The 3-litre

S0 many readers are interested in the Bentley marque, and as every owner of an old-school Bentley we meet seems to enthusiastically read MoToR SPORT we feel that some notes on improving the performance of the old-model Bentleys may prove of interest. We are indebted to Mr. L. G. McKenzie, the well known Rolls-Royce and Bentley specialist, for the information that follows. This article follows on those about the tuning of the M.G., Talbot and FrazerNash cars, which appeared in. MOTOR SPoRT in January, February and March, 1936, back issues of which are available. The first Bentley model was the famous 3,-litre 80 x 149 min., four-cylinder car, designed just after the War by W. 0. Bentley and the late F. T. Burgess, and put into production in 1921. There were two editions, the Standard and the Speed Model, usually known to-day as the ” Blue Label” and ” Red Label ” models respectively, by reason Of the colour of the centre portion of the winged radiator badges. The former, in all forms up to 1924, with open bodywork, would do 46 m.p.h. on third gear and 70 to 73 m.p.h. in top and the Speed Model 67. on third and 80 to 82 m.p.h. in top without exceeding the maker’s rev, limits of 3,000 r.p.m. and 3,500 r.p.m. respec tively. Naturally, tuning a ” Blue Label ” resolves into first converting it into ‘ a ” Red Label,” so that it is permissible to deal with the variations in specifications between the two types. The Standard car had a wheelbase of 10 ft. 10 ins., the Speed Model one of Oft. 01 ins., the respective weights being 24f cwt. and 221cwt. in chassis form. The weight limits under guarantee were, respectively, 35 and 281 cwt. The. Standard car has a single Smith-Bentley five-jet carburetter, the Speed Model twin SIT. instruments. The Standard chassis had the ” B” type gearbox with ratios of 16.2, 8.$, 6.5 and 4.23 to 1 (reverse 16.2 to 1). The wide gap between first and second resulted in a slow change, but the 16.2 to 1 bottom gear is useful for mountainous touring and is Sometimes loaned for this reason by Mr. McKenzie to Speed Model clients, The Speed Model normally had the “A ” type box giving ratios of 10.0, 6,18, 5.05 and 3.78 to 1 (reverse 10.0 to 1) and a rapid change from first to second if a clutch stop is fitted, a straight through change from second to third, and a slower change from third to top. Naturally, higher speeds on the indireets are possible with the ” A ” type box. Other differences include D.N. shock-absorbers on the long chassis and Hartfords on the Speed Model, a single, three-baffle silencer on the Speed Model and a double expansion-box with fantail on the long chassis, and a lower tail pipe on the Speed Model. The long chassis cars had a compression-ratio of 4.3 to 1 and developed 70 b.h.p. at 3,800 r.p.m., while the Speed Models up to 1925 had a compression-ratio of 5.8 to 1 and gave 80 b.h.p. at 3,200 r.p.m. The 1926-28 Speed Models were designed for a 4.8 to 1 compression-ratio but an error was made in the drawings and they actually have 5.6 to 1. The twin carburetters have been mentioned ; they were Smith single-jet on the 1924 Speed Models and S.U.s thereafter. Tuning a ” Blue Label” long chassis model resolves itself into first of all converting to” Red Label “specification. The wheelbase can be shortened by cutting the side-members in front of the rear spring anchorage and before the compensator cross-shaft and shortening the propeller shaft, but a better method is to obtain a new, short frame which will be unstressed and in good order. Twin carburetters with the manifolding to belt them in place without drastic alteration can be obtained from Bentley Motors, Ltd., for about 1l3 10s. Actually, two different

sizes of carburetter are available. The ” B ” type gearbox can be fitted but if this is done the higher axle-ratio should

be used. The Speed Model normally had the 3.78 axle, but a few cars had a top gear of 3.53 to 1, useful for speed work. Turning now to the Speed Model, additional modifications can be carried out for speed work, but it is a unit that does not show a great increase in horse power. If the valves are stamped ” S ” they must be scrapped and replaced with K.V. 9-05 or Hadfield’s Heckler valves. The compression-ratiO can be put up to 6.5 to 1 without resort to special fuels, although the 3-litre is more prone to pinking than the 4Hitre. This ratio can be obtained by machining 5 ni.m. from the bast of the cylital which is the limit permissible with the standard pistons (low-compression type). If this is done the top piston rings must be omitted and it is essential to deepen the con-rod relief slots by 11 ni,m, as even with the standard compression-ratio the rods are apt to foul if big-end clearance is too liberal. A better method of raising the compression is to use special pistons, particularly as modern thought suggests that the older designs use too narrow ring lands. liecrankshaft balance is satisfactory but the eon,-rods should be balanced and the piston weights equalised, which can sometimes be done

by changing round the rings. These modifications, carefully carried out, will increase the output to about 88 b.h.p., which was the power developed by Bentley’s own Le Mans cars. Mr. McKenzie has had 100 b.h.p. out of the 3-litre, but only in highly experimental form. The road performance can be further enhanced by using the ” ” type gearbox from a late 1927 Or 1928 Standard Six Bentley or from a 4k-litre. The ” A ” and ” D ” type boxes have ratios of 2.64, 1.63, 1.23 and 1 to 1, whereas the” C “type box has ratios of 3.364, 1.823, 1.375 and 1 to 1. Before leaving the 3-litre there are some servicing matters of interest to those developing this power-unit. Particularly is it desirable to pressure-test the cooling system whenever the block is removed, for the oil-drain tubes passing oil from the camcase to the sump are apt to corrode and leak water into the crankcase, and a tiny trickle will pass unnoticed, yet can dilute the lubricant sufficiently to thin the engine bearings in three months. A nother important point is that the long chassis cars had camshafts needing .004″ and .006″ clearance on inlet and exhaust respeetively, with the engine hot. Speed Models for 1.926 and subsequent years require a clearance of .019″ all round. Frequently the earlier cars have been fitted with the later-type camshafts and new owners, on the advice of an early instruction book. set to the small clearances, whereupon the exhaust yalvs-s burn and shed their heads. The trouble cen be avoided by remembering that BM 2391 camshafts require the small clearances and BM 6800 camshafts the larger setting. The very early engines (up to No. 222) require a setting of .015” and this usually applies if they have hard ened valve caps. The st andard carburetter setting recommended by Bentley’s for the Smith-Bentley carburetter was :— 45 Well jet, 45 No. 1, 75 No. 2, 50 No. 3 and 35 No. 4. Standard ignition advance

was 450 before t.d.c. fully advanced on both magnetos. The thermostat setting should maintain 75 to 80°C. The ‘correct oil-pressure is 12 lb. at 30 m.p.h. in top gear with the oil thoroughly hot. Firing order 1, 3, 4, 2. The 3-litte Bentley engine, although designed with a big safety margin, nevertheless developed almost its maximum safe power output in production form. To raise this by 8 b.h.p. is a very satisfactory accomplishment and tuning chiefly resolves itself into improving the torque to improve acceleration or to enable higher gear-ratios to be used with a consequent increase in speed. Performance increase is largely dependent on reduction of weight and head resistance. Even so, we believe that a spccial-bodied 3litre guaranteed to do 100 m.p.h. was put on the market about 1925 and early racing examples driven by Dr. Benjafidd and Capt. Woolf I3arnato lapped Brooklands at 102 to 103 m.p.h. and about 104 m.p.h. respectively. H. P. Bowler’s four-seater lapped at over 95 m.p.h. As the year of the car affects subsequent hotting-up modifications, it may be noted that 1924 engines are numbered 368 up and 376 up, 1925 engines from 747-749 up, and 1926 engines from 1,200 up–engine. numbers on the front cross-member. The 3-litre was virtually discontinued in 1927 but about twenty-five long chassis and some five Speed Models were made that would rank as 1928 cars. Although the Speed Model wheelbase has been quoted as 9 ft, 91 ins., ten special cars were made with a wheelbase of 9 ft. Concerning the chassis, front brakes were added to both chassis about September 1923. 1922 chassis have geared up dynamos running

at three times camshaft speed. Late 1923 chassis have a Hardy coupling in the carden shaft. One final note : if new pistons, etc., have been fitted as part of the tuning operation, it is well to remember that Bentley Motors Ltd. used to specify a running-in period of 200 miles for new pistons and 500 miles for new bearings.

The 44-litre

The 41-litre, 100 x 140 m.m. fourcylinder Bentley was produced as an experiment for the 1927 24-Hour Race at Le Mans and followed the general spedfication of the s-litre so closely as to be a

virtual enlargement. It was put into production in 1928. All years had twin S.T. carburetters. The closed models usually had a compression plate to give a ratio of 5.1 to 1 and the open-bodied ears had a compression-ratio of 5.3 to 1. The output was about HO b.h.p. The 1928 cars had a cone-clutch, later models a plate

clutch. A refinement is to fit a plate clutch to the 1928 cars, and Mr. McKenzie can now supply Borg and Beck plate clutch assemblies for £30, which are not only a great improvement but which save 60 lb. over the Bentleyplate clutch, being only about 6 lb. heavier than the 1928 cone assembly. Tuning operations largely follow those applying to the 3-litre. The crankshaft only requires balancing for racing work, but con.-rods should be balanced and piston weights equalised. The flywheel should be lightened, which also applies to the 3-litre. 1.7p to 14 lb. can be removed but further weight reduction is difficult to achieve. The lubrication system can cope with the additional loads, but a highgeared oil-pump may be used for racing purposes. The valve springs can be left unaltered. The compression-ratio can be raised by milling the base of the block. For speed work 6 m.m. may be removed, and pump fuels can still be used. Special pistons are desirable. Specialoid can supply suitable ones, with a saving of about 3 oz., but Mr. McKenzie has a design of his own which has several considerable advantages and a weight con servation of 4 oz. each. The standard B.H.B. pistons are prone to head collapse at speeds above 3,500 r.p.m. With 6 m.m. off the block the output will be in the region of 125 b.h.p. at 3,500 r.p.m. Normally this is the highest output permissible and in any ease torque may be expected to fall away above 3,750 r.p.m. When the block is milled the cam shaft drive is sometimes disturbed. The best remedy is to bore the taper and allow the vertical shaft to protrude higher than before into the head assembly. The standard S.U. carburetters, either 45° or vertical, Cannot be bettered, nor should the standard settings be changed for ordinary tuning purposes. As with the 3-litre cars, much can be done by variation of gear-ratios and reduction -of chassis weight. The top gear ratios are usually 3.78 or 3.53 to 1, with 3.3 to 1 the highest ratio, sometimes used on two seater examples. These are with the ” C ” type and ” D ” type boxes. The standard chassis weighs 25 cwt. and the four-seater about 37 cwt. Weight reduction is naturally dependent on the shortening of the wheelbase from the standard length of 10 ft. 10 in. The car built by Windrum and Garstin for M. Chambers to drive weighs about 27 cwt. with a shortened frame (actually a 3-litre chassis) but the fuel tank when full adds 4 cwt. which is valuable from the viewpoint of wheel adhesion under trials conditions. There is one very important point to be considered before hottingup is commenced. Certain early 41-litres had unsatisfactory connecting rods, most of which have been replaced, but about twenty cars remain in service with the original rods. This applies to S.L. and S.T. series engines and these rods should. be replaced before any attempt is made to increase speed and output. Incidentally, these chassis number pre(‘ xes. were changed about every twenty-five cars, without there necessarily being any change in specification. The unblown 41-litre, normally capable of 85 to 90 m.p.h., can be develow_d into a very fast

car. For example, the short chassis, car aforementioned laps Brooklands at over 91 m.p.h. with a 6 to 1 compressionratio, and it has extremely good acceleration. Miss Margaret Allan’s four-seater lapped Brooklands at over 98 m.p.h.,. and Baker-Carr’s short two-seater at over 104, while a 41-litre Bentley engine developed for racing purposes and installed in the single-seater Pacey-Hassan chassis, has lapped at over 128 m.p.h. The tappet-clearance is .019″ all round.

The Blown 41-litre

A development of the standard 44litre, to which a Villiers-Root supercharger driven from the nose of the crankshaft was added by C. Amherst-Villiers for Bentley Motors Ltd. and on which much experimental work was carried out on behalf of the late Sir Henry Birkin, Bt., at his Welwyn shops. The 19P.0 engine was blown at 10 lb. per square inch and the 1931 unit at 12 lb., the latter having a ribbed blower casing. The compressionratio is believed to have been the same for both years but was later reduced to Continued on page 391 obviate plug troubles, by fitting a com pression-plate. Afterwards suitable plugs were found and the compressionratio raised to the former ratio. The output was in the region of 175 b.h.p., and the Standard 10 ft. 10 in, wheelbase 1930 four-seater did -about 98 m.p.h. The special short-chassis four-seater raced by Birkin gave about 240 b.h.p., the port areas being increased and larger valves

used. Robertson-Roger, the present owner, is using a standard cylinder-block, but these blocks can be Modified in the manner described without recourse to fresh casting. This four-seater could do about 130 m.p.h., while it is common knowledge that the famous blower singleseater was capable of speeds approaching 150 m.p.h., taking the B rookl ands Lap Record in 1932 at 137.96 m.p.h.

The Sixes

The Standard Big Six 100 x 140 Mill. 6k-litre model was introduced at the end of 1926 and went out of production in 1929. With single-carburetter it gave 130 b.h.p., the compression-ratio being very low. Tuning may be carried out on almost identical lines to that of the 4i-litre. If the block is milled to

The 8-litre Bentley. put up the compression-ratio it may be necessary to readjust the triple-throw

system of camshaft drive. With 1929 standard modifications these engines gave 140 b.h.p. at 3,200 -r.p.m., and by raising the compression-ratio and fitting twin or triple S.U. carburetters the output can be brought into line with that of the Speed ,Six (175 b.h.p.) which was first produced late in 1928 and continued to 1931, most of the sales being Speed Sixes for the later period. The Speed Six has achieved over 130 m.p.h. on Brooklands in racing form, and the Barnato-HananSpecial with the 8-litre engine held the Lap Record for a time at 142.6 m.p.h.

Three chassis lengths of the Standard Six were made. 1930 models had larger inlet ports than the earlier cars,. friction shock-absorbers at the front and hydraulic at the rear and dual in place of double magneto ignition. Also larger fuel tanks. The normal top gear was 4.16 to 1, and the speed about 85 m.p.h. and weight with open body 45 cwt. The Speed Six had a top-gear ratio of 3.53 to 1, a chassis weight of 34 cwt. and a speed of about 90 m.p.h. The 8-litre Six-cylinder could exceed 100 m.p.h. in standard trim. The top-gear ratio was 3.53 to 1, and rev. limit 3,500 r.p.m. Same of Mr. McKenzie’s tuning operations on the latter type were described last February. The compression-ratio may be raised from 5.1 to 5.6 to 1, and the axle-ratio raised to 3.3 to 1. With balanced Crankshaft and conrods, lighter pistons and a better exhaust system 4,300 r.p.m. may be realised. Twin Delco &ins will give easier starting and better low speed running than dual

besides being easier to synchronise.

The Ricardo o.h. Inlet Engine

About fifty of these 4-litre cars were produced after the o.h. camshaft Bentleys had gone out of production. The bottom half of the engine is beautifully made and would permit of a greatly increased output. The head design does not allow any increase of compression-ratio and increasing the port areas and fitting triple carburetters only results in an output of 150 b.h.p. against the standard

output of 120 b.h.p. Supercharging is the most likely development, but Mr. McKenzie installs 61-litre old-type engines in the 4-litre chassis for clients who require improved running while retaining the advantages possessed by the later chassis.