The following is the second in a series of articles from the 1923 edition of Modem Mechanical Engineering, on the subject of gas engines. The original articles were sent to us by Jan van der Gugten, 2633 Ware Street, Abbotsford, B.C., Canada V2S 3E2, who thought our readers would find them of interest.
CHAPTER III: Methods of Ignition
In the evolution of the gas-engine, ignition was for long a source of trouble, and the solution of the problem of igniting the charge with certainty and regularity proved long and difficult.
In the earliest non-compression of low-compression engines, as, e.g., those of Street (1794) and Wright (1833), a crude 'touch-hole' flame method was used. The piston near the commencement of its working stroke uncovered a small hole through which an external flame was sucked, thus exploding the fresh charge; the hole was sometimes so small as to cause no perceptible loss of pressure on explosion, or was fitted with a valve closed by the rise of pressure in the cylinder or ignition.
In engines compressing their charge before explosion this simple method was obviously inapplicable; but in 1838 W. Barnett introduced the celebrated 'Barnett Igniting Cock', which successfully solved the problem of ignition by flame in compressing engines, and was extensively used from that date down to about 1892. An illustration of this igniting cock is given in Fig. 4. It comprises a hollow open-ended single-ported plug (A) ground into a casing (B) containing two ports (C) and (D), of which (C) communicates with the atmosphere and (D) with the engine cylinder. A gas-jet (E) burns continuously outside the cock, while (F) is a gas-pipe terminating in a small burner within the hollow plug (A). The plug (A) is rotated by the engine in working, and in the position illustrated the external flame (E) ignites the jet within the plug. The rotation next causes the plug port to register with the port (D), whereupon the charge within the cylinder is immediately exploded; the explosion extinguishes the jet in the plug, but when next it opens to the atmosphere it is relighted by the flame (E) in readiness for the next explosion; the action is thus positive and regular. The later improved flame-igniting devices of Hugon, Otto and Langen, and Dr. Otto were based upon the principle of the Barnett cock.
In 1799 Lebon proposed to use a machine worked by the engine to produce electric sparks for exploding the charge; in 1850 Stephard suggested the use of a magneto-electric machine for this purpose; in 1860 Lenoir employed a Bunsen battery, Ruhmkorff coil, distributor, and sparking-plug, substantially as was largely used in automobiles between 1898 and 1908; the art of constructing reliable coils and plugs was, however, but little developed in Lenoir's day, and great trouble was always experienced with the ignition of his engines. In Hugon's Lenoir-type engine an important improvement was the replacement of the electrical ignition by a flame device of modified Barnett type. Ignition by magneto-electric machines is now all but universal in gas engines, but the unfavorable impression created by the early Lenoir arrangement probably retarded its development in Great Britain.
A method that has been very largely used since about 1883, and is even now (1922) not quite obsolete, particularly in small gas-engines of up to about 10 b.h.p. is that known as 'hot tube' ignition.
As early as 1855 Dr. Drake, of Philadelphia, described an igniter consisting of a hollow cast-iron thimble projecting into a recess in the cylinder side, and maintained constantly at a red heat by an external blow-pipe flame. Early in the working stroke the piston uncovered the recess, and the red-hot thimble then exploded the charge. Hot-tube ignition of the modern type was however, apparently first employed in the 'Stock-port gas-engines' in 1883. The principle of this mode of ignition is illustrated in Fig. 5. The lower diagram shows the very simple method of ignition by 'open tube' wherein the hot tube (A) is always in free communication with the combustion chamber (C); firing occurs when the compressed fresh mixture reaches the red-hot part of the tube, the Bunsen-flame lamp (B) being adjusted in position by trial until the heated zone of the tube causes firing to occur at the correct instant. Open-tube ignition proved quite satisfactory in engines of up to about 20 b.h.p., especially when using town's gas; for engines exceeding this power the possibly serious consequences of pre-ignition rendered it necessary to regulate positively and exactly the instant of ignition. This was effected by introducing between the hot tube and combustion chamber a small cam-operated valve termed a 'timing' valve, as illustrated in the upper diagram of Fig 5. In this diagram (D) is the timing valve normally held on its lower seat by a spring, but raised by a cam during the compression stroke so as to seat at (E) and thus cut off communication between the hot tube and the combustion chamber. Near the end of the compression stroke the cam released the valve, which at once returned to its lower seat by the action of the spring, thus permitting the compressed fresh mixture to enter the hot tube, where it was immediately ignited.
Timing valves, exposed to a constant rush of flame, gave a good deal of trouble and required frequent renewal; with this system also the ignition was necessarily initiated at some little distance from the combustion chamber, which was often unsatisfactory, especially with large cylinders; the now almost universal electrical methods permit ignition to be effected simultaneously at two or more points actually within the combustion chamber, thus causing a quicker firing of the charge.
The hot tubes are of wrought iron, porcelain, or nickel alloy. Wrought-iron tubes are cheap, but oxidize rapidly, and have but a short working life. Porcelain tubes are readily heated and resist the chemical action of the flame, but are brittle and at once crack if wetted. Nickel alloy tubes resist the flame action satisfactorily, and have been very largely used. The best material of all is platinum, but its costliness prohibited its use except in the small petrol-engines of the earliest motorboats, motor cars, and motor-bicycles.
Electrical ignition. It has already been stated that electrical ignition was attempted quite early in the evolution of the internal-combustion engine, but its general adoption was impracticable until recent years owing to the unreliability of the available apparatus. Within the limits of the present article a brief reference only can be attempted to the development of electrical methods of ignition.
After Lenoir the battery-coil method went entirely out of use for many years, but improvements in manufacture caused its revival about 1898, and from that date until about 1908 it was very extensively employed, particularly in the small engines of motor vehicles. An important development of this method of ignition is that known as the 'Lodge system', which has been widely used, particularly in large gas-engines. The Lodge system involves a 4-volt or 8-volt accumulator, induction coil, and sparking-plug; the induction coil is the special feature of the system owing to the interposition of a pair of 'Leyden jars' between the high-tension terminals of the coil and the sparking-plug, whereby an unusually violent oscillatory discharge is caused to occur at the sparking-plug gap, thus producing a very effective ignition especially valuable when poor fuel, as, e.g., blast-furnace gas, is used.
Magneto Ignition. Accumulators for ignition purposes proved a source of constant trouble, and an important advance was accordingly effected by deriving the igniting spark from a small electromagnetic machine worked by the engine itself. Such machines are conveniently classed as of 'low-tension' and 'high-tension' type. For stationary gas-engines of all but the smallest sizes low-tension magneto ignition is now (1922) almost universally employed, although recent advances in the design and construction of that very perfect little machine, the high-tension magneto, will probably result in an extension of its employment to large gas-engines in the near future.
A widely-used type of low-tension magneto by Bosch is diagrammatically illustrated in Fig. 6; it comprises permanent horseshoe magnets (A) between whose pole-pieces a fixed shuttle-wound armature (B) is fitted. Between (B) and the pole-pieces is a split soft-iron sleeve (CC) partly surrounding the armature; this sleeve is turned through an angle of about 45 degrees by an engine-driven cam (D), and on release is suddenly 'flicked' back to its initial position by the strong springs (E), (E). The terminal (K) is connected by an insulated lead with the insulated electrode (H) of the make-and-break ignition plug in the combustion chamber of the cylinder, the other end of the magneto circuit being 'earthed' to the engine frame; during the flick back of the sleeve (C) the light connecting-rod shown causes the 'hammer' (F) to break contact suddenly with the central electrode or 'anvil' (H) within the combustion chamber, as indicated, whereupon a strong spark leaps across the gap thus made, and ignites the charge. In the earliest designs the armature itself was rocked, but its inertia was found soon to cause trouble from disintegration and wear; with improvements suggested by experience, however, modern practice has largely reverted to the use of the rocking armature. The light soft-iron rocking sleeve was adopted in order to minimize inertia effects; it gave a good firing spark and enabled the current to be taken from a stationary armature, thus avoiding the necessity of collecting it by brush contact from a moving surface.
An important advantage of the low-tension magneto method of ignition is that a good 'fat' firing spark is obtained no matter how slowly the engine may be running, a point of great value when poor gas is used. Starting is also often facilitated by 'flicking' the magneto by hand after first inducing a partly compressed explosive charge in the cylinder by injection of gas or petrol, or by barring round. The make-and-break device is contained in a small casting usually attached to the combustion chamber by a pair of studs or nuts as indicated, and is thus readily removed for adjustment or repair; in starting from 'all cold' it is frequently found necessary to remove and warm the make-and-break casting in order to dry off any condensed moisture. The sparking current produced on break of contact is of high voltage, and the effective and durable insulation of the anvil (H) has occasioned considerable difficulty. Asbestos, steatite (soap-stone), mica, porcelain, enamel, etc., are all used as insulators; the material employed must be not only a good insulator, but must also be capable of withstanding a high temperature and of resisting the disintegrating tendency of the constant impacts of the hammer (F) on the anvil (H). To meet the somewhat rapid burning away at the gap, replaceable contacts of nickel, nickel-steel, and other alloys are provided.
In large cylinders two make-and-break ignition plugs are often fitted, each independently served by its own magneto; this is useful in case of the failure of one system, and also reduces the time of explosion of the charge, which is an important point when using weak fuels, as, e.g., blast-furnace gas.
In 1922 low-tension magneto ignition was employed by the following prominent British gas-engine builders: Anderson-Grice, Brotherhood, Campbell (above 10 b.h.p.), Crossly (in all larger sizes), Premier, Ruston 6k Hornsby, Tangye, and Vickers; normal engine speeds in these cases range, roundly, from 100 to 300 RPM.
The National Gas Engine Company in 1922 employed high-tension magneto ignition (p.48) in all their engines except the smallest, usually with two sparking-plugs per cylinder; for engines of up to about 10 b.h.p., however, they still used tube ignition, largely to diminish first cost. Messrs. Browett-Lindley also used high-tension magneto ignition, as did the Campbell Gas Engine Company in their smaller quick-speed engines.
The high-tension magneto, universally employed for the ignition of the innumerable small high-speed engines of motor vehicles of all kinds, is so well known that a very brief reference to it only is necessary. In the high-tension magneto the shuttle armature rotates continuously between the pole-pieces of the permanent field-magnets and carries two windings, viz. a stout-wire primary and fine-wire-secondary; the primary current is automatically broken twice per armature revolution by a contact breaker forming part of the magneto, and the induced high-voltage secondary current is 'collected' by a carbon brush and transmitted to a highly insulated 'distributor', by which it is served to the cylinder sparking plugs in proper order. In its modern form the high-tension magneto is extremely reliable, and will often run for years without any more attention than an occasional oiling; it is usually made water-and dust-tight. Modern high-tension sparking plugs also are exceedingly reliable and involve no moving parts as in the low-tension system, while the regulation of the instant of ignition or 'timing of the spark' is very simply and easily arranged. In 1922 the British Lighting and Ignition Company introduced a high-tension magneto having a stationary armature and rotating magnet, resulting in a considerable saving of weight, and claimed to possess several mechanical advantages over the usual design.*
*The Automobile Engineer, November 1922.