:5!) X 75.791 457.11 29. 91.106 660.52 M 76.184 4C.I si; i ( J1.1!I ( .> 066.23 % 76.576 4(16.64 Y4 91.892 671.96 8 76.969 471.44 :; s 92.284 677.71 -* 77.362 476.26 ', 92.677 683.49 ; , 77.754 4S1.11 .-, 93.070 689.30 % 78.147 485.98 '% 93.462 695.13 Y* 93.855 700.98 CIRCUMFERENCES AND AREAS Continued. Diam. Circum. Area. Diam. Circum. Area. 30. 94.248 706.86 35. 109.956 962.11 % 94.640 712.76 Ys 110.348 969.00 V* 95.033 718.69 % 110.741 975.91 % 95.426 724.64 '% 111.134 982.84 H 95.819 730.62 ft 111.527 989.80 n 96.211 736.62 II 111.919 99K.78 % 96.604 742.64 74: 112.312 1003.8 jl 96.997 748.69 % 112.705 1010.8 81. 97.389 754.77 36. 113.097 1017.9 y. 97.782 760.87 H 113.490 1025.0 '4 98.175 766.99 g 113.883 1032.1 < 98.567 773.14 \7 78 114.275 1039.2 s 98.960 779.31 ft 114.668 1046.3 jl 99.353 785.51 || 115.061 1053.5 % 99.746 791.73 ^ 115.454 1060.7 % 100.138 797.98 % 115.846 1068.0 32. 100.531 804.25 37. 116.239 1075.2 i/ 100.924 810.54 ^ 116.632 1082.5 '-1 101.316 816.86 % 117.024 1089.8 \ 101.709 823.21 % 117.417 1097.1 H 102.102 829.58 u 117.810 1104.5 ''s 102.494 835.97 XH 118.202 1111.8 i 102.887 842.39 % 118.596 1119.2 % 103.280 848.83 % 118.988 1126.7 38. 103.673 855.30 38. 119.381 1134.1 % 104.065 861.7!) T-/ 119.773 1141.0 & 104.458 86S.31 % 120.166 1149.1 % 104.851 874.85 % 120.559 1156.6 X 105.243 881.41 % 120.9ol 1164.2 % 105.636 888.00 'V 121.344 1171.7 % 106.02!) 894.62 % 121.737 1179.3 7/8 106.421 901.26 % 122.129 1186.9 34. 106.814 907.92 39. 122.522 1194.6 H 107.207 914.61 i/ 122.915 1 202.3 i 107.600 921.32 4 123.308 1210.0 ' ' s 107.992 928.06 % 123.700 1217.7 K 108.385 934.82 1 124.093 1225.4 ''s 108.778 941.61 124.4X6 1233.2 :! 4 109.170 948.42 j2 124.878 1241.0 7 .s 109.563 955.25 % 125.271 1248.8 213 CIRCUMFERENCES AND AREAS- -Concluded. Diam. Circum. Area. Diam. Circum. Area. 40. 125.664 1256.6 45. 141.372 1590.4 H 126.056 1264.5 % 141.764 1599.3 15 126.449 1272.4 & 142.157 1608.2 % 126.842 1280.3 % 142.550 1617.0 YL 127.235 1288.2 A 142.942 1626.0 % 127.627 1296.2 78 143.335 1634.9 % 128.020 1304.2 % 143.728 1643.9 /8 128.413 1312.2 % 144.121 1652.9 41. 128.805 1320.3 46. 144.513 1661.9 H 129.198 1328.3 Ys 144.906 1670.9 i| 129.591 1336.4 % 145.299 1680.0 /9> 129.983 1344.5 >8 145.691 1689.1 J^Z 130.376 1352.7 ft 146.084 1698.2 ftZ 130.769 1360.8 II 146.477 1707.4 % 131.161 1369.0 ft 146.869 1716.5 % 131.554 1377.2 '/8 147.262 1725.7 42. 131.947 1385.4 47. 147.655 1734.9 X 132.340 1893.7 Ys 148.048 1744.2 I/ 132.732 1402.0 H 148.440 1753.5 ? 133.125 1410.3 ' ! s U8.X33 176:2.7 Iz 1:58.51 <S 1418.6 % 149.226 1772.1 /8 133.910 1427.0 % 149.618 1781.4 3% 134.303 1435.4 % 150.011 1790.8 % 134.6% 1443.8 Ys 150.404 1800.1 43. 135.088 1452.2 48. 150.796 1809.6 Yi 135.481 1460.7 i,' 151.189 1819.0 i/ 135.874 1469.1 ft 151.582 182S.5 3^ 136.267 1477.6 % 151.975 1837.9 1^ 136.659 1486.2 % 152.367 1847.5 S 137.052 1194.7 '\s 152.760 1857.0 /4 137.445 1503.3 % 153.153 1866.5 % 137.837 1511.9 H 153.545 1876.1 44. 138.230 1520.5 49. 153.938 1S85.7 % 138.623 1529.2 Y* 154.331 1X95.4 is 139.015 1537.9 i 154.723 1905.0 '&/ 139.408 1546.6 ''s 155.116 1914.7 \y 139. SOI 1555.3 ft 155.509 1924.4 7 140.194 1564.0 Ys 155.902 193 J. 2 74 140.586 1572.8 % 156.294 194:5.9 % 140.979 1581.6 % 156.687 195:5.7 2T4 INDEX. PAGE Areas of circles, table of 209 Automobile engines 103 Automobile engines, cooling cylinders of .... 105 Automobile engines, power of 104 "Backfiring" 37 Balance-weights 167 Balance-weights, formulas for 169 Balance- weights, location of 169 Barometer, use of in testing 182 Batter, in foundations 175 Battery, care of 30 Battery cells 38 Beau de Rochas' propositions 1 Brake arm, determination of effect of 200 British thermal unit, definition of ....... 198 Camshaft, formula for 177 (Jam, laying out a 78 ( -amshaf t, position of 7s Carbon, deposits of in cylinder 37 Carbureters 42, 44 Carbureter, defects of 51 Carbureter, heating a 50 Care of an engine 27 Cells for battery 3s Circumferences of circles, table of 20!' Clearance, percentage of 19 1 Clerk's two-cycle engine 5, <> 215 PAGE Coal, for gas producer '. . 18 Condenser, use of 03 Connecting-rod 151 Connecting-rod, formulas for 152 Connecting-rod, proportions of . . 150 Con tact points, destruction of 6:5 Cooling tanks, evaporation in, per I. H. P 177 Cooling tanks, size of. 177 Crank-pin, formula for -. . . 154 Crankshafts 153 Crankshafts, formulas for lf>;',, 154 Crankshaft, proportions of 155 Cylinder, formulas for diameter of .... 123, 126, 127 Cylinder, proportions of 129-134 Cylinder head 134 Day engine 6, 7, 70 Diagrams 108 Diagrams, examples of actual. fc . 116, 117 Diagrams, formulas for 110, 111 Diagram from gasoline engine 120 Diagram, size of 181 Diesel cycle, comparison with others 15 Diesel cycle, principles of operation of the . . . 9-12 Dimensions of a gas engine 121 Efficiency, thermal ' 199 Electric lights, fluctuation of light in 205 Energy, storage of 101 Engine, shaking of 20^ Exhaust gases, heat carried off by 19s Explosion, cause of weak , 36 Explosions, in exhaust passages 36 Explosions, premature 37 Flame, color of 3.> Flame, trouble with 38 Flywheel, computing weight of ... 162 Flywheel, diameter of 166 Flywheel, formula for 163 Flywheel, proportions of 165 216 PAGE Flywheels; 1M Foundation bolts, formula for 176 Foundations 171 Foundations, formula for weight of 17:'! Foundations, location of 172 Foundations, materials for 171 Four-cycle, comparison with others 18 Four-cycle engine, principles of operation . . . 2, 3, 5 Frame 157 Fuel, power derived from 17 Fuels 17-20 Fuels, table of -'I Gases, specific heat of 195 Gases, weight of 19.~> Gasoline engine, power of a 19, 41 Gasoline engines 41 Gasoline engines, attachments for 42-48 Gasoline engine, handling a 49 Gasoline engine, starting a 49 Gasoline engine, time of ignition in 50 Gasoline fires, how to extinguish 53 Gasoline, power derived from 19 Gasoline, precautions in handling 51 Gasoline, storage of 52 Gasoline tanks, construction of 52 Gasoline tanks, explosions in 52 Gas producer 18 Gas valve, adjustment of 33 Gears, types in use 8:-! Governors 87 Governors, adjustment and care of 33 Governors, examples of 92-% Heat absorbed by jacket water 197 Heat absorbed in work 1.97 Heat carried off by exhaust gases 198 Heat values, table of 21 Hornsby-Akroyd engine - r >9 Horsepower, formulas for 123, 200 217 PAGE Igniter, adjustment of :\t\ Igniter, match 100 Igniter, setting of 2s Igniters 54 Ignition, methods of 54 Ignition, premature 37 Ignition tubes . 31 Indicators isi Jacket water, flow of . . 177 Jacket water, heat absorbed by 197 Jacket water, management of 32 Jacket water, measurement of isr> Jacket water, range of temperature in 194 Jets 44, 4<;, 59 Jump-spark 67 Kerosene engines, vaporizer for 50 Leaks, remedies for :'.9 Make-and-break igniter . . 65, 67 Manometer 1*2 Marine engine, formula for size of 17.S Mean effective pressure, determination of .... 199 Meter for testing 1*1 Muffler, formula for 177 Nash engine > Otto cycle 2. Oil, crank-case 29 Oil, cylinder 29 Oil, lubricating 28 Pipe, table of gas and water 20s Piston, care of - 3:i Piston-pin, formula for 151 Piston, proportions of 1 IS Piston rings 1 17 Piston-rods in gas engines 1 1<> Piston, trunk 1 17 Plauimeter, use of 19'.' Pounding, cause of 39 Power, loss of :>9 2lS PAGE Premature explosions 37 Pressure, coefficients for reduction of 197 Prony brake isi Propellers, formula for 178 Pyrometer 182 Katio of gas to air 194 Resistance, non-inductive 63 Scale, testing a platform 181 Selection 20:1 Slowing down, cause of ;;6 Smoke, cause of 39 Soot, in exhaust passages 29 Spark, cause of weak 38 Speed, formulas for 12.~> Speed variation Kid Speed variation, allowable 16.5 Springs, adjustment of 30, 3<; Starting, rules for 22 Starters 97 Stopping, rules for 25 Suction valve 71 Table of areas and circumferences of circles . . . 209 Table of capacity of cylindrical vessels 207 Table of gas and water pipe 208 Table of heat values 21 Tanks, table of capacity of cylindrical 207 Temperature, formula for reduction to standard . 196 Temperature, reduction to standard 196 Test, apparatus for 181, 183 Test, customary determinations in a factory . . . 202 Test, log for 188 Test, measurements for 190 Test, method of conducting a 187 Test, number of runs in a 189 Test, objects of f , ISO Test, report of ! 192,19:; Testing 180 Thermal efficiency 199 219 Thermometers 1 s - Timing valve 5s Tubes, material for ignition 31 Tubes, proper temperature for ignition 31 Troubles, gas-engine 35 Two-cycle, comparison with others 14 Two-cycle engine, principles of operation ... 7, i> Valves, arrangement of 141-145 Valve boxes, examples of 141-145 Valves, leaks in :!0, :' . Valve mechanisms 73 Valves, proportions of 139, HO Valves, size of 136-139 Valve-stems, care of 34, 40 Vaporizers 4-J, 4f> Water, flow of in jacket 177 Water, formula for weight of ]!>! Water-jacket, deposit in 29 Water, measurement of jacket 1*5 Water pipe, table of 20s Weak spark, cause of :->* Wipe break 63,07 Work, heat absorbed by 197 220 1 **, The... "WATKINS" GAS AND GASO- LINE ENGINES* *jf Only Engine using cMagneto Generator No Hot Tubes. No ^Batteries. Descriptive Catalogue on application. F. ML WATKINS, 309 West Fourth Street, CINCINNATI, OHIO. GASOLINE ENGINES.. STATIONARY, CARRIAGE AND LAUNCH ENGINES. VL to 4 H P * Flange-Cooled and r Water-Jacketed Cylinders. CARBURETORS, GASOLINE MIXING VALVES, PROPELLER WHEELS and SHAFTS. Also complete sets of Castings of Gasoline Engines, with Forgings, Screws and Working Drawings, either in rough or partly ma- ehined, as required. LOWELL MODEL CO., P. O. BOX 292, - - LOWELL, MASS. GAS If you have a Gas Engine, send for a sample of QJXON'S No. 635 GRAPHITE. It will lubricate your Gas Engine Cylinder better than oil. It is not affected by heat. JOSEPH DIXON CRUCIBLE CO. JERSEY CITY, N.J. PATENTS secured in the United States and foreign coun- mmmmmmmmmmi^^fmmmi^i^t trieS. Investigations aS to novelty and validity. Litigation conducted in the Courts and Patent Office. Trade Marks registered. Members of the bar of the United States Supreme Court and various Circuit Courts. BALDWIN, DAVIDSON & WIGHT, 25 Grant Place, WASHINGTON, D. C. 141 Broadway, NEW YORK. THERE/A LITTLE BOOK-TELL/ WHV " WHICH WAY" POCKET LEVEL.. TELLS in an instant "WHICH WAY" your work is out. See? It is the size of a silver dollar and three - eighths thick. Nicely nickeled and polished. To in- troduce it, will mail one for 70c. in stamps or three for $2.00. Caliper catalog free. E. G. SMITH, Columbia, Pa., U. S. A. Attractive Catalogues ^ Carefully Edited^ "WTE offer a service that is W unsurpassed, including p r i n t i n g , half tones, wood engravings and the best typographical display* A me- chanical engineer of demon- strated ability, who is a for- cible writer on mechanical subjects, and of wide experi- ence, is specially engaged to do the editing* We make a specialty of catalogues that will catch the attention of the reader at the start and retain his interest until he has read the entire pamphlet* Make your catalogue not only a work of art, but put informa- tion in it that the recipient will wish to keep* Write us for estimates* We know that we can please you* Address The Ohio Printing Co*, 330 W. 9th Street, - - Cincinnati, Ohio. The Mietz & Weiss KEROSENE ENGINE [Patented.] COMMON KEROSENE ITS FUEL. % CENT PER HORSEPOWER PER HOUR. IGNITES BY COMPRESSION. MOST ECONOMICAL AND SAFEST POWER KNOWN 128-132 MOTT ST., NEW YORK. MARKT &, COMPANY, LTD., EUROPEAN AGENTS. LONDON. HAMBURG. PARIS. WORKING DRAWINGS FOR SALE, 0a$ and Gasoline Engines Horizontal, Stationary, - - - - 12 sizes. Vertical and Marine, ."-."- 4 sizes. Steam Engines. Corliss, simple and compound, - - 12 sizes. Center Crank, 4 sizes. Ant, High-Speed, - - - - 10 sizes. QIDDINQS & STEVENS, Mechanical Engineers, ROCKPORD, ILL. There are 10,000 Laundries in the United States and Canada. THE STARCHROOM LAUNDRY JOURNAL Write for rates. Covers the Entire Trade. The Starch room Publishing Co., Cincinnati, Ohio. Cbe 6as Engine magazine. Stationary, marine, flutomobik. Devoted 10 the interests of an up-to-date power. illustrated. edited by G. 01. Roberts, the 6a$ engine contains tbe latest news relating to tbe das engine and tbe automobile. Special Teatum, '^ e e( ^^ " a ^ s 8 lve a monthly review of the gas engine situation in general. Special Articles, s P edal articles a pp ar each month which deal with the gas engine and kindred subjects. foreign Correspondence, We hav * just made arrange- ments with a well-known expert in Europe to keep our readers informed on the latest developments across the Atlantic. TtettlS. e industrial columns contain information re- garding new enterprises. Automobile news, This column gives se lected items of news, relating to automobiles. The inquiry column is at the service of our subscribers, and is one of the most valuable features of the magazine. All questions are answered carefully and in full. The gas engine is, without a shadow of doubt, the power par excellence of the Twentieth Century. Engineers and mechanics, the world over, are studying its developments with interest. The Gas Engine is the only publication in the English language devoted to this subject. It will keep you informed, and at the insignificant outlay of $1.00 each year for the twelve issues. Send for a sample copy to THE GAS ENGINE PUBLISHING CO., Goodall Build- ing, Cincinnati, Ohio. i"J< THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW AN INITIAL PINE OP 25 CENTS WILL dE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL INCREASE TO SO CENTS ON THE FOURTH DAY AND TO $!.OO ON THE SEVENTH DAY OVERDUE. SEP 27 1939 UNIVERSITY OF CALIFORNIA LIBRARY