BMW 328 Engine

BMW's six-cylinder car engines began with Schleicher. His entry in the Neue Deutsche Biographie has him designing a first six-cylinder engine in 1932 and a series of up to 3.5 litres following from 1934; the Frazer Nash Archives credit him and Fritz Fiedler, who had worked together at Horch, with the six-cylinder cars that made BMW's name as a car maker. In the 319, the 328's forerunner, the six had upright valves worked by pushrods and rockers, and Jenkinson put the limit on its power down to breathing.

Racing practice offered two cures, twin overhead camshafts and a supercharger. Jenkinson explains why BMW took neither: any change had to suit the production lines, and nothing radical could be accepted. BMW's own history gives money as the reason.

Jenkinson started his count from a different engine, the 319's sports six, whose bore was a millimetre smaller and whose stroke was the same, so in his account the new head lifted the output from 55 bhp, where BMW's starts at 50. Jonathan Wood dates Schleicher's decision to build a new head for an existing block to 1935, and names a BMW engineer, Rudolf Flemming, as having had a large part in its design; BMW and Jenkinson credit Schleicher and his team.⁠source 5​source 6​source 1​source 2​source 3

The new head was cast in aluminium alloy and bolted to the standard block, and it kept the old camshaft, low on the left of the engine and driven by a short chain, a "silent running Duplex roller chain" in Motor Sport's 1937 guide to AFN's car. The old engine had 12 upright valves worked by 12 pushrods from a single rocker shaft. The new head had two rocker shafts and two rows of valves set at an angle to each other, the inlets on the left and the exhausts on the right. Six pushrods rose to the inlet rockers. The other six worked a second set of rockers, which pushed six short rods across the top of the head, inside alloy tubes, to the rockers over the exhaust valves. Motor Sport's survey of engine design in February 1938 put it in a line: the BMW's single camshaft "operates the distant valves via rockers and horizontal push-rods".

The Bristol Owners' Club counts 18 pushrods in all and borrows L. J. K. Setright's phrase for them, "a forest of pushrods". Motor Sport's 1937 guide and the Frazer Nash Archives give the angle between the two rows of valves as 90 degrees, which gave each combustion chamber its hemispherical shape; the BMW Historic Motor Club's 45 degrees is each valve's lean from the vertical. The last idea, as Jenkinson saw it, was in the ports: the inlets were cast to run straight down into the head, not horizontally as was usual for a hemispherical head, which made room for longer inlet tracts and for three downdraught carburettors on top.⁠source 1​source 7​source 8​source 9​source 10​source 11

BMW, AFN's listing of 1937, Cecil Clutton and Jonathan Wood all give the road engine the same output; where they part is the engine speed at which it came, and the specifications set out which source gives which.

Racing pushed the output up quickly. Jenkinson traced the climb in three steps: an easy first gain, 110 bhp by 1938, and more again from the factory's racing engines by the end of 1939. For the Mille Miglia of 1940, BMW says, the race engines had stronger crankshafts with nine counterweights, among them a reinforced central weight that took away the risk of the shaft bending.

For the 1940 cars themselves BMW's own texts do not agree, and Motor Sport's report of the race and Mike Lawrence, writing in 1986, add figures of their own; Lawrence gave the works cars 140 bhp that year. BMW 328 Mille Miglia and the Touring coupé show BMW's and Motor Sport's figures in their specifications.⁠source 12​source 13​source 7​source 14​source 3​source 1​source 2​source 15​source 16​source 4

BMW's 327, a coupé and cabriolet that began in 1937 with the 326's engine, could also be had with the 328's. The car is the 327/28 (Wood writes 327/80), and German Wikipedia says it can be told from the ordinary 327 outside only by its wheels, held by centre-lock wing nuts, and a different speedometer. Wood gives it a top speed of about 140 km/h (87 mph). AFN sold it in Britain as a Frazer Nash-BMW, and Motor Sport's notes on AFN's range for 1939 gave the 327 with the 328's engine a higher cruising speed and a higher maximum than the 327 with its own.

On how many were built, the sources differ. Wood counts 580, coupés and cabriolets together; the production table in German Wikipedia's article on the 327 gives 569, built from 1938 to 1940, where Wood has the car introduced for the 1939 season. They disagree over the bodies as well: Wood says the coachbuilder Autenrieth of Darmstadt made both bodies, German Wikipedia that Ambi-Budd of Berlin supplied the 327's.

The 327/28 has a place in the Bristol story too. H. J. Aldington drove one to Filton in 1945, Wood writes, and George White of the Bristol Aeroplane Company told him it was exactly the kind of car he had in mind.⁠source 3​source 17​source 18

Bristol had planned for cars long before the war ended. In October 1941 the Bristol Aeroplane Company set out, in a post-war planning document, a car with a six-cylinder engine of two to 2.5 litres, and named Lancia and BMW as the makers to follow: their cars, the document said as Wood quotes it, were "'designed' in the best sense of the word", not merely assembled.

AFN, the importer of the Frazer Nash-BMW 328, had something Bristol wanted. Andrew Blow counts its BMW licence of 1934, which carried a right to manufacture, among the reasons Bristol was keen to talk. Serious work began on 9 May 1945, when Don Aldington called on Eric Storey at Filton, and by July, Blow writes, Bristol had agreed to buy AFN outright, once the licence's legality and its transfer to Bristol had been examined. Wood describes a majority shareholding, taken that July, with Reginald Verdon Smith as AFN's chairman and H. J. Aldington as its managing director.

Wood tells the Munich side as two journeys. On the first, soon after the war in Europe ended, Aldington went in uniform to find the car he had raced, and crashed, at Hamburg in 1939 (Wood) and left in Germany for repair (Blow). Wood says he could not find it, and that on the advice of BMW's sales director Fritz Troetsch he came back with one of the 1940 Mille Miglia roadsters instead. The second, later in 1945, was Bristol's. Aldington flew to Munich in a Stirling bomber and brought back six BMW engines, listed on a BMW delivery note dated 31 August, with the drawings of the 326, 327 and 328. Blow, writing from Bristol's records, puts the drawings, the prototype engines and Aldington's own car together in October 1945: in his account the car came home too.

Aldington also brought news of BMW's newer engines. Wood names an experimental two-litre twin-cam, the 318, meant to succeed the 328's engine, and the 3½-litre six of the 335 with a 328-type head; the 328's six won because it was available and proven. Lawrence has Aldington trying to bring a 2.5-litre twin-cam six to Bristol, too late: the company had gone too far with the engine it had to be interested.

Fritz Fiedler, the 328's chassis designer, worked on the engine too. The Frazer Nash Archives, which call him a consultant to Bristol on the 400, say he did most of his work at Isleworth, where he took charge of the post-war Frazer Nash and developed the FNS engine for it.

The partnership did not last. In April 1947, Wood writes, the agreement with Frazer Nash was unscrambled, and control of AFN went back to the Aldingtons. Lawrence gives the engine's part in it: AFN wanted up to 50 per cent more power than Bristol would give the 400, so the two agreed to part, and Bristol supplied AFN with FNS engines developed with Fiedler. Lawrence spelled the letters out as Frazer Nash Special; the Bristol Owners Heritage Trust writes Frazer Nash Sports.⁠source 3​source 19​source 4​source 6​source 20

How the engine came to Bristol is told two ways, and so is what it became. Mike Lawrence wrote in Motor Sport in 1986 that Aldington went to Munich with one of the teams that took German technology away as war reparations, and the Bristol Owners' Club's guide says Bristol "liberated" the remains of BMW at the end of the war. Wood answers with Rainer Simons's BMW 328: From Roadster to Legend, which says that "in compensation quantities of cash in English currency were sent to Munich". Blow, from Bristol's records, thinks a payment "likely, but cannot yet be proved", and writes that the deal was no part of war reparations.

On the engine itself the voices are closer. Jenkinson, describing the 328's head in 1986, saw the Bristol in it.

Lawrence, in the same issue, gave the Bristol engine "the same dimensions and layout as the BMW 328" but warned that one "has to be careful in describing the lineage": the ideas and drawings came from BMW, then passed through a drawing office in Bristol with ideas of its own. The Bristol Owners' Club says Bristol put the engine back into production with "only minor changes". In the Trust's film below, a Bristol voice puts it more bluntly: "People say, oh yes, it's a BMW made in England. It certainly wasn't."⁠source 4​source 3​source 19​source 1​source 11

The first Bristol-built engine was running in May 1946, by Bristol's own historians' account, and went for testing into a Bristol director's own 1938 327, which had been brought up to the 328's engine specification in 1939. It gave more power than Bristol thought a production car needed, and the production engines were set lower. The Bristol Owners' Club lists what changed: Whitworth threads instead of metric ones, better materials (the cylinder liners were of Brivadium, a very hard alloy from Bristol's aero engines) and, according to Setright, a higher standard of build all round. Wood adds 10 mm sparking plugs, changes to the camshaft and the carburation, and three SU carburettors, by all accounts at Aldington's insistence, on the 85A engine that most 400s had, though a good number kept three Solexes. Bristol also copied the Hurth gearbox (transmission) from the later 328s.

Then Bristol developed it. The club's 100 series, for the 403 to the 405, had larger valves, more valve overlap, larger main bearings, a different oil pump and lighter valve gear. The 110 series of 1958, for the 406, grew to 2,216 cc with a larger bore and a longer stroke, and gave its power lower down. AFN's FNS engines had more overlap and larger inlet tracts; the BS series, the "Bristol Sports" engines, was built on the 100-series block. For the Bristol 450 racing cars of 1953, Lawrence writes, the compression was raised, the valves were enlarged and a head with six ports replaced the standard head's three siamesed ones. How far it went is disputed.

The Bristol Owners' Club is one of those that quote the higher figure, for the 450s that raced at Le Mans.⁠source 3​source 11​source 4

Every Bristol from the 400 of 1947 to the 406 used the engine; the 400 itself, Wood writes, had a 326 chassis and a body after the 327 coupé. All but two post-war Frazer Nashes used it too, from the Le Mans Replica, named after the fourth car built, which finished 3rd at Le Mans in 1949, to the Sebring of 1954; the Le Mans Replica's front and rear suspension were still based on the 1930s BMWs'. By the Frazer Nash Archives' list the exceptions were one Targa Florio with an Austin four, shown in 1952, and the last car the factory built, the Continental of 1957, with a BMW V8. Blow counts 85 post-war Frazer Nashes sold by AFN.

The engine reached Formula 2 as well, which was the World Championship formula in 1952 and 1953. Lawrence tells how an FNS engine "had somehow gone through the back door at Bristol" into the Cooper-Bristol of 1952, which Mike Hawthorn drove, and how a rift opened between AFN and Bristol. ERA used a Bristol six in its G-type, meant only as a stop-gap, and Stirling Moss drove it from the European Grand Prix at Spa on 22 June 1952; Bristol bought the G-type and its spares when ERA's owner, Leslie Johnson, left the business, and its own 450 grew from them. A 450 finished 5th overall and won the two-litre class at the Reims 12 Hours in 1953; at Le Mans in 1954 the three 450s finished 7th, 8th and 9th, the first three places in the class, and in 1955 they did it again.

AC fitted the 100-series engine, in its sharper forms, to the Ace and the 110 series to the Greyhound, and the BS engines went, the club says, into Arnolt-Bristols, Coopers, ERAs, Lotuses, Listers, Tojeiros and a number of one-offs. Lawrence's list adds Kieft.⁠source 3​source 21​source 19​source 4​source 11

In Germany the engine stayed in its cars, and the cars went racing again. Veritas was founded in 1947 by Ernst Loof, BMW's former head of racing, to build sports cars on the 328, and it often began with a used one and gave it a more streamlined body; by the BMW Historic Motor Club's count, 32 Veritas cars had been built by 1950, all but one with 328 engines, and the firm stopped building cars in 1953. BMW says the 328 and the cars built from it were still dominating the two-litre class into the 1950s.

German Wikipedia's article on the 328 gives reasons for so many conversions and specials: the engine was almost the only competitive one left in Germany, there were enough of them, and its size suited the new Formula 2. The same article says Veritas and AFM made their first cars from existing BMWs, and that at Eisenach, by then across the border in the east, development of high-performance cars based on the 328 went on under the EMW name, for the state racing collective, and that Veritas, AFM and the collective all built smaller versions of the engine, with roller-bearing crankshafts, for 1.5-litre racing. It names other builders too: several 328s converted for competition in the workshop of Reif, the former BMW agent in Chemnitz; the Grosser Werkmeister, built by Georg Werkmeister in Dingelstädt; and, outside Germany, Maurice Monnier's Monnier Spéciale of 1947, built with a 328's running gear and engine parts.⁠source 22​source 2​source 23​source 3

BMW's own post-war six went back to the 326. When Jenkinson visited Munich in 1955, guided by Fiedler, by then head of BMW's design department, the six in the cheaper BMW was, he wrote, "virtually a re-designed pre-war Type 326 engine". Bristol carried its version on to the 2.2-litre 110 series of the 406. As the 1950s ended, Lawrence writes, Bristol planned a 3.5-litre six to replace it, bought a Chrysler automatic gearbox to try, and found it "came with a V8 engine attached"; its own engine work stopped, and from 1961 Bristols had Chrysler V8s.

Some of the engine's working life survives on paper. In 2019 the Bristol Owners Heritage Trust was given about 500 of Bristol's test-bed cards from the 1950s, from Tony Crook's estate, most of them for sports and racing engines: Cooper and ERA engines, the FNS, the BS series and the 450's. </content>⁠source 24​source 11​source 4​source 3​source 20