In 1957, Enever, O’Neill, and draftsman Peter Neal began developing concepts for a new MGA-based coupe. Apparently unimpressed with these early efforts, BMC managing director George Harriman commissioned Turin’s Carrozzeria Frua to develop an alternative concept, also based on the chassis of the MGA. The Frua concept was quite striking despite a heavy-handed grille/bumper treatment, but it found little favor in Abingdon. According to Hayter, the main objections to the Frua car concerned its size and weight, which would have translated into high production costs, but David Knowles speculates that MG engineers’ disdain for the Frua concept had as much to do with its origins as its practical merits. Only three years earlier, John Thornley had finally persuaded BMC to transfer MG design from Cowley back to Abingdon and no one at MG was eager to undermine that victory by endorsing an out-of-house design.
O’Neill and Hayter subsequently developed a new in-house concept, the EX205, again based on concepts suggested by Enever. This progressed as far as the full-size mock-up stage, but even O’Neill didn’t like the awkwardly proportioned results. The EX205 was rejected.
By the time the EX205 mock-up was complete, Abingdon had decided that the MGA’s successor should abandon the platform frame in favor of monocoque construction. Ironically, Thornley had argued against unit bodies back in 1955, concerned about their high production costs, but he subsequently did a volte-face, in part because unitized construction offered both greater strength and superior space efficiency.
Since there was no longer any need to maintain the MGA’s chassis hardpoints, Enever gave Hayter permission to create an entirely new design with a new type number, EX214, taking as its starting point the EX181, a highly aerodynamic one-off that he had recently developed for record-breaking. In August 1957, Stirling Moss drove the EX181, powered by an experimental supercharged DOHC engine, to five International Class F speed records on the Bonneville Salt Flats, reaching speeds of more than 245 mph (395 km/h).
Hayter’s EX214 design, completed in June 1958, bore only the loosest resemblance to the EX181, but it was low-slung and very sleek, wearing a lower, wider modern interpretation of the traditional MG grille. Although the EX214 was a roadster, not a coupe, Thornley liked the design and approved it for production. Thornley still hoped to add a fixed-roof coupe, but he decided that in the short term, the roadster was a stronger bet for the crucial American market.
(Historian Wilson McComb suggests that some of the EX214’s final detailing was massaged by Pinin Farina, whom BMC had engaged to design its bread-and-butter sedans, but Farina’s contributions to the MGB roadster, if any, were apparently minor. In later interviews, Don Hayter indicated that his final EX214/12 rendering, completed in early 1959, was very close to the production MGB.)
V4 ENGINES AND FOUR-LINK SUSPENSIONS
While the exterior styling of the new roadster, now known by its BMC project number of ADO23, was largely finished by early 1959, its mechanical specifications were still in flux. Both Abingdon and BMC management hoped to make the ADO23 a much more sophisticated package than the MGA, although very few of those ideas ultimately came to fruition.
BMC’s plans revolved around an entirely new family of V4 engines. Inspired by contemporary Lancia practice, the V4 had a very narrow bank angle of only 18 degrees, making the engine quite compact. Valve actuation was still by pushrods, but the camshaft was driven by a then-novel rubber timing belt, as in the later Glas and Pontiac OHC engines. BMC envisaged two four-cylinder versions, one of about 1,100 cc (67 cu. in.), replacing the smaller A-series engine, the other about 2,000 cc (122 cu. in.), replacing the B-series; there were also tentative plans for a derivative V6. The V4 was a corporate project, developed at Longbridge rather than Abingdon, and Syd Enever was never particularly happy with it. V4 engines were tested in MGA mules for possible use in the ADO23, but BMC management finally shelved the project in 1960, concluding that it involved too much unknown territory for a mass-market engine.
Abingdon’s ambitions for the ADO23 involved updating the MGA’s somewhat rustic suspension. The A’s front suspension was still basically that of the 1947 Y-type sedan (itself based on a Morris design by Alec Issigonis), with old-style lever-action shock absorbers and Hotchkiss drive: a simple live axle on leaf springs. Few contemporary drivers seriously criticized the the A’s handling, but its ride comfort was another matter. Since the rear springs had sole responsibility for axle location, they were quite stiff. That was fine on a smooth racetrack, but less so on real-world roads, where the unyielding suspension and high unsprung weight made for a jarring ride. There was substantial room for improvement.
Enever and chassis engineer Terry Mitchell explored a variety of alternative suspension layouts in hopes of smoothing out the ADO23’s ride without sacrificing handling. The ideal solution would have been to reduce the rear unsprung weight with either independent rear suspension or a De Dion axle like that of the EX181, but cost considerations precluded either. As a compromise, Roy Brocklehurst, who had designed the original MGA chassis, came up with a four-link design that traded semi-elliptical leaf springs for coils, locating the axle with four trailing arms and a Watt’s linkage, later exchanged for a cheaper Panhard rod. The new suspension provided a much better ride, but the Panhard rod mounting caused structural problems and erratic handling.
Given time, those problems could undoubtedly have been resolved, but with the project already running well behind schedule, Brocklehurst and Mitchell decided to retain Hotchkiss drive. The production car’s sole concessions to improved ride quality were redesigned spring shackles and longer (and thus softer) leaf springs, the addition of which required the tail to be stretched about an inch (25 mm).
ENTER THE MGB
John Thornley very much wanted the ADO23 to be ready before the end of the decade, but the various false starts delayed construction of the first full prototypes until the spring of 1960 and they were not ready for testing until that fall. In the meantime, Abingdon launched the MGA 1600 Mk 2, a facelifted A with an unfortunate new grille. As Thornley had feared, MGA sales lost ground rapidly after 1960 as buyers gravitated to newer rivals like the Triumph TR4.
There was still one final obstacle to overcome: the high manufacturing costs of the ADO23’s new monocoque shell. The normal procedure in the automobile business is to pay all tooling costs in advance and amortize that expense over the course of the production run. Amortization schedules are something of a juggling act, but the goal is to pay off the entire expense before the end of the model’s lifespan while still allowing a reasonable per-car profit. If volume is high enough (or costs low enough) to repay the tooling costs before the end of the model run, late models will obviously be even more profitable; conversely, a model that costs too much to tool and/or that sells in only modest numbers usually needs a long production run to pay off its initial costs.
When BMC’s finance people saw the projected tooling costs for ADO23, they concluded that there was little hope of the car making any profit even over what was then expected to be a seven-year run. According to Wilson McComb, BMC solution was to convince supplier Pressed Steel Ltd. to manufacture each body shell for a flat per-car fee; in that way, if the ADO23 didn’t sell well enough to recoup its tooling costs, the loss would be Pressed Steel’s, not BMC’s. At the time, this seemed like an eminently sensible solution, although we have to wonder if anyone had second thoughts after BMC acquired Pressed Steel a few years later.
The production ADO23 was named MGB, a designation Thornley originally wanted to use for the short-lived MGA Twin Cam. Pilot production began in May 1962 and the B made its public debut on September 20.
The MGB was somewhat smaller than the MGA, but its monocoque body made it significantly roomier and noticeably more rigid. Like many early American efforts at unitary construction, the MGB was actually heavier than its body-on-frame predecessor, particularly after an additional support member was added between the A-pillars to solve a cowl shake problem. In compensation, the MGB’s body shell was extremely stout, although owners later discovered the bane of many early British unit-body roadsters: a tendency for the complex sill structure to trap moisture, creating expensive rust traps.
While early prototypes had used the same 1,622 cc (99 cu. in.) four as the Mk 2 MGA, the 1.6-liter engine was hard pressed by the ADO23’s greater curb weight. To compensate, the engine of the production MGB was expanded to 1,798 cc (110 cu. in.), giving 94 hp (70 kW) and 107 lb-ft (145 N-m) of torque. As in the MGA, the engine was linked to a four-speed gearbox with unsynchronized low and the option of Laycock de Normanville overdrive, usable in both third and fourth gears. Abingdon originally hoped to offer a high-performance model powered by the DOHC four from the MGA Twin Cam, but that engine was costly and had developed a reputation for fragility, so it was canceled well before before the ADO23 went into production.
MG advertising trumpeted the MGB roadster’s newfound weather protection and interior conveniences, which was perhaps overstating the point. The B did have wind-up windows rather than the MGA’s side screens, but the standard soft top was a cumbersome “pack-away” affair with a separate framework that had to be assembled by hand. A permanently attached convertible top cost extra, as did a heater, something many contemporary reviewers saw as an anachronistic touch. Although the B was a low-slung roadster, outward visibility was not great either; the seats were set low enough that some drivers found the car a trifle claustrophobic.
If the smaller exterior dimensions and lower spring rates suggested a cushy boulevardier, a brief test drive soon erased that notion. While the MGB was softer than the MGA, the driving experience was still decidedly rugged. The B was quite agile, but its controls were heavy and its ride was stiff-legged. Noted racing driver Ken Miles, who tested the MGB and several other affordable sports cars for the September 1966 issue of Car and Driver, concluded that the B sacrificed too much comfort for its capabilities. Nonetheless, the magazine’s editors ranked the MGB second overall, ahead of the Triumph TR4, Sunbeam Alpine, Fiat 1500 Spider, and Datsun SPL311.
If the MGB had a single, defining virtue, it was directness. It wasn’t particularly sophisticated or even very civilized, but it was free of gimmicks and it had no significant dynamic vices — it simply went where it was pointed with a pleasing alacrity. By comparison, rivals like the Sunbeam Alpine were more comfortable, but often felt more sedanish than sporty. The MGB behaved the way buyers expected a sports car to behave and made no pretense of being anything it was not.