The Rowell 60 Story

by Adrian Duncan



Click on images to view larger picture,
hover over the images for a description.
    Background
    Establishment of the Company
    Rowell Production Commences
    The Rowell 60 Mk I Described
    The Rowell 60 Mk I on Test
    Further Development of the Rowell Range
    Final Fling—the Rowell 60 Mk II Series 2
    The Rowell 60 Mk II Series 2 on Test
    Gathering Clouds
    The End of the Road
    Production Figures and Serial Numbers
    Conclusion

Here we take an in-depth look at the Rowell 60, a near-legendary 10 cc racing engine from Scotland which was highly respected by British competition modellers in its day and is something of a Holy Grail for present-day collectors of classic racing engines. The development of this fine powerplant was inspired by its designer's interest in tethered model car racing, but the Rowell found application in model aircraft and tethered hydroplane racing as well.

I may as well start right out by making it very clear that my personal claim to any credit for this article is strictly limited! A significant proportion of what follows is synthesized from information published on the outstanding "On the Wire" (OTW) website, which is devoted to tethered hydroplane and car racing, past and present. This is a truly fascinating site which can be highly recommended to modellers in general and model engineers in particular. The primary author of the material on the Rowell 60 was Lynn Blowers.

Any article on a subject of this nature is entirely dependent on the quality of the information that is made available by others. Lynn was most punctilious in acknowledging the help that she received from a number of primary sources. In turn, I hereby acknowledge my own inherited indebtedness to all these individuals. In particular, most of the company details that were provided for the original article were by courtesy of the recognized authority on Rowell Motors, George Blair. All of us are greatly indebted to George for his efforts to preserve the history of this company.†

I'm deeply grateful to Lynn for her gracious permission to draw upon her earlier work. In addition, I wish to acknowledge the kindness of both Lynn and Hugh Blowers in reading the following text in draft form and offering many helpful suggestions and much encouragement. On top of all this, these two fine people and kindred spirits have been most generous in providing a number of the attached images and much of the reference documentation. Without such assistance, this article would have little interest or credibility.

Although I have drawn very heavily upon the sources acknowledged above, I have incorporated many of my own comments and observations into the following text as well as reorganizing the material somewhat for MEN readers. I must therefore stress that any errors or omissions in what follows are entirely my own responsibility.

Having clarified and acknowledged our sources, let's look at the background to the introduction of this very interesting engine.

Background

Prior to the refinement of dependable multi-channel radio control systems beginning in the early 1950's, the all-out use of large high-performance model engines was a practical proposition from a safety standpoint only in applications where the model remained under restraint at all times. In the absence of reliable radio gear, this meant models which followed a circular path at the end of a centrally-secured wire. This was referred to as tethered operation of the models in question. Rail-guided operation was also possible in the case of model cars, but the principle remained the same—the vehicle was constrained to follow a fixed path when running.

Tethered operation of model aircraft had its roots prior to WW2 in the form of round-the-pole flying. However, it was not until 1942, when Jim Walker patented his two and three-line U-control systems which allowed full in-flight control of a tethered flying model, that this form of aeromodelling commenced its rapid rise to prominence. Under the name control line, it became immensely popular in Britain and elsewhere during the years immediately following WW2, maintaining this popularity right through to the 1970's. Although radio control has dominated the model aircraft scene since that time, control line flying continues to retain a dedicated group of adherents (including myself) who are drawn by the closeness of the pilot to the action as well as the sense of direct tactile feedback from the model that control line flying alone can impart.

Tethered hydroplane racing was well established in Britain prior to the onset of WW2. Indeed, the first national model boat racing event, the inaugural "Model Engineer Speedboat Competition", was held as early as 1902, the winner's speed being an awe-inspiring 5 mph! Progress thereafter was rapid, and by the mid 1930's speeds were approaching 50 mph using both flash steam and internal combustion (I/C) engines. The sport maintained a steady growth in popularity right up to the onset of WW2 and resumed thereafter, continuing to draw a small but dedicated group of adherents up to the present day. Speeds of over 135 mph are now being achieved—quite a step up from 1902!

By contrast, the parallel sport of tethered car racing was a relatively late starter in Britain. Led by the famous Dooling brothers, much of the pioneering work took place in the USA during the latter half of the 1930's and early 1940's, with ever-increasing commercial interest being displayed in relation to the supply of purpose-built engines such as the Hornet 60 from Fresno, California, as well as cars and accessories. Activity in Britain during that period was largely confined to free-running clockwork and rubber-powered models.

Oddly enough, it was the advent of WW2 that triggered the growth of interest in I/C-powered model car racing in Britain. During that unhappy period, the flying of I/C-engined model aircraft was officially banned. The effect of this was to ground a sizeable number of perfectly serviceable model aero engines, leaving a group of frustrated power modellers very much open to alternative suggestions for use of these engines.

Somewhat illogically in view of the severe rationing of petrol which prevailed at the time, the operation of model I/C engines for purposes other than flying was not prohibited. Aeromodeller publisher DA Russell was not slow to recognize the pent-up frustration which existed, and in September 1942 the magazine featured the first of a series of articles promoting the use of the grounded model aero engines to power model racing cars instead. A set of rules was established for the use of such cars in competition. The ensuing inaugural competition was well supported, the winner being Gerry Buck, whom we shall meet again during the telling of the Rowell story. This and subsequent successes led Russell to go one step further in late 1944 by establishing the British Model Car Club.

During this period, British enthusiasts necessarily had to work with chassis, accessories and in some cases engines of their own making, since no such items were commercially available during the war years. However, following the conclusion of hostilities, model car racing experienced a huge upsurge of interest in Britain, with new clubs and tracks appearing almost on a weekly basis.†This activity in turn drew the attention of commercial interests to the hobby, with the result that firms supplying goods related to model car racing quickly began to appear. At the peak of the model car boom, it's estimated that over 50 such firms were active in Britain! The growth of the hobby was also sufficiently rapid to promote the establishment of Model Cars magazine, which published its first issue in September 1946. A second publication, "Model Car News", soon followed.

The 10 cc class was one of the Blue Riband categories in all branches of competitive power modelling, causing several British manufacturers to decide more or less simultaneously in 1948 to enter the field of 10 cc model racing engine production. The North Downs Engineering Company of Whytleafe, Surrey, introduced their Nordec series which included both car and aero versions of their McCoy-inspired 10 cc offering. Ten-Sixty-Six Products Ltd. of Worcester added the 10 cc Conqueror to their already-established range, while our main subject company, Rowell Motors Ltd. of Dundee, Scotland, released the Mk I version of their Rowell 60 model. In contrast to the Nordec, which was intended all along as a multi-purpose unit, both the Ten-Sixty-Six and Rowell models were designed primarily for model car use.

It's interesting to observe that all three of the above-mentioned designs appear to have been inspired by a genuine passion for the modelling hobby on the part of their instigators. The Ten-Sixty-Six and Rowell ventures were unquestionably both established as a result of their respective founders' well-documented interest in model car racing. The makers of the Nordec series were already well-established in the full-sized automotive business, thus having no economic incentive for entering the relatively limited field of big-bore model racing engine manufacture. Their entry into this field was entirely due to the fact that their works manager and chief designer John Wood was a keen participant in control line speed model competition. It was undoubtedly due to Wood's enthusiasm that the company ownership was persuaded to become involved in this field.

The instigators of these three competing models were not alone in promoting the development of large model racing engines in Britain. Ted Martin, chief designer for the Anchor Motor Company of Chester, manufacturers of the AMCO range, also tried unsuccessfully to persuade his employers to sanction the development of a 10 cc AMCO racing engine. In hindsight, given the eventual fate of the other three British 10 cc racing models, it was almost certainly a wise decision on the part of AMCO management to divert Ted's attention away from racing engines towards the development of what became his most famous design, the AMCO 3.5 BB diesel.

Having set the scene, it's now time to look at the genesis of the Rowell Motors business. Those interested in learning more about the tethered car and hydroplane racing scenes both past and present are highly recommended to visit the outstanding OTW website to which reference has already been made.

Establishment of the Company

Wilfred George Rowell, always known as Wilf, was a professional engineer whose main field was the design of industrial cranes. Contrary to a widespread belief, he was not a Scotsman but was born in 1915 at Wall in Northumberland, England, a little to the south of the border with Scotland. He served his apprenticeship with Hexham iron founders Tyne Metal Co. Ltd., subsequently working for various industrial crane makers including Cowan Sheldon of Carlisle, Herbert Morris and finally Coles Cranes.†By 1947, at the age of 32, he was living and working in Dundee, Scotland. He was to spend the rest of his life in Scotland.

Wilf Rowell was a talented hands-on engineer whose technical skills were unusually wide-ranging in scope, extending very much into his hobbies. His personal projects included building 8 mm silent and sound cine-projectors, about which he wrote an article which appeared in the June 1943 issue of Model Engineer, going on to publish a book in 1948 entitled "How to build your own projector". Some measure of the range of his design and technical skills may be gained from the fact that he also constructed a camera, a full sized Wurlitzer organ and a 9" screen television set that he finished just in time to watch the Coronation broadcast in 1953!

Wilf was also a skilled model maker. In 1930 at the age of only 15 years he completed a large live steam loco named "Helen of Troy", and in 1951 constructed a smaller-gauge Class V locomotive to run round the garden of his home at Invergowrie, Dundee...

In July 1947, a letter appeared in Model Car News announcing the formation of the Dundee and District Model Racing Car Club, the secretary being one WG Rowell. The Club's first official race meeting was held in March 1948 and the report of the event shows that WG Rowell competed using a car powered by a 10 cc Hornet engine which had been sent over to him by a relative in America. Brian Sherriff, who was soon to become an integral part of the Rowell operation, ran a 2 cc E.D. powered Masco Kitten at this meeting.†

All of this makes it abundantly clear that, along with his many other interests, Wilf Rowell had by this time become a dyed-in-the-wool model car racing enthusiast. It was presumably this interest which led him to throw his hat into the model racing engine design ring in the summer of 1948. In taking this step, he was doubtless encouraged by a number of his fellow enthusiasts, notably the prominent competitor Gerry Buck, who was a passionate supporter of British-built cars and engines and had actively encouraged Wilf to develop an engine that could challenge the imports.

Rowell Production Commences

In the summer of 1948, advertisements began to appear for the Rowell 60 racing engine, then in its Mk I configuration as illustrated. This initial version of the engine clearly drew much of its design influence from the American Hornet 60 model—doubtless Wilf Rowell's direct experience with the Hornet design played a significant role here. The Mk I Rowell was most readily distinguishable from the Hornet by the tapered cylinder fins, unusually large-diameter head, one-piece main crankcase casting with detachable front housing and oversized exhaust stack. It actually looked a bit like a Hornet that needed to go on a diet!

With RRV induction through a Hornet-influenced venturi and a sturdy ignition timer using standard Lucas automotive parts, the motor was far in advance of anything else being manufactured in Britain at the time of its introduction. The Ten-Sixty-Six Conqueror and Nordec 10 cc models which became commercial rivals to the Rowell both appeared shortly thereafter, but the Rowell had already established a new benchmark for design and performance in Britain which the domestic competition would have to work very hard to beat.

The Rowell 60 was advertised as being available from Rowell Motors at 93 Victoria Road, Dundee, together with an ever-growing line of model car racing components and accessories. At first sight this would logically have been taken to be the factory location, but in reality it was that of a model shop owned by Wilf's previously-mentioned friend and fellow competitor Brian Sherriff, who distributed the Rowell products.

By 1948 standards, the engine was very far from cheap. The price quoted in the brochure for the standard spark ignition engine, complete with KLG plug, was a whopping £12 10s 0d (£12.50)—a small fortune in those far-off days in early post-war Britain when a person earning £8 a week would have been considered well-off! The engine was offered in glow-plug configuration as well, albeit at the same price. If you wanted to use the engine in a model aircraft application, you had to pay an extra 7s 6d (38 p) for a prop driver like that fitted to the illustrated example. A marine flywheel and universal coupling for hydroplane use added £1 2s 6d (£1.13) to the basic price of the engine. Worse yet, use of the engine in model car service required a further outlay of £3 0s 0d for a flywheel and centrifugal clutch. Half a month's pre-tax salary for one model engine—pretty steep!

Although undeniably high, these prices were directly comparable with those charged for one of the Rowell's chief British competitors, the Nordec R10. By contrast, the basic price of the competing Ten-Sixty-Six Conqueror in factory-finished guise was a mere £8 5s 0d (£8.25) for the spark ignition version and £8 2s 6d (£8.13) for the glow-plug model. However, most of the Conquerors were sold in kit form to model engineers. In that form, they were very much cheaper than either of their rivals at a very economical £2 17s 6d (£2.88), but required a great deal of highly skilled work to complete. In general, the Conqueror appealed to a somewhat different market sector. Like the Nordec, it also came nowhere near matching the performance of the Rowell.

Quite apart from the engines themselves, the ever-increasing list of chassis and drive train components offered to the car racing fraternity under the Rowell name raises the question of where all these products were made. The manufacturing certainly did not take place at Brian Sherriff's model shop at 93 Victoria Road, Dundee. In fact, it's probable that Rowell Motors never had their own manufacturing facility, instead contracting out most if not all of their manufacturing to others. The production figures for the engine and accessories were certainly low enough for this to have been a perfectly rational proposition.

A considerable body of evidence points towards the involvement of at least two local companies in the production of components for Rowell Motors. These were the Arbroath Tool and Gear Company, who may have manufactured the gearbox components and other precision parts for Rowell, and Kingdom Models at the quaintly-named Coaltown of Wemyss in Fife, who built the very well-made 1 cc Clan diesel motor (reportedly in a converted cowshed!) and may well have been involved in the manufacture of the Rowell 60 engines in addition. The OTW article on Rowell informs us that "a thrifty Scotsman who cycled to Coaltown of Wemyss to buy a Clan engine direct from Kingdom Models (thus cutting out the middleman!) reported seeing a number of Rowell engines whilst he was there."' It's a reasonable inference that the Rowell engines were assembled and tested at Wemyss, at least at this stage. It is certainly true to say that the early engines were quite crudely sand-cast, with very little attention being paid to quality of external finish or fettling of the castings. It was a clear case of quality where it counted.

Wherever the engines were made, it's almost certain that they were individually constructed rather than being assembled on a production line. The number of minor variations between examples makes this abundantly clear. A case in point is the configuration of the driving end of the crankshaft. Hugh Blowers reports that examples have been encountered with threads of 1/4 BSF, 5/16 BSF (as on my own examples) and 5/16 UNC! Moreover, the flywheel-prop driver attachment varies between taper-on-shaft, split collet and taper/grub screw systems! It appears that the makers began by manufacturing the key components to somewhat flexible specifications and then finished the rest to fit individually, using whatever materials, threading tools and fasteners they happened to have on the bench at the time. Not atypical of small-scale British model engine production during the early post-war period...

The engines carried a thirty-day warranty against defects encountered during normal operation. Failures due to accidents involving any model in which the engines were installed were specifically not covered, nor were problems caused by tampering with the unit. Reading the fine print, one may suspect that few of these guarantee cards were actually registered since they had to be returned to Rowell Motors within three days of the purchase date! It may be doubted that many modellers managed to comply with this requirement ...

By 1949 Rowell Motors had become a limited company.†The directors were Brian Sherriff, Wilfred Rowell and one T. Scott. The respective roles of these individuals are not completely clear. Brian Sherriff evidently had the trade contacts and the marketing premises, while Wilf was obviously the designer and motivator. The part played by the elusive T. Scott is unknown. However, the scale of the company's activities makes it clear that a sizable financial commitment would have been required from some external source. It's possible that Scott was the money behind the venture.

The Rowell 60 Mk I Described

The Rowell 60 was a more or less conventional racing engine of its day, displaying a mixture of Hornet and McCoy influences. It featured the standard combination of rear disc valve induction, big-bore venturi with surface jet needle valve, bolt-on front and rear housings, twin ball-race crankshaft, automotive-style timer, high-crowned aluminium baffle piston with two rings and cross-flow loop scavenging. Nominal bore and stroke were 0.940" and 0.875" respectively for a displacement of 0.607 cu. in (9.95 cc), exactly the same on all counts as the McCoy and the Hornet. However, the Rowell was far more massively constructed than either of its American rivals, hence being both bulkier and heavier. It weighed all of 546 gm (19.25 ounces) minus flywheel—well in excess of any of its competitors. It was of course intended primarily for car use, but even so this figure seems somewhat excessive. That said, a stiff structure is a well-demonstrated aid to dependable high performance.

Although weight was doubtless a factor even in a racing car or hydroplane context, the weight of the Rowell was not an over-riding impediment to its employment in those disciplines. However, it did represent a substantial objection to the engine's use in model aircraft applications, since power required to generate lift is unavailable to generate speed. It is likely for this reason that the name of the engine is conspicuously absent from the results of control line speed contests held during the period when it was in production. However, it did find such applications, although the only published record of any contest success that I can find is a top three placing in the 10 cc speed category at the South East Area Control Line Championship meeting held at Dover on Easter Weekend of 1949 (Aeromodeller, June 1949). To put this in perspective, it has to be noted that the winning speed at this meeting fell slightly short of the 100 mph mark.

That said, I retain some very clear recollections of a heavily modified Rowell 60 (a Mk II to the best of my recollection) running on glow-plug ignition which was still powering a control-line speed model in the hands of one of my club-mates in Sheffield, Yorkshire, around 1960. I actually acted as helper on this model on a few occasions, thus getting a close-up (and extremely noisy!) view of a Rowell 60 in action. My recollection is that the engine started and ran well and got the model into the air with no problems but was not competitive against the contemporary opposition.

In keeping with established large racing engine practise, the crankcase, front and rear housings, timer frame and cylinder head of the Rowell 60 were all sand-cast in a high-grade aluminium casting alloy which was subsequently heat-treated. Although the finish of some of the early castings was a bit on the rough side, the engines were machined to very high standards, with particular attention being paid to the achievement of outstanding working fits and finishes. In the view of many observers, the standard of construction exceeded that applied to the famous McCoy engines from America.

A great deal of design effort was clearly put into reducing crankcase volume in order to improve the engine's pumping efficiency. This extended to the use of a bypass passage which, like most others of the period, was rather on the small side, a feature which it shared with the Nordec and Ten-Sixty-Six opposition. This was likely due to the influence of the contemporary versions of the Hornet and McCoy engines, both of which also had somewhat undersized bypasses. McCoy were very shortly to address this issue later in 1948 with the introduction of the vastly-improved Series 20 version of the McCoy 60, but the Rowell (and indeed the Nordec and Ten-Sixty-Six models) had already been released by that time. Sadly, this improvement to the McCoy together with the eye-opening performance of the Dooling 61 made all of the British challengers obsolete in all-out performance terms before they had fairly got started.

A feature of the Rowell which was seemingly intended to further reduce crankcase volume was the fact that the bore in the upper cylinder casting for the liner was not carried all the way through to meet the axial bore through the crankcase interior for the front and rear covers and the crankweb. Instead, it was bored only deep enough to accept the full length of the cylinder liner and accommodate the piston skirt at bottom dead centre. This left a pair of wedge-shaped pieces of metal between the lower end of the bore and the upper perimeter of the crankcase interior. The con-rod actually operated though a slot cut through these wedges. The effect of this approach was to reduce crankcase volume by the space occupied by the residual metallic intrusions, at the cost of somewhat restricting gas access to the bypass passage. This arrangement precisely mirrored that seen in the design of the Hornet 60.

The porting of the engine was quite conventional by contemporary racing engine standards. The basically rectangular exhaust port was divided into 6 segments by narrow pillars intended to prevent ring snag. The transfer port was similarly divided into only three segments. Once again, both of these features reflected the design of the Hornet 60.

The lower entry to the bypass passage was extremely restricted in size. The presence of the previously-mentioned metallic intrusions left behind after machining didn't help in this regard either. To compensate for this, the transfer process was assisted by the presence of additional porting in the liner on the transfer side below transfer port level which corresponded at bottom dead centre with similar ports in the piston skirt. In addition to enhancing transfer efficiency by improving gas access to the bypass passage, the piston ports also encouraged improved piston cooling and lubrication by promoting the flow of cool incoming mixture through the piston interior. In terms of the engine's breathing capacity, such ports were very necessary in the Rowell design, just as they were with the Hornet. The Rowell piston looked in fact very much like its Hornet counterpart—even the rings are interchangeable.

The piston porting used in the Rowell 60 appears to have taken two distinct forms. Many examples used matching pairs of drilled circular ports exactly like those used in the Hornet 60 as illustrated above. However, my own illustrated example has a single port of oval "race-track" form in the piston skirt to go along with a matching oval opening in the cylinder wall. The effect of this is of course to further open up communication between the piston interior and the bypass passage—a good feature from the standpoint of the engine's breathing capacity. The appearance of illustrations of the Rowell 60 featuring the oval port both in Lawrence Sparey's July 1949 Model Cars test report (see below) and in a May 1950 Model Engineer article on engine maintenance by GW Arthur-Brand seems to prove that this was a factory variant as opposed to an owner modification. This appears to be a further example of the individual approach taken to the construction of each engine.

Like the competing Ten-Sixty-Six Conqueror, the port timing of the Rowell 60 was very much on the conservative side. The figures measured by Dick Roberts in the course of his 2008 test of a Mk I Rowell 60 (see below) were approximately 130 degrees for the exhaust and 110 degrees for the transfer, very close indeed to figures taken from my own Conqueror.

The light alloy rear disc valve of the Mk I Rowell was timed somewhat more conservatively than that of either the Conqueror or the Nordec R10, remaining open for some 170 degrees as compared to the 190 degrees of the Nordec and the even more aggressive 195 degrees of my own Ten-Sixty-Six model (which may well have been modified by an owner). The disc was bronze-bushed at the centre to provide a hard-wearing bearing surface. The steel mounting spindle had a working journal diameter of a nominal 1/4 in., which extended part-way through the backplate. The balance of the spindle was stepped down to a 4BA threaded length which engaged with a similar thread in the backplate. Once assembled, the spindle was secured by a lock-nut placed on the protruding spindle thread at the rear.

The 0.350" inside diameter of the screw-in intake venturi was well chosen, since in combination with the use of a surface jet needle valve of conventional racing engine design it provided sufficient suction to permit the operation of the engine without pressure feed while still ensuring an adequate supply of fuel mixture. To further assist suction, the jet was actually extended some distance into the venturi throat. The venturi bore was substantially larger than the standard venturis used on both the Rowell's Mk I Conqueror and Nordec rivals. Examples have been reported with 0.375" dia. venturis—these may either be owner modifications or represent a change made by the factory at some point.

The connecting rod was machined from light alloy. Somewhat unusually, this component was un-bushed at either end. The crankshaft had a fully circular web with a crescent shaped counterweight machined onto its rear face. The design of this component clearly reflected that of the Hornet in that the forward face of the crank disc was recessed, with additional counterbalance being provided through the inclusion of a crescent-shaped metal weight set into the recess on the counterweight side. Unlike the Hornet, which used a steel counterweight which was welded in place, that used in the Rowell was of brass and was secured using three rivets which passed through the crankweb.

The combustion chamber was more or less orthodox, the piston having a McCoy-style high domed baffle and the head having a matching contour. The plug was offset towards the transfer side, presumably to speed up the involvement of the pocket of gas behind the piston baffle in the combustion process. The head was secured by no fewer than eight slot-head screws. The compression ratio of the Rowell was cited as 12:1—very high for the period, but no doubt tailored to the use of a methanol-based fuel as opposed to gasoline. A volumetric measurement taken from my own example more or less exactly confirms this figure.

The owner was evidently expected to run the engine on a methanol/castor oil mixture, a fuel which was also specified for the Ten-Sixty-Six Conqueror with its even higher 13:1 compression ratio. Nitro-methane was virtually unobtainable in Britain at the time, so its use would not be anticipated, otherwise a somewhat lower compression ratio would likely have been employed. Even so, the manufacturers of the Rowell did note on the instruction sheet that a higher performance could be obtained through the addition of nitromethane to the fuel.

I must admit that I would be rather hesitant to run this engine for any length of time while leaned out on glow-plug ignition, especially on a high-nitro fuel. This is because glow-plug ignition timing at a given plug element temperature is mainly dictated by compression ratio. Plug element temperature in turn is heavily influenced by mixture strength—a lean mixture creates a higher element temperature regardless of operating speed. As a result, pre-ignition when leaned out using a 12:1 compression ratio would likely be a factor until relatively high speeds were reached. The engine is probably best operated on spark ignition. Indeed, the absence of any presently-known surviving purpose-built glow-plug examples suggests that it was more or less universally operated in spark ignition mode, at least during its heyday.

Another feature which gives rise to considerable hesitation is the complete absence of any form of knurling on the driving face of the bobbin prop driver, at least on my example as received. A high-torque engine of this displacement having such a high compression ratio undoubtedly requires some form of positive prop/driver keying system, if only for safety reasons. I found by direct experiment that even flipping a prop over compression was enough to cause the prop to slip unless it was tightened to what I'd consider a drastic degree. This was especially true once a trace of oil got into the prop-driver interface, as it almost inevitably will during operation.

Since I planned to test-run my example using an airscrew, I first tried installed a pair of 5 BA studs in the face of the prop driver, creating matching 1/8 inch holes in the rear of the prop. I've seen a similar provision applied to several examples of the Nordec RG10, and that engine had striations on the driving face as well as a lower compression ratio and significantly lower torque output. However, even this proved inadequate upon initial testing—the studs are of course very effective in transferring torque loadings between the driver and the prop hub but do nothing to support the prop hub as it resists the combination of centrifugal forces and cyclic torsional stresses imposed upon it by the prop blades. I shattered two wooden props during starting attempts before deciding that in the interests of self-preservation, further measures were required! Accordingly, I ended up knurling the driving face of the driver. This proved completely effective.

I've little doubt that anyone planning to use a Rowell in aircraft service would have taken steps to improve the interlocking between the prop and driver. In fact, Hugh Blowers informs me that some (but not all) examples did have spiral knurling on the driving face of their prop drivers. It's a matter for wonder that all did not do so—it's certainly not safe to try running a Rowell on an airscrew without some measures of this kind. Be warned!

The timer was of some interest, incorporating an effective adjustment feature for the point gap. The fixed point was mounted on a stamped metal bracket that rotated about the moving point pivot. It was restrained from rotating by a bolt which passed through both the timer frame and a slotted hole in the bracket, which was then secured to the frame by a nut. Clearance was adjusted by first slackening off the clamp nut and then rotating the fixed point assembly slightly. The clamp nut was then re-tightened.

A final oddity worth noting is the fact that the assembly screw threads used in many examples of the Rowell 60 are BSW rather than BA. My own two examples both use 1/8-40 BSW head-screws and 5/32-32 BSW front and rear cover screws. This is rather unusual for a British engine of the period, suggesting that the use of such threads was dictated by the fact that this was what was available at the manufacturing facility! This highlights a not-uncommon facet of early post-war British model engine production—materials and fasteners were scarce, forcing manufacturers to use whatever they had on hand at any given time. This actually resulted at times in different examples of the same model having different assembly threads.

The Rowell 60 Mk I on Test

As the most powerful model engine yet to emerge from a British manufacturer, the Rowell 60 naturally attracted the interest of the engine testing gurus of the day. Prominent among these was the redoubtable Lawrence H. Sparey, whose tests appeared in a number of contemporary modelling magazines.

The Rowell 60 Mk I was among the engines which was subjected to Sparey's attentions. It was in fact the second British racing engine to be tested by him, the first having been the Nordec RG10. Sparey's test of that model had appeared in the March 1949 issue of Aeromodeller magazine. It was in fact Sparey's first-ever test of a glow-plug model engine.

Given the fact that the Rowell was specifically designed for model car racing, Sparey's test of that model did not appear in Aeromodeller. Instead, it was published in the July 1949 issue of Model Cars as number 2 in a series of tests on what Sparey termed "High Speed Engines". It appears from this that Sparey's test of the multi-purpose Nordec must also have appeared in Model Cars, although I do not have direct evidence of this at present.

Sparey's report on the Rowell was generally very complimentary as regards its handling and performance. Once he had established a starting procedure which accommodated the flooding-prone gravity fuel feed system employed, he stated that he "could not speak too highly of its handling qualities". Sparey took advantage of the opportunity to make a direct comparison of the engine's performance both on spark and glow-plug ignition. On spark ignition, running was said to be "perfect at all speeds". On glow-plug ignition, running was reported to be "equally good" apart from a tendency to misfire "at speeds below 6,000 rpm"(!).

This latter comment highlights Sparey's chief failing as an engine tester—his mind seemed unable to keep pace with emerging engine design trends and technologies. In particular, he clearly failed to grasp the fact that glow-plug ignition timing is fixed to within relatively narrow limits by a combination of compression ratio, fuel mixture composition and plug heat range with its associated element temperature. While the effect of engine speed on element temperature certainly plays a role in establishing the timing, it has less primary influence than the other three factors. Consequently, the timing of a glow-plug engine having a set compression ratio and using a given fuel composition is relatively (although not of course completely) independent of the engine speed when fully leaned out, and the operator can do little to change it apart from using either a colder plug or a less "hot" fuel. The only way to keep such an engine happy is to run it at or near the speed at which the various factors combine to optimize the leaned-out ignition timing. In the case of a racing engine like the Rowell, this speed will inevitably be relatively high.

The timing of the Rowell when leaned-out on glow-plug ignition with its 12:1 compression ratio was guaranteed to be well advanced regardless of operating speed. This would be OK and even desirable at high speeds, but any attempt to operate the engine leaned out at speeds significantly below its design operating range would inevitably produce pre-ignition and elevated internal stresses. The concept of operating a Rowell 60 leaned out at below 6,000 rpm under load on glow-plug makes the blood curdle...! In any event, why Sparey would expect an engine clearly designed for high speeds (by the standards of the day) to operate happily at such pedestrian speeds is quite beyond rational explanation. However, he repeated this error in test after test, never seeming to grasp the futility of testing engines at speeds far below those at which they were designed to operate. The "misfiring" to which he referred was most likely detonation!

Sparey completed his test of the Rowell 60 on spark ignition, finding a peak output of 0.732 BHP @ 13,900 rpm using straight methanol fuel with 30% castor oil. This was some way below the figures claimed by the makers and was also considerably less than later testing has implied for the Rowell. Sparey's testing procedures may have had a bearing on this issue. Even so, the reported numbers significantly topped the figures of 0.48 BHP @ 11,200 rpm which had been reported by Sparey in March 1949 for the rival Nordec RG 10 model on glow-plug ignition using a straight 75-25 methanol/castor fuel mix. Peter Chinn's figures of just over 0.6 BHP @ 12,000 rpm (Model Aircraft, June 1949) for the RG 10 were probably more representative of the Nordec's true capabilities, but even so there's no question that the Rowell far outperformed its English rival.

Sparey's subsequent tests of the Rowell using glow-plug ignition were commenced using standard Mercury glow-plug fuel. In this form the engine was found to deliver marginally less power at a given speed than it had on spark ignition, at least at the speeds tested. This is almost certainly a reflection of over-advanced ignition timing at the lower speeds on glow-plug. However, as speeds passed the point at which the peak on spark ignition was reached, the output on glow-plug was found to be still climbing, with no sign of a peak being reached. Sparey actually commented that the engine seemed "much more lively" on glow-plug ignition.

Unfortunately, Sparey was unable to report just how much more lively, because as speed reached 14,400 rpm with power still rising steadily, the mounting spindle for the rotary valve disc sheared at the point where it was stepped down to the 4BA installation thread, rendering the valve inoperative and stopping the engine. Little damage was sustained apart from some scuffing of the con-rod, but this put an end to the test. The engine had however matched the output achieved on spark ignition, albeit at a somewhat higher speed.

Interestingly enough, my own example of the Rowell 60 Mk I suffered an identical failure to that experienced by Sparey during initial testing, fortunately again without significant damage to the rest of the engine. An identical replacement was easily made. It's clear that there was a fundamental design weakness in the rotor mounting system adopted. Sparey suggested that the 4BA thread by which the rotor mounting spigot was secured could with advantage have been made 2BA. This advice actually makes a lot of sense. The use of a left hand thread would also have been a wise provision against the possibility of the spindle becoming loose and unscrewing during operation. I used medium-strength Loctite when installing my replacement spindle. I would advise other owners to follow suit.

A far later test of a Mk I Rowell 60 bearing the serial number 150 appeared in the June 2008 issue of "SAM Speaks". Tester Dick Roberts described the engine in some detail and put it though its paces on the bench. He reported some difficulty in achieving the specified .006" point clearance using the adjustment system described earlier. The rotation of the fixed point bracket by only a microscopic amount makes quite a difference to the gap, rendering the establishment of precisely the right gap somewhat finicky. The setting also tends to shift while tightening the clamp nut. Dick tried inserting a feeler gauge and then making the adjustment, but this did not work as the moving point spring was not strong enough! I have actually found my example to be easy enough to adjust, but I must admit that the very clever and effective point adjustment arrangement seen on the rival Ten-Sixty-Six Conqueror is far more practical in use.

Dick's initial test runs were undertaken on glow-plug ignition to facilitate the establishment of a needle setting. Using a 12x6 APC airscrew, the engine hand-started quite easily on a straight 80/20 methanol/oil mix and responded well to the needle. Further runs were undertaken using spark ignition and an electric starter (try hand-starting a large fixed-timing racing engine!). In this form, the engine ran even better, with improved needle response.

Dick found that running qualities were excellent, vibration in particular being notably absent. This was no doubt due in large part to the attention paid to the counterbalancing of the engine. Everything stayed tight—even the mounting bolts needed only a minor tweak after a few runs, and no leaks developed anywhere.

Dick tried the engine on two different props, recording 10,560 RPM on the 12x6 APC (equivalent to around 0.97 BHP), and 11,780 RPM on an 11x6 APC (around 0.96 BHP). Since the output at the higher speed is marginally less than that at the lower figure, it would appear that the engine tested by Dick peaked at a speed lying somewhere between these two figures, probably producing around 1 BHP at some 11,500 rpm. This in turn implies the development of considerably greater torque than that found by Sparey. In fact, these are remarkably good figures for 1948, beating the original Nordec by a mile and coming very close to the reported output of the 1946-48 black-case McCoy 60. However, that version of the McCoy was almost immediately replaced by the Series 20 version which produced some 50% more power at far higher revs. So in real terms, the contest between the McCoy and the Rowell was over before it had begun ... and when it came to the Dooling, the Rowell was never in it aside from its superior durability.

I've tried my own example on the bench using glow plug ignition, keeping it on the rich side for the most part while doing so as well as using a low-nitro fuel containing 30% castor oil. The revised prop driver with its knurled front face performed well, and I had no further prop troubles. I started out trying an 11x6 APC prop, but actually found that the engine seemed happier at the somewhat higher speed permitted through the use of a 10-1/2 x 6 APC prop. This prop seemed to suit the engine very well indeed, and I feel comfortable in recommending its use to any other Rowell owner who may wish to give his engine a run or two.

I can confirm Dick's statement that the Rowell is very easy to start by hand despite the rather daunting compression ratio, also running very smoothly. For hand-starting, I find that my usual big-engine "reverse-bounce" technique is by far the safest—set the prop at "ten past four" with the piston at bottom dead centre, administer a healthy prime and then HIT the prop backwards from the bottom dead centre position with the plug activated. The inevitable backfire almost always starts the engine in the correct direction. Using this method, the Rowell is a one-flick starter practically every time. The compression seal provided by the rings is outstanding.

Fuel draw is more than adequate for operation on suction feed—I'm at a loss to understand why Sparey needed to use gravity feed with its associated potential for crankcase flooding during starting. I can also endorse Dick Robert's statement that vibration levels are remarkably low, especially for such a large engine.

After some 20 minutes of rich running to settle things down, I did try leaning the engine out very briefly, using an extremely cold plug with an oily 10% nitro fuel. In this state, I actually did a little better than Dick, seeing 12,000 rpm on the 11x6 prop and 13,300 on the 10-1/2 x 6, with very smooth running in both cases. This is actually quite impressive—Hornet 60 no. V197 tested at the same session using the same fuel and props only managed 12,200 rpm and 13,600 rpm respectively. Based on established power absorption coefficients for the props in question, the implied power outputs for the Rowell at the two speeds achieved are 0.98 BHP and 1.09 BHP respectively. These figures certainly appear consistent with factory performance claims, although the implied output falls well short of the levels then being achieved by the Dooling 61 and the Series 20 McCoy 60 which had been introduced more or less concurrently with the Rowell. The "bulge bypass" Hornet update which appeared in 1949 also handsomely beat these figures, although it too was no match for the Dooling or McCoy.

Further Development of the Rowell Range

Although its performance fell well short of matching that of the best contemporary American products, the Rowell 60 was undoubtedly the most powerful British-made 10 cc racing engine on the market. Furthermore, it enjoyed the significant advantage that, being British-made, it was readily available in Britain. During the Rowell's first year or more of production, the importation of American engines by any legitimate means was impossible. Anyone wishing to obtain an example had to persuade one of the many US servicemen still stationed in Britain to "obtain" motors for them, or had to rely upon equally philanthropic people Stateside sending them over as "gifts". Wilf Rowell's previously-mentioned 1947 acquisition of a Hornet 60 as a "gift" from a relative in America is a case in point.

Due in large part to its relative availability, the Rowell 60 quickly achieved a position of considerable prominence in the British model car racing world. The April 1949 issue of Model Cars carried a front page picture of W. (Bill) Armstrong of Dundee with his Rowell-engined bevel-geared Demon. The associated write-up noted that the car had reached 86 mph despite the fact that the chassis had not yet been pared of all its superfluous weight. It is necessary to place this performance in context by noting that in July of 1948 Gerry Buck had made the first 100 mph run ever recorded in Britain with "Topsy", his homebuilt car and engine. Even so, the Rowell was achieving highly competitive speeds right out of the starting gate, even before the tuners got their teeth into the engine.

In 1949 the company published a 40-page booklet entitled "Miniature Race Cars, their Design and Construction" priced at 2 shillings (10 p).†In addition to providing detailed information about all aspects of building and running a tethered model car, the booklet showcased the extensive range of products being offered at this time by Rowell Motors Ltd. Oddly enough, the cars illustrated included the Hellcat and Bill Armstrong's Demon which it was claimed were designed and built by Rowell Motors despite the fact that they both incorporated a significant number of components actually manufactured by Z.N. Motors of Willsden in London.

An interesting statement appeared in the introduction. This read: "... in order to provide high speed, we are compelled to offer designs and information for building cars on the American lines". More or less an open admission of the superiority of the very functional American approach as well as an apology for riding on the coat-tails of the Americans rather than ploughing their own furrow!

Having said this, they did include a plan for the Rowell Rapier, which was quite unlike any other car being produced at the time—definitely a Rowell original! The Rapier deviated from common practice in almost every aspect. Instead of a conventional chassis or pan, it had solid aluminium plate side members that tapered sharply towards the nose, being joined at that point by an aluminium casting that carried the front axle mounting.†Further bracing was supplied by the engine, crank support, gearbox and tank, all of which were bolted between the side plates and thus functioned as part of the structure. The body was a complex two-piece aluminium stamping. The design was expressly intended to be suitable for home construction.

All the items on offer were branded as Rowell "products", even though most or all of them were in reality produced by others. The list included the Rowell 60 Mk 1 engine with spark ignition at a slightly reduced price of £12, or £11 10s 0d (£11.50) in glow plug configuration. The complete absence of known survivors of the latter model implies that few if any were in fact sold. Also available were an American pattern bevel gearbox, spur gear drive unit, centrifugal clutch, "Hot Spark" ignition coil, direct-drive flywheels and gears and a Rowell-Barnard 4 volt accumulator. The Rowell wheels and tyres were offered in both driving and knife-edge configuration, being almost identical to the American Dooling designs. A range of parts for the Rapier car completed the list.

As the most powerful commercial British-made engine at the time, interest in the Rowell 60 was by no means confined to the engine testers of the day. Edgar T. Westbury commented upon it in an article, and the Rowell also appeared in both Col. C. E. Bowden's 1949 book "Model Glow Plug Engines" and the 1949 "Model Car Manual" by Model Cars Editor G. H. Deason. The "MCN Grand Prix Special" was designed around the Rowell, while the write-up for the "Stubbs Austin"' published in Deason's "Model Car Manual" suggested the motor as being suitable for high performance applications. In addition, G. W. Arthur-Brand joined Lawrence H. Sparey in putting it under the microscope in a published bench-test report. Minor problems were identified, but overall the Rowell 60 received very favorable reviews which doubtless provided very good publicity.

From the above comments, it may be seen that the Rowell 60 stood up well to a close examination in isolation. However, the Rowell was not competing against itself but rather against other engine designs. In R. H. Warring's 1949 book "Speed Control Line Models", the author compared the Rowell with the American Hornet, McCoy and Dooling 10 cc models as well as the British Nordec. In the glare of this comparison, the Rowell did not fare so well.

Warring's data showed the Rowell to be some 3 oz. heavier than the Hornet or Dooling and almost 4 oz. heavier than the McCoy and Nordec. This was bad enough, but when the power outputs were considered an even wider gulf was revealed. While the Rowell handily outperformed the original Nordec, the leading American motors showed themselves to be up to 50% more powerful. The fact that the very heavy Rowell was credited with an output of just 1 BHP compared with the contemporary Dooling's 1.5 BHP (matched by the McCoy 60 Series 20) simply could not be ignored by anyone seeking the ultimate in performance.

Warring's comparison showed that a great deal of development was still required if the Rowell was to meet its objective of providing meaningful competition to the Dooling and McCoy offerings. Although Wilf Rowell disputed some points in Sparey's Model Cars test of his engine, it's clear that he took the underlying message to heart. This was that the latest American models performed at significantly higher levels than the Mk I version of the Rowell 60, making further development essential if Rowell wished to remain in the game.

The direct consequence of this realization was the early 1950 announcement that a new Rowell 60 Mk II engine would be ready in March at the significantly reduced price of £9 17s 6d (£9.88). This revised model duly appeared, featuring a lighter crankcase having an enlarged bypass passage which was both wider and deeper in section. The revised case also sported a rounded exhaust stack as opposed to the square section stack featured on the Mk I. The other immediately obvious alteration was a new die-cast back-plate featuring both a down-draught venturi set at around 35 degrees from the horizontal and a longer disc bearing. The use of die-casting for this component as well as for the piston implied that more advanced foundry techniques or even a different producer were now involved.

Internally, the revised model used a die-cast piston of improved design along with changes to the head and cylinder liner. These revisions were claimed by the manufacturer to have raised power output to 1.25 BHP, more or less matching the output of the much-improved Nordec Special Series 2 which appeared in prototype form at more or less the same time but never seems to have made it into series production. These figures of course still fell well short of the measured performances achieved by the contemporary McCoy 60 Series 20 and Dooling 61 models. However, if this claim was accurate, the Mk II undeniably represented a real step forward. Moreover, the revised components could be retro-fitted to existing Mk I units, creating what might be termed a Mk 1.5!

So far so good, but the development of the Rowell 60 did not stop there. It appears that the original Mk II version of the engine remained in production for only a few months, since the June 1950 issue of "Model Engineer" carried an article written by Wilf Rowell on the subject of ignition systems which featured a new Rowell car design, the Sabre. This design differed markedly from that of the earlier Rapier, exhibiting very clear American design influence. It featured a cast pan and general layout which was similar in most respects to the extremely successful Dooling Arrow.

More significantly from the standpoint of our present interest, the engine in this car appeared to be the final version of the Rowell 60, the Mk II Series 2. Let's examine that model next.

Final Fling—the Rowell 60 Mk II Series 2

The Rowell 60 Mk II Series 2 illustrated here is a little-used and unmodified example which is missing its timer. The engine has clearly spent most or all of its seemingly short running life in glow-plug mode—while the combustion chamber and piston crown bear clear evidence of at least some running, the cam area of the crankshaft bears no indication of the residual effects of cam follower rubbing, nor does the exterior of the front bearing housing show any of the usual scuff marks resulting from timer rotation. It's possible that the engine was originally completed in its present form—we simply don't know.

As received, the engine lacked any prop mounting components. The set-up seen in the images consists of my own replicas of the components used in my example of the Mk I model. In addition, at some point in its life the lower edge of the backplate mounting flange had sustained damage of some kind. This had been very competently restored by welding, but the owner had never got around to cleaning up the repair. I completed this work successfully, to the point that the repair is now quite unobtrusive.

Somewhat unusually, the engine displays no serial number. Such an omission is normally due to dressing of the outer ends of the mounting lugs, but this example shows no evidence of this, actually retaining traces of the original milling marks on the ends of the lugs. The surfaces appear to have been smoothed with a piece of emery cloth, but nothing more.

Given the above observations, Hugh Blowers and I are in agreement that this example of the engine was most likely assembled after series production had ceased. It's known that considerable quantities of finished parts remained on hand when the company ended its manufacturing activities, and it seems likely that this engine was built up later from a set of left-over components, thus explaining the absence of a factory serial number. Other un-numbered examples of the Rowell 60 have been reported, raising the possibility that a number of engines were assembled after the fact in this manner.

If this is indeed how this engine came to be constructed, it's entirely possible that no timers remained on hand. Alternatively, the intention from the outset may have been to operate the engine in glow-plug mode. Either way, the absence of the timer would be explained. Another possibility associated with this scenario is that the only remaining Series 2 backplate casting was one which was flawed during the casting process and had been rejected. This would explain the need for the weld repair to the backplate, which is otherwise difficult to account for—how do you break a backplate mounting flange?

Turning now to the engine's design features, the only externally-visible difference that distinguishes the Series 2 from the original Mk II design (now referred to as the Mk II Series 1) is the even more steeply-angled venturi that required a new sand-cast back-plate with a curved intake stub. The main crankcase casting with its enlarged bypass passage (by comparison with that used in the Mk I) appears to be identical to that seen in the Mk II Series 1.

The revised model was if anything even heavier than the original Mk I variant. Weight of my own example as illustrated minus timer is 522 gm (18.41 ounces). The addition of the quite substantial standard Rowell timer would undoubtedly have raised this at least to the 546 gm (19.25 ounces) of the Mk I, and probably even a little more. The heavier backplate and oversized induction arrangements probably account for much of the apparent increase.

Apart from the revised crankcase and backplate castings, an examination of my own example of the Mk II Series 2 model during the course of a precautionary rebuild has revealed a number of significant departures from the design of the Mk I model. Beginning at the venturi intake, the throat of this component now has a monumental diameter of 0.410 inches. This would logically be expected to impose a requirement for the provision of either pressure or gravity fuel feed. The fact that this diameter is more or less identical to that used in the trend-setting McCoy 60 Series 20 is probably not coincidental ...

It might be asked at this point why the change was made to the more steeply angled intake venturi. Could this be a simple matter of slavishly copying the design of the all-conquering Dooling 61 which was then sweeping all before it? Well, Dooling influence may have played its part, but the fact is that the revised design permitted several distinct improvements to be made to the induction system. The most important of these was the fact that space considerations made it impossible to accommodate a near-horizontal or shallow-angled intake venturi of the size now proposed alongside the extended disc valve bearing. The upward angling of the venturi kept it completely clear of the disc valve bearing despite the fact that in plan view it encroached considerably into the space occupied by that bearing. The above illustration should make this clear.

The second advantage was that the revised intake orientation permitted the intake register in the backplate both to be made significantly larger and located closer to the base of the backplate, thus making more efficient use of the swirl effect within the crankcase to promote the smooth transition of incoming mixture from the intake to the bypass system.

The disc valve mounting set-up is completely different. Instead of the former screw-in fixed spindle with lock-nut (which had proved somewhat less than completely reliable), the cast aluminium alloy disc valve was now securely fixed onto a long steel spindle which ran in an extended bearing formed in the centre of the revised backplate casting. This feature appears to have been carried over from the Mk II Series 1 variant. The disc and spindle on my example are retained using a soldered-on washer, exactly as in the case of the Hornet. However, other examples feature a nut and washer to serve this function. This provision is actually somewhat redundant in any case since crankcase pressure keeps the disc in contact with its backplate during operation, with no tendency for the disc to move forward under these conditions.

In addition to this change, the intake timing was drastically altered, being far less conservative. The disc valve in my example opens at 30 degrees ABDC and closes at 60 degrees ATDC for a total induction period of 210 degrees. This is definitely a very high speed timing set-up—blow-back at low and even moderate speeds will obviously be pretty chronic. In fact, given the engine's greatly improved induction capacity thanks to its far larger venturi and backplate induction port, I would objectively question the need to delay the closure of the system to this extent. I would have been tempted to close the intake considerably earlier, thus gaining the benefit of increased crankcase pressure prior to the opening of the transfer ports. That said, there is no evidence whatsoever that this example has been modified in any way—all of the as-cast surfaces on both the rotary disc and the backplate remain undressed. Hence the timing appears to be as established during manufacture.

Inside the case itself, an immediately obvious change is the elimination of the alloy intrusions into the crankcase cavity which were created in the Mk I model by terminating the installation bore for the cylinder liner at the base of the liner. This change was achieved very simply by carrying the vertical bore for the cylinder liner all the way down into the axial main crankcase bore. It's clear that the designer's primary concern had become to minimize any obstruction to the free passage of incoming mixture from the intake to the piston ports. An increase in crankcase volume was obviously seen as an acceptable price to pay for the achievement of this goal.

The reason for this switch in priorities becomes immediately obvious when we look at the piston ports themselves and their corresponding cylinder wall ports—the engine's bypass capacity has been greatly enhanced. The manner in which the piston skirt ports registered with their counterpart openings in the lower cylinder wall represented a significant performance constraint in the original Mk I design. The design was such that the two sets of openings were only in perfect alignment (and thus fully open) at bottom dead centre. This corresponded to the point of maximum volumetric crankcase displacement by the descending piston, since it could of course descend no further. Ideally, with a free-flowing bypass/transfer system, a significant proportion of the transfer gas movement should already have taken place at this point.

Such considerations lead to the obvious conclusion that if the piston skirt ports are to be fully effective they should be completely open well prior to bottom dead centre. In terms of area, the bypass entry system should remain well ahead of the transfer ports at all times during the transfer cycle. To address this deficiency, the revised design featured a pair of cylinder ports which were extended upward, maintaining an oval "race-track" form with no sharp edges to create stress concentrations. At the same time, the corresponding piston skirt ports were extended downward, again retaining a matching oval form. The result was that when the transfer ports began to open (as shown in the attached illustration), the piston port system was already wide open and only got wider as the piston continued to descend. At all stages of the cycle, overall bypass entry area was substantially greater than the total available area of the transfer openings, thus presenting no undue constraint upon gas movement from the case to the cylinder. At first sight, a very rational and worthwhile improvement, which was actually a not-uncommon modification applied by knowledgeable users of the Hornet and McCoy 60 models.

Obviously, this modification combined with the greatly enlarged Mk II bypass passage and improved induction capacity significantly enhanced the ability of the bypass system to supply mixture to the transfer ports. In recognition of this, the number of transfer openings was increased from three to four. The far wider Mk II bypass passage could easily accommodate this. It seems probable that the earlier Mk II Series 1 model also featured four transfer openings.

Moving on upwards, both the piston crown and the cylinder head were substantially changed. The crown of the die-cast piston now featured a baffle having a wedge configuration in plan view, with the interior of the wedge facing the transfer ports. The idea was clearly to promote a more focused movement of transfer gas towards the upper cylinder. Interestingly, the top surface of the piston crown was lightly ribbed, presumably to gain greater strength.

The head was of course reconfigured to match the revised piston crown contour. As part of this reconfiguration, the engine's compression ratio was significantly reduced. My example of the Mk II Series 2 has a measured compression ratio of only 8 to 1 as opposed to the confirmed 12 to 1 of the original model.

I suspect that this was the designer's reaction to the fact that by mid 1950 the general switch from spark ignition to glow-plug operation was well underway, even among users of large racing engines. Indeed, as noted earlier, my own example of the Mk II Series 2 has apparently been used exclusively as a glow-plug motor. An 8 to 1 compression ratio is far more suitable for such operation, especially if significant amounts of nitro are to be employed.

The exhaust porting was unchanged—indeed, the timing of the cylinder ports was essentially identical to that used formerly. In addition, the bore and stroke remained unaltered, as did the crankshaft and main bearing assembly. However, it will hopefully have been made clear by now that the Mk II Series 2 version of the Rowell 60 represented far more than a tune-up of the Mk I model—it was in fact a completely different engine. Looking at the changes objectively in the light of long practical experience with classic racing engines, I would expect a significantly enhanced performance from this model at the expense of some of the earlier design's fine handling characteristics and mid-range power.

Since I do not have access to a Mk I Series 1 example, I'm unable to determine how many of these changes were carried over from the Mk II Series 1 as opposed to being introduced in the Mk II Series 2 model. I suspect that the four transfer openings were common to both the Series 1 and Series 2 variants, as was the revised means of mounting the rear disc valve. However, given the fact that the crankcase casting also appears to be the same, the magnitude of the claimed performance increase for the Mk II Series 2 over the Series 1 is sufficiently large that many of the other observed changes must surely have been introduced in that model. Rowell claimed that this version of the engine produced no less than 1.75 BHP at 18,000 rpm—way up there in Dooling and McCoy territory if true! This is a claim that merits testing, as we shall see in the following section of this article.

At this stage at least, the engines were supplied in sturdy cardboard boxes with a red label bearing white lettering and a picture of the Rowell 60 engine. Apart from the engines themselves, the company continued to offer a full range of accessories for model car racing. The somewhat unique Rapier car remained available as a complete set of parts, or built to order for £29 without an engine.†The far more conventional Sabre with its clear American heritage also continued to be offered.

The Rowell 60 Mk II Series 2 on Test

Given the claimed performance improvement for the revised design, it was clear to me that a test of this model would be of immense interest. It must be said that the results achieved in the field with this unit do not bear out the makers' claim when compared with those of the Dooling and McCoy. Even a fast Hornet could seemingly give the updated Rowell a run for its money. The Mk II Series 2 variant was never the subject of a published test in the contemporary modelling media, leaving it up to later researchers to test the manufacturer's claims.

I was actually not all that keen on putting my own example of this very rare engine though the testing wringer, but decided that if I didn't conduct a test, no-one else would! Moreover, the engine appeared to be in first-class mechanical condition, while the design of the Rowell is about as sturdy as one could hope for in a racing engine. So there was really no reason to believe that the engine would not stand up to a fairly serious test.

Accordingly, having gathered my usual set of classic 10 cc racing engine test props (and my industrial-grade ear muffs!), I headed out to the flying field with the intention of carrying out the minimum amount of testing necessary to establish some kind of performance bench-mark for this model. I took along the Mk I which had been tested previously so that a direct same-day comparison could be undertaken. The fuel used was a commercial 10% nitro brew which had been brought up to an oil content of 28% by the addition of some extra castor oil.

As it turned out, I needn't have worried! Contrary to my initial doubts regarding its handling qualities, the engine proved to be just as easy to start as its Mk I sibling, merely needing a healthy prime followed by my usual reverse-bounce starting technique described earlier. Not only that, but it also confounded my expectations by turning out to be perfectly happy running on suction feed! I actually flooded it quite badly during my initial attempt using a gravity fuel system which was a challenge to manage. Things went very much better when I set the tank with the fuel surface level with the jet during starting. The engine picked up right away in this configuration. Needle response was outstanding without being in the least bit critical, making it very easy indeed to optimize needle settings.

As expected, blow-back was pretty chronic at the lower speeds. It was quite entertaining to look into the venturi while the engine was running rich on the 11x6 prop—the fuel splashed around in there like nobody's business, with as much spraying out of the venturi as going into the case! The rate at which the level in the tank dropped certainly reflected this—I'm glad that I don't have to pay this engine's fuel bills on a regular basis!

Running was dead smooth on all props tested, despite the blow-back at lower speeds. Even when leaned out all the way, there was no trace of misfiring or sagging at any time. As expected, the motor was far happier on the faster props—by the time the speed was pushed up over 14,000 rpm, the blow-back had been reduced to relatively minimal proportions, although it was still there until the speed was pushed into the 15,000 rpm range. The induction timing of this model is clearly set for speeds in the region of 16,000 rpm or higher. The fastest speed that I achieved was 16,100 rpm on the 10x4 APC, with the engine running as sweetly as one could wish. There's no doubt at all that the Rowell would run very well at even higher speeds than this, but I wasn't prepared to go any further with this very rare unit.

Thankfully, the engine came through this fairly strenuous test with flying colours. A subsequent check showed that things had stayed together very well indeed, with no signs whatsoever of any undue stress on the components. Compression seal remained outstanding, as did all bearing fits. I'd have to say that these engines come as close to being bullet-proof as any large racing units of my past and present acquaintance.

Next up was Mk I Rowell 60 no. 116 which had been tested earlier. It was its usual good-mannered self, starting very easily and running flawlessly. It more or less matched its previous performance on the two props which were common to both tests. For comparison purposes, I also tried it on two of the additional props which I had used with the Mk II Series 2. The results obtained were quite illuminating, as shown in the following table:

PropRowell 60 Mk I no. 116BHPRowell 60 Mk II Series. 2BHP
APC 11x611,9000.9610,8000.72
APC 10x7 12,9001.0512,7001.00
APC 10-1/2x613,2001.0713,3001.09
APC 10x613,8000.9714,8001.20
APC 10x4Not tested-16,1001.21

Interesting! The Mk II Series 2 is clearly an inferior torque producer at lower speeds, falling well short of matching the Mk I version. We would of course expect this given the induction timing. It actually doesn't catch up with the Mk I until around 13,200 rpm, but then starts to pull away to a healthy lead.

Although there's insufficient data here upon which to base a truly representative power curve, it's clear that the Mk I peaks somewhere in the vicinity of 13,300 rpm, at which it is producing a quite respectable 1.08 BHP or thereabouts. Given that it had clearly already run past its peak on the 10x6, I elected not to expose the engine to undue stress by trying the 10x4—no point in pushing these old classics past their inherent performance limits.

By contrast, the Mk II Series 2 is just getting into its stride at the peaking speed of the Mk I. In fact, power is still rising steadily as 15,000 rpm is approached. Although the curve is based upon a rather small data sample, the above numbers clearly imply a peak output of around 1.23 BHP at somewhere in the neighborhood of 15,800 rpm.

Two comments may be offered on the basis of these tests. Firstly, the above numbers unquestionably constitute a ringing endorsement of Wilf Rowell's claimed output of around 1 BHP for the Mk I Rowell 60. An output of 1.08 BHP at 13,300 rpm on 10% nitro is right on par with reported figures for the 1946-48 blackcase McCoy 60 and actually beats the claimed performance of the Hornet 60 in standard unmodified form. If Wilf Rowell's goal was to match the American models which were current when the Rowell 60 was being developed, it appears that he succeeded (if we set aside the Dooling 61). He deserves great credit for this. It was his misfortune that the Dooling 61 upset the apple cart by entering the mix when it did and that Dick McCoy's reaction to the Dooling in the form of the McCoy 60 Series 20 was both immediate and effective.

On a less positive note, the above figures do not provide any support for the claim of 1.75 BHP at 18,000 rpm for the Mk II Series 2 model. There's no question that the Mk II Series 2 is a more powerful engine than the Mk I, but on this showing it does not appear capable of reaching such heights. Admittedly, the 8 to 1 compression ratio suggests that a high nitro fuel might yield some very positive results. During their early experiments with the use of nitromethane, the Dooling brothers reported a power increase of over 30% for the Dooling 61 when operated on a fuel containing 37% nitro as opposed to a straight methanol/oil blend. If the same kind of ratio applied to the Rowell, we might expect to see results far closer to the manufacturer's claims. But then, how many modellers in early 1950's Britain could afford to feed large doses of nitro to such a thirsty engine as this?

I have to confess that I'm not altogether surprised at this finding. The contest results achieved by the Mk II Rowell 60 in either of its two guises certainly do not bear out the maker's performance claims. In particular, if the Mk II Series 2 had in fact developed anything even close to the claimed output, we would unquestionably have seen far more of it at the top of the contemporary results sheets. Even the flyboys would have sat up and taken notice of an engine having that level of performance, regardless of its weight.

Still, the Mk II Series 2 is a very stout performer in its own right, and one which seems sturdy enough to deliver its best performance over an extended working life. It's also a delightful engine to operate if you can deal with the noise levels and pay the fuel bills! I would still rate it as a very successful design on the part of Wilf Rowell. It's certainly by far the most powerful British-built commercial 10 cc racing engine from the "classic" era.

Gathering Clouds

Despite the reduction in the price of the Rowell 60 to go along with its improved performance, it can't be denied that model car racing at this level was an extremely expensive business by 1950 standards. By the time a customer could bring a competitive 10 cc car to the starting line, his investment in that single model might well be on the wrong side of £40, a small fortune in those days, being equivalent to at least three thousand dollars in today's money, over 60 years on. In effect, the pursuit of model car racing at a competitive level in the 10 cc class was becoming an activity open only to those having the right combination of money and technical ability. Naturally, such individuals were very much in a minority.

An exchange of correspondence which took place in mid 1950 serves to highlight the corner into which this branch of the modelling hobby was increasingly being driven. In a letter which appeared in the July 1950 issue of Model Cars, a reader complained about the high cost of components putting the building of a competitive car out of reach for the average enthusiast. This prompted what may be seen in hindsight as a prophetic response from Wilf Rowell. His main thesis was the fact that the relatively small demand for such components precluded their mass production and thus made high prices inevitable. Among other things, he wrote: "Demand for racing engines and other components is relatively small in this country, amounting to a few hundred people. The model car trade could not commit itself to large-scale production, although Rowell's had produced a batch of wheels approaching four figures and we are going to have to wait some considerable time before these are disposed of". He could have had no idea just how long and what form that disposal was actually going to take ...

Wilf Rowell's discouraging estimate regarding the potential numbers of buyers for his engine and accessories proved to be well founded. It was noteworthy that from 1950 onwards the company's advertising campaign progressively lost steam, presumably due to a growing and sadly accurate perception that the market was in retreat. As a result, the further development of the Rowell 60 described earlier was very much under-promoted. This may be an indication that all was not well commercially at Rowell Motors Ltd. even at this stage.

Despite this, a perusal of the race reports from 1950 shows that several prominent competitors adopted and raced Rowell Sabres, including Joe Riding and Gerry Buck. Gerry made the trip up to Dundee in August 1950 both to attend the Scottish Speed Championship meeting and to take delivery of his own Sabre fitted with the newly-introduced Mk II Series 2 engine. Out of 16 entries at this meeting, the fastest Rowell was fifth at 91.9mph, run by Bill Armstrong. The best performance by a Rowell at the following year's Scottish Speed Championship was put up by J Soutar, who came fourth at 88.94 mph running a Rowell-powered Sabre.

Following its introduction in mid-1950, the Sabre underwent some degree of further development. An alterative version of the car was created by giving the pan the "bobtail treatment". This involved sawing off the rear end and tether brackets and fitting a panhandle. A longer and more streamlined body was carved to suit the new shape. It is not known if new patterns were made for this version or indeed if it was ever marketed commercially. However, in the 1952 European Championships at Monza, Italy, Carlo Davario put in a run of 156 kph (96.94 mph) with a Rowell car that looked almost identical to the "bobtail" Sabre. A Rowell was also seen in Italian domestic events in 1953, run by Franco Rota.

All of the above-noted technical developments took place in a relatively short space of time during 1950. However, further development appears to have stagnated fairly soon after the mid-1950 introduction of the Mk II Series 2 engine and the Sabre car. Indeed, it was becoming ever more obvious that Rowell Motors Ltd. was entering a challenging period. The advertisements published in 1951 spoke of the company "...having difficulties in production owing to material and labour shortages" and of "... a difficult period (lying) ahead for us all". How true that proved to be ...

The End of the Road

As the 1950's got underway, the Rowell 60 remained in fairly widespread use. A few examples of the engine were used in model aircraft and tethered hydroplanes, but most found their way into tethered cars, exactly as their designer intended. Joe Riding of Bolton used one of these motors in his cars for much of his competitive career, becoming highly adept at tuning them for greater performance. Gerry Buck also continued to successfully campaign his Rowell Sabre car. However, he too was unable to beat the Doolings or McCoys in the Open class.

Joe Riding was unquestionably the most successful Rowell-powered competitor of them all, being the first Rowell user to break 100 mph and still to this day holding the record for the fastest tethered car speed ever recorded with a commercial British engine at 115.83mph, set in 1952 with his tuned Rowell. In an article published in Model Maker which included some tuning tips for the Rowell, Joe revealed that the engine used to set this record was a Mk II model which used a Mk 1 head and backplate together with other listed tuning measures, all of which combined to improve performance substantially. Even so, the engine used by Joe was no match for the best of the imports.

A similar situation existed in the tethered hydroplane world. Rowell supported the MPBA International Regatta in 1950, but the engines were not a great success in this discipline. Jim Hampton ran a hydro which was fitted with a Mk 1 Rowell engine that Jim claimed to be very reliable and capable of consistently producing runs of around 40 mph. However, this was very far from competitive by 1950 standards.

George Stone, who already held the outright British hydroplane record with his Dooling-engined "Lady Babs II" at 70.1 mph, did somewhat better than this. In the face of heavy criticism from other competitors for allegedly "buying" his success through his acquisition of the American-made Dooling, he fitted a Rowell into his hull "Rodney" in 1950 and issued his "try to beat me with a British motor" challenge. So equipped, "Rodney" achieved a speed of 54.74 mph, backing up Stone's challenge by becoming the fastest British-engined hydroplane at that time, albeit well short of matching the speeds achieved with the Dooling.

The ongoing failure of recognized experts to come close to the performances put up by the imports made it increasingly clear that the Rowell was simply unable to match the best of the American products despite its many excellent qualities. It was doubtless a combination of this factor together with prevailing economic conditions and the waning interest in model car racing that heralded the end for Rowell Motors Ltd. Like so many of his contemporaries, Wilf Rowell found that the production of engines, models and accessories for a very small and apparently shrinking specialist market was simply not commercially viable in early post-war Britain. Rowell Motors had virtually ceased trading by the end of 1952, with the final advertisements appearing early in 1953. By the time of the last advertisement, the company was offering components at less than cost price, indicating that production had already ceased and they were simply attempting to recover capital by selling off existing New Old Stock.

Based upon some comments in a surviving letter written by Wilf Rowell to Gerry Buck during the early years of the Rowell venture, it is a reasonable inference that Rowell Motors was not the only business activity with which Wilf was involved throughout this period. Prudently, he seems to have kept his powder dry by retaining some connection with his former line of business. Following the demise of Rowell Motors Ltd. in early 1953, he moved away from Dundee and continued working with industrial cranes, eventually setting up his own business on the Clyde. Here he remained until his death in November 1980 at the somewhat premature age of 65.

Brian Sherriff's shop continued to trade from the premises at 93 Victoria Road until the mid 1960's, when re-development forced a relocation of the shop to Cowgate in Dundee. Brian's son, also named Brian, had the job of clearing out the shop and cellar, which still contained boxes of Rowell parts. These met their fate by being consigned to a skip. A visitor to the Cowgate shop in the early 1990's reported that his "...jaw dropped on being told that a large box of Rowell tyres had only just been thrown out". Remember Wilf Rowell's 1950 comment about the disposal of the tyres ...?

The demise of Rowell Motors preceded that of the sport of tethered model car racing by only a few years. By 1955 the sport had virtually died in Britain, having enjoyed a heyday of only some 12 years or so. At one point there had been over 50 manufacturers supplying engines and accessories for this branch of the modelling hobby, but as of 1955 the vast majority of these had disappeared, including the only two companies producing both cars and engines for the 10 cc class, Ten-Sixty-Six Products of Worcester and Rowell Motors Ltd. Even the producers of the Nordec racing engines, the North Downs Engineering Co. of Whytleafe in Surrey, had abandoned model engine production well before this time and returned to their automotive roots. There were thus no commercial manufacturers left in Britain producing racing engines in the 10 cc class, nor as things turned out were there to be any in the future.

So the sport of I/C powered model car racing withered and finally died as a mainstream activity, not to return for three decades or so. When it did reappear, it was in a very different guise—R/C cars powered by soul-less new-era Schnuerle-ported glow-plug engines, generally made in China using CNC machinery as opposed to hands-on craftsmanship. Most of the cars now arrived in boxes, ready to run—there were no more craftsman-modellers, merely individuals who wished to race ready-built R/C cars. Little opportunity there for a revived version of Rowell Motors!

Control line speed flying continued to draw adherents however, and both control-line competitors and the small but enthusiastic group of tethered hydroplane aficionados who still soldiered on had no problem using engines from other manufacturers which were primarily designed for model aircraft use. But as events proved, the Rowell 60 was both the first and last of the purpose-built commercial British 10 cc racing engines.

Production Figures and Serial Numbers

Based upon the sample of engines reported so far, either with or without serial numbers, Lynn Blowers considers it probable that at most some 400 Rowell engines of all types were manufactured in total during the 4-1/2 year life of the company. Indeed, it may well have been somewhat less than that. By contrast, the contemporary Nordec 60 unquestionably sold in excess of 1100 units over a significantly shorter period of time, but that engine was designed by an aeromodeller (John Wood) and was always promoted more for model aircraft use than for car or hydroplane applications. It thus enjoyed a substantially broader customer base. These factors certainly explain the relative rarity of the Rowell compared with the Nordec.

The issue of serial numbers for the Rowell engines is somewhat confusing and remains open to further research. The engine numbers recorded up to this point by Lynn Blowers imply that the numbers for the Mk I version started at 100, with 190 being the highest number recorded so far. My own illustrated example is early in this series, bearing the number 116. Numbers for the Mk II Series 1 engines seem to have started at engine number 600, while the Mk II Series 2 model which followed shortly afterwards unaccountably appears to have started at number 500! Anyone having information on Rowell serial numbers is requested to contact Lynn through the previously-cited website.

Conclusion

We hope you've enjoyed this in-depth look at one of the rarer and more exclusive model racing engines of them all. The Rowell 60 was (and is!) a fine engine and a great credit to its designer and manufacturer. Original examples in complete and unmodified condition turn up very seldom these days, and when they do they command some pretty fancy prices. Accordingly, the laws of supply and demand dictate that most enthusiasts will never own one—those of us who do may count ourselves fortunate indeed. However, we can all enjoy contemplating one of the more sincere efforts on the part of a British company to produce a racing engine of real merit.

One gleam of light at the end of this particular tunnel came in 2012 from Ian Russell of Rustler replica engines fame. At that time, Ian was taking expressions of interest in a proposed replica of the final fully developed Mk II Series 2 version of the Rowell 60. My own expression of interest went in right away—if such a replica had been produced to Ian's usual standard, it would have been a very fine product indeed! However, it appears that the proposal did not attract sufficient interest to justify its implementation. A great pity ... a sad case of lost opportunity.

Our thanks once again to the individuals acknowledged at the outset who have contributed so much towards the preservation of this fascinating piece of British model engine history.