SYMPTOMS, CAUSES, AND CORRECTIVES

ENGINE-STARTING AND RUNNING DIFFICULTIES

1. Defective action is sometimes due to causes so apparent that explanations are unnecessary; hence, for the sake of convenience all these possible sources of trouble have been grouped under the headings Causes of Refusal to Start, Causes of Misfiring, and Causes of Weak Explosions. In each case, the cause of the trouble may generally be traced in the last analysis to faulty ignition, a faulty mixture or an insufficient supply of mixture. These broad, ultimate causes have been stated first, and the principal mechanical or electrical defects that produce the trouble are enumerated afterwards. It will be understood that these do not comprise all the possible troubles with engines. In particular, they omit entirely such matters as preignition, knocking, and overheating. The object of the following presentation is to enable the user to trace the difficulty when his engine refuses to give its normal power through some trouble, the nature of which is not immediately obvious.

2. It is a familiar fact that the internal-combustion engine is far more liable to stoppages and weaknesses, for reasons at first mysterious, than is the steam engine. The explanation of this is that, while the steam engine is purely a mechanical apparatus, the internal-combustion engine is partly mechanical, partly chemical, and generally partly electrical in its functions, and the chemical and electrical parts of its organism may go wrong through causes not connected with the visible mechanism, or, as in the case of a badly adjusted trembler, a poorly working timer, or a leaky float, through mechanical derangements so slight as to escape notice.

From this it follows that, to manage successfully, an internal-combustion engine (especially one that works under such a variety of conditions, often very severe, as the automobile engine) it is first of all necessary for the operator to make good use of his reasoning faculties. The symptoms of derangement, when taken singly, are often such as may be caused by any one of several possible defects; in nearly every case the defect, whatever it may be, will produce several symptoms a careful study of which will lead to the elimination of causes that do not tally with all the symptoms; as, for instance, causes affecting all cylinders when only one or two are misbehaving, or vice versa. When the user has reached this point, generally a short further investigation of the points at which he has found trouble of that particular sort is most likely to occur will lead him to the discovery of the true cause. The cause of loss of power, due to such faults as a loose battery connection, a sticking inlet valve, or a bit of dirt in the carbureter, will at once be recognized in its true character by the experienced operator. The only way to attain final proficiency in these things is by extended experience with the particular engine in hand; but, on the other hand, there is absolutely no excuse for the aimless groping of many inexperienced users, who will often send needlessly for a tow, or will pull an engine to pieces in their search for some simple fault that might have been located by intelligent diagnosis.

3. Causes of Refusal to Start, or of Sudden Stoppage. - The fundamental reasons for an engine refusing to run, or of a particular cylinder refusing to work, may be summed up as due to (1) no spark; (2) no mixture; or (3) wholly wrong mixture. These cover all the possible causes, which may be enumerated as follows:

• 1. Switch not closed.
• 2. Gasoline not turned on.
• 3. Carbureter not primed, or (rarely) primed too much.
• 4. Weak battery.
• 5. Gasoline stale or mixed with kerosene.
• 6. Gasoline too cold to vaporize.
• 7. Dirt or waste in carbureter or gasoline pipe.
• 8. Mud splashed into air intake.
• 9. Water in carbureter.
• 10. Soot on the spark plug or contact igniter.
• 11. Water on spark plugs.
• 12. Broken spark-plug porcelain.
• 13. Grounded wire (generally secondary).
• 14. Broken wire (generally primary), or loose connection.
• 15. Very bad adjustment of the coil tremblers.
• 16. Defective spark coil or condenser (rare).
• 17. Broken igniter spring.
• 18. Broken valve stem, spring, or key.
• 19. Valve cams slipped (rare).

4. Causes of Misfiring.-The principal cause of misfiring is irregular sparking, which may be due to a variety of causes. Irregular sparking may be caused by the following:

• 1. Soot on spark plugs or contact igniters.
• 2. Weak battery.
• 3. Broken wire, making intermittent contact through the vibration of the engine (generally found in the primary circuit.
• 4. Loose connection to binding post (generally found in primary circuit).
• 5. Wire occasionally grounded through vibration of engine. This is generally found in the secondary circuit, and it is not necessary for the bare wire to make contact with the metal into which this secondary current is escaping. If the insulation of the secondary cable is weakened, and the cable is lying loosely on a metal part, the spark will often jump through the insulation.
• 6. Timer contact surfaces roughened by sparking.
• 7. Wabbling timer.
• 8. Poor trembler adjustment.
• 9. Trembler sticking at high speeds, due to inertia of heavy armature.
• 10. Insufficient pressure on timer contacts.

A sticking inlet valve, which stays open when it ought to close, will cause irregular firing and occasionally back firing. Another possible cause is a very lean or rich mixture ignitible only by a strong spark. It can always be distinguished from ignition troubles by the fact that the explosion impulses, when they occur, are of much less than normal strength. If the mixture is too weak, the explosions are likely to occur every other cycle.

5. Causes of Weak Explosions.-The causes of the explosions being weak are as follows:

• 1. Mixture too lean or too rich.
• 2. Leakage of compression.
• 3. Mixture diluted by exhaust gases.
• 4. Spark timing later than, it should be, in one or all cylinders.

If the trouble is in the mixture, the explosions would be regular, unless the mixture is so far defective that it sometimes fails to ignite in spite of the spark occurring regularly. The same will be true in any case where, as is usual, the cause of the weakness is unconnected with any irregularity in sparking.

The principal causes of 'weak explosions may be enumerated as follows:

• 1. Dirt or waste in carbureter or gasoline pipe, causing weak mixtures.
• 2. Stale gasoline.
• 3. Air intake partially obstructed, causing rich mixture.
• 4. Bad carbureter adjustment.
• 5. Trouble with float.
• 6. Choked muffler.
• 7. Lack of oil on piston, or too thin oil.
• 8. Leak through valve (generally the exhaust valve).
• 9. Leaky spark plug.
• 10. Valve timing wrong. This is most likely due to the fact that the cam-shaft, etc., have been taken out and replaced with the gears in incorrect angular relation. It may, however, be caused also by wear of the cams, push rods, or valve stems, by spring in the cam-shaft or valve lifters, or by the slipping of cams.
• 11. Broken or worn piston rings.

6. A two-cycle marine engine may be running along smoothly and begin gradually to slow down. This condition may be caused by too much or too little gasoline; the ignition devices may have become disarranged; there may be too little cylinder or other lubrication or too little water circulating through the cylinder jacket; something may be caught in the propeller wheel; in cool or cold weather, the moisture in the atmosphere may have become frozen by the rapid evaporation of the gasoline, thus preventing the free flow of air or the proper seating of the valve in the vaporizer controlling the gasoline supply and the flow of mixture from the crank-chamber; the piston and rings may have been fitted too snugly, causing them to bind in the cylinder, which may have become distorted by the different temperatures to which it is subjected, there being a comparatively cold inlet on one side of the cylinder and a hot exhaust port on the other; the exhaust ports, piping, or muffler may have become partly stopped by water, carbon, salt, or other deposits; the exhaust may have been submerged by a different trim of the boat, or there may have arisen conditions such as could not have been foreseen or provided against, and that might never again be experienced. At any rate, such slowing down is a forerunner of trouble and should be investigated. If the cause of the trouble cannot be discovered, the engine should be stopped when it is safe to do so, the position of the boat being made such as not to endanger either boat or occupants through collision with passing craft.

7. The remedies for slowing-down troubles due to the causes just mentioned will in practice suggest themselves. In many cases, the cause of the difficulty can readily be determined and overcome. For instance, trouble due to an insufficient quantity of cylinder oil or circulating water might be attended to readily without stopping the engine, or a temporary stop might be made to remove a rope, grass, etc., from the propeller, or foreign matter from the sea-cock strainer or pump check-valves, or to adjust the ignition or replace a broken or weak valve spring. Structural troubles, such as tight pistons and distorted cylinders, would have to be attended to at some more opportune time.

If the vaporizer should freeze, it may be necessary to run the engine awhile and then give the accumulation of ice and frost a chance to melt. If the water supply is insufficient and the jacket becomes overheated, it may be possible in case of an emergency to continue running by using a hand pump connected with the supply; or, with the supply open water may be pumped through or poured into the water discharge. In such case, the transformation of the water into steam might make it a little dangerous for the operator, and should the cylinder be too hot the water might possibly crack the cylinder at its weakest part, or at the point where it is subject to the greatest stress.

When it becomes necessary to run a four-cycle marine engine with too little circulating water, the compression should be relieved, the cooling action of the large quantity of gas, a part of which is wasted, helping to cool the cylinder, while the smaller amount exploded does not heat the cylinder as much as would full charges at the usual high compression pressure.

8. Irregular running of marine engines is a condition rarely encountered, and its cause is problematical. The trouble may be caused by back pressure in the exhaust, or may be due to improper location, with reference to the exhaust port, of the transfer, or passover, port connecting with the crank-case; this could occur only in two-cycle engines. As a result of such improper location of the port, the engine cylinder might not be thoroughly scavenged of 'burned gases at high speed, when it would slow down to normal speed or slightly below, and, getting a better mixture at that speed, would speed up. It might also be caused by the exhaust ports opening too late or the inlet ports opening too early. It is well known that, with no thought of fuel economy, two-cycle engine ports should open much earlier when designed for high than for low speed, in order to more thoroughly get rid of the products of combustion. When it is discovered that the engine is being run at a speed in excess of that to which it is best adapted, the remedy is to make the ports open earlier, or hold the engine to slower speed by increasing the diameter, pitch, or blade surface of the propeller.

Should the engine, without missing explosions, begin to increase its speed, and then miss explosions and slow down, one would naturally be led to suppose the cause of the trouble to be insufficient length of contact of the sparking device as well as poor scavenging of the cylinder.

Trouble from loss of compression in the combustion chamber, whether in a two-cycle or a four-cycle engine, must be remedied before the engine can be made to run satisfactorily. If, in attempting to start, it is found that there is no compression, the valves should be examined to see if they seat properly and are timed correctly. Loss of compression may be caused by a leaky gasket, allowing the pressure to leak into the water-jacket, which is the first place to look for the cause of trouble after examining the valves. A leaky, gasket may sometimes be discovered by noting whether or not pressure escaping into the waterjacket shows at the water discharge.

KNOCKING OR POUNDING

9. Undoubtedly the sense of hearing is more useful in detecting irregularities in the running of an engine than any other sense. By means of the sounds produced, the engine talks to the operator, and with a little intelligent study he will soon understand the language. Even at a distance it is often possible to tell whether an engine is running regularly or whether, as indicated by the sound of the exhaust, some of the charges admitted to the cylinder are expelled without being exploded. Standing in close proximity to the engine, the operator may distinguish a variety of sounds indicating defects about the engine and calling attention to the necessity of applying proper remedies 'at the first opportunity.

A sharp, knocking sound in stationary engines may be due to any one of the following causes:

• 1. Lost motion in the bearings of the connecting-rod, either at the crankpin or the piston-pin end.
• 2. Lateral movement of a piston ring, the groove in the piston having become widened by wear.
• 3. A loose key in the flywheel or pulley.
• 4. Lost motion in the gears, causing the gear-shaft to be retarded in its revolution for a fraction of a second when the exhaust or inlet-valve cam hits the roller and lever.
• 1 6. Piston or cylinder worn to a considerable extent, causing an up-and-down movement of the piston.
• 6. The piston having worn a shoulder in the bore of the cylinder, and striking the shoulder if any play in the bearings is developed.
• 7. The piston striking any foreign body that may accidentally have been drawn into the cylinder.

10. Knocking in engines may also be due to looseness or rattle in some external part, owing to nuts having worked loose or to bolts being sheared off or being too small for their holes. Knocking due to such causes is readily detected by a careful inspection while the engine is running, and this inspection may be aided by laying the hands on parts suspected of being loose, when vibration will easily be felt; also by careful scrutiny with an electric flashlight for evidences of movement where two parts are bolted together.

About the most likely place to find looseness of this description is in the holding-down bolts that hold the engine to the frame on which it is mounted; but in certain horizontal engines it way also be found that the caps over the main bearings are loose, owing to the fact that they have not been properly tongued into the bottom halves or pillow blocks of the bearings. Looseness at either of these two points should be remedied at the repair shop, as it always necessitates the substitution of larger bolts, aided perhaps by dowel-pins; and in the case of the bearing cap it may be necessary to make a wholly new cap, with proper tongues fitting into grooves that must be machined or chiseled in the pillow-block.

11. A more probable cause of knocking is looseness due to wear in the main-shaft bearings, crankpin bearings, or the wristpin bearings. In a four-cylinder vertical engine, the main-shaft bearings may be quite loose without causing a knock, because the weight of the shaft and flywheel holds the shaft down; but a horizontal engine will, under certain conditions of speed and load, pound with a small amount of looseness. Only a very limited amount of looseness should be permitted in the main-shaft bearings of any engine, both on account of the danger of springing the shaft and because a bearing worn beyond this extent is liable to begin cutting, as it is difficult to keep sufficient oil in it.

12. Looseness in the flywheel bearing of a vertical engine is disclosed by putting a jack under the flywheel and working it gently up and down. In the case of a horizontal engine it is necessary to move the shaft approximately in line with the pressure of the explosions, and a lever will have to be applied to the flywheel or shaft in whatever manner seems most practicable. Occasionally, looseness of the shaft can be detected by rocking the flywheel back and forth against the compression in the cylinder. If the pull of the sprocket chain comes on the engine shaft, it maybe possible to detect looseness in the adjacent bearing by alternately stretching and relaxing the chain, which can be done by grasping it midway between the sprockets and pulling it up and down as far as it will go.

A novice should not attempt to refit the main-shaft bearings, as this requires a good deal of skill and experience for its correct execution.

Wear in the crankpin bearings is disclosed by setting the cranks at about half stroke, and rocking the shaft back and forth.

13. Knocking in the wristpin, due to wear of the pin and its bushing, is not among the commoner troubles, and it does not need to be attended to at once unless aggravated. It is well, however, not to neglect it too long, as the bushings and the pin will be worn out of round, so that they cannot be used. When it is taken out, the wristpin should be calipered all around. If it is out of round, it should be ground true; or, if this is impracticable, a new pin will have to be supplied, and the bushing reamed or scraped to fit. This, of course, should be done in a repair shop.

14. A cause of knocking occasionally found is due to the wristpin and the crankpin not being quite parallel. This causes the connecting-rod to oscillate from end to end of the wristpin and crankpin bearings; and if, as is customary, there is 1/16 or more of end movement in these bearings, the knocking may be quite noticeable. If, as is likely to be the case, it is impossible to make the pins parallel, the only recourse is to take up the lost motion at the end of one or the other bearing, and possibly both bearings, by the use of washers or cheeks soldered to one end of the bushing and brasses. This is not a common cause of knocking, particularly in the better class of engines.

15. The best construction is to secure flywheels to short shafts by bolting them to flanges instead of keying them. Sometimes, however, a flywheel is held on by a common key, or by two keys 90' apart, and frequently it will work loose on its keys. This will inevitably result in a knock, which will be very loud if the engine has less than four cylinders. The crank-case should be opened and the cranks blocked so that the shaft cannot turn, and then force should be applied to the flywheel to disclose the looseness, if any. Sometimes the flywheel will be so tight on its shaft as to resist turning in this manner by using any ordinary force. In this case, it is best to take the engine to a repair shop if a thorough search has failed to disclose any other cause for the noise.

A sprung shaft will always cause knocking, and also rapid wear. and cutting of the bearings.

16. Besides the foregoing mechanical causes of knocking, there is a class of what may be called combustion knocks that are altogether distinct from the preceding, in that they may occur without appreciable looseness in the bearings, and are due to excessive rapidity of combustion, coupled generally with too-early ignition, the charge being completely burned before the piston has reached the end of the compression stroke. Combustion knocks are due to a variety of causes, the most obvious of which is simply too-early ignition, as when running a heavy load without suitably retarding the spark. A contributing cause is a slightly weak mixture, since such a mixture burns faster than a normal or overrich mixture. Pounding in particular cylinders of a multicylinder engine may be due to unequal rapidity of combustion, which itself may be due to unequal charges, as when the valves are unequally timed, or to irregular spark timing, such as may result from a wabbling timer or badly adjusted vibrators. If the timer contact surfaces have been roughened by sparking or by wear, they will cause the contact maker of the timer to jump when running fast, and therefore to make erratic contact, resulting in irregular firing.

17. The classes of combustion knocks just mentioned are easily traced to their causes. The knocks are not necessarily violent, and they may sound a good deal like the knocks due to loose bearings, except that, if caused by faulty action of timer or vibrators, they will occur irregularly instead of regularly.

There is, however, another and very common sort of knocking due to spontaneous ignition of the charge before the spark occurs. This may be caused by overheating of the engine from lack of water or other trouble with the circulation - a trouble at once indicated by boiling of the water in the radiator or by smoking of the exterior of the engine. It is a temporary phenomenon, and involves no harm to the engine if the latter is promptly stopped and allowed to cool.

18. Much more troublesome, and also more common, is spontaneous ignition, or preignition, as it is termed, due to a deposit of carbon in the combustion chamber or on the piston head. A carbon deposit of this nature may be caused by too much gasoline or by too much cylinder oil, and it will accumulate gradually even with the carbureter and lubrication correctly regulated. A small quantity of carbon will give no trouble, but as the deposit thickens some portions of it will remain incandescent from one explosion to the next, and will ignite the fresh charge at some point in the compression stroke, depending on conditions. The fact that the charge is not ignited until some time during compression is due to the fact that the more highly it is compressed, the more easily it ignites. True preignition results almost always, except at the highest engine speeds, in the charge being completely burned before expansion begins, and it is easily distinguished, especially if the engine is taking full charges, by the resulting sound, which is a sharp, metallic bing! bing! bing! closely resembling that produced by a hammer striking a block of cast iron. Usually, though not always, an engine that preignites in this manner will continue running by spontaneous ignition for some seconds after the igniter switch has been opened. The hammering due to preignition, as would be expected, is most marked when the engine is running slowly with the spark suitably retarded, and it will generally manifest itself under load, owing to the fact that the throttle is then wide open and the spark necessarily retarded to suit the slow speed of the engine.

19. In stationary engines, a heavy, pounding noise, such as is caused by premature ignition, may also be due to excessively high compression for the grade of fuel employed. In addition to its initial effect in producing a pounding noise, either preignition or a too-high compression pressure may cause the piston to expand unduly and to stick in the cylinder, which it would not do if the conditions were normal. This sticking of the piston would produce a knocking sound due to the small amount of play in the connecting-rod bearings necessary for smooth running. A coughing or barking sound is caused by the escape of pressure past the piston, and would indicate the necessity either of replacing any worn or broken piston rings or of reboring the cylinder and fitting a new piston.

With marine engines, a loose coupling may cause a pound, as may also a loose propeller wheel, but these pounds can easily be located.

CYLINDER AND PISTON DISORDERS

20. Scored and Leaky Cylinders. - One cause of scoring of the cylinder lies in the fact that the ends of the piston pin or wristpin when loose sometimes protrude through the hole or bearing in the piston. Some pins have their bearing in the piston itself, while others, being tightly secured in the piston, have their bearing in the upper end of the connectingrod. No matter which construction is employed, the ends of the pins should never come in contact with the cylinder walls. The pin must by some absolutely positive method be kept in place. While a loose wristpin is often the cause of a scored cylinder, there are three other causes, resulting from imperfections of design or of machine work, to which scoring can be traced; namely, loose core sand, imperfectly fitted piston rings, and loosening of the pins that are used to prevent the piston rings from turning in the slots in the piston.

21. Trouble from loose core sand Is due to sharp sand that usually comes from the cored passage connecting the crankcase with the inlet or passover port to the combustion chamber of two-cycle engines. With cylinder castings properly pickled in dilute sulphuric acid to remove the sand, this trouble would not be experienced; but with modern methods of cleaning castings by means of the sand blast, the cored passages are frequently neglected. Some engines are provided with a removable plate over the inlet port, for the express purpose of making sure that there shall be no core sand therein to cause trouble.

If, in an engine of the two-cycle type, the scoring consists of several parallel marks on the side where the inlet port is located, it is safe to ascribe the trouble to sand. If the scoring is on the exhaust-port side, it is usually an indication of insufficient lubrication; as the hot exhaust gases pass out they burn the oil off that side of the piston and cylinder, the exhaust side of a two-cycle engine cylinder being always hotter than the inlet side. Scoring may occasionally be due to the presence in the cylinder of pieces of the porcelain insulation of spark plugs. Cylinders have been practically ruined through rough dropping into the cylinder the pin or nut holding in place the spring on an inverted inlet valve.

22. Leaky cylinders (particularly in two-cycle engines) render the wristpin, crankpin, and main-shaft bearings subject to excessive wear, because the heat of the gases that pass by the rings into the crank-case tends to burn up the oil and heat the bearings. If the engine is of the two-cycle type, the leaking products of combustion not only foul the fresh charge of gas so that it is not so explosive, but the quantity of each charge is reduced.

If, in an engine in which the inlet and exhaust valves are tight and there is no leaky gasket, it is found that the compression has become materially reduced, the trouble is probably caused by leaks from distorted, scored, or imperfect cylinders, the pistons or piston rings being worn considerably or stuck in the slots in the piston. The only remedy is to remove the pistons for examination. If the cylinder is found to be out of round or scored, it will have to be rebored, and new pistons and rings fitted. If the rings are found to be rusted or stuck in the slots, they will have to be removed, even if to do so it is necessary to break them. They may have worn to such an extent that the openings at the points of parting are such as to allow a loss of pressure, the leaking charge passing either into the tight crank-case, if the engine is two-cycle, or into the atmosphere. If such leakage is not stopped, the heat of the escaping gases will burn the oil out of the crank-case, and the bearings will soon become badly worn, if not ruined.

23. The piston should be examined carefully for wear. The side on which the angular pressure of the connecting rod is exerted should, of course, show the most wear. If the front or rear side of the piston shows wear at top or bottom, with a corresponding amount of wear on the opposite bottom or top, it is proof that the hole through the piston for thepiston pin, to which is connected the upper end of the connecting-rod, is higher at the end showing wear at the top of the piston than at the end showing wear at the bottom. If this is found to be the case, and the wristpin is tightly secured in the piston, the connecting-rod bearing for the wristpin will be found to have worn badly and will be bell-mouthed, that is, larger at the ends than at the center. The remedy for this is to true up the bole carefully and bush it, or use a pin that is a trifle larger than the hole, increasing the size of hole in the tipper bushing slightly. This is a repair job that should be entrusted only to a thoroughly reliable machinist having the tools and means for doing accurate work. Side wear on the piston is much more likely to show in engines having the wristpin held securely in the upper end of the connecting-rod, the ends of the pin having bearings in the piston.

24. Piston rings become stuck in the slots in the piston from two causes; namely, from water getting into the combustion chamber, causing the rings to rust, and from the sides of the slots being slightly tapered instead of parallel. Where tapered sides are found, it is usually necessary to straighten them up in a lathe and use slightly wider rings, Piston rings should be renewed much oftener than is customary. As they become more and more open at the ends, the hot gases passing by the ends of the rings have a harmful effect on the polished, cylinder surfaces, and in two-cycle engines they foul the mixture in the crankcase.

25. Broken piston rings, particularly in engines with ports that are opened and closed by the pistons, are a source of annoyance, and frequently cause much trouble. Broken piston rings are frequently the result of insufficient care in putting the piston, with the rings in place, into the cylinder, but are more likely the result of getting a ring end caught in a port. To prevent this, two-cycle engine rings are usually pinned to prevent them from turning until the ends can get into the port.

The breaking of a piston ring is rather an unusual occurrence; it will cause loss of compression, that may be distinguished from leakage due to the rings being worn by the fact that the broken ring will make a distinct clicking sound at the end of every stroke. It will also be found that oil squirted on the piston when a ring is broken will not stop the leak. If the engine has more than one cylinder, it is probable that loss of compression due to lack of oil would affect all the cylinders, whereas a broken ring affects one only. If a piston ring is broken, it becomes necessary to take off the cylinder without delay and put in a new ring.

26. Piston rings are supposed to be held in position by small pins, one in each ring, so that the joints of adjacent rings are diametrically opposite. If for any reason these pins break, a ring may slip round until its joint is in line with that of the next ring above or below. This will cause loss of compression that may be very puzzling; it is an unusual occurrence, and it may be necessary to take off the cylinder to locate the trouble.

27. A good method for pinning piston rings is shown in Fig. 1 (a) and (b) Fig. 1 (a) is a diagram of a piston head, the dotted lines showing the bottom of the ring slot, while Fig. 1 (b) is a sketch of a portion of one side of the piston. With the piston square on its lower end, drill, at a, a point about half way between the inlet and exhaust ports, through b, c, and d, a hole large enough for clearance for a small tap, continue the hole into e with a tap drill, tap the hole, and screw into it a slotted screw to extend into the slot for a distance not quite one-half the width of the slot. Then tap and plug the hole through b, c, and d with screws dipped in muriatic acid to rust them in place, the screw plugs being in each case below the surface of the slot faces. At another point, where it would not come opposite a port, drill a hole through b and c and tap into d, plugging the clearance holes, as before. Drill at another point a hole through b, tapping into c. The slotted screws extend one-half or less the width of the slots from the bottom, so that, if the rings be parted as in Fig. 2 (a) one of the ends could be cut off slightly to receive the pin, or, if parted diagonally, as in Fig. 2 (b), a space could be cut out for the pin. With this method of (a) pinning the rings, there is no way for the pins to work out to score the cylinders. While it is customary to pin the piston rings for two-cycle engines, pins are rarely found necessary in four-cycle engines, as such engines have no ports to catch the ends of the rings, except when an auxiliary exhaust is employed.

28. Cylinder-Packing Troubles.-The joints between the cylinder head and the cylinder of stationary gas engines are kept tight by packings usually cut out of asbestos sheet about 1/32 inch thick. When the packing is damaged by overheating or excessive pressure, water from the jacket leaks either to the outside or into the cylinder. The latter is the more serious leak of the two, as it interferes with the running of the engine by corroding the points of contact on the igniter and the valve seats and stems, and prevents proper lubrication of the piston and cylinder. Leaking toward the cylinder i's generally indicated by splashing of the cooling water at the overflow pipe when the explosion takes place.

In most cases, the blowing out of a packing is caused by the combustion pressure opening the joint between the packing surfaces, the packing being heated and partly destroyed, and allowing water to enter the combustion chamber. A partial or complete stoppage of the coolingwater supply or the clogging of the water spaces with lime or similar deposits will also result in the overheating of the cylinder and consequent damage to the packings.

As soon as a leak of water from a faulty packing develops, preparations should be made to renew the packing at the first opportunity. If the leak is to the outside, which may not interfere with the operation of the engine, although it will cause inconvenience through having to catch the water in buckets, it is not necessary to shut down the engine until the day's work is done. If the leak is toward the combustion chamber, the engine will generally stop in a short time.

29. Many engines have the cylinder heads and cylinders in one piece; but a few engines have copper or aluminum water-jackets. There are, however, some old engines with separate heads still in service. In some cases, the cylinder beads, when separate, are made a ground fit on the cylinders, but they are commonly made tight by asbestos gaskets. Leakage through these may be detected some. times by the sound, and sometimes by putting a little oil over the suspected place and noting the resulting bubbles when the crank is turned.

In case a cylinder-head gasket leaks, it will be necessary to put in a new gasket. The head should be taken off, the old gasket removed, and the iron surfaces in contact with it should be carefully scraped clean. The new gasket may be of sheet asbestos, and it should be sprinkled evenly with powdered graphite to prevent it from sticking. It may be cut to size by laying it on the cylinder and tapping it lightly with a small hammer to indicate the outlines. Care should be taken not to let inwardly projecting edges interfere with the valves or igniters; and, also, if there are openings through the head for the passage of water, it should be seen to that these are not closed by the asbestos.

A good packing for cylinder heads is sheet asbestos with woven brass wire embedded in it. This packing is much stronger than ordinary sheet asbestos, and will not blow out unless the cylinder-head bolts are loose or the head is a bad fit. In replacing a cylinder head, the bolts should be tightened gradually and evenly, each being tightened a little at a time, and the round being made three or four times, so as to avoid localizing the stress on any one bolt.

There is, of course, but one remedy for leaky gaskets, namely, renewal. The old gasket should be carefully and completely removed, and by means of a straight edge a careful examination should be made to discover, if possible, why the gasket gave way at a particular point. There may have been insufficient surface or too little holding-down pressure to keep the packing in place; the studs may have been too far apart at the point of rupture, or the nuts may not have been tightened after the engine had become heated.

30. Leaky Inlet and Exhaust Valves.-Trouble from loss of compression in the combustion chamber, when the spark plug is tight and there is plenty of oil on the piston, is generally due to leaky valves. In order to determine whether the leak is in the valves or in the piston rings, a moderate quantity of oil may be squirted through the compression relief cocks and the crank turned two or three times, which will temporarily check whatever leakage there may be around the piston. If the compressed charge still escapes, the inlet valve, if located over the exhaust valve, may be taken out and examined. The leak, however, is much more likely to be in the exhaust valve.

To take out the exhaust valve, turn the engine over by hand, with the switch off and the compression relief cocks open, until the valve is opened. Then prop up the valve spring with two pieces of wood or brass a, a, Fig. 3, cut to the proper length to go between the spring collar b, and the upper end (or lower end, if this is more convenient) of the push-rod guide c, and turn the engine again until the push rod d is down as far as it will go. Push the I exhaust valve down; the key at e may now be slipped out If the props have been made accurately to length, the valve may be slipped up and out, leaving the spring and the collar in place. Inspection should show the valve seat to be of uniform appearance all the way around, and dull - not glossy. If the seat of either valve is pitted or rough, or if it is worn bright on one side, showing that it has been seating only on that side, it should be reground.

31. The remedy for leaky valves is to regrind them to their seats. If badly scored and worn, which will be shown by a blackening of the seat and valve, it may become necessary to reseat and true up the valve, but if the engine has had ordinary care and attention, grinding should be sufficient. For this purpose, the exhaust valves may need emery and oil, finishing up with powdered oilstone, ground glass, silex, or the dirt that accumulates under a grindstone. The valve should not be rotated its whole circumference (as is frequently done, using a brace or breast drill with a bit screwdriver) but should be rotated a little, first in one direction and then in the opposite direction, raising it off the seat very often, and using oil freely, until a dull surface appears on both the valve and the seat throughout their bearing surfaces. Rotating the valve rapidly is very likely to cause grooves and ridges that are extremely hard to remove and make the valves tight.

While there is little or no danger of getting emery or other abrasive substance into the cylinder when grinding exhaust valves, ordinary care to avoid doing so should be exercised. The passage of the products of combustion being outwards, such matter would be carried away from the cylinder. Grinding the inlet valves is a very particular operation, for any particles of abrasive substance left behind to be drawn into the cylinder are liable to cause trouble. All traces of grindstone dirt, which will be found well adapted for grinding and may be mixed with water instead of oil, should be wiped off carefully.

The valve stems should be inspected, and, if rusted or rough, should be cleaned and smoothed, a few drops of kerosene being applied to cut any deposits that may have accumulated in the guides.

32. Weak or Broken Inlet-Valve Spring - Sometimes the inlet-valve spring, especially if the valve is of the automatic variety, will weaken from becoming overheated. This is almost sure to occur if the engine has been allowed to overheat from lack of water. In time, a spring loaded too near its elastic limit will break from the jarring to which it is subjected. The symptoms in either case are loss of power at high speeds (although the power may still be ample at low speeds) and clattering of the valve and blowing back in the intake pipe at high speeds. The latter may easily be detected with a single or double cylinder engine by holding the fingers close to the air intake, when the backwards puffing will be very perceptible. If the engine has four cylinders, it may be possible for the inlet valve springs to be slightly weak without the mixture blowing back at the intake, owing to the fact that one or another cylinder is aspirating all the time, and the air expelled from one cylinder is drawn into the next. One way to get around this difficulty is to block open the exhaust valves of two cylinders - the first and fourth or the second and third while the others are tested. It will probably be simpler, however, to experiment with the valve-spring tension. If the valve spring is weak, and if it is temporarily increased in stiffness by putting washers under it to compress it, a marked increase in the power of the engine at high speeds will be observed. The proper remedy, however, is to put in a new spring, or, if this cannot be done, to stretch the old spring, For a valve lift of 1/8 inch, and for average engine speeds, the tension should not be less than 1 pound per ounce of the weight of the valve, washer, and key. The engine will work better if the springs are a ttle too stiff than if they are not stiff enough. There will also be less danger of breakage of the valve stems and keys.

33. Unequal Tension of Automatic Inlet-Valve Springs.-The effect of unequal tension in the springs of automatic inlet valves is to permit one cylinder to take more gas than another. Consequently, at slow speeds the cylinder whose valve spring is weak will get the larger charge; and at high speeds part of the charge will be blown back through the valve whose spring is weak, so that the other cylinders will get stronger impulses. A quick way to test the equality of valve-spring tension without taking out the valves is to run the engine slowly with the throttle almost closed. This will cause the cylinders whose springs are stiffer to receive scarcely any gas, and the cylinders whose valve springs are weak will do most of the work. It is possible, however, to go to excess in a test of this sort, since, when an engine is running light with the minimum quantity of gas, one cylinder is almost sure to get more gas than another, if the inlet valves are automatic, even with the most careful equalizing of the springs. If the tension of the valve springs is under suspicion, the valves should be taken out and the springs tested by compressing the valve stems together.

34. Excessive Lift of Automatic Inlet Valve.-The lift of an automatic inlet valve should be proportionate to the spring tension and to the weight of the valve, so that the spring will be able to overcome the inertia of the valve, and close it before the piston has started so far on its compression stroke as to expel any of the mixture through the open valve.

The symptoms of too great a valve lift are loss of power and blowing back at high speeds. A valve 2 inches in outer diameter should not ordinarily lift more than 1/8 inch and a lift of 3/16 inch would be excessive for almost any valves found on high-speed engines. An excessive lift, like a weak spring, is likely to result in breakage of the valve stems and keys through unnecessary hammering of the valve when opening and closing.

35. Broken Inlet-Valve Stem or Key.-Trouble from a broken inlet-valve stem or key is more likely to occur with automatic valves than with those mechanically operated. The result, if the valve opens downwards, is to let it stay open all the time, causing that cylinder to cease work, while the sparks from the plug ignite the mixture in the intake pipe and cause explosions there and in the carbureter. If the valve, whether automatic or mechanically operated, opens upwards, it will clatter on its seat and permit much of the mixture to be expelled during the first part of the compression stroke.

36. Weak or Broken Exhaust-Valve Spring.-Owing to the heat to which it is subjected, the exhaust-valve spring is more likely to weaken than that of the inlet valve. The symptoms are loss of power, owing to the valve lingering open at the end of the exhaust stroke, and clattering when the valve closes.

37. Broken Exhaust-Valve Stem or Key.-As there is nothing to prevent the valve from being sucked wide open on the suction stroke, an accident of this kind will generally cause that cylinder to go out of action entirely. The clattering, if the engine continues running by virtue of other cylinders, is likely to be marked.

38. Slipped Valve Cams.-Some cheaply constructed engines have the valve cams held on the shaft by taper pins that in time shear partly or wholly through, permitting the cams to turn on the shaft. The cams may turn a short distance and then be jammed by fragments of the taper pins. The symptom indicating trouble due to this cause is partial or complete loss of power in the cylinder affected, when nothing is wrong with the ignition, valve-spring tension, etc.; and it will be equally marked at all speeds. If a cam is pinned on its shaft, the proper way to secure it is to add another pin, or, better, to add a key to take the torsional stress, and depend on the pin only to keep the cam from slipping endwise on the shaft.

39. Lack of Cylinder oil.- The symptoms of lack of cylinder oil are manifested in a sudden laboring of the engine, a dry or groaning sound, and partial loss of compression, followed by probable seizing of the piston. If the piston does not seize, it and the cylinder walls will at all events be scored.

Among the causes of lack of cylinder oil are clogging of lubricator by dirt or waste, obstruction in oil pipes, leaky check-valves, leaky pump packing, broken oil pipe, oil too cold to feed, lack of oil in crank-case, etc.

The remedies for trouble from this source will become obvious on inspection. The engine should be stopped and allowed to cool, and a liberal quantity of oil should be put in the crank-case before starting again. Squirt a little oil through the compression relief cocks to insure lubrication of the pistons, without waiting for oil to reach them from the regular sources. Remove the obstruction or repair the break as soon as possible.

40. Lack of Oil In Bearings.-A slightly loose main or crankpin bearing will sometimes be cut badly as a result of a temporary stoppage of oil feed, and yet give no noticeable symptom until the bearing is so badly cut that knocking begins. If a well-fitted bronze-bushed bearing becomes dry, it is more likely to stop or at least retard the engine. A babbitted bearing will melt out and let the shaft settle as far as other supports or bearings will allow. The result may be a violent pounding, a bent or broken shaft, or cut bearings generally, according to the particular conditions. There is no real safeguard against lack of oil in bearings except in the vigilance of the operator, combined with a system of oiling not liable to go wrong. It is not safe to depend on detecting a dry bearing by the sense of touch, because often the metal adjacent to bearings is sufficient to carry the heat away.

Generally, trouble from this cause is due to neglect to supply oil or to see that the sight feeds are working properly. It may also be due to a broken pipe, cold oil, etc.

There is no excuse for neglect to clean the oil strainer, or failure to inspect the oil pipes, unions, etc., or to know when starting out how much oil is in the crank-case. A badly cut bearing should be sent to a repair shop, and should be attended to without delay; but a bearing only slightly cut may be kept in service by the addition of a small quantity of flake graphite to the oil. If possible, the shaft should be taken out and polished with emery cloth and oil, else bronze from the bearing is likely to cling to it and aggravate the cutting. A bearing supplied with oil from a well beneath it, and a chain running over the shaft, may occasionally fail to receive oil owing to the chain catching on some internal roughness or projection in the oil pocket. It is always safest to keep a more or less regular supply of oil passing through bearings of this sort when in use, and depend on the oil well only as an equalizer.

41. Improper Oil In Cylinders.-The trouble symptoms produced by the use of oil unsuited for lubricating the piston are white or yellow smoke in the exhaust, rapid fouling of spark plugs, partial clogging of inlet and exhaust valves, and rapid accumulation of carbon on the valves in the combustion chamber and about the piston rings.

To remedy the trouble empty out all the unsuitable oil if possible, and substitute oil known to be good. Inject kerosene freely through the compression relief cocks to loosen the carbon deposit on the piston rings, and use kerosene to free the valves if they stick. Drain the crank-case, and, if possible, open it and clean out any carbon that may have worked down past the piston and mingled with the oil. Change all the spark plugs, and clean them when opportunity offers. Put in plenty of fresh oil before starting, and see that oil is supplied to the pistons so that they will not go dry before oil begins to feed from the cylinder lubricator.

42. Too Much Oil on Pistons.-Too much oil on the pistons is indicated by white smoke in the exhaust, fouled spark plugs and valves, substantially as when inferior oil is used, though the symptoms will not be so pronounced. An examination of the combustion chamber through the inlet valve or spark-plug hole, using a mirror and electric flashlight if necessary, will show an unnecessary amount of oil around the top of the piston. With the oil correctly regulated, it should not accumulate on the piston head in any great quantity.

Trouble from this source is remedied by drawing off part or all the oil from the crank-case, if it contains more than is necessary for running the engine, and reducing the oil feeds to the cylinders if necessary.

43. Lack of Water. - Lack of water in the radiator of the cooling system for engines is indicated by the rapid emission of steam, if there is sufficient water to enter the engine jacket; the bottom of radiator being cold; the overheating and smoking of the engine, followed by laboring, groaning sounds, owing to the oil being burned away faster than it is supplied to the pistons; and, if the engine still continues running, expansion and seizure of the pistons in the cylinders.

Trouble from lack of water is due to carelessness in not filling the tank before starting; leakage in radiator or piping; accidental opening of the drain cock at the lowest point of the circulation system; breakage of drain cock, etc.

The remedies for such trouble are apparent on inspection. If the engine becomes overheated so that the water boils rapidly away, and there is reason to think that the upper portion of the water-jacket is dry, the engine should be allowed to cool before water is added; otherwise, the sudden contraction may warp or even crack the cylinders, or it may cause the cylinders to contract and seize the pistons. The crank-case should be liberally supplied with oil to prevent the pistons from becoming dry, or, if a sight-feed oil cup is put on the cylinder, it should be set to feed quite rapidly. The engine should be stopped at the first sign of distress, as indicated by a groaning sound, turning with difficulty, or knocking caused by preignition due to hot cylinders.

44. Obstructed Circulation.-An obstruction to the circulation of the cooling water elsewhere than in the radiator will cause the bottom of the radiator to remain cool while the top is, probably, boiling hot.

Among the causes of obstructed circulation are a broken pump, broken driving connection to pump, or slipping belt or friction pulley, if the pump is driven in that manner; waste or the like lodged in the pump or piping.

The remedies for this trouble will become obvious on inspection. If the belt or friction pulley has oil on it, gasoline may be used to clean the pulley, as well as the flywheel if it drives the pulley.

45. Scale or Sediment In Radiator.-The presence of scale or sediment in the radiator is indicated when the whole radiator becomes hot or when steam formed in the jacket forces water out of the upper pipe to the radiator, there being no oil on the inside or dirt on the outside of the radiator.

Scale will deposit from hard water if the temperature of the water is allowed to approach the boiling point. A similar scale, almost impossible to eliminate, will crystallize from calcium-chloride non-freezing mixture; if these are allowed to become supersaturated.

A radiator badly choked with lime scale is practically hopeless, although, if it is made entirely of brass and copper, it may sometimes be helped by the use of a dilute solution of hydrochloric acid in the proportion of about one of acid to ten of water. This should be left in the radiator long enough only to loosen the scale, and should then be drawn off, and the radiator washed out. It is better in doing this to disconnect the radiator from the engine, in order to confine the effects of the acid. Another method is to use washing soda. Ordinary dirt maybe cleaned out by a strong, hot solution of lye, which should be used with care, as it burns the skin badly. Rainwater should be used wherever possible, and all the water should be strained.

46. Dirty Radiator.-When the whole radiator is hot and it is impossible to run without boiling the water, the circulation being good, it is evident that the radiator is dirty.

Flying oil about the engine may lodge on the air surfaces of the radiator tubes, and gather dust, which forms a nonconducting covering. Oil sometimes gathers on the water surfaces by gradual escape from the pump bearings, or may remain after an attempt to substitute refrigerator oil for water as a cooling medium in freezing weather. The film of oil, preventing the water from coming in contact with the metal, acts practically as an insulator.

To remove the oil from the radiator use kerosene, or a mixture of 'kerosene and mineral-oil soap. Dissolve the soap in water and add it to the kerosene, fill up the radiator with the mixture, and run the engine for an hour or more until the radiator gets well heated. The soap and kerosene will form an emulsion with the oil, and when the mixture is hot it may be drawn off and the radiator washed out with cold water. For the removal of the external oil and dirt, use gasoline, with a brush or swab.

A simple trouble, but one likely to be mistaken by the novice for radiator or circulation trouble, is slipping of the fan belt. The belt should be tested occasionally, and not allowed to get so loose that the fan pulley can spin inside it. It does not need to be tight.

PROPORTIONING OF MIXTURE

47. Overrich MIxture. -If a mixture is very rich, that is, if there is an excessive amount of gasoline in the charge, the fact will be manifested by black smoke in the exhaust. If the mixture is not rich enough to produce smoke, it will still produce an acrid odor in the exhaust, and will cause overheating of the radiator, unnecessary sooting of the plugs, accumulation of carbon in the combustion chamber, and unnecessarily rapid comsumption of gasoline, with diminished power.

The causes of an overrich mixture are: faulty carbureter adjustment; leaky float; leaky float valves; float too high on its stem or too heavy; spray nozzle loosened or unscrewed by vibration; and dirt on the wire-gauze screen over the mouth of the air-intake pipe. For float troubles, see paragraphs. 52 to 55, inclusive.

48. Flooding is the most common source of trouble in marine engines using vaporizers. It is caused by leakage of gasoline into the vaporizer, from which in a two-cycle engine it readily runs into the crank-chamber; the resulting mixture is too rich in gasoline, and, not having sufficient oxygen, is unexplosive. When trouble from flooding is suspected, turn the engine over two or three times, with the gasoline valve and the switch closed. If there is an explosion, note the color of the flame at the relief cock, or priming cup, which should be left open for the purpose, If no explosion occurs, leave the cock or cup open and slowly turn the flywheel to a point just before the exhaust port opens, thus drawing air into the cylinder through the priming cup to dilute what is thought to bean overrich mixture. Now revolve the flywheel in the opposite direction rather rapidly until the spark occurs. If there is no explosion, try again, and repeat the operation two or three times if necessary. If an explosion then takes place, it is evident that flooding is present.

To remedy this in a two-cycle engine, open the draw-off, or drain cock in the lowest part of the crank-case, and draw off the contents, taking care, however, to replace with a fresh supply the lubricating oil thus drawn out. If there is no draw-off cock, it will be necessary to turn the flywheel many times to exhaust the excess of gasoline in the crank-case, leaving the switch closed and the compression relieved as much as possible. Af ter a while, an explosion should take place, then another, gradually becoming more frequent, until finally the engine may run with an explosion at every other revolution or so. The gasoline valve should be kept closed until the charges explode regularly and the red tinge to the flame at the relief cock and smoky exhaust disappear, after which the gasoline may be turned on and regulated at the needle valve in the vaporizer, closing it slightly at first; and, if the engine slows down somewhat, open it slightly until it is possible to tell whether it is getting too little or too much gasoline.

In case of flooding in a four-cycle engine using a vaporizer, two or three revolutions of the crankshaft will usually dispose of any excess of gasoline, for there cannot be as large an amount in the exhaust piping of a four-cycle engine as could accumulate in the crank-case of a two-cycle engine. Trouble from flooding in a two-cycle engine is the first thing to be suspected when an engine of that type refuses to start readily.

If the cause of a failure to start is found to be an insufficient supply of gasoline, due to dirt in the needle valve, or to a small amount of water in the gasoline piping, lift the valve in the vaporizer from its seat and let a little gasoline run through to clear the obstruction or get a drop or two of the water out, being sure to catch the drip for examination. If there is any water it will show in globular form at the bottom of the vessel. In case water is found, the pipe must be disconnected and drained, and any water in the tank must, if possible, be removed, for a single drop of water will completely close the aperture in the seat of a needle valve.

49. Weak Mixture.-Among the symptoms produced by a weak mixture are insufficient power, although the explosions are regular; a tendency to preignite or to burn very rapidly if there is the slightest carbon deposit; the engine sometimes will miss every other explosion. There is likely also to be difficulty in starting the engine. It is not always easy to distinguish between lack of power due to an overrich mixture and that due to a weak mixture, but the tendency of the former is to produce black smoke and of the latter to preignite and Miss explosions. Some experimenting with the carbureter adjustment will often be necessary to settle the point.

Nearly all the causes named in paragraph 47 will make a mixture richer at some speeds than at others, and if the carbureter has been readjusted, for example, in the attempt to correct trouble due in reality to a heavy float, the result will be to make the mixture faulty again at certain other speeds. Special causes of weak mixture are dirt or waste in the gasoline pipe or strainer; stale gasoline; carbureter too cold to vaporize; dirt in the spray nozzle; float too light or too low on its stem.

For float-trouble remedies see paragraphs 52 to 55, inclusive. Experimenting with the carbureter adjustment should be very cautiously done, with the original setting or adjustment marked so that it can be restored if necessary. The carbureter should then be adjusted slightly in one direction or the other, and the effect noted before further change is made. Very often a combination of adjustments will be necessary, but it is best to make them one at a time. If a radical change is made it may be very difficult to start the engine at all, and this would leave the experimenter completely in the dark as to what was required.

DIRT IN CARBURETOR AND GASOLINE PIPING

50. Dirt In Carbureter.-If there is dirt in the float valve, it will prevent the latter from closing and will cause the carbureter to flood. This will produce an overrich mixture, especially at low speeds, and is highly dangerous on account of the liability to fire. If the dirt is in the spray nozzle, it will produce a weak mixture. If the dirt has been splashed into the air intake, it will produce an overrich mixture, especially it high speeds.

The remedies for trouble due to dirt in the carbureter will become obvious when the nature of the trouble is located. A carbureter that has previously worked well and that suddenly begins to leak has in all probability dirt in the float valve. A carbureter that suddenly gives a very weak mixture has dirt probably in the gasoline pipe, strainer, or spray nozzle.

51. Dirt or Waste in Gasoline Pipe.- Dirt or other obstructions sometimes accumulate, especially if the gasoline has not been properly strained. The symptom is a sudden or gradual weakness of the mixture, necessitating readjustment of the carbureter in order to keep the engine running. The most probable place of lodgment for obstructions of this sort is in the gasoline line pipe where the latter connects to the carbureter, or in the strainer, through which the gasoline generally passes just before it enters the float chamber. Disconnecting the gasoline pipe or the union exposing this strainer will generally disclose the obstruction. Sometimes it may be necessary to disconnect the gasoline pipe at both ends, and blow it out. This is necessary only when the pipe has been disconnected near the carbureter and gasoline does not flow freely from it when turned on at the tank.

FOAT TROUBLES

To remedy the trouble grind in the valve with pumice or fine sandstone.

53. Float Too High.-By the expression float too high is meant that the float is set too high on its stem so that it is not lifted by the gasoline sufficiently to close the float valve before gasoline escapes from the spray nozzle.

When this trouble is present, the carbureter drips when the main gasoline valve is opened; but the float valve is soon closed by the float if the spray orifice is covered by the finger. The float valve closes tight when manipulated by the fingers, or when the float is lifted by a pair of bent wires, When the trouble is due to a high float, it will be found that the float itself is empty, and, if of cork, that it not gasoline soaked.

Unless the float is adjustable on its stem, the easiest remedy for this trouble is to bend the levers by which the float acts on the float valve. If this cannot be done, shift the float 1/16 inch lower on the stem by the use of a soldering iron.

54. Float Too Heavy. --The same symptoms are present when the float is too heavy as when the float is too high, but they are caused generally by a leak in the float or by its being gasoline-soaked.

If the float is hollow, it will sometimes be found that there is present in it a minute leak due generally to some oversight in soldering. If the float is taken out and shaken with the hand, the presence of the gasoline inside of it will at once be apparent. The float should be immersed in warm water until all the gasoline in it is slowly boiled away and its vapor has been expelled through the aperture in the float, By holding the float under water, the escape of bubbles will indicate this aperture. Care should be taken that the vapor escaping from the float does not cause fire. When the leak has been located it should be marked with a pencil, and after the float has become cold the leak may be closed with a minute drop of solder. If the float is of cork, it may be saturated with gasoline. It should be taken out, allowed to dry slowly, and given a coat of shellac care being taken that the shellac enters all the holes on the surface.

55. Float Too Light or Adjusted Too Low. - By the expressions float too light or adjusted too low is meant that the float is lifted by the gasoline in the float chamber when the gasoline level is still some distance below the orifice of the spray nozzle.

Among the symptoms produced by a light float or a low adjustment are a weak mixture at slow speed, and, probably, difficulty in starting the engine, owing to the fact that considerable suction is required to lift the gasoline to the mouth of the spray nozzle. The height of the gasoline in the spray nozzle can generally be determined, with the aid of an electric flashlight, by a little experimenting with the float, pushing the latter down for an instant after it has closed the valve.

To remedy the trouble, the float must be weighted slightly, so that the gasoline Will rise higher before the float closes its valve. The weight may take the form of a few drops of solder carefully distributed over the float so as not to over-balance it on one side; or, if this is not sufficient, a ring of sheet brass may be soldered to the top of the float.

FUEL TROUBLES

56. Stale Gasoline.-If an engine has been left standing for some time unused, more or less of the gasoline in the tank will evaporate, and it may get too stale to give a correct mixture without readjustment of the carbureter. The usual symptoms are difficulty in starting the engine, and insufficient power owing to a weak mixture. The best remedy is simply to fill up the tank, when the mixture of old and fresh liquid will probably work satisfactorily. It may be necessary, however, to readjust the carbureter or to throw away the stale fuel.

57. Water In Gasoline.-Water maybe found in gasoline taken from a barrel standing out of doors. The water, being heavier than the gasoline, will always settle to the bottom, and by close observation it may be seen before it is poured into the tank. If the gasoline is strained through a piece of chamois skin or several layers of cheese cloth, or even through very fine brass-wire gauze, the strainer will hold the Water while permitting the gasoline to pass through.

The user should make it an invariable rule to strain his gasoline in this manner.

The symptom of water in the gasoline will be immediate stoppage of the engine when the water reaches the spray nozzle, in spite of the fact that the timer, coils, battery, spark plugs, etc., are in perfect order, and the gasoline tank is known not to be empty. The only remedy is to unscrew the wash-out plug at the bottom of the carbureter, and let the water and gasoline run out until it is certain that all the water has escaped. Sometimes it may be necessary to disconnect the gasoline pipe entirely mid blow it out in order to expel the last drop of water. It is well also to look into the tank with an electric flashlight and see if any drops of water can be discovered on the bottom. If so, it may be well to drain the entire tank. Extreme care should betaken to avoid fire while gasoline is being run off.

58. In stationary practice, besides using gasoline of proper quality, it is of course supposed that the storage tank contains a sufficient quantity of fuel to run the engine. This appears to be a superfluous precaution, nevertheless it has frequently happened that an expert has been sent several hundred miles, on complaint from the purchaser of an engine that he was unable to start it, only to find that there was no gasoline in the tank. In other cases it was discovered that, instead of gasoline, almost pure water was pumped to the engine. The explanation was that fuel purchased from a local dealer contained a considerable quantity of water, which of course settled to the bottom of the tank, and accumulated gradually until with the tank about one quarter filled, nothing but water would be delivered to the engine. To avoid this, the contents of the tank should be examined at regular intervals or when the supply is low, and the tank drained whenever there is any doubt about the quality of the liquid that settles in the lower portions.

BACK FIRING

59. The cause of back firing in stationary engines is in most cases due to the delayed combustion of a weak mixture containing an insufficient amount of fuel. The result of such a mixture is a weak explosion and slow burning, so that, during the entire exhaust stroke and even at the beginning of the suction stroke, there is a flame in the combustion chamber. The fresh charge will therefore be ignited by the flame of the delayed combustion of the previous charge; and, as the inlet valve is open at that time toward the air-supply pipe or passage, a loud report will be heard in the air vessel or in the space under the engine bed whence the air is taken. The remedy for this condition is to increase the fuel supply until the explosions become of normal strength and the back firing ceases.

Another cause of back firing may be the presence of an incandescent body in the combustion chamber, such as a sharp point or edge of metal, A projecting piece of asbestos packing, soot, or carbonized oil and similar impurities accumulating in the cylinder. To stop back firing from these causes, any projections of metal or other material should be removed with a suitable tool, and the walls of the combustion chamber made as smooth as possible, or the cylinder should be cleared of any deposit of soot or carbonized oil that may have gathered there.

Failure of the igniter to fire all charges admitted to the cylinder, or improper composition of the mixture resulting in the same way, will be indicated by heavy reports at the end of the exhaust pipe. One or more charges may in this manner be forced through the cylinder into the exhaust pipe, and the first hot exhaust resulting from the combustion of a charge will fire the mixture that has accumulated in the pipe and the explosion will be accompanied by a report similar to that of the firing of a heavy cannon.

60. On account of the shorter time between the opening of the exhaust port and the admission of the new charge in a two-cycle engine, there is much greater liability to back firing in an engine of that type, than in a four-cycle engine. In a four-cycle engine back firing will occur only when the inlet valve is off its seat; hence, in marine practice, back firing is more of an element of danger in four-cycle than in two-cycle engines. If there is no check-valve in the carbureter or vaporizer, and there is no direct opening to the atmosphere, the column of flame that would be blown into a boat through a carbureter or auxiliary air supply on account of back firing would be particularly dangerous because accumulations of gasoline vapor, especially in cabin boats, might thereby become ignited.

To be absolutely safe, a marine four-cycle engine having a float-feed carbureter not supplied with a check-valve should take its supply of air from some point outside of the cabin or from the top of the engine, rather than from a point near the base. As the use of a check-valve in the carbureter would materially reduce the efficiency of the engine, it is rarely used. If a float-feed carbureter is used, and indications point to imperfect carburization, the carbureter should be examined carefully. If the float leaks, so, that the height of gasoline is constantly above the desired level, or if the float does not cut off the supply where it should, it will be necessary to take the carbureter apart to ascertain the trouble, which may be due to a stopped-up needle valve or nozzle.

61. Explosions in the muffler and exhaust piping are usually caused by the ignition of the gas accumulating from missed explosions due to weak mixtures or faulty ignition. They are not usually dangerous unless the muffler is large and is weakened by rusting inside or out, as from saltwater passing through it or from damp salt air, against which it seems almost impossible to protect it in a boat.

62. Explosions in the carbureter are sometimes caused by the inlet valve sticking open and permitting the flame to communicate from the spark. More often it is due to improper mixture, which bums so slowly that flame lingers in the cylinder even af ter the exhaust stroke is completed and the inlet valve begins to open. Either a weak or a rich mixture will produce this result, though not always both in the same engine. Carbureter explosions are often attributed to the exhaust valve closing after the inlet valve opens, or to simple leakage of the inlet valve; but these are seldom the real causes.

PREIGNITION

63. Definitlon.-Premature ignition, or preignition, while somewhat similar to back firing in its nature and origin, manifests itself in a different way and has a different effect on the action of the engine. Premature ignition, as usually understood, is the firing of the partly compressed mixture before the time fixed by the igniting mechanism. Its causes are similar to those that result in back firing, the effect being different in that the charge is ignited later than when back firing takes place, but before the end of the compression stroke. Preignition will cause the engine to lose power on account of the maximum pressure being exerted on the crank before it reaches the inner dead center and thus having a tendency to turn it in the wrong direction, against the momentum of the flywheels.

64. Causes of Preignition.-Besides the causes cited in connection with back firing, preignition may be due to any one of the following defects: Insufficient cooling of the cylinder, due either to shortage of cooling water or to the fact that portions of the water-jacket become filled with lime deposits or impurities contained in the water, thus interfering with proper circulation; compression too high for the grade of fuel used; imperfections in the surfaces of the piston end or valve beads exposed to the combustion, such as sandholes or similar cavities in which a small portion of the burning charge may be confined; electrodes or other parts of the engine exposed to the burning charge too light; or the piston head or exhaust-valve poppet insufficiently cooled and becoming red hot while the engine is running under a fairly heavy load.

65. Premature ignition manifests itself by a pounding In the cylinder, and, if permitted to continue, a drop in speed, finally resulting in the stopping of the engine. It will also put an excessive amount of pressure on the bearings, especially the connecting-rod brasses, and cause them to run hot even when properly lubricated. After a shut-down due to premature ignition and a short period during which the engine is idle, allowing the overheated parts to cool off, it is possible to start again without difficulty and run smoothly until the conditions of load will cause a repetition of the trouble.

66. The remedies to be applied, according to the source of the difficulty, are as follows; Increase the water supply until the cooling water leaves the cylinder at a reasonable temperature, which may vary with the fuel used, but which should never be over 180 degrees F. Clean the water space and ports of any dirt or deposit so as to insure free circulation of the cooling water. Reduce the compression by partly throttling the air and fuel supply. Plug any sandholes or blowholes in the piston or valve heads, and make these surfaces perfectly smooth. Replace electrodes or other light parts with more substantial ones, capable of absorbing and carrying off the beat without becoming red hot. If necessary, arrange for cooling the piston by blowing air into the open end of the cylinder.

If the head of the exhaust valve becomes too hot, it is a sign that it is not heavy enough, and it should be replaced by one with a head of sufficient thickness to carry off through the valve stem the heat imparted to it by the combustion. If a small particle of dirt lodges in a remote portion of the combustion chamber, the richer part of the charge may not reach it until the piston has traveled over a considerable portion of the compression stroke, and the resulting self-ignition may properly be called preignition. It is advisable, therefore, to examine thoroughly every part of the combustion chamber and remove any dirt that may have lodged there.

67. Preignition in engines is Indicated by early ignition with a retarded spark. Usually, the engine will continue running for several seconds after the switch has been opened. The knock due to preignition has a sharp, metallic ring, easily distinguishable from other knocks in the engine. Even if ignition is not actually started by hot carbon or other cause, the first increase in pressure after the spark occurs may produce spontaneous ignition of the mixture near the heated object, so that the charge burns from two or more points at once, thus spreading the flame far mom rapidly than usual.

If the engine has two or more cylinders, and only some of them incline to preignition, the result is that it is impossible to time the ignition correctly for all cylinders. The cylinders having a tendency to preignition must receive a late spark to prevent combustion from being completed too early, while the other cylinders will require an early spark. It follows from this that it is impossible to get the engine to develop its full torque, or turning moment, unless it is running so fast that the tendency to preignition may be neglected. As the effect of preignition is to cause combustion to be completed before expansion has begun, it is dangerous to run the engine slowly, and this is true even if only one cylinder is preigniting. If the engine is running at good speed, with an early spark, the symptoms will be those of rapid combustion in the cylinders affected; namely, a hardness in the sound of the explosion, without actual knocking, while in the other cylinders, if any the explosion will be soft. As the speed of the engine is reduced, and the spark retarded to suit, the hard sound of the explosions gives place to unmistakable knocking. A good test for preignition due to carbon is to start the engine with everything cold, and run it under load before the water in the radiator has had time to get hot. The bing! bing! bing! then is a sure sign.

In automobile as in stationary engines, preignition is brought about by incandescent carbon deposits in the combustion chamber, on piston head, or on valves, or by bits of loose carbon left after scraping out, etc. It is sometimes due to small, accidental projections on the inner wall of the combustion chamber or head, due to defects in casting. If these are located in the path of the hot gases, it will take wry little carbon deposit on them to overheat. Preignition is also caused by lack of water, resulting in general overheating.

It must not be supposed that all carbon deposits are due to neglect. Even the most scrupulous regulation of the best possible oil, and even the most efficient carbureter, will not wholly prevent a gradual accumulation of carbon, but it ought not to become troublesome in less than a season or two. A high-compression engine will, other things being equal, preignite sooner than one with low compression.

The only remedy for carbon deposit that amounts to anything is to scrape it out. To do this it may be necessary to take off the cylinders, but it may also be done in some cases by the use of special forms of scrapers that will much into the combustion chamber through the inlet-valve or spark-plug hole.

If it is impracticable to scrape the cylinders at once, the trouble may be evaded after a fashion by running throttled and by running on a lower gear at the first symptoms of a pound. Increasing the richness of the mixture will also prevent pounding by making the charges burn more slowly, but this brings its penalty by adding to the carbon already present, If this trouble is due to chance projections in the combustion chamber, then may generally be disclosed by an electric lamp and mirror and when the cylinders are taken off, the projections can be cut away with a cold chisel.

BATTERY TROUBLES

68. Weak Battery.-Missed explosions may result from a weak battery. An open-air test of the spark, by disconnecting a cable from one of the plugs or laying a screwdriver on the plug binding post, will show a weak spark when the battery is weak. It is sometimes difficult to determine whether the explosions are missed because the battery is weak or because of a loose connection or broken wire somewhere in the ignition circuits. The only reliable way to determine this point, unless one has a fresh set of cells in reserve, is to carry a battery tester and test the cells as soon as skipping occurs. The battery strength required will depend on the character of the coil, but it is not often that a dry cell showing less than 5 amperes on short circuit is worth retaining.

If both sets of dry batteries are so far exhausted, that neither will work the coil, the two may be coupled in series, which will generally make it possible to run the engine for some time longer.

A wet (or a dry-cell) battery for supplying the current will be exhausted after a certain period of time, and, if handled carelessly, its life may fall far below what may reasonably be expected. If a wet battery becomes exhausted through long service or accidental short circuit in its parts or connections, the contents of the jars must be emptied and the charge renewed. The manufacturer or dealer in electrical supplies furnishes full printed instructions with every set of renewals for batteries. It is generally false economy to try to use part of the old charge. In almost every case it is far better to throw away all of the original zincs, oxide plates, and solution, rather than to try to rejuvenate the cell by adding to or replacing part of its contents.

69. Current Leakage.-Sufficient leakage of current to make trouble (but not enough to be observed without testing with a magneto) may be due to moisture in the mica insulation of the insulated electrode or to a bridge of carbon.

When it is suspected that the trouble is due to either of these causes, it is a good plan to dry out the insulation thoroughly and clean the lower end with a brush or piece of waste and a little gasoline.

These troubles are more liable to occur when the batteries have become weak from use, or so far exhausted that they will not give sufficient current for ignition.

70. Testing Batteries. -- BY using a small electrical buzzer or bell each cell may be tested separately, and by the tone or sound it can readily be observed whether or not the battery needs renewing, as is often the case. A small pocket ammeter or voltmeter is very convenient for the purpose of testing batteries, but each cell should be tested separately, as the pocket apparatus will rarely stand the voltage or amperage of more than one cell. Occasionally the buzzer, bell, or voltmeter will show one of the cells of the battery exhausted or dead, and on its removal the battery will show sufficient strength for ignition purposes.

71. Reserve Battery Power.-While four or five dry cells, when new, will furnish sufficient current for ignition, it is customary to install six or even eight cells, so that, when they become partially exhausted, or it becomes necessary to remove one or two from the circuit, there will be a sufficient number left to supply the necessary current. It is, however, never safe to depend on a single battery. A reserve set of dry cells, carefully wired up, should always be carried in a dry box, for frequently when used in a boat the bottoms of the dry cells may become damp or the switch is liable to be left closed, with the electrodes in contact, with the result that, through the short circuit thereby produced, the battery will be exhausted and ruined in a very short time.

SPARK-PLUG DISORDERS

72. Broken Spark-Plug Porcelain.-The breaking of a spark-plug porcelain usually results in complete failure to ignite the charge in that particular cylinder, owing to the secondary current shorting, that is, short-circuiting, through the break The outer end of the porcelain will generally be loose when tried by the fingers.

The usual cause of breaking is screwing the bushing down too tight. If the asbestos packing is of uneven thickness, it may be necessary to screw the bushing quite tight to prevent leakage. Overheating and splashing of water on a hot porcelain will also cause breaking. Remedies for such trouble are found in using new asbestos packing and in providing protection from water, etc.

73. Soot on Spark-Plug Porcelain.-Soot on the spark-plug porcelain will cause misfiring, or total failure to ignite, when the battery is of proper strength and the vibrators on the coils are working properly. If the engine has more than one cylinder, probably one or more will be found to be working properly, and the one with the defective spark plug may be located by holding down one coil vibrator after another, thus stopping explosions in each cylinder in turn, until the vibrator feeding the inactive cylinder is reached. By listening carefully to the exhaust, when it Is known that one cylinder is misfiring, it will be observed that, when the vibrator of an active cylinder is depressed, it will cause a noticeable break in the cycle of explosions. When the vibrator of an inactive cylinder Is depressed, no such break will be noticed. It is, of course, necessary to know which cylinder is fed by each vibrator. A spark plug may be sooted to the extent of short-circuiting when in the cylinder, and yet spark properly in the open air, as the electrical resistance of air increases greatly when the air is compressed. If a plug is slightly sooted, and there is uncertainty as to whether the trouble is due to the soot or to something else, insert a fresh plug, substituting one from another cylinder, if there are no spare plugs at hand, and note the result. A primary sparker coated with soot will act nearly the same as a sooted plug; the extra current producing the spark will leak away to a considerable extent through the carbon instead of producing an effective spark.

The causes of sooting are too much lubricating oil, inferior oil, or a too-rich mixture. The overrich mixture will deposit pure black soot, whereas an excessive quantity of lubricating oil will produce, a rusty-brown deposit. Inferior oil may produce almost any sort of a deposit, according to its quality. A great excess of either good or bad oil will not burn completely before it reaches the plug, and will deposit on the latter a greasy mixture of carbon, tar, and oil. An engine receiving oil in such quantities as this will foul the plugs quickly, and energetic measures must be taken to get rid of the surplus oil.

If the sooting is not excessive, and if the cause is removed, the plug may be kept in action without cleaning by the use of an auxiliary spark-gap device, which may be connected to the binding post of the plug. The soot will then be gradually burned off.

74. Leaky Spark Plug.-If the leak is between the plug shell and the cylinder, it will be denoted by the hiss of escaping gas on the compression and power strokes. The plug may be screwed tighter or a new gasket used. If the leak is through or past the packing inside the plug, the same hiss will be beard, and in addition the outer end of the porcelain will show traces of soot after the gases have been leaking for some time. If the bushing of the plug has been screwed as tight as is prudent, with regard to the safety of the porcelain, it will be necessary to repack the plug. A plug allowed to leak to any noticeable extent will overheat, cracking the porcelain or burning the screw threads.

MAKE-AND-BREAK IGNITER TROUBLES

75. Poor Contacts In order to obtain a spark of sufficient size in the combustion chambers of engines equipped with the make-and-break system of ignition, it is necessary that a good contact be made between the two electrodes of the igniter plug before they separate. The current passes through the bearing of the movable electrode, and, if the contact between the bearing and the stem of the electrode is poor, only a weak current can find its way to the point of contact resulting in a feeble spark that may be too weak to fire the compressed mixture. Poor contact of the electrode may be caused by an inferior quality lubricating oil forming a thin layer of carbon (which is a poor conductor) on the stem, or it may be due to wear of the bearing and a loose fit of the stem. To prevent wear on the stem and bearing it is important that the seat of the electrode be kept tight, so as to prevent the heat of the burning charge from reaching the stem and to keep it as cool as possible. This will aid in keeping the stem well lubricated, as the oil cannot be burned and form the objectionable carbon deposit. At the same time, the electrode will move easily without sticking, which is essential to a prompt separation of the two contact points.

76. Short Circuits.-A ground or short circuit of the current is often responsible for difficulties or failures of the igniter. This maybe caused by carbonized oil on the exposed surface of the insulators, or by dampness between the mica washers if these are used for insulation. By placing the igniter plug in a warm place and drying it thoroughly, a short circuit of this kind can often be remedied.

77. Short-Time Contact.-The length of time during which the electrode points are in contact has a decided effect on the size of the spark. To test whether the contact is of sufficient duration, hold the two points together by exerting pressure by band on the movable electrode. If this is found to cure the trouble, it is a sure indication that the contact is too short, and the parts that make the contact must then be adjusted so as to prolong the time of contact. This is accomplished in some igniters by increasing the tension of the igniter contact spring, while in others the adjustment is made by changing the relative positions of the interrupter lever of the movable electrode and the blade of the igniter lever that operates it and presses it against the fixed electrode.

78. Dirty Contact Points.-The contact points must be kept free from rust or moisture, both of which will interfere with the making of a bright spark. An occasional cleaning of the points by the use of emery cloth is advisable. Moisture on the electrode may be caused by condensation of the exhaust gases if the electrodes are very cold, which is likely to be the case in freezing weather before starting. The remedy is to heat the igniter plug thoroughly before attempting to start the engine. If moisture deposited on the electrodes is the result of a leaky packing or gasket, or of a defect in the cylinder, allowing water to enter the combustion chamber from the surrounding jacket space, it is possible to overcome this temporarily by wiping the interior of the combustion chamber dry with cotton waste or similar material. In this way the water way be kept away from the igniter long enough to get the engine started; but the real source of the trouble should be remedied at the earliest opportunity.

COIL DERANGEMENTS

79. Vibrator Out of AdJustment.-If the vibrator sticks, the symptoms will be erratic firing; few or no explosions will be missed, but the impulses will sometimes be very weak because the sticking causes a very late spark. Too light a pressure of the contact screw will cause the engine to run weak and fitfully; too much pressure will exhaust the battery rapidly. Either condition will manifest itself to the practiced ear by the sound of the vibrator. Poor firing may be caused also by pitting of the contact points. This may be remedied by filing the contact points, which should bear squarely against each other, and readjusting the spring and contact screw.

80. Defective Condenser.-A condenser short-circuited or having one of the connections broken will show it by sparking at the trembler and timer contacts, and by rapid burning of the metal where the spark occurs. The only remedy is to send the coil to the factory for repairs.

81. Short-Circuited Coil.-A spark coil may short circuit from breakdown of the insulation in either the primary or secondary winding. The symptom is a poor spark or none at all, and refusal of the vibrator to work, even with a good battery. The only remedy is to send the coil to the factory for repairs. The spark coil must be kept in a thoroughly dry place, as moisture will surely cause trouble and will interfere with the current passing through the coil to the engine. If the spark coil is found to be moist, it can generally be put in serviceable condition by drying it in an oven.

WIRING TROUBLES

82. Break In Primary Circuit.-The symptoms produced by a break in the primary circuit, which includes all wiring except from the coil to the plug, or from a secondary distributor to the plugs, are intermittent or complete failure to spark, according to whether the connection is intermittently restored by vibration or is wholly broken, and failure of the vibrators to work.

The almost Invariable cause of breaks in the primary circuit is vibration, which will loosen nuts on binding posts and break wires in places sometimes quite unexpected.

The first step to be taken in remedying the trouble is to test every binding post, usually by shaking the wires with the fingers. If this does not disclose the trouble, hunt for a break in the wiring. It will generally be found close to a binding post, switch terminal, or other connection, where the bending due to vibration is most severe. As a last resort, close the switch, open the compression relief cocks, retard the spark, and turn the crank so as to make contact at the timer; then with a length of spare wire shunt successively each wire in the primary circuit by touching the ends of the spare wire to the ends of the regular wire until you have found the one with the break, The spare wire thus bridges the break in the regular wire and causes the igniter to operate. Then hunt down the break in that particular wire, or take it out and put in a new one. If the wire has a soldered joint, it will be brittle at that joint and may have broken; or, it may have been fastened in such a manner as to strain it; or a badly made and twisted joint my have worked loose. Note that the break may be between the timer and the coil, in which can it will affect one coil only. A wire is quite likely to break inside its insulation, or just at the point where the insulation has been stripped off. A troublesome kind of break is that which is opened only by the vibration of running, and is closed by the elasticity of the wire or insulation, or by the weight of the battery cells or other connected members, when the engine is stopped. A great deal of patience is sometimes needed to trace a break of this sort.

83. Short Circuit or Ground in Primary.-A short circuit or ground in the primary conductor is not a common trouble, and it can be avoided by the most ordinary care in insulating the primary. The symptoms are much like those due to a broken wire, but an ammeter test close to the battery will show that current is flowing. It is most likely to occur by the chafing through of the insulation of poorly supported wires, or by neglect to insulate properly some home made attachment in the circuit. It may be due to contact of the dry primary cells or bolts passing through the battery box. A little patience is all that is needed to locate the trouble.

84. Broken Secondary Cable. - As the secondary cables are short and thick, a break in them is an unusual fault. If the break is not too great, the current will jump it, and the sparking there will at once disclose the trouble.

85. Grounded Secondary Cable.-A grounded secondary cable, which is indicated by failure to spark when the vibrator is working, is generally due to the chafing through of insulation on a badly supported cable. Sometimes it is due to rotting of rubber insulation by heat and oil. If the secondary cable has been spliced and taped, the current will go through the tape unless the cable is well out of the way of grounded metal work near the splice. Such a cable may give a spark at the plug as well as at the ground, which will soon exhaust the battery.

The temporary remedy for a grounded secondary wire is to tie the cable clear of the metal work. The permanent remedy is to put in a fresh cable, adequately protected by fiber tubes or other insulating supports. A cable with a varnished exterior is the best, as it resists oil. A rubber-covered cable exposed to oil may be protected by a coat of shellac or a layer or two of tape.

86. Loose Electrical Connections. - To obviate failure to start because of loose or defective electrical connections, the ignition mechanism should be tested carefully. With the make-and-break system of ignition this is done by disconnecting the wire from the binding post or nut of the insulated electrode while the electrodes are in contact, and then snapping the end of the wire across the binding nut of the insulated electrode. If a good fat spark is produced when the wire slips off the nut, thus breaking the circuit, it is evident that the circuit is not defective beyond the igniter and that the contact between the electrodes is good.

If, with the wire connected to the insulated electrode and with the igniter contact points separated, a screwdriver were placed so as to make contact with the binding nut of the insulated electrode and with a capscrew studbolt, or some bright part of the engine, the production of a spark when the contact between the screwdriver and the nut of the insulated electrode is broken would indicate that no short circuit exists in the igniter. If, however, no spark should be produced on breaking contact with the screwdriver, it would indicate the existence of a short circuit that should be found and eliminated. Should a spark be produced on breaking contact with the screwdriver when the two electrodes are in contact, it would be evidence of poor contact between the points. No spark will appear on breaking the circuit when the contact between the points is good.

The break of a wire inside the insulation, while not of frequent occurrence, is harder to locate than a loose electrical connection. In cases where it appears impossible to find the trouble, the existence of the broken wire may be determined by running a temporary wire from the coil to the engine, spark coil, switch or battery, as the broken wire may be so situated as to show occasionally either an open or a closed circuit.

A loose rocker-arm fastened to the movable electrode will sometimes give considerable trouble that will be found difficult to locate. A very little lost motion where the shaft is small is increased rapidly; and, as soon as the shaft becomes the least bit loose, the pounding to which it is subjected will cause it to loosen very quickly.

Switches should have good, clean contact points, otherwise leaks will affect both systems of ignition.

TIMER TROUBLES

87. Tiimer Contacts Roughened by Sparking. Trouble due to roughening of the timer contacts by sparking is likely to occur in any timer in which the contact segments are inserted flush with the insulator barrel or internal ring, instead of projecting therefrom.

The symptom produced by roughened contacts is irregular firing, due to jumping of the contact roller or fingers. This is not noticeable at low speeds, but becomes marked as the speed increases. The remedy is to true the insulator ring and segments in a lathe, and, if necessary, put in a new roller or contact fingers.

88. Wabbling Timer.-Some timers have their stationary portion supported on the shaft by a very short bearing that quickly wears loose and allows the stationary portion to wabble out of its correct plane. This will cause irregular, firing or even misfiring. One may easily determine whether the cause of the misfiring is here or elsewhere by steadying the tinier with the hand. The remedy is to bush the bearing, and, if possible, to make it longer.

89. Incorrect Timing. -- With marine engines having make-and-break ignition mechanism, even if the current is sufficient and there are no leaks, the time of contact may be too short, may be made at the wrong point in the stroke, or may be broken when it should not be, owing to incorrect timing. The timing may be tested by turning the flywheel carefully in the proper direction, and noting when the contact is made and at what point the spark occurs. By scratching the flywheel at these points, when the engine is running satisfactorily, it is always a simple matter to correct any trouble in the time of sparking. Raising or lowering the igniter pin without following any particular rule or without knowledge of what one is doing is very bad practice, and is more likely to aggravate than to remedy the difficulty. It is evident that, in multicylinder engines, it is quite important that there should be for each cylinder the same relative time of making and breaking the contact, with the same length of time in contact.

CLOGGED MUFFLER

90. Habitual feeding of an excess of lubricating oil to the engine will gradually clog the muffler with a mixture of carbon and half-burned oil, which will reduce the power of the engine and be very difficult to remove.

The symptoms produced are loss of power and inability to speed up the engine when the mixture, compression, valve timing, and ignition are known to be good; if the exhaust pipes can be disconnected, the engine gives its full power at once.

To remedy the trouble, take Off the muffler and saturate the interior with kerosene, after which the deposit can usually be knocked, scraped, or shaken out.

GASOLINE LEAKS

91. Probably the most dangerous trouble experienced with marine engines is due to leaks in the gasoline tanks or piping. They are more likely to occur at unions than anywhere else, and all joints and fittings should be soldered or brazed, as well as screwed. Hence, the piping is not liable to be broken at the threads, reinforced as they are with solder. Unions should be very heavy, and should be examined for leaks carefully and often. Do not use a light or match, but rub the finger around the joint, when, if there is a leak, it may be detected by the odor that will remain on the finger. Small leaks may be stopped temporarily by means of cloth and shellac or soap. Insulating tape will be found useless for the purpose, as the gasoline is a solvent for the Insulating material.

A good cord closely and tightly wound will be found serviceable. Shellac and cloth bound on tightly and allowed to dry with no gasoline in the pipe will be found very effective in stopping leaks. It is necessary to be extremely careful of fire in the presence or suspected presence of gasoline, particularly when in the form of vapor and mixed with air.

WATER IN EXHAUST PIPE OR MUFFLER

92. The exhaust gases from stationary gas or gasoline engines contain a certain amount of moisture, part of which is condensed and deposited in the exhaust pipe or muffler, where it may become a source of trouble if no provision has been made to drain these connections properly or if the draining devices accidentally fail to perform their functions as expected. Especially during cold weather, when the condensation in the exhaust connections is greater then at more moderate temperatures, it is advisable to inspect closely the condition of the drain cocks. If neglected, the level of the water in the muffler may rise to such an extent as to prevent the exhaust gas from being expelled, first causing loss of power and finally stopping of the engine.

In engines in which the governor acts on the exhaust valve, and this valve is kept open while running under light load, the trouble from water in the exhaust, when no charges are admitted to the cylinder, is naturally intensified, on account of the fact that a portion of this water is drawn into the cylinder while the valve is open during the suction stroke. The presence of water in the exhaust connections is usually indicated by steam or water spray issuing from the end of the exhaust pipe.

As before stated, water is frequently used for deadening the noise of the exhaust by introducing it in a small steady stream into the exhaust pipe and allowing it to be carried off in the shape of vapor or spray with the exhaust gases. In such cases, the draining devices require particular attention, because, in the case of failure to have a free outlet to the drain for any part of the water not carried off with the exhaust, the accumulation of water would in a short time be sufficient to stop the engine.

WATER IN ENGINE CYLINDER

93. An accumulation of water in the cylinder - a condition encountered more or less frequently in marine practice - will effectually prevent a gas engine from starting. The water may get in through the exhaust pipe because the installation is faulty, because the exhaust extends below the surface of the water, or because there is a leak due to a crack in the cylinder or to a broken and imperfect gasket between the cylinder and the water-jacket. Running the exhaust cooling water into the engine exhaust is a frequent source of such trouble.

Provided the trouble from water in the cylinder is not due to leaks the remedy is to remove the water entirely, by means of absorbent materials, through any openings there maybe in the cylinder. The insulated electrode should then be carefully dried out, the defect in installation remedied by changing the exhaust piping to drain outboard, and, if exhausting below the surface of the water, a vent provided in the highest part of the exhaust piping.

FAILURE TO GOVERN

94. If the connection between the governor and the throttle is too long, the throttle may fail to close until the governor balls have been moved out to an excessive extent by the speed of the engine. In an old engine, wear of the connecting links may produce the same result. Sometimes there is an adjustable screw and nut connection between the governor and the throttle, and this is easily adjusted. Sometimes, however, it may be necessary to bend the rod connecting the two The throttle should be opened, and its position when barely open should be marked in such a way that it will be known when the throttle is reassembled. Then the engine should be run idle and the position of the governor lever noted when the engine is running at the speed at which it is desired that the governor should act. With these particulars known, it is easy to shorten the rod to bring the throttle to the desired position. It should be remembered that a very slight opening of the throttle is sufficient to keep the motor running.

REFITTING PISTON AND PISTON RINGS

95. It is practically impossible to turn a piston in a lathe so as to fit the cylinder in such a manner that the engine will run properly even under a partial load. The best that can be done is to have the cylinder bored slightly larger at the end nearest the crank-shaft, so that the piston can be pushed in easily from this end and will fit rather Snugly at the other end near the combustion chamber. To put the piston and cylinder in condition to stand constant running under load necessitates filing the surface of the piston by hand, as follows; See that both cylinder and piston are thoroughly clean and free from dust or filings. Apply a liberal amount of lubricating oil, place the piston in the cylinder, and attach the connecting-rod to the crank-shaft. Start the engine, and let it run idle for a while. As soon as the heat of the explosion causes the piston to expand, it will begin to stick in the cylinder, as the water-cooled walls of the cylinder do not expand to the same extent as the piston. The sticking is manifested by a pounding or knocking sound caused by the very slight amount of play that necessarily exists in the bearings of the connecting-rod at both the crankpin and the piston end. As soon as this pounding appears, apply more lubricating oil to the piston, and let it run for a few minutes in this manner, without any load. Then stop the engine, take out the piston, and wipe it dry. The portions of the piston that bear hard against the cylinder will be indicated by glossy spots, which should be carefully filed with a smooth, flat file, removing only a little at a time. To facilitate filing, remove all traces of lubricating oil by means of kerosene. After filing the piston surface in this way, clean the piston, put it back in the engine, and start up again. It will be noticed that it is now possible to run the engine for a longer period without any pounding in the cylinder and perhaps to be able to put on a light lend for a short time. Do not keep the engine running with any load for any length of time, so long as there is any pounding noticeable. This operation may have to be repeated from four to six times, depending on the skill of the operator, before the engine can Tun steadily with the usual maximum load.

These instructions apply also to cylinders that have been rebored and fitted with new pistons, as the conditions in this case are the same as in a new cylinder.

96. The piston rings also require fitting in a similar manner, and in this connection the following points must be observed: Before placing the rings in the grooves, each ring should be tried, to ascertain that it fits in the groove for which it is intended. If the ring is found too thick, place it on a straight board, and hold it in place by fastening three or four nails within the ring, driving them down until the heads are slightly below the top of the ring. Having thus secured the ring on the board, file it carefully and reduce its thickness so as to get an easy sliding and uniform fit in its groove.

The rings can now be put in place by opening them and slipping them over the piston from the closed end. In doing so, the rings should be expanded and twisted as little as possible. The first ring must be placed in the groove farthest away from the closed end of the piston, the others following in order. If, after running the engine with new rings for a short time, the rings show that they bear hard and unevenly, the hard-bearing portions must be touched up with a fine file. Should it become necessary at any time to replace a broken ring located between other rings, the use of small pieces of thin sheet tin will be found of advantage. They are slipped in between the inside of the ring and the outside of the piston, at a convenient point of the circumference, so as to keep the ring evenly expanded and enable it to be moved laterally over other rings already in place to the groove for which it is intended. Having reached its groove, the pieces of tin are withdrawn, and the ring is allowed to enter the groove.

A ring that, from undue expansion or twisting, has lost its original diameter will not bear evenly and will wear out the cylinder in a short time, causing leakage and loss of power.

REPAIRING CRACKED WATER-JACKET

97. Neglect in draining the cylinder jacket when stopping the engine after the day's run may result in cracking the outer shell in cold weather, owing to the freezing of the water. It is very seldom that the inner cylinder is damaged in such a case, but if it should happen to be injured, the casting is generally rendered useless and must be replaced with a new one. The outer shell, being much lighter than the cylinder itself, provides a safeguard against damage to the latter, and in most cases, if the cylinder and jacket are, cast in one piece, it will be possible and economical to repair the cracked shell.

The following directions are intended to cover repairs for various kinds of cracks, and apply to cracks in cylinder jackets proper, as well as to cracks in the outer shell of cylinder heads or valve casings of larger sizes. In large castings it will pay to repair the part, rather than replace it with a new one; but with small castings it may be found to be more convenient and cheaper to replace the heads or casings with new ones.

Fig. 4 (a) and (b) shows a cylinder, the outer shell of which has been burst by frost. The crack a b extends only a portion of the entire length. After the ice has been thawed and the jacket emptied, the first thing to do is to drill two holes a and b, about 1/4 inch in diameter, at the ends of the crack. The purpose of these holes is to prevent the crack from extending any farther on account of the chipping necessary in the next operation. Then take a chisel about 3/16 inch to 1/4 inch wide and cut a groove along the line of the crack, dove tailed as shown at c in the sectional view of the cylinder and jacket, Fig. 4 (b), the groove being widest at the bottom.

Next secure a piece of 1/4 inch round copper wire, well annealed, and hammer it tightly into the groove. By careful calking, a crack of this nature can be made perfectly tight.

98. Fig. 4 (a) also shows a crack d e extending from one of the water ports to the outer end of the cylinder. In such a case, it will be necessary to shrink a steel band f on the end of the cylinder, before the crack is chipped out and calked in the manner just referred to. Use a flat steel band about 1/2 inch by 3/8 inch, and be sure that the finished end of the cylinder projects about 1/8 inch beyond the band when in place.

If the crack extends over the entire length of the jacket, as shown at g h, it will require additional bands i and j as shown. If the cylinder has finished collars at the ends, as is frequently the case, it will not be possible to slip the ring j over the end of the cylinder into its proper place, unless an auxiliary band k, open to the extent of about 1/4 inch as shown at l, is first placed on the cylinder. This band k must, of course, be thick enough to make up the difference in diameter of the cylinder body and the finished collar. In shrinking rings on a cylinder, they should be heated to a dull red beat and must be handled dexterously, as the cooling takes place rapidly and the ring may shrink so as to stick before it reaches its position if not applied quickly. After the bands have been put in place and have been found to be tight, the cracks should be grooved and calked as directed.

If a crack should develop in the surface of a joint between the cylinder and one of the valve casings attached to it, and if this crack crosses the port through which the entering charge or the exhaust gases pass, as shown at m n, Fig. 4 (a), it will be practically impossible to repair the casting in such a manner that a packing can be made to stand, and the only remedy is to replace the damaged part with a new one.

99. Another method of repairing a short crack in the surface of the jacket wall consists in applying a piece of steel boiler plate, about 1/8 inch thick. Before putting on the plate, two 1/4 inch holes should be drilled at the ends of the crack, to prevent it from going farther, and a V-shaped groove cut along the crack from end to end. The plate must be bent so as to conform to the shape of the cylinder jacket. A packing of thin asbestos wick soaked in white-lead paste is now put in the V-shaped groove, after which a packing of sheet asbestos the size of the plate and dipped in water is placed over the surface to be covered by the plate. Now apply the plate, which is held in place by a number of 1/4 inch to 3/8 inch screws, the size of the screws depending on the thickness of the water-jacket. The screws should be about 1 inch apart, 1 inch on each side of the crack; and, if possible, the tapped holes in the jacket, in order to prevent water from leaking past the screws, should not be drilled all the way through.

If the jacket is so thin as to make it necessary to drill the holes all the way through, each screw head must be packed with hemp or asbestos soaked in white lead.

100. An engine water-jacket split by freezing is also sometimes repaired by the following methods: If the crack is very small it may be rusted up. For this purpose, a saturated solution of salammoniac is made and poured into the jacket. A plug, screwed into one of the water openings, is drilled and tapped for a small tube, by which air pressure is put on the liquid in the jacket by means of a tire pump. The cylinder is so laid that the crack is at the bottom, and after several hours it will be found that the edges of the crack have rusted solid from the action of the salammoniac.

Another method of closing a crack is that shown in Fig. 5. The process is to drill and tap a series of 1/8 or 3/16 inch holes as close together as practicable for the entire length of the crack, the first and last holes being at the extreme ends of the crack, in order to prevent it from extending farther. These holes are plugged with cast-iron plugs turned and threaded for the purpose, and the job is completed by rusting in with the salammoniac solution as just described. When brazing facilities are available, it is much better to braze a cracked cylinder than to try rusting it, as the chances of securing a permanent repair are much better.

REPAIRING BROKEN ENGINE BED

101. The breaking of the studs or bolts that hold the connecting-rod box to the rod will often wreck an engine, involving the breaking beyond repair of the piston, cylinder, and even the bed. As the bed is usually a rather costly part to replace, it is frequently found possible to repair it with the aid of a strong steel rod properly applied.

A break repaired in this manner is shown in Fig. 6. It is possible to make this kind of repair only when there is a clean separation of the casting in two pieces; if the bed is broken into a number of small pieces, it must be replaced with a new casting.

To repair a bed, as shown in Fig. 6, first be careful to preserve the two pieces so that they will fit exactly when put together, using every precaution against careless handling and further damage to the surfaces that form the joint. Then investigate and find the best way in which the steel rod should be run so as to take hold of the strongest available part of the bed.

The figure shows the rod running inside of the double wall casting, a 2-inch rod being used in a space 3 inches wide, and being secured by two nuts at each end. The line a b indicates the break of the bed casting. At c and d are cast iron washers made to conform to the shape of the casting and providing a straight surface for the nuts of the bolts e to rest on. It is important that the nuts should bear squarely against these washers to avoid any excessive stress on the bed casting. Jamb nuts or some other locking device must be provided to prevent the nuts that hold the bed together from becoming loose as a result of the shocks and jars to which the casting is subjected while the engine is running. A frequent inspection of the tightness of the nuts is advisable.

REGRINDING VALVES

102. It is not often that inlet valves must be reground, because they remain comparatively cool under the influence of the incoming charge, and, moreover, the seats are not exposed to the erosion of burning gases. Exhaust valves, on the other hand, require regrinding at intervals, depending somewhat on the temperatures in the cylinder, and to a large extent on the material of which the exhaust valves are made. Ordinary mild-steel valves must be reground quite frequently. A much better material is an alloy of nickel and steel containing a high percentage of the former metal, usually about 25 per cent. Such an alloy as this has a very small coefficient of expansion, and is less subject to erosion due to the heated gases. Moreover, it is not liable to warp out of shape.

For large engines, and occasionally for small ones also, cast iron has been found to be a very good material for the exhaust valves. If cast iron is used, the stems and heads are made separate; the stems are made of steel, and the heads are riveted on the stems. The only drawback to cast iron for this purpose is that it has not the strength of steel, and the valve head must be of unusual thickness, which, of course, adds to the weight and inertia of the valve.

103. Inlet and exhaust valves are reground with emery. If an exhaust valve, the spring is first slipped off to make sure that there is no sidewise pressure on the stem to prevent a true bearing of the valve on its seat. The emery is mixed with oil until it forms a paste, and is applied freely to the surface of the valve and its seat. Extreme care must be taken to prevent any of the emery from getting into the interior of the cylinder, where it would quickly ruin the piston and the cylinder walls. In some cases, a plug of waste can be thrust into the valve chamber between the valve and the piston; but, if the chamber is not long enough for this, the work will have to be watched carefully, using an electric light, if necessary, to see that none of the paste works away from the valve toward the piston.

104. If the valve seat is badly out of true, the operation of grinding may be begun with emery of medium coarseness; but this is seldom necessary, for the reason that, before the valve had reached such a condition, the cylinder in question would have lost almost all of its power. In any case, the work is finished with fine flour of emery. The emery being applied, the valve is set into its place in the valve seat, and a screwdriver is used in the slot in the valve head to rotate the valve, which should be worked by quarter-turns back and forth with moderate pressure, and should be lifted at frequent intervals to allow the paste to work in between the valve and its seat. In order to grind the valve evenly all around, it should occasionally be advanced a quarter-turn, and the grinding-in process continued. When the grinding is almost finished, the pressure should be comparatively light.

If the valve has been pitted, it will not be necessary to grind it until the pits have entirely disappeared, so long as there is a good bearing around them.

When the work is finished, the ground portion of the valve should have a smooth, dull appearance, and neither the valve nor its seat should at any point be bright, as this would indicate that metal had been rubbing on metal without emery between.

105. After the valve has been reground several times, it is likely to have settled so much lower in its seat as to cause the valve stem to remain in contact with the push rod when the valve is supposed to be seated. When the valve is closed, the clearance between the valve and the push rod should be fully equal to the thickness of an ordinary visiting card. If the distance is less than this, any slight irregularity in the cam, or some slight springing of the metal parts when the engine is running, might bring the valve stem and the push rod together and cause the valve to be opened slightly.

106. In an old engine, it may be found that the bushing or sleeve in which the valve stem runs is worn to such an extent as to permit considerable sidewise movement of the stem. A valve in this condition will still operate if it has been carefully ground, but it is likely to need grinding much oftener than if it were truly guided by its bearing. It should never be ground with the spring washer merely blocked up; the spring should in each case be wholly removed.

RENEWING BABBITT-METAL LINERS

107. When a babbitt-lined bearing becomes overheated and the trouble is not noticed in time, the soft metal of the lining, which may have a tin or a lead base, will melt and run out of the box. While in some engines the Babbitt metal is cast directly in the rough bearings of the engine bed, it is the general practice in a first-class engine to bore out the bearings in the bed and fit them with cast-iron or bronze boxes lined with Babbitt metal.

If the journals of the shaft are in good condition after the metal has been melted and run out of the box, the method of rebabbitting the bearing is the same as was followed at the time the box was made at the factory. To reline the box in such a case, proceed as follows: Remove all traces of the original lining from the box. While melting the new metal in the ladle, place the box on its end on a flat-finished surface, and insert an arbor, from 1/8 to 1/4 inch smaller in diameter than the journal, in the center of the box, being careful to have an evenly divided space all around the outside of the arbor. The box a being made in halves, as shown in Fig. 7, place shims b, b made of cardboard 1/32 inch thick between the joints, having the shims extend well into the space around the arbor so as to allow only a thin strip of the Babbitt liner c to connect the halves of the lining, in order to facilitate the breaking apart of the lining after it has been cast. It is well to heat the box to a moderate degree before the Babbitt metal is poured into it, as it will aid in the free flow of the molten metal. After being put in position, apply clamps d that hold the halves of the box together. Then pour the Babbitt metal into the box until it is full, and, after allowing it to cool, take off the clamps and separate the box. The metal is then peened with the ball-shaped end of a machinist's hammer, until it adheres firmly to the box. The joints are then dressed so as to be flush with the parting of the box. After clamping the halves together, the box is bored in a lathe to the size of the journal, or, rather, to a small fraction of an inch smaller than the journal, so as to allow making a perfect fit by scraping the box by hand.

108. If the journals of the shaft are damaged during the heating and melting of the Babbitt metal, the shaft must be taken to a lathe and the journals trued up before the new lining is prepared. The same should be done if the journal has become unevenly worn or if it shows flat spots. If the wear of the shaft is limited to unevenness of the diameter at various points of the length of the bearing, it may be sufficient to pour the Babbitt metal around the shaft itself, instead of using the arbor as just explained. In such a case, it is best to place the shaft in the engine bed and close up the ends of the space around the journal with wet fireclay, so as to prevent the Babbitt metal from running out at these points while it is being poured. The same method should be pursued if the Babbitt lining is poured directly in the engine bed. Care must be taken in all cases that the box or bearing is perfectly dry before any hot metal is poured, to avoid sputtering of the metal and possible injury to the operator.

PACKING RENEWALS

109. To replace a packing, loosen the nuts or bolts that hold the two parts together, take the valve chamber or cylinder head off, and carefully clean the packing surfaces of any parts of the old packing, using a scraper or similar tool, but being careful not to mar the metal surfaces. If the packing has blown out repeatedly, it is advisable to examine the condition of the packed surfaces and ascertain whether they are true and have not been warped by heat, or drawn out of shape by excessive tightening of the bolts or studs.

If it is found that the metal surfaces are not true, they must be put in good condition by filing, or, better, by scraping, using a surface plate covered with a thin layer of red lead and oil for determining the high spots. In case studs are used in the cylinder, they must, of course, be removed before the surface on the cylinder can be trued up. The scraping must be continued until the test with the surface plate shows that the packing will have a perfectly straight and evenly bearing metal surface around the water ports a and combustion-chamber port b, Fig. 8, to rest against. The portion of the surface immediately surrounding the combustion-chamber port b, where the pressure is highest, should bear a little harder than the rest of the packing surface.

110. After the packing surfaces have been put in proper condition, cut the new packing from a suitable thickness of asbestos sheet, being careful to have the opening that corresponds with the port communicating with the combustion chamber large enough to allow no portion of the edge of this hole in the packing to project over the joint. Such projections of the packing material will become incandescent and produce back firing or premature combustion of the charge. Put back in place the studs that hold the valve casing to the cylinder, apply a coat of flake graphite to the side of the packing that is to rest against the valve casing, and carefully slide the packing over the studs until it bears against the metal surface. Then attach the valve or cylinder head to the cylinder, screw the nuts on the studs, and tighten them gradually and evenly. After everything has been put in order, start the engine and run it under a light load or idle, until it begins to warm up, when it is found that the nuts can be tightened up still more. This should be done promptly, as neglect to take up any expansion by the heat of the combustion may cause the new packing to become leaky soon after it has been put in.

111. While the packing surfaces must be true and straight, it does not follow that they should be as smooth as glass. Experience has shown that a grooved packing surface gives much better results than a perfectly smooth one, although many manufacturers seem to take great pains to make the packing surfaces as smooth as possible. In many cases, troublesome joints have been permanently cured by the judicious application of grooves in the metal surfaces. The packing fills the grooves and prevents the escape of gas between the packed surfaces. Fig. 8 shows, in dotted lines, the positions of the grooves c, which in small surfaces may be 1/32 inch deep and 1/16 inch wide. On circular surfaces, such as the packing surface between the cylinder and the cylinder head, shown in Fig. 9, the grooves should be cut concentric, and should not come opposite each other; but, when placed together, the groove a in the cylinder b should be half way between the grooves in the head c, as shown.

112. Whenever possible, the edge of the packing should be protected against the pressure by a projecting rim d, that enters the end of the cylinder, as shown in Fig. 9. If not originally provided by the maker of the engine, it will pay the user to have the rim attached by riveting it to the cylinder head, in case of persistent trouble with the packing of this joint. The depth of the projection d should be about 1/4 inch, and it should fit rather snugly in the bore of the cylinder, but not so that much force will be required to insert the head.

113. As the material employed for gaskets is usually asbestos alone, or asbestos, wire gauze and graphite or similar filler, a knife or pair of scissors makes very little impression on it; but it can be cut out very readily if laid on the cylinder head and carefully cut around on the outside with a light, flat-faced, round-peen hammer. The holes can then also be cut with the round peen. Great care should be exercised not to pull out any wires from gaskets in which wire gauze is used. The wires should be cut off very carefully.

If the material used is ordinary asbestos paper 1/100, 1/64, 1/32, or even 1/16 inch thick, it should be thoroughly soaked with linseed oil, either raw or boiled, and dusted carefully with powdered or flaked graphite, or with graphite foundry facing that contains talc, etc., which is a very good substitute. It is a good plan to let this dry a little while in the air, when it becomes much tougher. It should not, however, be allowed to get too dry. When put in place, the holding nuts should be screwed down carefully, going over them several times and screwing down opposite nuts instead of adjoining ones. The engine should then be started and run a few minutes, with the compression relieved and the circulating water turned off, in order to heat up the engine and assist in drying out the oil or any dampness in the gasket. The nuts should then be tightened carefully, when the water may safely be turned on. If these directions are followed closely, and the gasket is not defective, it should last a long time. The oxidation of the linseed oil will make the gasket tough, and if it is dusted with graphite every time the cylinder head is removed it should be very durable.

In using a gasket of asbestos and wire gauze having material on one side to make it adhere to the cylinder top, the opposite side being treated with graphite, there is no need of treating the gasket with linseed oil. A gasket of this sort is almost indestructible when care is exercised in tightening the holding nuts when the gasket is new.