In 1962 and 1963, Oldsmobile offered a short-lived turbocharged version of the compact F-85. Called F-85 Jetfire, it used a high-compression aluminum V-8 engine with a complex, troublesome fluid injection system. Chevrolet also developed a simpler turbocharger installation for the air-cooled flat-six engine of the rear-engined Corvair. The Corvair offered turbocharged engines from 1962 to 1966.
The Pros and Cons of Passenger Car Turbocharging
Even with high-temperature alloys to provide the necessary heat tolerance, turbocharging still had another serious limitation for automotive applications: what we now call “turbo lag.”
Because a turbocharger is driven by exhaust gases, the speed of the turbine and impeller depend less on engine rpm than on load. Under light load, the exhaust stream may not turn the turbine fast enough to produce much if any boost, and the presence of the turbine increases back pressure in the exhaust system, reducing power. When load increases, it takes some time for the rotating assembly to accelerate enough to provide useful boost, creating a noticeable delay in response. For industrial, marine, or aircraft engines that mostly operate at or near maximum load, turbo lag is seldom a significant concern. Until fairly recently, it has been much more troublesome for passenger cars and light trucks, which are frequently required to accelerate from idle and which spend much of their lives operating at less than 50 percent of maximum load, resulting in big disparities between on- and off-boost performance.
On the other hand, in this era, that was also true of contemporary crankshaft-driven superchargers. While we now tend to think of superchargers as offering immediate boost with no lag, that generally wasn’t the case prior to the eighties. A crankshaft-driven centrifugal supercharger of the late fifties or early sixties generally produced maximum boost at something between 25,000 and 30,000 rpm, and typically had to reach at least two-thirds of that speed to make even half its rated boost. Positive-displacement superchargers, like the Roots-type Detroit Diesel (GMC) blowers beloved of drag racers or the vane-type superchargers offered in this era by Judson and Shorrock, were a little stronger than a centrifugal supercharger at low speeds, but suffered substantial air leakage that greatly limited their volumetric efficiency at lower rpm. Moreover, with any crankshaft-driven supercharger, the need for step-up gearing forced a tradeoff between boost at lower engine speeds and the real danger of overspeeding the compressor at higher rpm. The larger Shorrock superchargers of the fifties, for instance, were redlined at only 5,000 rpm! McCulloch tried to mitigate this problem by using variable-speed or variable-ratio drive systems to provide the necessary step-up gearing, but their complexity made them maintenance-intensive and rather fragile.
With a turbocharger, the relationship between engine speed and boost is more flexible. The boost generated by any centrifugal compressor depends largely on the tip speed of the impeller, which is a function of the impeller diameter and the square of its angular velocity. A larger-diameter impeller produces more boost at any given speed, so crankshaft-driven centrifugal superchargers for passenger car applications generally use relatively large impellers to keep rotational speeds (and the need for step-up gearing) within reason. Since a turbocharger has no mechanical connection to the crankshaft and no need for step-up gearing, it can produce the same maximum boost with a smaller-diameter impeller rotating at higher speeds; minimizing the diameters of the impeller and the turbine significantly reduces the polar moment of inertia of the rotating assembly, which makes the turbocharger more responsive to changes in throttle position.
Turbocharging also offered another compelling advantage over mechanically driven superchargers: It was possible to maintain constant boost over a fairly broad range of engine speeds by controlling the flow of exhaust gas to the turbine. There were several ways to accomplish that, but the most common control system was a wastegate, a valve that would allow some of the exhaust gas to bypass the turbine if manifold pressure or boost pressure exceeded a certain threshold. Using a wastegate obviously didn’t prevent turbo lag, but once the turbine reached the speed necessary to produce the desired maximum boost (a point known as control intercept), the wastegate would hold boost pressure at that level all the way up to the turbine’s maximum operating speed, with far less mechanical complexity than a variable-speed drive mechanism for a crankshaft-driven supercharger.
Oldsmobile and AiResearch
Gil Burrell clearly considered all these points and saw much greater potential than Boegehold had three and a half years earlier. By using a wastegate-controlled turbocharger and limiting maximum boost to a modest 5 psi (0.34 bar), it would be possible to set a very low control intercept — as low as 2,000 rpm at full throttle — and maintain that boost through almost half the engine’s useful rev range. This approach wouldn’t provide a huge gain in peak horsepower, but it would produce a healthy increase in torque at the engine speeds more commonly used in normal driving, and the relatively low boost requirements could easily be met with a small, responsive turbocharger. Better still, a turbocharger installation along these lines would impose only modest penalties in fuel economy and engine weight. In principle, therefore, a turbocharger could be made to provide many of the advantages of a larger-displacement engine in a lighter, more compact, more fuel efficient package.
Using a turbocharged engine in this way is now very common, but it was an unusual idea in 1959, although there was some precedent in Studebaker-Packard’s recent use of a 289 cu. in. (4,737 cc) Studebaker V-8 equipped with a McCulloch VS57 supercharger to replace the departed Packard V-8. Unsurprisingly, none of the turbochargers then available commercially in the U.S. — all or almost all of which were intended for diesel engines — were really suitable for what Burrell had in mind. Oldsmobile therefore approached AiResearch about developing a new turbocharger for this application.
The Garrett Corporation had originally established its AiResearch Manufacturing subsidiary (styled until 1942 with a lowercase “r”) in 1939 to develop cabin pressurization systems and other related products. In 1950, the parent corporation, seeking to diversify beyond the aerospace industry, had directed AiResearch to develop turbochargers for commercial applications, beginning with Caterpillar Tractor diesels. In 1955, Garrett reorganized the turbocharger business line as the AiResearch Industrial Division, headed by Wilton E. Parker, who had previously worked on Garrett’s gas turbine program. The new division got off to a rocky start: Developing a commercially viable turbocharger turned out to have a painful learning curve, and the early Caterpillar turbos had been troublesome, requiring considerable work (and no small amount of patient client cooperation) to resolve. Turbocharger sales were modest, and the division didn’t turn a profit until late 1958. Nonetheless, AiResearch remained prominent in the field, thanks in no small part to engineers like turbine research director Werner Theodor von der Nüll (sometimes anglicized as “von der Nuell”), who was considered a leading expert on supercharging and had presented a number of much-cited papers on that subject.
We don’t know what kind of production volume the Oldsmobile proposal envisioned, but it appears it wasn’t much — Wilton Parker was dubious that the project would do much for his division’s still-shaky bottom line, even though GM was offering to underwrite the engineering and tooling costs. GM also insisted that AiResearch provide an extremely thorough breakdown of production costs, including not only the costs of parts and tooling, but also factors like the amount of machine time involved in each manufacturing operation, a level of detail AiResearch had never attempted to calculate before. However, GM was willing to pay for a comprehensive cost analysis study to develop the data they wanted (perhaps in anticipation of eventually bringing turbocharger production in-house, although that never happened), and the prospect of gaining even a small foothold in the auto industry was attractive, so Parker finally agreed.
An essential step in any successful supercharger application, and one where most aftermarket superchargers were necessarily a compromise, was properly matching the blower to the engine. To give AiResearch a head start in this area, Oldsmobile arranged for the GM Engineering Staff to conduct a series of bench tests on a Rockette V-8 fitted with a Roots-type supercharger. Armed with this data, AiResearch was able to deliver the first prototype turbocharger in March 1960. Oldsmobile engineers tested the turbo on the dynamometer and in a test rig in April and May. In June, they fitted the turbocharger to an early production aluminum V-8 and installed it in an F-85 prototype for testing at the GM Proving Grounds and on the road.
The rotating assembly of the AiResearch T-5 turbocharger was quite small, with a turbine just 2.4 inches (61.0 mm) in diameter and a 2.5-inch (63.5-mm) diameter impeller. As a result, its polar moment of inertia was only about one-tenth that of the Schwitzer D4U turbocharger used on the Research Laboratories’ experimental Oldsmobile Super 88, which had a 4-inch (101.6-mm) impeller. (For comparison, the impeller of the various McCulloch and Paxton superchargers had a diameter of 5.78 inches (146.8 mm).)
To withstand exhaust gas temperatures that could reach up to 1,550 degrees Fahrenheit (843 degrees Celsius), the turbine of the AiResearch turbocharger was cast of Haynes Stellite Alloy No. 31 (HS-31), an extremely hard alloy of cobalt, chromium, nickel, and tungsten that had been used for aircraft turbo-superchargers since around 1941. The compressor impeller, whose temperature requirements were less extreme, was die-cast aluminum. Turbine and impeller were mounted on a 0.5-inch (12.7-mm) diameter shaft 5.75 inches (146 mm) long, running on two aluminum sleeve bearings, with the turbine and impeller sides separated by a cast iron housing that acted as a heat shield.
Boost pressure was regulated by a poppet-valve wastegate, controlled by spring-loaded diaphragms in the boost control assembly that sensed compressor inlet and outlet pressures. The spring preload determined how much outlet pressure could exceed inlet pressure before the pressure on the primary diaphragm would begin to open the wastegate bypass valve, progressively reducing the amount of exhaust gas entering the turbine. This didn’t cut off boost completely; rather, the wastegate served to relieve excess boost pressure, allowing boost to equal but not exceed the desired maximum value.
While crankshaft-driven superchargers typically placed the compressor “upstream” of the carburetor(s), in what was sometimes known as a blow-through arrangement (because it blew pressurized air into the carburetor), Oldsmobile decided to position the turbocharger impeller “downstream” of the carburetor (a draw-through arrangement). In this way, the pressure drop through the carburetor would help to keep the impeller’s idle speed high, and the compressor outlet could be positioned as close as possible to the intake manifold so that pressurized air-fuel mixture would have a relatively straight shot into the intake runners. Using a draw-through impeller also avoided the need for an air box to seal the carburetor.
For the turbocharged engine, Oldsmobile eschewed the downdraft carburetors of the normally aspirated Rockette engines (and most contemporary American V-8s) in favor of a side-draft carburetor, selected mostly for packaging reasons. Existing horizontal carbs didn’t have enough venturi area, so GM’s Rochester Products Division developed an all-new 1.50-inch (38.1-mm) single-barrel side-draft carburetor, the Rochester Model RC. This was of more or less conventional design except that it had no throttle butterfly valve. Instead, there was a separate horizontal throttle body, connecting the carburetor to the compressor inlet, which contained primary and auxiliary throttle valves, the latter used under certain conditions as an additional boost limiter.
We should reiterate at this point that although the GM Research Laboratories had conducted the preliminary research and the corporate Engineering Staff had provided some technical support, the Oldsmobile turbocharger program, known internally as XP-200, was very much an Olds project, not some corporate initiative awarded to (or foisted on) the division. Indeed, the only reason Oldsmobile pursued the idea at all was that Gil Burrell saw promise in it and Olds chief engineer Harold Metzel agreed that it was worth investing some of the division’s not-unlimited development resources.
It eventually fell to Oldsmobile general manager Jack Wolfram to convince corporate management to approve the XP-200 program. Wolfram was not a notably progressive executive, and he bore no small responsibility for the staid character of contemporary Oldsmobile cars. However, the turbocharger project promised to expand the utility of the Rockette V-8 for a fraction of the cost of developing and tooling an all-new engine, and perhaps reclaim some divisional pride in the bargain. Oldsmobile staff and executives were apparently rather sensitive about the aluminum V-8’s Buick origins (having their own engines was an important part of each division’s identity in those days), and Oldsmobile’s reputation for advanced engineering had been slipping of late. The most recent noteworthy innovations the division had fielded were the J-2 triple-carburetor setup, discontinued in 1958 in the wake of the AMA racing ban (and wholly overshadowed by Pontiac’s similar but better-promoted Tri-Power option), and the short-lived and troublesome “New-Matic” air suspension system, which had also been dropped. Oldsmobile had some more interesting projects in development — the division’s work on front-wheel drive began in this same period — but none was ready for public airing.
Another factor was the recent debut (in May 1960) of the Corvair Monza. This sporty version of the rear-engine Corvair featured bucket seats, a novelty in those days, and was shortly offered with a higher-output engine (RPO 649) and a new four-speed gearbox. The Monza had been an immediate hit, taking even Chevrolet by surprise and demonstrating that there was a largely untapped market for sporty compacts. Oldsmobile’s immediate response would be the F-85 Cutlass, which would arrive in April 1961, with bucket seats and a four-barrel premium-fuel Rockette V-8. Judging by Burrell’s subsequent comments, we suspect that Wolfram used the popularity of the Monza to help “sell” the turbocharger project, presumably pitching it as a kind of sportier Super-Cutlass.
The Oldsmobile turbocharger program received production approval in the summer of 1960, but testing of what Oldsmobile marketers dubbed the “Turbo-Rocket” V-8 continued through early 1961. Repeated full-power dynamometer runs found that the aluminum V-8 withstood the additional heat and pressure of turbocharging relatively well, but Oldsmobile made some changes for greater durability, fitting engines slated for turbocharging with heavy-duty pistons, aluminumized intake valves and valve seats, longer bolts for the main bearing caps, and Moraine M-400 aluminum inserts for the main and connecting rod bearings. (These parts were added during Oldsmobile’s engine assembly process and were not shared by Buick, which never offered a turbocharged version of its aluminum Fireball V-8.) To deal with the greatly increased blow-by that accompanied boosted operation, Oldsmobile also equipped turbocharged V-8s with positive crankcase ventilation, not yet a universal feature on American engines (although newly required in California and shortly to be mandatory in New York as well).
Other refinements to the production turbocharged engine included a higher-capacity cross-flow radiator; a bypass fuel filter, larger fuel lines, and a fuel return line to handle the greater fuel flow requirements; a higher-voltage ignition coil; cooler spark plugs; and a revised distributor advance curve. All Turbo-Rocket engines also had a vacuum-controlled throttle stop (normally fitted only to F-85s with Hydra-Matic) and a damper on the throttle linkage to slow the closure of the primary throttle valve.
There were also a variety of preproduction changes to the turbocharger itself, which was revised to incorporate an integral wastegate, reverse the direction of shaft rotation (mostly to allow the hot exhaust pipes to be mounted farther from the underside of the hood), and adopt a reshaped diffuser with a water jacket for better cold-start performance. AiResearch finalized the turbo design that spring.
11 CommentsAdd a Comment
Pleased to see this. Impressive, as usual.
There may have been an antidetonant for other applications at the time. If so, I’m betting that the manufacturer never explicitly marketed it to Jetfire owners.
I’m not sure if Thompson was still selling Vitane by this point, but it wouldn’t have been ideal for the Turbo-Rocket V-8 anyway. At any rate, Oldsmobile literature most emphatically discouraged using substitute fluids, which was understandable, since using a different fluid risked either diminishing the effectiveness of the system or risking it freezing in cold weather.
There were a number of technical papers published toward the end of WW2 summing up wartime experience and best practices with regard to fluid injection, which I suspect the engineers at Oldsmobile (or perhaps Rochester, who designed the final injection system) read and took to heart, as they followed those recommendations quite assiduously.
I remember reading that the F85 – in its earliest development stage – was to have a transverse 60 degree V6, and an automatic transaxle that used the hardware from the Roto-5 automatic but with three chains to connect engine to gearbox to wheels. It was killed early in the development stage – due to cost, of course – but we did get something useful from that project. The 60 degree V6 developed by Olds engineers sat on the shelf until the late 1970s when a narrow V6 was needed for the upcoming X car. The Olds design was handed to Chevrolet to get them started and then they finalized the engine for production. But if you look carefully at the 2.8/3.1 V6 from the early 80s, you can see details in the block – notably timing chain area – that look a lot like the small block Olds V8. In fact, it looks a lot more like an Olds motor than a Chevy.
What hit the showrooms in late 1960 was fairly conventional, but looking at the MSRP that would be in the high $20k in 2023 dollars, there wasn’t much engineering magic they could include without pricing the car out of the market or losing a lot of money on every one they sell.
It’s true that Oldsmobile did experiments with FWD for a car the size of the F-85, but the timetable makes it unlikely it would have been for 1961. The first FWD test mule wasn’t built until early 1960, when the initial production F-85 was very close to pre-production. At that point, the FWD project was at a rather nascent stage (the test mule weighed about 600 lb more than a RWD F-85, the chain drive was still quite crude, they were still evaluating whether they needed two CV joints per side, and I think it was using pieces of an older four-speed Hydra-Matic), so even if the division had been enthusiastic and corporate management had signed off on it for production, I think it would likely have been for a second-generation F-85. There simply wouldn’t have been time to get it into producible shape for 1961.
(There is some confusion on the timeline, stemming in part from the fact that both Oldsmobile Advanced Design Group and the corporate Engineering Staff were working on the project concurrently in different ways. (Oldsmobile asked Engineering Staff to develop a gear-drive transfer unit, which involved a Buick Dual Path Turbine Drive two-speed automatic.) However, Andy Watt, who was head of Advanced, said unequivocally that until early 1960, Oldsmobile FWD prototype development was still only at the stationary test rig stage.)
I have seen Oldsmobile engineers attribute the Chevrolet 2.8-liter V-6 to the abortive Oldsmobile FWD project. However, I’m very skeptical of the idea that it “sat on the shelf” until the seventies or that an almost 20-year-old design for an unproduced Oldsmobile experimental engine would be “handed” to Chevrolet. Weird things happen sometimes, so I suppose it’s not outside the realm of possibility, but it seems more like a piece of internal folklore born of old divisional rivalries. I’ve never found any substantive details about the V-6 used in the FWD mules (neither the SAE paper nor the GM Engineering Journal articles about Oldsmobile FWD development even mention its displacement), and since that engine never got close to production, it’s not a claim that seems particularly verifiable. Also, the “small block” Oldsmobile V-8 — by which I assume you mean the 330/350/307 rather than the aluminum Rockette engine — didn’t yet exist as such in 1960 (the bulk of its initial development was in 1962 and early 1963), although it’s possible that the experimental V-6 previewed certain details later used on that engine. To the extent that the 2.8-liter V-6 looks “like an Olds motor,” it seems more plausible that the source of inspiration was the smaller Oldsmobile V-8, which WAS a production engine, and a very familiar one by the time the FWD X-body cars were in development.
(At a glance, the most obvious resemblance between the 2.8-liter V-6 and the Oldsmobile V-8 is the way the block forms a kind of integral shroud for the timing chain. Oldsmobile said they did that because they wanted to be able to use a flat timing chain cover, while Chevrolet’s explanation was that it allowed them to avoid using a steel backing plate for the cam drive.)
Great article as usual! Thanks.
Not only did turbocharged Corvairs have a cylinder head temperature gauge, there was also an under-dash buzzer that alarmed if the head temp get too high. I’ve read that Chevrolet engineers were worried about sustained high loads like pulling a trailer or climbing a long, steep mountain pass and wanted an audible alarm to get the driver’s attention to back off the throttle.
Some of the period testers noted that while there was a cylinder head temperature gauge, the gauge didn’t provide any specific indication of how high was too high, so supplementing it with a buzzer was prudent. That notwithstanding, adding the new instruments rather than simply an amber “HEAD TEMP” light in the existing panel was a worthwhile move, and suggested that despite the comparative simplicity of the Spyder package, Chevrolet engineers had a clearer idea of its potential market and what those buyers might want.
I think it had all three: a gauge, an idiot light, and a buzzer. I’ve never driven a Corvair turbo but I’ve had several Corvairs and all had a combination head temp/oil pressure idiot light. As far as I know the turbo kept that and added the gauge and buzzer.
Your extensive follow up on later turbocharged cars was fascinating, especially the point about Porsche not producing a 911 turbo until 1975, 13 years after the Corvair turbo. It seems as if the Corvair wasn’t so much a poor man’s Porsche, rather, the 911 turbo should rightly be considered a rich man’s Spyder.
Porsche’s 1978 SAE paper on their turbocharging development makes clear that they were aware of the Spyder and Jetfire (as one would expect), but were not terribly familiar with them, asserting, for example, that they were made only in 1964 and 1965.
At any rate, what’s distinct about how Porsche approached turbocharging was that it was an offshoot of their racing efforts; their first turbocharged 917 was for the 1972–73 Can-Am series, and the development of the 930 and 935 followed that. Competition has shaped a lot of automotive turbocharging development, and so it’s notable that it WASN’T a factor in the creation of the Jetfire or Spyder. Oldsmobile didn’t develop the Jetfire as an homologation special, and while Chevrolet created the Spyder in large part to try to bolster the credibility of the Corvair Monza as a sporty car, it was neither developed for or as a street version of any racing project.
Marvelous. As usual.
I am–again–impressed with both your research and your ability to write about it.