FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1914
1914 - 0047.PDF
JANUARY 10, 1914. (TOCHf) AERO ENGINES. By GRANVILLE £. BRADSHAW. (Concludedfrom lent week.) As far as the engine designer is concerned, the shape of the engine makes quite a lot of difference, chiefly in the balance of moving .parts. Unbalanced parts are not often found in aero engines, as the number of cylinders adopted is usually large, and it is easy to arrange the cylinders in such a manner that the moving parts balance one another. There are, however, quite a number of engines in which balance weights are fitted, namely, the stationary cylindered radial, and although simple rotary balance weights will correct the unbalance of the pistons and connecting rods in this particular type of engine, to design an aero engine that requires balance weights at all seems to be deliberately burdening oneself with an altogether unnecessary handicap. In one well-known radial engine, a balance weight of something in the neighbourhood of 60 lbs. is fitted. An engine with rotating cylinders does not require balance weights, and with the exception of its lack of economy with conse quently increased weight for long flights, it is an excellent engine for aviation purposes in every way. The criticisms levelled at this engine with regard to gyroscopic force when turning do not appear to have been discovered in practice. The six-cylinder vertical engine is being favoured by many at the moment, but it is not a design that lends itself to lightness of con struction, on account of the length of crank-case and crank-shaft. A four-cylindered engine also has the same disadvantage, the weight of crank-case, crank-shaft, cam-shaft, &c, being large in proportion to the cylinder capacity. A " V " type engine is superior in this respect, as twice the number of cylinders can be fitted to the same length of crank-case and crank-shaft, with a consequent reduc tion in weight per brake horse-power developed. Three, four, and five cylinder air-cooled semi-radial engines appear to be disappearing altogether, and on account of their lack of balance, their need for heavy balance weights, their uneven torque, and their head resistance, they are not likely to survive for any length of time. Turning to the question of materials, we find that whilst the car •engine designer, in order to obtain better balance and more power, turned to fitting steel pistons in his cast-iron cylinders, the aero engine •designer was not long in discovering that the exact opposite is superior both in theory and in practice. It was found in car practice that in fitting steel pistons in place of cast-iron ones, more clearance between the piston and cylinder walls had to be allowed in order to prevent the piston seizing, owing to steel having a greater coefficient of •expansion than cast-iron. Aero engine practice is to fit the metal which expands most on the outside, and this in practice gives a remarkable freedom from tight pistons and seizure. With piston clearances cut down to the smallest working fit, the author has found that it is almost impossible to detect any tightness of the pistons, or any tendency to seize. More particularly is this evident in the case of air-cooled engines —the hotter the piston and cylinder become the more the cylinder expands away from the piston. One of the greatest difficulties the aero engine designer has to contend with is the tendency for cylinders and other details to warp when heated, owing to their extreme lightness and the thinness of the walls. The cylinders have to be proportioned with the utmost care andldesigned so as to be of exactly equal thickness round the bore. For this reason side valves such as are general in automobile engine practice are now practically discarded in favour of overhead valves, as the latter allow the cylinder to be machined accurately both inside and out. In the case of water-cooled engines, the direction of flow of the cooler water from the radiator through the cylinder is an item of no small importance, and any inaccuracy in this very •quickly shows itself in uneven wear of the cylinder. In the Gnome revolving cylindered engine, it is interesting to note that the cylinder bore and the circular radiating fins round the outside are eccentric, in order to compensate for the unequal amount of cooling air that comes into contact with the different parts of the cylinder. Built up parts in any shape or form are undesirable, and because these have often resulted in failure, we now see elaborate details carved out of the solid forging. Cylinders are invariably machined from the solid billet, with the cooling fins turned or milled on them, whilst in the case of water-cooled engines the valve ports are milled out and spun coppers, or electrically deposited copper jacl»ets are fitted. In other cases acetylene welded steel jackets are used with great success; the whole cylinder can in this case be carefully annealed and treated and the bore reamered out last. It is very important to provide for the unequal expansion and contraction of the cylinder and water jacket, and this is done by using either a corrugated jacket or flexible flanges on the cylinders. Heat treated steels which obtain their tensile strength and other qualities by being so treated, are unsuitable for the manufacture of cylinders, as these cylinders are more liable to distortion, and should the engine run really hot by any chance, they are robbed of their qualities. The steel must show a high tensile strength, with preferably a percentage of elongation that is not too high, the most important item being to have a steel which gives a hard bright surface after running. , The aero engine valve is on the average much thinner and lighter than that used in automobile practice, although in both cases the exhaust valves run red hot the whole time. The author has found that extremely long valve springs, so much advocated by long stroke designers, are directly opposed to valve gear reliability, and the one feature of the long spring, i.e., the small amount of increase in pressure of the spring when the valve is lifted, is at the best a disadvantage. In aero engines valve heads are unusually large, being generally in the neighbourhood of 3 ins. in diameter, and owing to extra weight and complication it is im practicable to fit two smaller valves. The great disadvantage of a large valve is its tendency to dish in the head during a long run on full throttle, unless made of considerable thickness which is pro hibited on account of the weight. The strength of a valve spring is usually made slightly in excess of that force required to accelerate the valve and valve parts in a given time, which time is determined by the speed of the engine and the shape of the cam. The force required to bring the valve to rest immediately it touches its seating on the return is very much in excess of the force required to acceler ate the valve, because it is decelerated in a very much shorter time, unless the cam is arranged so as to give a slow drop. It is practically impossible to calculate the length of time taken by the valve stem to decelerate, because this depends upon so many unknown factors, such as spring in the head of the valve, &c, but there can be no doubt that the greatest stress in any valve comes either in the neck just under the head, or in the centre of the head itself (whichever is the weaker), just at the instant of the closing of the valve. From this it will be seen that the more metal there is in the valve stem, cotter, collar and spring, the greater is the stress in the valve stem under the head, as a greater amount of weight has to be brought to rest in the same period of time. The author's experience is that the chief contributing factor to valve breakage is length of valve stem. This would apparently increase in a greater ratio than directly as the length of the stem, because as the length of stem (and consequently the weight) is increased, so the weight and strength of the spring necessary to operate it are increased. The average length of valve stem in the automobile engine is some where in the neighbourhood of seven or eight inches, and sometimes even more than this, whilst in aero engines the stem is cut down to the shortest possible, and one highly mccessful aero engine that the author has come into contact with, has a 3-in. exhaust valve with an extremely light head, the stem of which is only IJ ins. long. Of course it is quite agreed that a long spring is in itself more reliable than a short one, but it would appear that that design is best which enables both spring and valve to have the highest possible freedom from undue stresses. It appears that theoretically a valve stem ought to have its smallest diameter just below the cotter (in the case of an overhead valve) with the diameter increasing all the way, and running into a big radius under the head of the valve ; such, however, is imprac ticable, the best compromise being to drill the valve stem up from the cotter end in order to reduce the weight of this part. As regards the ratio of cylinder bore to stroke, aero engines do not appear to be following automobile practice. It is not usual to find an engine with very long stroke, as it has been found that a heavier engine results. Increasing the stroke necessitates a larger diameter crank-case, which requires to be much heavier in order to obtain the necessary stiffness. The length of the connecting rod increases at least twice the rate of increase of the stroke, whilst the power increases only as the stroke. Thus to increase the stroke of an engine two inches the connecting rod must be four inches longer, as must be the cylinders, valve gear, &c., whilst the crank-case must be increased in diameter by two inches and considerably more in weight, as the walls cannot remain the same in thickness with safety. A long stroke engine has undoubtedly lighter pistons and slightly shorter crank-case, but the balance is in favour of the shorter stroke. Perhaps a greater drawback to long stroke than any of the above is the great increase in crank-shaft weight in order to obtain the necessary stiffness, the crank-shaft being generally the heaviest item in any but the radial type of engines. The remarkable lightness of the crank-shaft in some aero engines is a matter of no small amount of surprise to many car engine 47
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
