US2324071A - Method of fueling solid fuel injection engines - Google Patents

Description

July 13, 1943. M. G. FIEDLER METHOD OF FUELING SOLID FUEL INJECTION ENGINES Original Filed Nov. 12, 1937 4 Sheets-Sheet 1 July 13, 1943. M. G. FIEDLER 2,324,071

METHQD 0F FUELING SOLID FUEL INJECTION ENGINES Original Filed Nov. 12. 1937 Q r21 1 Zia/ 290E012 My 4 Sheets-Sheet 2 112mg E pressure 4 00 Zg gg 57 I2.

case-a /a8 July 13, 1943- M. G. FIEDLER METHOD OF FUELING SOLID FUEL INJECTION ENGINES Original Filed Nov. 12, 1937 4 Sheets-Sheet 3 WSW 4 Sheets-Sheet 4 M. G. FIEDLER METHOD OF FUELING SOLID FUEL INJECTION ENGINES Original Filed Nov. 12, 1937 July 13, 1943.

Patented July 13, 1943 mum) on FUELING soup FUEL misc-nos suomss Max George Fiedle'r, Media, Pa., assignor to Fledler-Sellers Corporation, Philadelphia, Pa., a corporation of Pennsylvania Continuation of application November 12, 1937, Serial No. 174,238. This application January 2, ,915. In Canada April 3 Claims. (01. 123-21) This invention relates to compression ignition 1941, Serial No. 372 1937 engines of the oil-burning type, or that type usually referred to as Diesel engines, and more particularly to a method of operating theseengines in a manner such that they may operate on what is known as the Otto cycle; may be constructed and operate in the automotive speed sizes and ranges; and will operate without detonation throughout such speed ranges, and to provide a method of fueling the enginecontributing principally to the attainment of the foregoing .objects.

It is well known that the combustion of hydrocarbons may be either a direct oxidation or a decomposition followed by oxidation of the destruction products. In practice, there is a race between the two processes, the conditions being more favorable to hydroxylation when a fuel has been properly .divided and mixed with air before it is burned, at which time the flame is blue and has no tendency to soot. The'oonditions are more favorable to destructive combustion when the fuel is exposed very suddenly and in a highly vaporized condition to flame temperatures, the fuel particles decomposing rapidly before they can find oxygen, and under these conditions there is a yellow radiation caused by the burning carbon and a tendency to form soot.

In the ordinary Diesel engine, air is compressed to the greatest possible extent to prevent ignition lag, and a readily ignitible oil is employed for the same reason. I have found that in high speed Diesel engines having a high compression ratio the first fuel entering the combustion space does not ignite but meets with the turbulent compressed air. Part of the entering fuel will mix with more or less turbulent air until, in some part of the chamber suitable mixture for self-ignition has been established and combustion will start. If the duration of injection is continued after combustion has occurred this fuel will meet air which is not only strongly turbulent but mixed with combustionresidue. As injection continues the air becomes saturated with the products of combustion until the interference with further combustion is so serious that free carbon will be generated. In the ordinary Diesel this occurs at a point where approximately 50% of the available air has been consumed. The time necessary to establish the proper mixture for the first autoignition is equal to the ignition lag and influenced to a great extent by turbulence during injection, but also by the shape of the combustion chamber, the spray characteristics, duration of in-' pressure. To a large extent the ignition lag in the engine is directly dependent upon the degree of turbulence, increasing with such turbulence due to the fact that the air usually rotates at high speed around the cylinder axis throwing the entering fuel into the coldest zone adjacent the cylinder wall. This can only be counteracted through high compression and it follows that previous engines having high turbulence must work under very high compression pressures, often as high as 40 to atmospheres, and even under these circumstances detonation frequently results.

Analyzing the standard spray characteristics as produced today in the usual solid fuel injection engines, it is found that the fuel, through being Zubjected' to high injection pressures (3,000 to 0,000 lbs.) and through being forced through very small orifices 15- substantially in a vapor stage.

In addition to that, as has been photographically demonstrated, the spray itself is very compact and cannot be broken up even by the most violent turbulence. high compression temperatures without the ability to mix with air and the result is, inevitably, cracking. The combustion will, therefore, follow the second type mentioned above and since part of the combustion will be a hydrogenoxygen or oxygen-methane reaction at the high period necessarily, and of course purposely under the Diesel system, continues after combustion has actually begun and, obviously, the later injected fuel will crack producing further hydrogen-oxygen reactions.

I have discovered that proper operation may be provided by observing the following precepts:

1. In small bore, short stroke engines the use of a relatively large combustion space and the maintenance of the air in this space in ahigh state of turbulence in order that the mixture may be as complete and rapid as possible.

2. The instantaneous, or substantially instantaneous injection of the fuel into the combustion chamber, and the use of low compression pressures (approximately 240 pounds at the time of starting injection) thereby causing a combustion lag, in order that all of the fuel may be delivered thereto before combustion begins.

3. The introduction of the fuel spray into the combustion chamber in a form in which it is jection, and, to some extent, by the compression loosely bonded and under relatively low inJection This spray is, therefore, exposed to pressures (preferably less than 1200 lbs.) thereby enabling the fuel to readily combine with the air in the combustion chamber and establish an autoignition mixture as rapidly as possible.

This application relate particularly to the method of introducing th fuel spray into the combustion chamber, and i a continuation of my prior application Serial No. 174,238, filed November 12, 1937, for Fuel nozzle for solid fuel injection engines and method of operating the same."

These conditions may be satisfied by use of the methods hereinafter described and the apparatus disclosed in the accompanying drawings wherein:

Fig. 1 is a sectional view through an engine suitable for operation in accordance with my invention;

Fig. 2 is a cycle diagram of the engine operation;

Fig. 3 is a conventional injection nozzle of the type at present in use;

Fig. 4 is a sectional view through a nozzle modiiied for use in accordance with my invention;

Fig. 5 is a diagrammatic view illustrating the type of spray produced by the nozzle of Fig. 4;

Fig. 6 is a sectional view showing a preferred type of nozzle and the pump connection thereto;

Fig. '7 is a highly enlarged sectional view through a portion of the nozzle of Fig. 6;

Figs. 8 and 9 are card diagrams showing operation of the engine with the ordinary type of injection; and

Figs. 10 and 11 are card diagrams taken under identical conditions with those in Figs. 8 and 9, but with use of the new type of injection and operation.

In attaining the first of these precepts it is possible, and preferable, to employ, particularly where small-bore, short-stroke engines are being utilized, a structure such as shown in Fig. 1 comprising, briefly, a working cylinder ID, a supercharging cylinder H, a common crank shaft I2 controlling the operation of the piston i3 and M of these cylinders, and a common crank case l5 for the cylinders. Communication between the upper ends of the cylinders is through a valve i6, disclosed as of the rotary type, preferably so arranged that the valve itself acts as a storage reservoir for the final compression of the supercharging cylinder and subsequently delivers the stored pressure to the working cylinder at the beginning of the compression stroke thereof. As shown in the drawings, the compression space I! is relatively large as compared with that of the usual compression ignition type engine, and the supercharging cylinder i of such size that compression ignition pressures may be provided in this chamber but of a comparatively low order. Scavenging of the working cylinder is attained by injecting air compressed in the crank case l5 by the simultaneous downward movement of the pistons l3 and I4, there being a port It! which, when the piston 14 is in its lowermost position, places in communication, through an opening i9 in the piston of the supercharging cylinder, the interior of the crank case and the interior of cylinder l0. Relatively highly compressed air provided through compression by both pistons swirls violently in the cylinder i0, setting up a high degree of turbulence which is maintained until the working piston has again passed through its compression stroke with the supercharging and the charge has been admitted as hereinafter described. Thus, while a large volume of air is retained in the cylinder, a charge ous complete mixture of the air with' the injected fuel.

The second precept may, obviously, be obtained in a variety of fashions as, for example, by multiplying the number of nozzles employed and thereby increasing the effective injection area so that the period necessary to injection of a predetermined amount of fuel may be reduced to the desired point. I have found that the injection period should not exceed 12 of crank travel, and preferably should be confined to approximately 7 thereof. Since the engine is now to operate as a constant volume engine, all of the fuel being injected before ignition, it is, of course, desirable that thi fuel be entered prior to the arrival of the working piston It at top dead can ter, the arrangement being preferably that illustrated in the cycle diagram forming Fig. 2, in which the injection is illustrated a occurring in advance of the top dead center, a distance corresponding to the average combustion lag. While the injection period might be delayed beyond the point illustrated, this will, obviously, result in a loss of efficiency, the operation of the engine under such circumstances being substantially that of gasoline engines operating with a retarded spark. The compression pressure of the engine must be kept sufficiently low to insure a combustion lag enabling complete injection of the charge prior to initial combustion, and must be sufficiently rapid to insure against vapor formstion. It will be noted that this is directly contrary to Diesel practice in which the production of highly vaporized sprays is sought and in which the compression pressures are carried to the highest possible point in order to avoid ignition lag. My invention, as so far described, is disclosed in my prior Patent No. 2,202,761, issued May 28, 1940, for Internal combustion engine.

Such an arrangement as that already suggested will result in a highly improved operation of the engine, but to insure complete elimination of detonation, a smooth operation of the engine over a wide speed range, and an economical fuel injection system, the construction should be restricted to a single injection nozzle of peculiar characteristics. As is well known to those familiar with Diesel construction as hitherto practiced, the usual nozzle comprises a valve lifted by the injection pressure and thus permitting the escape of the fuel to the cylinder through orifices II (see Fig. 3). Due to the relatively large area of the lifting surface 2| provided on such valves, with the attainment of injection pressure, 'the valve is violently thrown upwardly against a stop 22 and there remains throughout the injection period, the restricted orifices providing sumcient back pressure to maintain it in this position.

I have found that by materially reducing the lifting area of this valve, as at Ila in Fig. 4, and increasing the diameter of the discharge openings 20a. to an extent such that the pressure beneath the valve is constantly relieved and at the same time providing the valve with a relatively heavy spring of high frequency, the valve will chatter against the seat or rather upon the fluid passing over this seat, reducing the fuel to a foam which is discharged through the openings 200. at a much reduced pressure and in relatively large particles. The increase of the size of the openings 20a requires a thickening of the wall 23 through which these openings are formed and insure correct distribution thereof.

in order that they may properly guide the fuel The fuel then has a spray characteristic typified in Fig. 5, and is in the form of loose particles which may be readily taken. up and surrounded by the air in the combustion chamber, with the result that a thorough commingling is obtained, insuring proper proportioning of air to fuel at the time the mixture attains the flash point. As indicated above, the openings should be much larger than those ordinarily employed, and in an injector found to give excellent results 6 openings were employed of a diameter of .9 millimeter as compared to 6 openings having .3 millimeter utilized in the usual injection nozzle; in other words, an

increase of approximately 900 per cent in injection area.

In general, it may be stated that the area of the openings must exceed the maximum area afforded by the valve seat 24 and the valve 25 during the injection period. It will be obvious that the valve, during the injection period,'acts not only as a valve but likewise as a flexible oriflee, through which the fuel may enter, exercising upon the fuel because of its flexibility an agitating action producing a relatively loose foam. The valve seat itself should be of relatively large diameter in order to obtain large impact relief areas with a minimum lift and the valve spring should be strong enough so that at minimum pressure it will counteract the inertia effect of the valve stem under impact and continually attempt to reseat the valve and thus,

produce the chattering action on the fuel passing over the seat. To this end, as previously stated, a heavy valve spring having a high frequency is preferable.

While the injection nozzle of Fig. provides a highly improved operation, it was found in practice that this nozzle, after heating to approximately 700 caused detonation. It was finally determined that the foam formed in advance of the nozzle tip by the chattering of the valve upon its seat tended to vaporize at the nozzle tip with resulting detonation in operation. For this reason, the valve of Fig. 6 was developed. In this valve the lifting area is transferred to the center of the valve, ports 28 communicating with a passage 28 opening through the bottom of the valve and into a chamber 30 at the nozzle tip. Thevalve seat is divided into two sections 3| and 32 by a groove 33 aligned with the discharge ports 34. By this construction all foaming fuel is discharged from the valve and the tip of the nozzle is kept cool by maintenance of a solid body of the fuel thereagainst.

Tests with an injection nozzle of this type have proven conclusively not only its value in improving the explosion characteristics of the engine but likewise that for maximum efficiency certain definite characteristics and proportions should be employed in the injection system exterior to the nozzle. I have found, for example, that the injection at the nozzle is responsive not so much to the injection to the system by a measuring pump of a predetermined amount of fluid as to the impact resulting on the fluid line from the initial opening of the fuel line to receive the fuel displaced by the pump. This is evidenced by the fact that the injection period does not appreciably vary through a considerable range of injection as regards the amount of fuel which passes through the nozzle. With the same nozzle it is possible to inject to 100 cubic millimeters of fuel to the cylinder in the same interval. This is apparently due to the fact that with a greater fuel injection a greater residual pressure exists in the line connecting the pump and nozzle and, accordingly, the impact blow resulting on initial injection is transmitted with greater force to the valve. The injection apparently immediately follows the closing of the intake openings of the pump by the piston 21 thereof. The'speed of operation of the pump piston apparently has little effect on the injection although it is found that a greater fuel injection can be obtained through use of a relatively slowoperation of the piston. In actual use, sharp, medium and eccentric cams have been utilized, and it is found that the eccentric cam gives much the best resuits, the sharp cam tending to upset equilibrium test, .533 meter. It is found that a. shorter line causes double injection; that is to say, there are two definitely separated injection periods for each operation of the pump, apparently caused by the fact that the shorter line does not provide the dampening effect necessary and the impact blow becomes too great for the'system. On the other hand, a longer line causes double injection, for

the apparent reason that there is too great a volume of liquid in the system causing a lag. It might appear that the presence of too much fluid in the system occasioned by the use of a long line might be overcome by a reduction of the diameter of the discharge line, but this is not the case, for it is found that unless the diameter of the line is kept quite large the impact blow creates such high speed in the column that it cannot be controlled by the valve or flexible oriflce. As a matter of fact, I have found that a relatively large line as compared to the standard line of 0 to 2 millimeters in diameter is essen-- tial and in practice employ a line at least 3, and preferably 4 millimeters in diameter. In any case, the line must be large enough to prevent too great an impact and, in general, the larger the amount of fluid to be injected, the larger the line should be. It has also been found that the amount of fuel injected through a given period can be varied within given limits by variation of the size of the pump employed. For example,

a pump having a piston 10 millimeters in diameter provides an injection range between 10 and cubic millimeters, while with the same line and conditions a pump having a piston of 13 millimeters in diameter provides an injection iange of between 10 and 160 cubic millimeters in the same injection period.

I have determined that the temperature range at which injection should take place in order to permit the proper ignition lag enabling all of the fuel to be injected before combustion starts is provided in an engine of usual construction using commercial fuel by a compression pressur having a minimum of pounds per square inch and a maximum of 400 pounds per square inch, the most emcient range being between 330 pounds and 360 pounds to the square inch. In such an engine this range of temperature is sufficiently high to cause rapid heating of the fuel and is at the same time below the decomposition temperature when the fuel is delivered to the cylinder in the form of a coarse spray.

However, the maximum compression pressure and, as a matter of fact, the most efficient operating range may vary considerably, for these pressures are dependent upon the end compression te'mperature attained in the engine. Obviously, in a large engine where the heat transfer to the air is relatively small as compared with that in a' small engine the higher pressure can be employed. In engines of the same size, the engine characteristics will determine the most efficient operating pressure; for example, an engine of the ordinary construction such as discussed above having cast iron pistons and head will operate at a much higher end temperature than an engine of the same size having special alloy pistons and head, and the cooler the engine the higher the pressure which may be employed therein. Furthermore, the pressure-ignition temperature depends to a large measure on the character of the fuel employed and, obviously, a fuel having a sufficiently high cracking point might be injected at a much higher compression pressure than the ordinary commercial fuel which has been considered in the specific example given. Additionally, the normal operating speed of the engine must be taken into consideration because of the time factor which affects the end compression temperature, a slow engine producing higher temperatures than a fast engine of the same construction. I have found that with a cold engine and with the injection of cold air into the engine maximum compression at the time of injection may be increased to as much as 600 pounds.

The vast improvement in operation of an engine in fuel injection under the above-outlined circumstances as against the normal injection methods may be visualized by comparing the indicator cards forming Figs. 8 to 11. The indicator cards of Figs. 8 and 9 are cards taken from an engine utilizing the ordinary nozzle and developing a perfect highly atomized compact spray in accordance with the principles of present-day Diesel practice.

It will be noted that a succession of detonaabove described operates neither on the Diesel nor the Otto cycle differing from the former in that the fuel injection occurs through an extremely short period and entirely prior to combustion; that the injection employed is a spray of loose particles rather than a highly vaporized injection; thatan ignition'lag is deliberately sought for, to enable the injection to be made prior to combustion, and that volumetric capacity of the engine is greatly increased from the ordinary Diesel, with the result that much higher engine speed can be obtained. It differs from the Otto cycle both in the fact that the volume of air introduced is materially increased, and that the fuel is separately injected.

Since both the method illustrated and the construction described are capable of considerable modification without departing from the spirit of the invention, I do not wish to be understood as limiting myself thereio except as hereinafter claimed.

I claim:

1. The method of fueling variable speed solid fuel injection auto ignition engines of the reciprocating piston type comprising, supplying air to the engines cylinders in such quantities that the maximum temperature due to compression is below the decomposition temperature of the fuel, instantaneously injecting fuel into the cylinder after the attainment of compression ignition conditions within the cylinder at a rate such tion peaks D appear even when the engine is I idling, as in Fig. 8. These peaks are much exaggerated when the engine is operating under load, as indicated at D in Fig. 9. On the contrary, when the type of injection just described is employed in the same engine and under the same conditions, these peaks immediately disappear and when idling or under load the engine shows no tendency to the detonations responsible for these peaks. (See Figs. 10 and 11.)

Through the use of this method of fueling the engine and the nozzle construction hereinbefore described, I am able to produce an engine which is particularly adapted for use in automotive fields. In an engine from which the indicator cards forming Figs. 10 and 11 were taken when using the new fueling system, such engine having 4 cylinders and a bore and stroke of 3 and 4" respectively excluding the exhaust port area, when operating at 1800 revolutions produced 150 H. P., and at 1,000 revolutions produced 80 H. P. It is, however, operable in the higher automotive ranges; that is to say 3,000 to 4,000 revolutions.

It will be noted that an engine operated as that, regardless of the quantity of fuel or of the engine speed, the entire charge is injected during the ignition lag period and at a time such that the injection of the fuel is completed prior to the attainment of maximum compression in the cylinder and prior to initial ignition of the fuel, and injecting the fuel charge through an orifice of such size that it does not constitute a restriction to the passage of fuel and at a pressure preventing atomization, whereby the fuel enters the combustion space in the form of wet globules of a size sufficiently large to prevent vaporization during the period of injection and ignition lag.

2. The method of fueling variable speed solid fuel injection auto ignition engines of the reciprocating piston type comprising, supplying air to the engines cylinders in such quantities that the maximum compression pressure is between 330 and 360 pounds per square inch whereby temperature due to compression is below the decomposition temperature of the fuel, instantaneously injecting fuel into the cylinder after the attainment of compression ignition conditions within the cylinder at a rate such that, regardless of the quantity of fuel or of the engine speed, the entire charge is injected during the ignition lag period and at a time such that the injection of the fuel is completed prior to the attainment of maximum compression in the cylinder and prior to initial ignition of the fuel, and injecting the fuel charge through an orifice of such size that it does not constitute a restriction to the passage of fuel and at a pressure preventing atomization whereby the fuel enters the combustion space in the form of wet globules of a size sufliciently large to prevent vaporization during the period of injection and ignition ing.

3. The method of fueling variable speed solid fuel injection auto ignition engines of the reciprocating piston type comprising, supplying air tn the engines cylinders in such quantities that the maximum temperature due to compression is below the decomposition temperature of the fuel, instantaneously injecting fuel into the cylinder after the attainment of compression ignition conditions within the cylinder at a rate such that, regardless of the quantity of fuel or of the engine speed, the entire charge is injected within 10 of crank travel and during the igni tion lag period, injecting the charge at a time such that the injection of the fuel is completed prior to the attainment of maximum compression in the cylinder and prior to initial ignition of the fuel and injecting the fuel charge through an orifice of such size that it does not constitute a restriction to the passage of fuel and at a pressure of the order of 1200 pounds per square inch preventing atomization whereby the fuel enters the combustion space in the form of wet globules of a size sufllciently large to prevent vaporization during the period of injection and ignition lag.

MAX GEORGE FIEDLER.

Images