AT THE HEART OF THE ROAD TRANSPORT INDUSTRY.

Call our Sales Team on 0208 912 2120

The Character of the Spark in Relation to Power.*

7th March 1907, Page 32
7th March 1907
Page 32
Page 33
Page 34
Page 35
Page 32, 7th March 1907 — The Character of the Spark in Relation to Power.*
Close
Noticed an error?
If you've noticed an error in this article please click here to report it so we can fix it.

Which of the following most accurately describes the problem?

By W. Watson, D.Sc., F.R.S.

That a strong or " fat " spark is better than a weak one, when used to ignite the charge in a petrol motor, appears to be a commonly accepted opinion, while there can be no doubt that in the ease of most engines the power developed is considerably greater when the voltage of the ignition battery is well up to its normal value, than when the battery has run down, although the engine continues to fire regularly. Not feeling satisfied as to the cause of the above, and not being able to find a record of any direct experiments, the author has made the measurements recorded below.

The engine employed is a double-cylinder one having a bore of 3.5in., and a stroke of 41n. The inlet valve is mechanically operated, and both valves open into a pocket on one side of the cylinder head, as shown in Fig. I. The sparking plug (S) screws into the cap used to close the hole

over the inlet valve, the spark points being well inside a recess in this cap. All the experiments were made on one cylinder only, the other cylinder working with the trembler coil ordinarily employed with the ordinary four-volt battery. The speed at which the engine was run varied between 950 and loon revolutions per minute. The primary of the coil used to fire the charge was connected to a commutator on the two-to-one shaft, a battery of to volts, an adjustable resistance, and an ammeter. The commutator used was either a wipe contact, consisting of a steel wheel running over a fibre cylinder in which is im bedded a brass strip or a make-and-break supplied by the The reason why a fat spark improves the working of a petrol engine may be either (1) the development of a greater pressure in the cylinder owing to the quicker ignition of the charge, or (2) the inure regular firing produced by the timing of the spark being more uniform. Experiments using a trembler coil at once showed that when the current in the primary of the coil is reduced, the time of firing is delayed, but that on advancing the spark more than usual, the mean pressure during the stroke can be brought back to the value obtained when the usual current is employed. As this indicated that the delay might be

caused by the coil, the ordinary wipe contact was replaced by a make-and-break contact, while the trembler coil was replaced by a non-trembler one. As a result the delay, which, with the trembler coil took place when the current in the primary was reduced, no longer took place, this point being illustrated in Figs. 2 and 3. In each figure* the indicator diagram marked (a) is taken with a strong current in the primary of the coil, and that marked (b) is taken with such a weak current that any further reduction In strength is followed by entire cessation of firing. It will be observed that the timing and the pressure developed is the same in either case, except that in Fig. 3 (b) an explosion has been missed, owing to the spark failing to pass, and the next explosion is much more violent than usual, due to the scavenging which took place during the idle stroke. In Fig. 2 the sparking points were .mm. apart, and in Fig. 3 .5mm.t

In Fig. 4 is shown the corresponding pair of diagrams when the mixture is very weak, a large excess of air being used. Here again, with the exception of some abnormal strokes following mis-fires, there is no elTect produced by reducing the intensity of the spark.

The above experiments indicate that it is not the weakness or fatness of the spark—at any rate, in my engine—which causes the loss of power which occurs when a trembler coil and a weak current is used.

The kind of effect which is obtained with a trembler coil when the current is reduced, the carburetter being adjusted to give the best mixture, which for shortness may be described as a full mixture, is shown in Fig. 5. In Fig. 6 the irregularity which may occur when a weak mixture is employed is shown, the diagram obtained with the normal current being omitted for clearness. It will be observed that the maximum pressure occurs at all sorts of points of the stroke between the normal and about 4/5 stroke. The delay which occurs when a weak current is used with a trembler coil may be due to sluggishness of the trembler or to the fact that the trembler fails to act, the spark which ignites the charge being produced when the current in the primary of the coil is broken at the wipe contact. The kind of delay which may be produced by lag of the trembler is shown in Fig. 7 for coil A and in Fig. 8 for coil B, and it will be seen that although the delay is appreciable the amount is quite regular, so that all the explosions occur at the same point in the stroke, and so the effect of the delay could, in the case of a single coil, be eliminated by slightly advancing the spark.

As the current actuating a trembler coil is gradually reduced there is at first a small lag, such as that shown in Figs. 7 and 8, due to lag in the trembler, and then there is a sudden great increase in the lag. This increase is due to the trembler ceasing to act, so that the spark, in place of occurring shortly after the primary circuit Is completed at the commutator, does not take place till the current is interrupted at the contact. When the current is below the critical value, holding the trembler against the upper stop produces no effect on the timing of the spark, showing that the trembler has really ceased to act. The change in current which at the critical value is required to pass from the state when the trembler acts to that in which it fails is very small, in most cases less than a tenth of an ampere, the absolute value of the current depending on the adjustment of the trembler. Figs. g and to show the diagrams obtained with coils B and C when the current is first just above, and in the second case just below, the critical value.

That the size of the spark has no effect on the power de

veloped was further shown by taking a number of indicator cards on which coil B was used, and the current in the primary was adjusted so that (a) the mean current was 0.6 ampere and the equivalent spark length in air was 5.1rnm.,* the trembler of the coil acting, and (b) when the current was reduced to 0.3 ampere the trembler no longer acting, but the delay caused by the fact that the spark is now produced when the commutator breaks the current being compensated by advancing the spark lever. The equivalent spark length in air was in this case o.31nrri. The mean i.h.p. for four cards for each arrangement, taken alternately, was the same within the limits to which the diagrams could be measured; the fact that the speed of the motor was unaltered by the change in spark also indicating that the power developed was the same whether a large or small spark is employed.

From the results of the above measurements it would appear that the trembler, although it undoubtedly is an advantage when starting, is apt to introduce very considerable variations in the point of the stroke at which the charge is tired. Hence, on this account a plain coil with a make and break is to be preferred, particularly in the Case of multicylinder engines having separate coils for each cylinder. The ill-effects of the trembler will also be felt when a single coil with a high-tension distributor is used unless care is taken to keep the electromotive force of the battery well above the value required to give the critical current below which the trembler ceases to act. There is one disadvantage in the make and break, besides that due to the loss of power of starting on the switch, namely, that with some forms, unless the adjustment is just right, the blade may act as a trembler, giving a spark before the cam on the engine shaft finally breaks the circuit, so that pre-ignition is produced. This effect is shown in Fig. ii. Since the power developed is not improved per se by the use of a fat spark, there are many advantages in using a coil which only gives a comparatively small spark, so long as the working of the coil is regular, for such a coil can be designed so that it only consumes a small current. The advantages of a small current are that, in addition to the economy in current, a point of importance where facilities for charging accumulators are not at hand, the wear on the points where the current is interrupted, due to sparking, can be reduced to a vanishing point. Thus, with a trembler coil the trembler, or with a plain coil the points of the make and break, need seldom be touched. The author feels confident that by. suitably designing the coil and the commutator absolutely regular firing can be produced when the mean current does not exceed a tenth of an ampere.

To show that a fat spark, and hence a large current consumption, does not necessarily imply a good coil, a comparison was made between coils B and C, coil B being that which gave the fattest spark, and C that which gave the thinnest of the three coils tested_ The two coils were so arranged that by means of two switches either coil could be used at will, and the current was adjusted in each case to the smallest value at which the coil would give regular firing. The results of the comparison are shown in Figs. i2 and 13. Fig. 12 shows the diagrams obtained with a full mixture, and Fig. i3 those obtained with a weak mixture. In either figure diagram a is that obtained with coil B, the mean current being .6 amp6re, while diagram b is that obtained with coil C, the mean current being .15 amptre. It will be noticed that coil C fires more quickly than coil B, probably owing partly to having a lighter trembler, and partly to the fact that its time constant is smaller.

The conclusions to be drawn from the above experiments appear to be as follow :— (1) As far as a petrol engine of the type used is concerned, the character of the spark which ignites the charge has no appreciable influence on the power developed.

(2) With a trembler coil the time at which the spark occurs is liable to vary greatly, and on this account the power developed may be considerably reduced.

(3) The variation in the time of firing obtained with trembler coils is different for different coils, and hence a multi-cylinder engine in which a separate coil is used for each cylinder is unlikely to develop its maximum power, particularly at high speeds; the reason being that although the tremblers of the coils may possibly be so adjusted for some particular voltage that each cylinder fires at the same point of the stroke, yet this adjustment will no longer be true if the voltage of the battery alters, particularly if it falls much below the value for which the tremblers were adjusted.

(4) When a single coil is used in combination with a hightension distributor, it is of very great importance that the current in the primary should never be allowed to fall to a value near what has been called above the critical value for the particular coil. In this connection it may be mentioned that, in the author's experience, when the trembler is so adjusted for any given voltage of the battery, i.e., for a given current, that the note produced is very clear and " pure," then a very slight decrease in current, due to a small fall in the voltage of the battery, will cause the timing to be defective, owing to the region of the critical current being approached. Hence, with the normal current passing—i.e., with the battery fully charged—it is advisable to adjust the trembler so as to give a somewhat harsh and shrill sound, for then the current may be considerably reduced before the critical value is reached.

(5) When selecting a coil, regularity in the working of the trembler for considerable variation in the current passing in the primary is of more importance than length or fatness of spark. Further, a coil taking a small current is to be preferred to one takinga large current, since trouble with the

adjustment of the trembler blade will be decreased, owing to the reduced sparking at the platinum points with a small current.

(6di Except for the fact that the engine cannot be started on e switch, the plain coil with a rapid break on the two-toone shaft seems preferable to a trembler coil, since over a very large range of current—in fact, whenever the current is large enough to cause the passage of a spark in the cylinder —the timing is exactly the same. The advantage of the trembler might be retained by using a switch, so that after the engine is started the trembler can be cut out, allowing the coil to act as a plain coil, a second condenser being provided.

The two diagrams shown in Figs. 14 and is are somewhat outside the scope of the paper, but may be of interest to drivers. They illustrate the advantage, so far as economy is

concerned, of advancing the spark more than usual when employing a very weak mixture—that is, when driving with the extra air valve as far open as possible. .Fig. 14 is that obtained when the spark is as much advanced as is advisable when using a full mixture, and the i.h.p. at 1,opo revolutions is 2.36. In Fig. 15 the spark has been considerably further advanced, so as to allow for the slow burning of a weak mixture, and as a result the i.h.p. at Loon is 2.76, an increase of nearly 17 per cent, in power, the consumption of petrol remaining the same.

The author desires to express his thanks to Mr. Orde for the loan of two trembler coils, 13 and C.

Tags

People: Orde, W. Watson

comments powered by Disqus