Tugging at this thread of engine development, we bump into two Giants of combustion theory who also had the capability of translating research into devastatingly fast motorcycles and cars. Kusmicki’s ‘uncle’ in this regard must be Harry Weslake (above and below, behind riders)
, who began his career tuning motorcycle engines for Brooklands racers, using the theories of Sir Harry Ricardo
, who we’ll meet later.
Harry Weslake was a child (b.1897) of relative privilege whose father was the director of Willey&Co., manufacturers of gas meters. As a young man, he became obsessed with motorcycles and adept at riding and modifying them. His first mount at 16 was a Rudge
which he set about improving for competition in his local Exeter hillclimbs, by fitting an NSU two-speed gear hub to the crankshaft (below)
. Thus, his appearances at competitions often resulted in embarrassment to older and more experienced riders. Weslake became notorious for his fast riding style, and should be considered certainly a Veteran Café Racer!
The Rudge used a unique carburetor, the ‘Senspray’, which had several good ideas, but required the constant manipulation of two levers to give good performance. Harry’s first patent in 1914 was an improvement on the Senspray principles, and led to his production of the ‘Wex’ (Weslake EXeter - see below) carburetor after WW1.
The Wex was certainly one of the best performing carbs available in the late ‘Teens, requiring only a single lever to operate, and gave significantly better fuel economy with a cleaner gas flow, which also happened to boost power. Soon the Wex could be seen on all sorts of competition machinery, including Sunbeams
, Brough Superiors
, and Zeniths
. George Brough (below, with ‘Spit and Polish’, which used a Wex carb – note Harry behind GB
), in his usual effort to have that something ‘extra special’ from his parts suppliers, insisted at one point that Weslake personally road test SS100s fitted with the Wex carb, to see for himself how the machines performed.
Gordon Cobbold, a die-hard Sunbeam racer in the early 1920s, pulled Weslake into the world of track racing at Brooklands
, asking Harry to tune his engines and sponsor his machines to retain Cobbold’s amateur status. Weslake quickly became an indispensable figure for serious racers, and plenty of Vintage-era photographs from that hallowed track show him standing behind a victorious motorcycle which used a Wex carb. In short order, John Marston
, owner of the Sunbeam factory, contracted with Weslake to tune their competition engines. In 1923 he was given three 500cc OHV Sunbeam engines to tune, and while externally identical, he found their power output varied by almost 20%. After giving some thought to the matter, he decided to investigate what must be a logical reason for these differences. Previously (and for many years to come actually) factories tested their supposedly identical production engines, sending the best ones to the competition shop, the worst to their ‘cooking’ road models, without ever investigating systematically the reasons why such variation should occur.
And this is where the family business comes in; remembering the testing facilities at Willey&Co which measured the volume of gas metered by coin-slot pay devices, Weslake reckoned that it would be very useful to measure the volume of fuel-air mix which passed through his motorcycle cylinder heads, to see if there was any correlation between the volume of gas passed and their power output. [And for my US readers; in England it is common to pay for your cooking/heating gas via a coin meter on your gas line... no meter readers or mailed-in bills; if you don't pay, you have no gas!]
Weslake quickly found a positive correlation between the ‘good’ and ‘bad’ Sunbeam cylinder heads, and how efficiently they moved a given volume of gas. His hastily set-up test rig (see above)
soon saw considerable action testing other cylinder heads, and remained in continuous usage right into WW2. The Sunbeams he tuned using gas-flow tests were shortly gaining plenty of Brooklands ‘Gold Stars’
(for laps at over 100mph during a race), and winning TTs and GPs across Europe. Weslake experimented endlessly with shaping inlet and exhaust ports, and within a short time was the foremost expert on cylinder head tuning in the world.
Such talents do not go unnoticed, and Weslake was soon approached by W.O. Bentley
himself to ‘see what he could do’ with some troublesome engines. His results were so spectacular, regularly gaining 20%-40% power increases, that Bentley too used his services to tune their racing 4.5liter and 6.5 liter engines. Thus, in 1929, four Bentley engines which were Weslake-tuned gained the first four places at the LeMans race
, and won again in 1930
. His services were soon in demand for designing cylinder heads for various automotive manufacturers, including A.J.S.
, and S.S. (later Jaguar).
Weslake kept his hand in motorcycle development as well, and in the late 1930s encountered Joe Craig of Norton at an Earl’s Court Motorcycle Show
. These two no-nonsense gents took an immediate and intense dislike for one another when Craig, notorious for his biting sarcasm, made a ‘remark’ to Weslake about his prowess with automotive tuning. Whereupon Weslake proceeded to relate all the relevant tuning details – cam timing, inlet bore, valve size, compression ration – of a new Works racing Norton sitting beside them on a plinth, after only a quick exterior observation. Craig was visibly shaken that the tuning ‘secrets’ of his unbeatable racer were so easily summarized, and despite their personal animosity, Craig hired Weslake in the late 1940s to help improve gas flow characteristics in the Manx engine, which included reshaping the inlet and exhaust ports, and improving the turbulence of the fuel/air mix into the combustion chamber. At this point, Weslake must have worked with Leo Kusmicki on the development of the new ‘Squish’ engine for the 1950 Manx. Their methods differed, and Kusmicki, the former academic, knew that Weslake’s ‘testing rig’ should theoretically be unnecessary, since gas flow characteristics could be calculated mathematically. Of course he was correct, but a computer had yet to be invented which could provide results as quickly as Weslake’s experienced hand at port-shaping. That ‘John Henry
‘ moment would not come until the late 1980s!
Kusmicki and Weslake worked in parallel again on the Vanwall
racing project (with Sterling Moss, above)
. Tony Vandervell knew Weslake since the 1930s, as Vandervell had been one of the Brooklands ‘boys’, racing (of course) a Norton at the track with good results. Thus, when the Vanwall F1 team was formed, Weslake was hired to modify their Ferrari cylinder heads. When Tony Vandervell hired Kusmicki to design a British engine for the Vanwall, Weslake was on hand to help shape ports and look after cylinder head design. In a curious side note, while Vandervell was a Board member at Nortons, he ‘took’ Kusmicki’s design for the four-cylinder, watercooled DOHC racing 500cc Norton engine with him when he left the Board, during the A.M.C.
takeover of Norton in 1955. This design was expanded and used in one of Vanwall F1 cars…. so in effect, the legendary 4-cyl Norton engine WAS built, only in a 2.5 liter form.
Weslake’s accomplishments in the motorcycle and automotive industries are simply too numerous to mention, and range from LeMans-winning D-type Jaguars
to Gurney-Weslake GP
cars, and special engines for the Weslake-Metisse racer (below)
, with four-valve cylinder heads (above)
, which became popular tuning accessories for Triumph twins (and was the basis of the Triumph 8-valve TSS
model). Not to mention the Weslake speedway
motorcycle which took several World Speedway Championships in the 1970s.
I cannot recommend highly enough the Jeff Clew biography of Harry Weslake, ‘Lucky All My Life’ (Foulis, 1979)
, from which much of this information is taken. In my opinion, this work is Clew’s masterpiece; completely readable and compelling from start to finish, sufficiently technical to satisfy the connoisseur, with a personal narrative fascinating enough to read in one sitting. Which I’ve done four times.