The Evolution of Modern Small Car Engines

Last year the I.M.E. instituted a Symposium on the Design of Small Mass-Produced Motor Car Engines, at which eight papers were presented. The following is a summary of these.

B.M.C. The first paper was presented by W. V. Appleby, Chief Designer, Power Units, B.M.C., and dealt with the A-series power units. The A-series B.M.C. engine originated as a push-rod o.h.v. 803-c.c. unit, to power the Austin A30 of 1952. It adopted a bore and stroke of 58 x 76 mm. in conformity with the successful 65.6 x 89 mm. dimensions of the existing 1,200-c.c. engine. The A30 engine was developed to a logical 950 c.c. and then the advent of the ADO 15 created a new design situation.

Dealing with unusual aspects of the A30 type engine and its derivatives, Mr. Appleby cited inlet and exhaust ports on the same side, to avoid possible leaks of oil and water at the cylinder head tubes that would otherwise have been necessary to enable the push-rods to pass the plugs. This compelled the use of siamesed inlet ports and a central siamesed exhaust port. No disadvantage was found from the former; in fact, the lower induction system volume, B.M.C. claim, gives better accelerator response. The disadvantages of the siamesed exhaust port were overcome by using 21-4N exhaust valve material.

Electrical equipment is all on the manifold side of the engine to avoid petrol drips or exhaust heat affecting the components, so the drive from camshaft to distributor has to be taken across the engine, and to eliminate another gear the oil-pump is driven from the rear of the camshaft. This necessitates a comparatively high location, above sump level, but the port arrangements are such that, once primed, re-priming isn’t required unless the engine has been stripped and re-built.

B.M.C. insist on flexibility over a wide speed range, achieved by large valves, low valve lift, conservative valve timing and a comparatively heavy flywheel. They employ four rings per piston, to ensure consistently good oil consumption from one engine to another; blow-by is also lower than with most other engines. All B.M.C. crankshafts and con.-rods are of 55-ton alloy-steel stampings, whereas 40-ton steel is more usual. This spells shorter die life and costs much money but enables lighter moving parts to be used and engine length kept to a minimum. The hardened-steel camshaft with chilled-iron tappets has been extremely trouble-free.

The unusual, very compact combustion chamber shape is covered by Weslake patents. As only Pool petrol was available, the original c.r. was 7.2 to 1.

The A30 engine had a Zenith d.d. carburetter, the Morris Minor version an S.U. carburetter. The road-test drivers found the constant-vacuum carburetter smoothed out many induction troubles. The oft-criticised clamped gudgeon-pin is used, as in America, as it has many manufacturing and servicing advantages. After 556,000 B.M.C. 803-c.c. engines had been built the bore was increased to 62.9 mm. to give 948 c.c. and, premium petrol being again available, the c.r. was put up to 8.3 to 1. This increased power from 28 to 37 b.h.p. for the Austin, from 30 to 37 b.h.p. for the Minor 1000, torque increasing from 40 to 50 lb. ft. The big-end dia. was increased from 1 1/16 in. to 1 5/8 in. and the material changed to lead-indium, which necessitated a full-flow instead of a by-pass filter.

The first of the 948-c.c. engines suffered from excessive bore distortion, ascribed to the cylinders being siamesed in pairs, but actually caused by the use of a thick c. and a. gasket instead of the former thin steel gasket, which itself had not been entirely trouble-free. A 1/32 in. thick c. and a. gasket was the answer, maximum bore distortion then being 0.0009 in.

Yet another B.M.C. speciality is a surface finish of from 25 to 60 micro-inches, obtained by first boring with a single-point tool, wire-bushing and then rolling. To increase bore life a chromium-plated top piston ring was tried, but oil consumption was excessive, calling for much experimentation. The cure was to increase radial depth to d/24, making top and bottom rings of D.T.D. 485.

For the Austin-Healey Sprite, introduced in May 1958, the 948-c.c. engine was given twin S.U. carburetters and gave 42.5 b.h.p. at 5,000 r.p.m., and 52 lb. ft. torque at 3,300 r.p.m. For the Sprite Mk. II of June 1961 a new head with a c.r. of 9 to 1 and 1 5/32 in. instead of 1 3/32 in. valves was designed, which gave 46.6 b.h.p. at 5,500 r.p.m.

By this time Issigonis had had the inspiration of Minis with a transverse power unit driving the front wheels. Some consideration was given to the use of air-cooled two-stroke and watercooled four-stroke twin cylinder engines but, as in 1923 for the original Seven, all were discarded in favour of the conventional, more refined engine—but Mr. Appleby did not say that such engines were actually built. The 948 c.c. engine was regarded then as too potent for an 11½ cwt. car, so the stroke was reduced to 68.26 mm., to give 848 cc. Cylinder block, head, valve and timing gear were the same as for the 948 c.c. engine, so the same power was developed, but at 500 r.p.m. more, with torque reduced to 44 lb. ft. “When we were well on the way to production we discovered that Dr. Giacosa of Fiat had produced a similar layout of transverse power unit in 1947 but had abandoned it two years later,” says Mr. Appleby’s paper.

A step forward was being able to use a c.r. of 8.3 to 1 for both regular and premium-grade fuels, varying only the distributor characteristics and ignition timing. The transverse engine has a primary gear and the clutch between the rear main bearing and the flywheel, reducing torsional frequency of the crankshaft to 24,300 c/min., compared with 29,400 c/min. for the 948 c.c. engine. Amplitudes of vibration were fortunately too low to justify a damper, being of the order of ±0.4° (sixth order) at 4,000 r.p.m.

John Cooper’s suggestion of a more powerful disc-braked Mini resulted in the 997 c.c. power unit, using the 948 c.c. block but with 0.020 in. smaller bore and 81.3 mm. stroke. The head and valve gear were Sprite Mk. II but with stronger valve springs to give bounce at just over 6,000 r.p.m. and a stronger crankshaft and lead-indium bearings were used. 55 b.h.p. was developed at 6,000 r.p.m., giving a car speed of 88 m.p.h. The longer stroke necessitated a torsional vibration damper to prevent the timing gear from breaking up.

When Issigonis mooted the ADO 16, the problem confronting the B.M.C. Power Unit dept. was whether to design a completely new power unit or keep to A-series overall dimensions. On the drawing board they did both but practical considerations decided them on the latter course. After consideration a bore and stroke of 64.58 x 83.72 mm. was adopted, retaining 948 c.c. cylinder centres. To B.M.C.’s surprise, although this necessitated only 6.5 mm. of metal between the bores, this was entirely successful. Pistons were the next problem. The block and con.-rods were standard A-series height and length but the stroke had been increased, so the distance between gudgeon-pin centre line and piston crown was insufficient to permit a clamped gudgeon-pin. A pressed-in pin, fitted by heating up the small-end, proved successful but fears of the Morris 1100’s reputation being ruined by loose pins scoring the bores led to the big production complication of fully-floating pins retained by circlips.

As the new car was likely to weigh approx. 16 cwt. and had to out-perform the Mini, a c.r. of 8.5 to 1 for single-carburetter ADO 16s and 8.9 to 1 for twin-carburetter versions were used. The Sprite Mx. II/Cooper Mini head was used for the former, a new head with 1 7/32 in. valves for the M.G. 1100. Respective performance figures were 48 b.h.p. at 5,100 r.p.m./60 lb. ft. torque at 2,500 r.p.m. and 55 b.h.p. at 5,500 r.p.m./61 lb. ft. at 2,750 r.p.m. Valve springs gave a bounce speed of just over 6,000 r.p.m. “which the A-series engine will stand quite happily.” Again, the crankshaft required a torsional vibration damper.

The 848 c.c. sufficed at first for the Wolseley Hornet and Riley Elf but these cars weighed some 80 lb. more than the Mini, so performance suffered and the demand for them was disappointing. A bigger engine, giving greater torque but little more power than the Mini engine, was decided upon, using the 1,100 c.c. block and a new 3 in.-stroke crankshaft, giving a capacity of 998 c.c. To restrict maximum power the standard Mini head and valve gear were used. Limited crankshaft production capacity had precluded using the Cooper Mini bore and stroke; B.M.C. had ample facilities for producing 3 in.-stroke crankshafts. This engine, with a single S.U. carburetter, gave 38 b.h.p. at 5,250 r.p.m. (compared to 34 at 5,500 of a Mini) and 52 lb. ft. torque at 2,700 r.p.m. (compared to 44 at 2,900 of a Mini).

The Cooper-Mini S engine is not mass-produced, is the direct result of work carried out on B.M.C.’s Competitions Department in F.J. racing, and the materials are the best for the job they have to do. The crankshaft is of nitrided steel which does not wear, bearing clearance is therefore maintained at a minimum during the life of the engine, and initial balance is maintained within close limits-very important, as 7,000 r.p.m. is normal, 7,800 r.p.m. when racing. Nimonic 80 is used for both inlet and exhaust valves, the stems tipped with stellite. A bore and stroke of 2.781-2.7815 in. x 2.6875-2.6925 in. gives a capacity of 1,070 c.c. For the first time in A-series history the cylinder centres have changed but quantity production was not, of course, envisaged. This engine develops 67 b.h.p. at 5,700 r.p.m., and a Maximum torque of 62 lb. ft. at 4,500 r.p.m.

Well over a million 948 c.c. engines had been built when Mr. Appleby delivered his I.A.E. paper at the end of 1963, and 2¼-million A-series B.M.C. engines had been produced overall.

W. B.

[More recent Cooper-Mini engine developments were described in Motor Sport last month.—Ed.)

(to be continued)