It sounded so promising at the time. After years of dismissing imported compacts as cars for kooks, GM was finally going to build an attractive, sophisticated subcompact featuring the latest advances in manufacturing technology. To follow that, Chevrolet was going to offer a sporty version with a racy twin-cam engine built by the legendary English firm Cosworth. It was the car that was going to save America for American cars — that is, until it all went wrong. This is the story of the 1971-1977 Chevrolet Vega and 1975-1976 Cosworth Vega.
DÉJÀ VU ALL OVER AGAIN
It all had to sound very familiar. Back in the late 1950s, the Eisenhower recession had provoked a sudden flurry of interest in economical compact cars, of which American manufacturers of the time offered very few. The market for imported cars, previously negligible, suddenly climbed to a worrisome 5% or so of the U.S. market. Each of the domestic automakers (except AMC, which already had compacts in the form of the Rambler and Rambler American) hastened to develop their own homegrown compact cars in hopes of driving the foreign invaders from American soil.
These not-terribly-small domestic compacts sold well, but they failed to reduce the growth of imported cars, which by the late sixties included Datsun and Toyota as well as the ubiquitous Volkswagen. By the middle of the decade, Detroit appeared to have given up and ceded the low end of the market to foreign competitors.
By Detroit standards, that was an eminently logical decision. In America, goodness was synonymous with bigness, particularly in the collective minds of the Big Three. They had considered compacts back in the forties, but abandoned the idea when they realized that they couldn’t build small cars any cheaper than they could full-size models. Other than AMC’s George Romney, most auto executives couldn’t imagine why anyone would want a smaller car if they could afford a big one. Even when Detroit finally, reluctantly introduced smaller cars, the resultant products tended to have a distinct loss-leader vibe. That too was customary Detroit logic; each step up the ladder meant a substantial increase in per-car profits, so anything you could do to convince a customer to choose a bigger car meant more money in your pockets. Who needed small cars, except perhaps to entice young buyers who might be enticed to later buy an Impala or a Caprice?
The flaw in that reasoning was that the fringe market kept getting bigger. By 1969, small cars accounted for 29% of the U.S. market and the imports’ total market share had risen to more than 11%. Volkswagen alone was now selling more than 500,000 cars a year in the U.S. Like it or not, the Big Three were going to have to respond — and soon.
In October 1968, GM chairman James Roche announced to the press that in two years, GM would build an all-new subcompact car codenamed XP-887, which he promised would feature new advances in engineering and manufacturing, but would be priced to compete directly with the Volkswagen Beetle. The XP-887 would weigh about 2,000 lb (907 kg) and be powered by a small (by American standards — less than 150 cu. in./2.5 liters) OHC four. The XP-887, Roche declared, would show the world what General Motors could do.
XP-887: ED COLE’S CORPORATE COMPACT
To put this announcement in context, we need to step back about a decade to the development of what had previously been GM’s most complex and sophisticated small car: the Chevrolet Corvair. Shepherded by then Chevrolet general manager Edward N. Cole, the Corvair was a radically engineered compact featuring an air-cooled, rear-mounted aluminum engine; swing-axle independent rear suspension; and unit construction.
Some of the Corvair’s original technological ambitions had proved infeasible, but even in final form, the rear-engine compact was complex and expensive to build, which had prompted a last-minute cost-cutting program to bring the list price closer to more conventional rivals’. Despite those cuts, the Corvair didn’t sell as well as expected and some of the cost-reduction measures had exacerbated the design’s inherent tendency toward final oversteer. A series of accidents involving Corvairs resulted in lawsuits against GM, attracting unwelcome attention from consumer advocates like Ralph Nader.
Later iterations of the Corvair, although much improved, were overshadowed by Ford’s less-innovative but far more popular Mustang and the cheaper Chevy II, hastily rolled out in 1962 as a more conventional alternative to the Corvair. The second-generation Corvair was left to languish, finally expiring in 1969. Although it had sold almost 1.8 million units in 10 years, GM considered the Corvair something of an embarrassment, and for several years it became the corporate equivalent of an Orwellian “unperson,” officially forgotten.
The Corvair’s woes did not hamper the career of Ed Cole, who also had many other achievements to his credit, including the development of GM’s two most successful modern V8 engines. Cole was promoted to group VP of car and truck operations in November 1961, executive vice president in July 1965, and president and chief operating officer in October 1967.
Cole still had strong ideas about small car design, and as executive vice president and then as president, he had continued to pay close attention to compact car ideas being developed by the corporation’s central Engineering Staff. Some of those had involved rear engines and appear to have been conceived as at least nominal successors to the Corvair, although by early 1967, development and styling work was also being done on front-engine designs. Work on the last rear-engine concept, coded XP-892, continued until mid-1968, probably ending around the time the XP-887 project began. It seems safe to assume, therefore, that the XP-887 was developed with a great deal of Cole’s direct input.
Like the XP-892 and the earlier front-engine XP-873, the XP-887 was not born as a Chevrolet. Its initial engineering package was developed by the Engineering Staff while the initial styling direction was set by the corporate Advanced studio, then led by Clare MacKichan (a former Chevrolet chief stylist who had previously led the development of the Opel GT as chief designer for Opel in Germany) and directly supervised by Styling VP Bill Mitchell. We don’t know to what extent Chevrolet engineers and designers were even aware of the XP-887 project prior to Roche’s announcement; during the tenure of general manager Pete Estes, Chevrolet had been pursuing its own plans for a subcompact, a smaller counterpart to the Chevy II/Nova powered by an all-new cast iron four.
According to John DeLorean, who took over as Chevrolet’s general manager in February 1969, the XP-887 was at a very nascent stage at the time of Roche’s speech. Most of the specifications, both technical and financial, were still essentially theoretical, created by extrapolating from various foreign-made subcompacts. This was more or less customary for corporate engineering projects in those days. The role of the Central Staff was not to develop production cars, but rather to come up with ideas and inventions that the divisions could then adapt for production. A variety of other, very successful products had originated in precisely that way, including the original Chevrolet Powerglide and Buick Dynaflow automatic transmissions.
However, in previous cases, the decision of whether or not to adopt a particular idea had usually rested with the divisions, subject to the ultimate financial oversight of GM corporate management. (The corporation would occasionally order multiple divisions to collaborate on a single project, but in this era, that was still the exception rather than the rule.) Furthermore, each division generally had a great deal of leeway when it came to adapting a particular concept to its own needs, facilities, and requirements, which is why different divisions’ versions of a particular invention or concept often differed significantly.
In this case, the corporation ordered Chevrolet to discard its own project — and canceled outright a similar one at Pontiac — in favor of the XP-887. Unsurprisingly, that order was greeted with considerable resentment from Chevrolet’s own engineering department, which was effectively told — by their former boss, no less — that their own work wasn’t good enough. Moreover, Roche’s announcement had publicly committed the division to building what had been conceived as basically a pie-in-the-sky advanced engineering project, the promised specifications of which would not be easily achieved.
PASSING THE BUCK
As they struggled to come to grips with the XP-887, DeLorean and his disgruntled staff quickly recognized that much of what the central Engineering Staff had just handed them was more wishful thinking than realistic proposal. Some of the estimates were outdated, while others were off-target or simply inaccurate. Worse, the corporate design specifications didn’t necessarily even reflect GM’s own current policies; for example, there was no provision for the new corporately mandated side-guard door beams.
While Chevrolet chief design engineer Jim Musser (succeeded in early 1970 by Lloyd Reuss) worked to correct these shortcomings, Henry Haga’s Chevrolet Advanced styling studio revamped the exterior design to reduce the (coincidental) resemblance to Ford’s new Maverick and provide a stronger styling link to other Chevrolet models — particularly the new second-generation Chevrolet Camaro. Styling ended up being the Chevrolet subcompact’s most successful feature; if it didn’t quite qualify as sexy, it was certainly pleasant enough to look at, with overtones of both the Camaro and the Fiat 124 coupe.
The styling changes did not alter the fact that the XP-887 was now more than 10% over its original design weight and cost targets. Since that would be hard to fix without a major redesign that the corporation was not about to permit, Chevrolet’s marketing staff recommended positioning the XP-887 as an upscale subcompact in the manner of the early Nash Rambler, adding more features to help to justify a higher price.
The corporate Engineering Policy Committee recognized the logic of this proposal and gave their tentative approval. The plan was consistent with Chevrolet’s past experience; buyers of previous domestic compacts had shown a definite preference for plusher trim levels and the pricier Corvair Monza had consistently outsold the price-leader 500 by a substantial margin. However, shortly before launch, the Pricing Review Committee ordered Chevrolet to delete most of the additional features in hopes of bolstering the XP-887’s profit margins — an unwelcome echo of the launch of the Corvair.
The crowning indignity, so far as Chevrolet was concerned, was that the division was not even allowed to select its own name for the XP-887. As is customary, the Chevrolet marketing department had considered an assortment of names. One possibility that popped up in early press reports was “Chevette,” which Chevrolet would adopt a few years later for a different model, but the internal and consumer favorite was “Gemini,” which evoked NASA’s successful Project Gemini space missions of a few years earlier. Cole, however, would hear none of this and insisted on the name “Vega,” after a large white star in the constellation Lyra, although DeLorean said that name had tested poorly with consumers. Again, Cole got his way despite the division’s objections.
To build the Vega, Chevrolet invested heavily in expanding and modernizing its factory in Lordstown, Ohio. As Roche had promised, the modernization included a high level of automation, with Unimate robots performing most of the welding for the Vega’s unit body. The plant update revealed the corporation’s ambitious expectations for Vega sales: Lordstown now had an annual capacity of 400,000 units.
THE FIRST CHEVROLET VEGA
In size, the original Chevrolet Vega was about halfway between the Ford Maverick and a Volkswagen Super Beetle, stretching 169.7 inches (4,310 mm) on a 97-inch (2,464mm) wheelbase. In contrast to most import rivals, the Vega was quite low-slung; the two-door sedan had an overall height of 51.9 inches (1,318 mm), only half an inch (13 mm) taller than a 1969 Camaro, and the coupe was a mere 50 inches (1,270 mm) high.
The Vega had unitized construction and a conventional front-engine/rear-drive layout with rather slow recirculating ball steering (22.8:1, giving 4.4 turns lock to lock in standard form). Suspension was a conventional double wishbone layout in front and a live axle on coil springs in back, located by four trailing arms — essentially a scaled-down version of the Chevelle’s rear end. Front disc brakes survived the last-minute de-contenting (both the Volkswagen and Ford’s new Pinto had four-wheel drums), but the Vega’s standard transmission had three speeds rather than four; a four-speed manual was an extra-cost option, as was two-speed Powerglide automatic.
With a full tank of fuel, an early Vega weighed a bit under 2,300 lb (around 1,030 kg), which was about 210 lb (95 kg) more than a Pinto and over 300 lb (say, 140 kg) more than a Super Beetle. The Vega’s weight reflected Chevrolet engineers’ efforts to beef up the body structure as compared to the original XP-887 design, prototypes of which had continually broken during durability testing. Nonetheless, the Vega’s body was still not terribly robust and its corrosion resistance left much to be desired, in part because the cost-cutting spree had left the Vega with unlined fenders that were easily damaged by road salt.
In most respects, however, the Vega was a perfectly conventional small Chevy, not terribly different from its larger Nova and Chevelle cousins in design or mechanical layout. Its most unique feature — and its real Achilles heel — was its engine.
THE VEGA’S ALUMINUM ENGINE
The Chevrolet Vega’s engine, known as the 2300, was an all-new inline-four with a single overhead camshaft — a first for Chevrolet and a rarity for American engines of the period. With a displacement of 140 cu. in. (2,286 cc), the 2300 was big for the economy-car class, with undersquare dimensions for strong low-end torque. More significantly, as GM had promised and Cole had insisted, the new engine, which was made on the Massena, New York, line that had previous produced Corvair engines, featured an aluminum cylinder block, something GM hadn’t offered since it sold its all-aluminum V8 to Rover in 1964, barring Chevrolet’s rare and very expensive ZL-1 performance engine.
While aluminum cylinder heads were relatively common on non-U.S. engines — the press initially expected Chevrolet to use an aluminum head on an iron block — aluminum blocks were still very rare. The main reason was cost; although aluminum is significantly lighter than iron (a very attractive quality for passenger vehicles, in which the engine is usually the largest single mass), it is also substantially more expensive than iron to produce, buy, and machine. Furthermore, cast aluminum is significantly softer than cast iron, creating problems with premature wear of the cylinder bores. Adding iron or steel cylinder liners mitigates the wear issues, but raises costs even further.
Since the late fifties, GM’s Engineering Staff had been looking for ways to avoid that extra cost with aluminum cylinder blocks that would not require cylinder liners. To that end, GM and Reynolds Aluminum had devised a new aluminum alloy called A390 that, when properly finished, provided a much harder bore surface durable enough to make cylinder liners unnecessary. Cole had been interested in this process since the late fifties, drawn by its potential cost and labor savings; he had actually wanted to use it for the Corvair’s engine, although at the time, that had proved impossible. On the Vega, GM claimed the elimination of pressed-in or cast-in iron liners saved a useful $8 per engine, although each block still cost more than an equivalent cast iron one.
There was nothing conceptually wrong with the Vega’s linerless cylinder block; Cosworth later found that it wasn’t strong enough for serious racing use, but it was adequate for street use and, except for some early casting problems, bore wear was normally very low. Indeed, linerless aluminum blocks are very common today and generally experience few problems. However, the aluminum block made the Vega’s engine more sensitive to overheating, particularly in combination with the cast iron cylinder head, specified for cost reasons. (Curiously, the Vega’s iron head actually weighed more than the engine block, making the engine somewhat top-heavy.)
Among their differences, iron and aluminum have very different heat-conduction and expansion rates. If an aluminum/iron engine overheats, the aluminum side will expand faster than the iron side, putting considerable stress on the head gasket (which mates the head to the block) and eventually causing it to fail. Since aluminum is softer than iron, severe overheating can also cause the aluminum to warp, resulting in permanent damage.
That issue was by no means unique to the Vega — engines with aluminum heads and iron blocks run the same risk — but the Vega’s aluminum cylinder block was also vulnerable in another way: Severe overheating would break down the silicon content of the A390 alloy, leaving soft areas in the cylinder bores that would be quickly scuffed and scored by the pistons.
As long as the engine remained within normal operating temperatures, none of this was likely to occur. However, the cost-cutting binge had left the Vega with an undersized radiator and no coolant overflow tank. Coolant temperatures would climb quickly in strenuous conditions, such as pulling a heavily laden car up a long mountain grade on a hot summer day. If the engine did begin to boil over, it would also lose coolant, eventually running itself dry. As a result, any serious cooling system problem, like a failed thermostat, was a recipe for serious engine damage.
This problem was compounded on early engines by excessive oil consumption. The thirst for oil usually had little to do with the aluminum block per se; the main culprits were the valve stem seals, which would crack with age or wear and leak oil into the cylinders. Since engine oil is partly responsible for engine cooling as well as lubrication, the resulting tendency to run a quart low only exacerbated the 2300 engine’s vulnerability to overheating.
Beyond that, the Vega engine simply was not a very pleasant companion. It had adequate power, but it was disturbingly noisy when revved and quite rough. Four-cylinder inline engines have an unbalanced coupling force that causes the engine to rock up and down; the bigger the displacement (in particular, the longer the piston stroke), the greater the shake. Modern fours generally quell these forces with twin counter-rotating balance shafts, but that technology was not well-developed when the Vega was designed. (Although English inventor Frederick Lanchester had first patented the balance shaft concept back in 1912, Mitsubishi’s well-known “Silent Shaft” system wasn’t patented until 1973). Even if it had been available, Chevrolet would probably have deemed it too expensive, particularly if it involved patent license fees. Instead, Chevrolet used very soft engine mounts, hoping to isolate the shaking from the passenger compartment.
The soft engine mounts may have contributed to another early Vega problem: loose carburetor mounting bolts that could allow raw fuel to leak into the cylinders, potentially resulting in either dramatic backfiring or an engine fire. In April 1972, Chevrolet recalled about 130,000 Vegas to correct that problem, which cost the division both money and credibility. So too did a larger recall of 350,000 early Vegas to fix another, unrelated carburetor problem that could lead to the throttle linkage jamming in the part-throttle position.
A LATE INTRODUCTION
Thanks to the cost and weight overruns, the Chevrolet Vega was introduced to the public in September 1970 with base prices starting at $2,091, about 10% more than the Volkswagen Super Beetle that had been its original target. The Vega was also $101 more than a two-door Datsun 510, $173 more than a Toyota Corolla 1600, and $172 more than the new Ford Pinto. Despite its high price, the Vega was also one of the most Spartan cars in its class, sacrificing many of the minor amenities that were standard on its import foes, like carpeting and even a glove box.
In other quantitative respects, the Vega made a decent case for itself. Even in base form, it was capable of 0-60 mph (0-97 km/h) in less than 14 seconds, quicker than most direct rivals other than the bigger and considerably thirstier AMC Gremlin; top speed was around 90 mph (145 k/h). The Vega was also capable of a respectable 25–27 mpg (8.7–9.4 L/100 km), handled quite well despite its slow steering, and had a reasonably civilized highway ride.
All that sounded great on paper, but the Vega was less impressive in the real world. The admirable economy was achieved by combining widely spaced gears with a yawning 2.53:1 rear axle that left the engine easily overtaxed by steep hills, and the comfortable ride was achieved at some cost in axle control, particularly in panic stops. The interior ambiance was rather dreary in standard form and not all that roomy considering the Vega’s size and weight. You could mitigate some of these complaints with the options list — a GT with the available four-speed transmission and 3.36 axle was a good deal quicker than the base car, if less economical — but that pushed the price even further above the subcompact pack.
Motor Trend inevitably named the Vega its 1971 Car of the Year, but Chevrolet soon found that capitalizing on that praise — and the Vega’s multimillion-dollar ad campaign — would not be easy. Shortly after the Vega’s public debut, GM was hit by a lengthy UAW strike, provoked in part by internal union conflicts following the death of UAW leader Walter Reuther in a plane crash earlier that year. As a result, production halted for weeks and Chevrolet dealers had fewer than 25,000 Vegas to sell until mid-January 1971. The strike limited first-year Vega sales to 277,705 units, which was well below the 400,000-unit target. Sales for 1972 climbed to 394,592, which was more what Chevrolet had in mind.
Whatever their design flaws and engine-casting issues, early Vegas were reasonably well assembled by American standards, but quality took an ugly turn in 1972. The Lordstown plant had previously been run by Fisher Body, but in the fall of 1971, GM turned over control to the GM Assembly Division (GMAD), looking to reduce costs. GMAD immediately laid off some 700 workers — including the quality-control inspectors DeLorean had added — while upping production from 60 cars an hour to 100 an hour.
Predictably, assembly quality began to drop. Workers protested that they no longer had time to do a good job, but it was to little avail. Accusations of deliberate sabotage and GMAD managers’ heavy-handed approach to discipline (which allegedly included sending workers home for minor infractions like unauthorized bathroom breaks) led workers to call another strike in early 1972, but it ultimately earned them nothing. The Vega soon became one of the worst-built models in the Chevy line-up and owners complained that Chevrolet dealers often seemed unprepared to fix the cars if something went wrong.
Still, once production was up to speed, neither these woes nor customer complaints about rust, overheating, and backfiring seemed to affect sales. Chevrolet sold 427,300 Vegas for 1973 and an impressive 460,374 for 1974, bolstered by the OPEC oil embargo that began in late 1973.
THE COSWORTH VEGA
Whatever its other foibles, the Chevrolet Vega coupe was a fairly sporty piece, particularly with the optional L11 engine, four-speed transmission, F41 handling suspension, bigger tires, and Positraction limited-slip differential. It was clear that the basic package could handle more power than the standard engine provided.
In March 1970, John DeLorean and Chevrolet assistant chief engineer Lloyd Reuss (who, as mentioned above, had previously been Vega’s chief project engineer) contacted Britain’s Cosworth Engineering about the possibility of developing a special racing version of the Vega engine. Established in 1958 by Mike Costin and Keith Duckworth, Cosworth was one of the world’s preeminent builders of racing engines, prominent in Formula One and other forms of motorsport. Up until that time, Cosworth had been most closely associated with Ford, which offered the DOHC Cosworth BDA engine in some British and European models for homologation purposes, but Duckworth was amenable to the Chevrolet proposal, seeing it as a way of broadening Cosworth’s options.
Reuss assigned the project to veteran Chevrolet performance engineer Cal Wade, who worked closely with Cosworth engineer Mike Hall to develop a dual overhead cam 16-valve version of the Vega 2300 engine that could be reduced to less than 2,000 cc (122 cu. in.) displacement for Formula Two. Although the project was originally intended strictly for competition, Wade suggested to DeLorean that a street version of the new engine would turn the standard Vega — which at that point had not yet been released — into a formidable sports coupe. Chevrolet chief engineer Don McPherson was extremely skeptical, but DeLorean was intrigued and authorized Wade to keep working on the idea.
The first running prototype of the 16V DOHC Cosworth Vega engine, completed in June 1970, used dual carburetors, but Wade wanted to substitute electronic fuel injection, which would provide greater flexibility and better emissions performance than either carburetors or mechanical injection systems while offering Chevrolet additional technological bragging rights. To that end, Wade approached Bendix, which had briefly introduced and then hastily dropped its advanced “Electrojector” system about 15 years earlier. Bendix in turn pitched the idea to DeLorean.
Both DeLorean and Bendix were enthusiastic about the idea of a fuel-injected Vega engine, which they saw as a stepping stone to offering the electronic injection system on other Vegas and eventually other Chevrolets — big business for Bendix. Thus inspired, DeLorean ordered a marketing study of the potential viability of a performance Vega powered by the fuel-injected twin-cam engine. Marketing consultant Gertrude “Jimmy” McWilliams (former advertising manager of Triumph’s U.S. organization and wife of British-Leyland executive Bruce McWilliams) studied the problem and concluded that if the performance was up to snuff, there would be a potential market of at least 30,000 units a year.
In the spring of 1972, DeLorean got Ed Cole and Corvette chief engineer Zora Arkus-Duntov to drive a prototype of Wade’s super-Vega and secured approval for an initial run of 5,000 cars. DeLorean also got permission for a license agreement with Cosworth Engineering that would allow the production cars to carry the Cosworth name.
It was about then that the trouble began. Cosworth hit the wall in its efforts to produce a viable racing version of the Vega. The standard aluminum block castings turned out to have a safe limit of 270 hp (201 kW), 20–30 hp (15–22 kW) shy of what a Formula 2 version of the Chevrolet engine would need. (By this time, Cosworth’s Ford-based BDA twin-cam engine was making around 240 hp (179 kW) in Group 2 rally tune.) The only solution was to make extensive changes to the Chevrolet’s block casting process, which the Massena engine plant was neither able nor willing to undertake; Cosworth finally gave up in the spring of 1973.
Meanwhile, Chevrolet’s street version, originally slated for a mid-1973 introduction, was running behind schedule. Wade had hoped for 185 hp (138 kW) in street form, but noise and emissions regulations made that impossible. Even at 130–140 net horsepower (97–104 kW), he and McPherson were still not sure the twin-cam engine could pass the EPA’s 50,000-mile (80,000-km) durability test. (McPherson, for his part, maintained that the whole project was a hopeless boondoggle.) They finally opted to delay emissions certification rather than risk failing the time-consuming durability test and having to start over.
Eager to bring the Vega some positive publicity, Chevrolet general manager Jim McDonald (who had replaced DeLorean in October 1972) decided to announce the Cosworth engine anyway, even allowing some magazine editors to drive the preproduction prototypes. The announcement generated great excitement in the enthusiast press and drew interest from GM’s Australian and German subsidiaries, Holden and Opel, as well as Jensen’s Kjell Qvale. Dealers started taking customer deposits for Cosworth Vegas and waiting lists began to form in anticipation of McDonald’s claimed mid-1974 launch date.
Unfortunately, the 1974 model year came and went and the car was still not ready. As Wade and McPherson had feared, the twin-cam engine failed its EPA durability test. Not only did that mean starting over, the engine would now have to meet the more stringent 1975 standards, which would require adding air injection (something Wade had hoped to avoid), a catalytic converter, and changes to support the use of unleaded gasoline. The revised engine didn’t complete its durability testing until January 1975 and didn’t receive its EPA certification until March. As a result, the Cosworth Vega’s public debut was delayed until April 17, late in the model year and almost two years after the original announcement.
The Cosworth engine’s technical specifications were impressive. It retained the standard Vega’s aluminum block, but had a forged, hardened crankshaft with a shorter throw that reduced stroke to 3.16 inches (80 mm) and total displacement to 122 cu. in. (1,995 cc). The entirely new head was aluminum, with pentroof combustion chambers and dual camshafts in a detachable aluminum carrier. The valves, which were smaller in diameter but more numerous than those of the standard eight-valve Vega 2300 engine, were actuated via inverted bucket-type tappets, and had sintered iron seats to facilitate the use of the unleaded fuel the catalytic converter demanded. The breathing improvements provided by the greater total valve area allowed a milder cam profile than that of the regular Vega GT engine, which, combined with the shorter stroke and smaller displacement, made the Cosworth engine both more powerful and significantly smoother than its 2.3-liter sibling.
In addition to Bendix electronic fuel injection, the twin-cam Cosworth engine featured high-energy electronic ignition and special low-restriction stainless steel headers with a novel “Pulse Air” air-injection system that consumed less power than did a conventional air pump. Even with those changes, the complete engine had a dry weight of only 305 lb (138 kg), compared to about 345 lb (157 kg) for the iron-head Vega engine.
For all its technological credentials, the Cosworth Vega engine’s output was disappointing. Chevrolet initially claimed 120 hp (90 kW), but that was hastily revised to 111 hp and finally 110 net horsepower (82 kW) at 5,600 rpm with 107 lb-ft of torque at 4,800 rpm. (On paper, the Z09 Cosworth sounded no more powerful than the Vega GT’s carbureted L11 engine of 1971, but the latter’s 110 hp was an SAE gross rating; by 1975, the L11 was advertised at a more realistic 87 net horsepower (65 kW).) The Cosworth’s 110 hp was certainly not bad — in fact, it was almost spot on the similarly sized engine in the BMW 320i, which appeared the following year — but it was not the junior Supercar Chevrolet had been promising.
A DECLINING MARKET
Despite the mediocre output and a hefty curb weight of 2,760 lb (1,252 kg), the Cosworth Vega was reasonably quick, capable of 0-60 mph (0-97 km/h) in around 9 seconds and a top speed of up to 112 mph (180 km/h). Those were not muscle car figures, but they were quite good for the performance-challenged mid-seventies. Thanks to a new rear suspension with three trailing links and a Panhard rod (borrowed from the new H-body Chevrolet Monza and added to other Vegas for 1976), the Cosworth’s handling was even better than the standard Vega’s and a match for almost any sporty coupe of its time.
Unfortunately, the Cosworth Vega was not priced like other sporty coupes. With a starting price of $5,979, it cost twice as much as a regular Vega coupe, over $2,000 more than a V8 Camaro, and $300 more than a Datsun 260Z. Chevrolet twisted the knife by advertising the Cosworth Vega as “One Vega for the Price of Two” — not exactly an enticing or politic tag line for owners of early Vegas that had ended up needing warranty engine replacements.
Chevrolet built a total of 2,061 Cosworth Vegas for 1975. At first, dealers found them to be effective showroom traffic builders, but interest evaporated quickly when customers learned that the list price was almost $6,000. Many early Cosworths ended up being sold at steep discounts and dealers were more cautious about ordering cars for 1976. Total ’76 orders amounted to only 1,446, about half what Chevrolet had expected. At that point, Chevrolet general manager Bob Lund (who had replaced McDonald in December 1974) decided that enough was enough. The Cosworth Vega was discontinued in July 1976, and many unused engines were scrapped.
All Vega sales were on the decline by 1975, partly because of the car’s increasingly spotty reputation and partly because of internal competition from the Monza and its H-body siblings, the Buick Skyhawk, Oldsmobile Starfire and Pontiac Astre. The H-bodies shared the Vega’s basic body shell, but had fresher styling and could be ordered with V6 or V8 engines, which were less economical but more reliable than the Vega’s four. Overall Vega sales slumped to 207,764 for 1975, falling to 160,524 the following year, when Chevrolet introduced the new subcompact Chevette.
By that time, Chevrolet had at least mitigated many of the Vega’s more serious problems. By 1974, the major recall campaigns were over and rust protection had progressively improved. The four-speed manual transmission belatedly became standard for 1976 along with an updated “Dura-Built 140” engine featuring improved cooling and oil circulation, a new coolant expansion tank with low-coolant warning light, a stronger head gasket, new valve stem seals, hydraulic valve lash adjusters, and iron-coated aluminum pistons. Chevrolet even offered a five-year, 60,000-mile (96,600-km) engine warranty in an expensive attempt to regain consumer confidence.
Still, sales fell to 78,402 for 1977, the Vega’s final year. The derivative H-bodies survived through 1980, but in their last three years, the previously standard Vega four was replaced by Pontiac’s 151 cu. in. (2,471 cc) “Iron Duke” engine.
Stillborn — perhaps mercifully so — was Ed Cole’s last great ambition: the GM Rotary Combustion Engine. The two-rotor RC2-206 version of the rotary engine, developed at staggering expense (including at least $50 million in patent licensing fees), was intended as the sole engine for the H-body cars and by some accounts was originally supposed to debut on the 1974 Vega. However, the GM RCE program had an extremely troubled development, and in September 1974, Cole announced that the rotary engine had been shelved, essentially admitting defeat.
As with the Corvair, any statements about the Vega’s failure have to be carefully qualified. Chevrolet sold more than 2 million Vegas during its seven-year lifespan, which is excellent by any standards. During the difficult period of the OPEC embargo — which briefly made big cars almost unsaleable — Chevrolet sold all the Vegas they could build. We’re not sure if the division actually made a profit on the Vega itself, particularly considering its high warranty costs, but if we factor in the sales of the derivative H-body Monza, they probably came out ahead. What it cost the division in customer good will is harder to measure.
The Chevrolet Vega story is eerily reminiscent of the Corvair’s history in many respects. Like the Corvair, it walked an uneasy line between high-tech sophistication and cheap-and-cheerful basic transportation. Like the Corvair, it embodied a number of perfectly sound concepts that were tarnished by clumsy execution. And like the Corvair, it had significant flaws that probably wouldn’t have been insurmountable had it not been for short-sighted, last-minute cost-cutting.
As of this writing, GM is preparing to launch the Chevrolet Volt, a plug-in hybrid intended to demonstrate GM’s engineering mojo and trump the Toyota Prius and Honda Insight. If you’ve followed the Volt’s somewhat torturous development, there are disconcerting echoes of the Corvair and Vega: cutting-edge technology, grand promises, and significant cost and weight overruns, all married with a curious sense of executive ambivalence. (Product czar Bob Lutz, at whose behest the Volt show car is being transformed into a production vehicle, has made no secret of his preference for thirsty V8 muscle cars and has publicly declared his disbelief in global warming.) The Volt is of undoubted technical interest, but given GM’s track record in this area, we only hope that its story has a happier ending.
NOTES ON SOURCES
Our sources for the story of Vega’s troubled development history included “A reminder on auto recalls,” Kiplinger’s Personal Finance Vol. 26, No. 8 (August 1972), pp. 14–16; the Auto Editors of Consumer Guide, Encyclopedia of American Cars: Over 65 Years of Automotive History (Lincolnwood, IL: Publications International, 1996); “Bye-Bye, Rotary Vega — Hello, V-8,” Motor Trend Vol. 26, No. 4 (April 1974), p. 24; Chevrolet Motor Division of General Motors Corporation, “Chevy’s New Little Car Is Open for Business” [brochure 1102], ca. September 1970; “1976 Vega: Built to take it” [brochure 3317], September 1975; “’74 Vega” [brochure 2677-Rev], January 1974; “The Little Car That Does Everything Well” [brochure 1619], September 1971; “The Little Car That Does Everything Well” [brochure 2229], September 1972; and “Vega 2300: Why and How It Came About” [dealer brochure], 1970; “Comparison Test Super Coupes ’74: Mazda RX-2, Open Manta Rallye, Toyota Celica GT, Capri 2800, Vega GT, Mustang II Mach I,” Car and Driver Vol. 19, No. 11 (May 1974), pp. 58–69, 86; Robert Cumberford, “Who Killed the Corvair?” Car and Driver Vol. 15, No. 2 (August 1969) pp. 34-35, 73; “Cole, Edward N.” (n.d., GM Heritage Center, history. gmheritagecenter. com, accessed 27 August 2009); Corvette Museum, “2011 Corvette Hall of Fame Clare MacKichan,” YouTube, https://youtu.be/KNkDFmTzUBo, uploaded 17 November 2011, accessed 8 April 2012; Mike Covello, Standard Catalog of Imported Cars 1946-2002, Second Ed. (Iola, WI: Krause Publications, 2001); Jim Dunne, “Detroit Report…” Popular Science Vol. 193, No. 6 (December 1968), p. 30; “Detroit Report…” Popular Science Vol. 194, No. 2 (February 1969), p. 48; “Detroit Report…” Popular Science Vol. 194, No. 5 (May 1969), p. 48; “Detroit Report…” Popular Science Vol. 195, No. 4 (October 1969), p. 26; and “Detroit Report…” Popular Science Vol. 196, No. 1 (January 1970), p. 36; Robert F. Freeland, The Struggle for the Control of the Modern Corporation: Organizational Change at General Motors, 1924–1970 (Cambridge: Cambridge University Press, 2001); John Gunnell, ed., Standard Catalog of American Cars 1946-1975 Revised 4th Edition (Iola, WI: Krause Publications, 2002); David Halberstam, The Reckoning (New York: William Morrow and Company, 1986); H-Body FAQ, n.d., h-body. org/library/hbodyfaq/hbodyfaq-1.html, accessed 27-28 August 2009; John B. Hege, The Wankel Rotary Engine: A History (Jefferson, NC: McFarland & Company, Inc., 2001); “Inside Detroit,” Motor Trend Vol. 21, No. 5 (May 1969), pp. 11-12; Michael Lamm, “Martyr,” Special Interest Autos #22 (May 1974), reprinted in Corvair Performance Portfolio 1959-1969, ed. R.M. Clarke (Cobham, England: Brooklands Books Ltd., ca. 1998), pp. 132-140; Richard Langworth, “Corvairs for the ’70s: What Chevy Might Have Built,” Special Interest Autos #68 (April 1982), pp. 20-27; Karl Ludvigsen, “GM’s Wankel: The $700 Million Miscalculation,” Motor Trend Vol. 27, No. 3 (March 1975), p. 53; Hirokazu Nakamura, Hikoichi Motoyama, Tadahiko Ito, and Seizo Iwasa, assignors to Mitsubishi Jidosha Kogyo Kabushiki Kaisha (Mitsubishi Motors Corporation), U.S. Patent No. 3,995,610A, “Four Cylinder Straight-Type Engine with Secondary Balancer System,” filed 27 September 1974, issued 7 December 1976; Jan P. Norbye and Jim Dunne, Pontiac 1946-1978: The Classic Postwar Years (Osceola, WI: Motorbooks International, 1979); the Old Car Manual Project Old Car Brochures website, oldcarbrochures.org; Ken Polsson, “Chronology of Chevrolet Corvettes,” 4 April 2012, kpolsson. com/vettehis/, accessed 18 April 2012; Arthur Pound, The Turning Wheel: The Story of General Motors Through Twenty-Five Years 1908–1933 (Garden City, NY: Doubleday, Doran & Co., Inc., 1934); Joe Sherman, In the Rings of Saturn (New York: Oxford University Press, 1994); Alfred P. Sloan with John McDonald, My Years with General Motors (Garden City, NY: Doubleday, 1964); Robert Spinello, “Complete Vega History 1970-1977,” H-body.org, n.d., www.h-body. org/library/ vegabob/vega-history-complete.html, accessed 27 August 2009; Daniel Strohl, “Might Mouse: The diminutive, but sporty, 1969 Opel GT 1.1L,” Hemmings Sports & Exotic Car #7 (March 2006); “Vega,” Sol Station, 2012, www.solstation. com/ stars/ vega.htm, accessed 14 June 2015; Mark Wan, “AutoZine Technical School: Engine Smoothness,” AutoZine.org, 1998–2000, www.autozine. org/technical_school/ engine/smooth1.htm; Joseph White and Paul Ingrassia, Comeback: The Fall & Rise of the American Automobile Industry (New York: Simon & Schuster: 1995); J. Patrick Wright, On a Clear Day You Can See General Motors: John Z. DeLorean’s Look Inside the Automotive Giant (Chicago, IL: Avon Books, 1980); and Wally Wyss, “Vega Rotary: Ford’s New Mustang II forces GM to debut its Wankel a year early,” Motor Trend Vol. 25, No. 7 (July 1973), pp. 50–52, 123.
We also consulted the following period articles: “Ford Pinto and Chevrolet Vega 2300: Detroit’s Compact Commitment,” Car and Driver Vol. 16, No. 3 (September 1970), pp. 25-30; “A Small World to Conquer: Six-Car Comparison Test: AM Gremlin, Chevrolet Vega 2300, Ford Pinto, Simca 1204, Toyota Corolla, Volkswagen Super Beetle,” Car and Driver Vol. 16, No. 7 (January 1971), pp. 20-29; “Road & Track Owner Survey: Vegas 2300,” Road & Track Vol. 24, No. 10 (June 1973), pp. 87-90; “Super Coupe Comparison Test,” Car and Driver Vol. 16, No. 6 (December 1971), pp. 25–32, 68–70; “1973 Chevrolet Vega,” Road & Track Vol. 24, No. 10 (June 1973), pp. 90-91; and John DeLorean, “Vega 2300”; Bill Sanders, “Vega: 2300”; Jim Brokaw, “Basic Sedan”; Bill Sanders, “GT Coupe”; and Jim Brokaw, “Wagon” all from Motor Trend Vol. 22, No. 8 (August 1970).
Additional information on the Cosworth Vega came from Arch Brown, “1975 Chevrolet Cosworth Vega: Twin-Cam, Quad-Valve Pace-Setter,” Special Interest Autos #129 (May-June 1992), pp. 38-45; Chevrolet Merchandising Department, Passenger Cars, “Chevrolet Introduces the Cosworth Twin Cam” [launch brochure], March 1975, and “Cosworth. One Vega for the Price of Two” [advertisement, ca. 1975]; Mike Cook, “Passing of a Pioneer,” Hemmings Sports & Exotic Car #20 (April 2007); John Dinkel, “Cosworth Vega for the Road,” Road & Track Vol. 24, No. 12 (August 1973), pp. 30–34; Don Sherman, “How to Hatch an Engine, Car and Driver Vol. 21, No. 4 (October 1975), pp. 73–83; Jeremy Walton, Escort Mk 1, 2 & 3: The Development & Competition History (Sparkford, England: Haynes Publishing Group, 1985); and “A Cosworth Vega History,” CosworthVega.com, n.d., www.cosworthvega. com/ cosworth_vega_history.html, accessed 28 August 2009.
The derogatory attitude of one Chevrolet executive toward import buyers — perhaps the most pointed expression we’ve ever heard of Detroit’s attitude towards imports during that period — was mentioned in Eric Dahlquist, “Declutching the BUG,” Motor Trend Vol. 20, No. 7 (July 1968), pp. 70-73. Bob Lutz’s comments about global warming were originally reported by Glenn Hunter, D Magazine 30 January 2008, frontburner.dmagazine. com; we originally read it in Frank Williams, “Lutz on Global Warming: ‘It’s a crock of shit,'” The Truth About Cars, 12 February 2008, www.thetruthaboutcars. com, accessed 12 February 2008.