"KNOCK" Its Causes
Page 49
If you've noticed an error in this article please click here to report it so we can fix it.
and Cure THREE separate, approaches were made to the contentious phenomenon of "knock" in internal-combustion engines, in papers read to members of the Automobile Division of the Institution of Mechanical Engineers, in London, last Friday.
Although great progress has been made in determining the underlying factors governing the phenomenon, which is now better understood, it is a problem of such complexity that investigations are being actively pursued with the object of increasing knowledge of the mechanism of knocking combustion.
Mr. D. Downs. B.Sc. (Eng.), A.M.I.Mech.E., and Mr. R. W. Wheeler, B.Sc., both of Ricardo and Co., Engineers (1927), Ltd., in their paper, "Recent Developments in • Knock Research," dealt at some 'length with experiments with the Ricardo E6 variable-compression engine.
Theory of Knock
• The earliest theory,said the authors, supposed that knock was caused by an actual mechanical blow, .between .two loose solid Partse Of the engine. Subsequent theories were that (1) it was akin to Ithe phenomenon of detonation observed during the propagation of flaine in closed tubes or (2) autoignition, Under the influence of high temperature and pressure, of a body of the mixture after the initiation of flame by the spark.
The broad facts relating to knock were simple. In a normal engine cycle, the flame was initiated at the spark and travelled in a fairly unifetim manner across the combustion chamber, compressing the unburned gas before it. This "end-gas," as it was termed, received heat due to compression by the expanding gases and by radiation from the advancing flame front.
If the temperature and pressure were below certain critical values, the flame front would move across in a regular manner to the farther wall of the chamber, the mixture being burned progressively. If, an the other hand, .the temperature and pressurecianditions were sufficiently severe, the rate of chemical reaction would exceed a certain critical value and, just before the flame reached the farther side of the combustion chamber, the unburned or partially. burned mixture in the "endgas " would be consumed at a high rate.
A Shock Wave This high rate of burning, and consequent momentary disturbance of the pressure equilibrium in the combustion chamber, were responsible for setting up a shock wave, which, impinging on the cylinder wall, gave the high-pitched knocking sound characteristic of detonation in an engine.
• The occurrence of knock depended on the temperature-pressure-time relationship of the mixture in the "endgas." Thus, any condition which raised the compression or pressure, such as by increasing the compression ratio, or the air-intake temperature, or any condition which increased the induction time interval, would encourage knock.
Much of the authors' work had been concerned with the chemical analysis of "end-gas.' Preliminary tests of samples taken from the engine cylinder under both knocking and non-knocking conditions showed that in all eases the qualitative results were similar. No substance peculiar to either condition was found in any of the samples taken over a range of timing extending from 70 degrees early to 20 degrees late.
Dealing with thea subject of autoignition, the authors said that this, when produced in a motored. engine having a high compression ratio, was in most cases less violent and the noise was not of such a high pitch as that produced by a knock in a spark-ignited engine.
Mr. H. J. Eatwell,
A.M.I.kTech.E.. and Mr. J. G.-Withers, B.Sc., A.M.I.I'vlech.E., research engineer and deputy research engineer, respectively, of the Anglo Iranian Oil Co., Ltd., in their paper dealing with the relationship of octane numbers to antiknock performance, said it was believed that knock was caused by auto-ignition of part of the charge.
Increases in compression ratio, spark advance, throttle opening, mixture temperature and cooling-water temperature, carbon deposits and supercharging. were mentioned as possible causes of engine knock,. but it was pointed out that changes in these, variables did not affect all fuels equally..
Explaining the reason for the introduction of the octane scale, the authors said that this was to provide a reproducible and permanent standard of comparison. The scale, they said, was based on the use of two pure hydrocarbon compounds—normal heptane and iso-octane—which were among the large number of hydrocarbons present in ordinary motor fuels.
Iso-octane had a high anti-knock value and was arbitrarily given an octane number of 100, whereas normal heptane had a low value and was defined as having a zero octane number.
The first widely accepted means for evaluating motor fuels in service was the Uniontown or Motor method. It was based on the aural estimation of knock intensity at a series of speeds during full-throttle acceleration. For laboratory tests, the C.F.R. variablecompression-ratio engine was . used. Both the Motor, and what was known as the Research method were still used to control fuel quality.
The octane requirements of individual vehicle engines varied widely and were governed principally by the basic ignition advance setting, the automatic advance mechanism, carburetter characteristics, the thickness of the castings, deposits in the cooling system, condition of the exhaust valves and the amount of carbon deposit in the cornbustion spaces.
Fuel Quality
Mr. J. D. Davis, M.A.. A.IVI.I.Mech.E„ engineer in charge of Automotive Gasoline Research. at the Thornton Research Centre of the Shell Organization, said that the spark' ignition engine, more than any other type of heat engine, depended for its power output and thermal efficiency upon the quality of the fuel which it was designed to use. His paper dealt with the factors affecting the use of the anti-knock qualities of. fuels in vehicle engines, and was an interim report on investigations conducted on a number of post-war engines.
Although no exact correlation was known to exist between octane requirements measured on the road and those obtained in the same engine on the , bench, bench-test results, said the author, were considered to give a satisfactory indication of road octane requirements.
The effects of combuStion-chamber deposits on the anti-knock requirements of an engine were customarily expressed in terms of the increase in octane number necessitated by the formation of deposit. This, however, had been found to be a misleading parameter, and the change in knock—limited spark advance with a commercial fuel was now considered to be. more apposite.
The results of tests showed that the increase in octane requirement of 6,000 miles of operation was 17.5 octane numbers with the engine having a basic ignition timing of top dead centre. It was natural to conclude that after 6,000 miles, either the ignition would have to be greatly retarded to cure knock when using a fuel which was just knock-free with a clean engine, or a fuel of much better quality would be required if the ignition were to be left untouched.
• Effect of Deposits
Expressed in terms of knock-limited snark advance of a test fuel, the results Showed, however, that although deposits formed during the first 1.000i miles reduced the knock-limited advance. there was a recovery until, at 4,500 miles, the original advance of 17 degrees was reached..
Thus, although at 6,000 miles the engine showed an octane requirement 'increase of 17.5 octane numbers, the knock-limited spark advance of a test fuel was slightly in advance of the timing when the engine was clean.
Whilst it would be untrue to suggest, said Mr. Davis, that no problem existed in connection with carbon deposits, it was contended that the effect of such deposits had. been exaggerated by the use of octane requirement increase.