Introduced for 1949, the famous Oldsmobile “Rocket” engine was the first of the modern postwar V-8s that kicked off the horsepower race of the 1950s. However, when Olds originally launched this engine, they emphasized gas mileage as much as power — and showed off the never-mass-produced, short-stroke, ultra-high-compression economy engine that Oldsmobile engineers hoped would one day replace it.
There’s a popular assumption that until the 1970s, gasoline was so cheap and plentiful in the U.S. that Americans didn’t care about gas mileage. Back in the 1940s, neither Detroit nor the oil industry saw it that way. The U.S. wasn’t long out of the Great Depression, and no one knew what the postwar economy was going to look like; the end of WW1 had been followed by a sharp economic downturn. Average retail gasoline prices were also on the rise after the war, like the prices of everything else. The auto industry’s technical papers of the time were very preoccupied with reducing fuel consumption.
In a highly influential 1947 SAE presentation, GM Research Laboratories chief Charles Kettering advocated strongly for raising automotive compression ratios into the same realm as contemporary diesels — between 12 and 17 to 1 — which he said could improve gas mileage by up to 40 percent. There were just two snags: No contemporary spark ignition passenger car engines had (or were capable of having) compression ratios that high, and no gas stations carried fuels with high enough octane ratings for such engines to use without severe detonation.
Until the early 1950s, the octane ratings of most automotive gasolines were fairly low. By 1947, typical U.S. regular gasoline averaged around 80 RON, while premium fuel averaged around 86 to 88 RON. There were higher-octane aviation gasolines, but motorists normally couldn’t buy those. Most had no reason to — the octane requirements of most passenger car engines weren’t that high.
Kettering’s experimental high-compression engine was a very different story. A 181 cid (2,973 cc) inline six with overhead valves and a 12.5 to 1 compression ratio, it had 95 net horsepower and produced very respectable fuel economy even in a 4,110 lb test car, but Kettering admitted that the engine pinged a bit even on experimental test fuel rated at 99 RON.
(“RON” stands for Research Octane Number, which is calculated using the CFR or ASTM Research Method. Gasoline is sometimes also rated using the ASTM Motor Method, which involves a different test procedure. The Motor Octane Number (MON) of a given fuel is typically 5 to 10 points lower than the RON. Today, U.S. automotive gas pumps list each gasoline-based fuel’s Anti-Knock Index (AKI), or “pump octane,” which is the average of RON and MON. If typical late ’40s gasoline were sold at modern U.S. gas stations, regular grade would be rated at around 76–77 octane while premium would be rated at 83–84 octane. The fuel used by Kettering’s experimental engine was rated at 99 RON and 87 MON, so it would be 93 pump octane on today’s scale.)
Even if oil refiners had immediately increased the octane ratings of their fuels by 10 points or more, the compression ratios of most postwar passenger car engines were already close to their practical limits. Increasing compression ratio increases peak firing pressures, so if an engine doesn’t have a very rigid bottom end — which most production engines of the 1940s did not — engine roughness and internal friction quickly become unacceptable at higher compression ratios. Also, many engines of the time were side-valve (L-head) designs, whose volumetric efficiency was low enough that increasing compression past a certain point produced no additional gains in efficiency. Conventional engineering wisdom was that about 8 to 1 compression was all most contemporary engines could handle, and even that was sometimes pushing it.
You probably already know what happened next: U.S. automakers who could afford it, starting with Oldsmobile and Cadillac, began developing all-new V-8 engines with provision for compression ratios up to 12 to 1. Oldsmobile was the first to actually announce its new engine, in September 1948, although Cadillac followed about a month later; both engines were developed more or less in parallel. Cadillac was extremely annoyed by the Oldsmobile V-8 program and had tried to convince senior management to squelch it, but they were ultimately unsuccessful.
The new Olds and Cadillac engines both featured overhead valves for volumetric efficiency; hydraulic lifters to keep the valves from chattering; wedge combustion chambers with a quench area over the piston to promote turbulence (an important detonation-control method); lightweight “slipper” pistons; and a very rigid crankshaft with five main bearings in a block with much greater bottom-end strength than before. Mechanical and thermal efficiencies were greatly improved.
Oldsmobile originally intended to call their V-8 the Kettering Engine, but GM management objected because Kettering was still living (in 1948, he had only recently retired, and was still a GM consultant with a seat on the board), so the new engine instead became the Rocket. In initial production form, the Rocket V-8 displaced 303.7 cid (4,977 cc) and had a dry weight of 745 lb with flywheel and all accessories, about 46 lb heavier than the new 331 cid (5,425 cc) Cadillac V-8. With its single two-barrel carburetor, Oldsmobile advertised the Rocket at 135 gross horsepower, but they also published as-installed net ratings of 122 hp and 240 lb-ft of torque, healthy increases over the 99 hp and 194 lb-ft of the 257 cid (3,942 cc) straight-eight engine used in 1948.
The Rocket dramatically improved Olds performance. Top speed of the big B-body Futuramic 98 increased from 88 to 95 mph, and Olds claimed the 1949 car could accelerate from 10 to 60 mph 4.6 seconds faster than the ’48. Tom McCahill of Mechanix Illustrated trimmed more than 8 seconds off the lackluster 0 to 60 time he had recorded with a 1948 98 sedan. The smaller A-body 88, which was 300 lb lighter, was faster still, earning it an early reputation as a formidable stock car competitor as well as a favorite of amateur hot-rodders.
However, Oldsmobile chief engineer Jack Wolfram emphasized that the new engine’s greater efficiency “has been utilized to increase both performance and gasoline economy.” The Rocket’s brake specific fuel consumption was over 20 percent better than the less powerful inline engine it replaced. In the 98, Olds said that was good for an extra 1 mpg at 50 mph. The lighter 88, which had a taller (lower numerical) axle ratio, returned 1.75 mpg more than that.
That was commendable, but it wasn’t anything like the 30 to 40 percent improvements Kettering had suggested. The reason, of course, was that the development of the new engines didn’t magically produce any big increases in the octane ratings of contemporary gasolines. Since both engines had to run on the same premium fuels as their predecessors, the new V-8s debuted with only a quarter-point more compression than the engines they replaced. The Olds was launched at 7.25 to 1, requiring 86 RON fuel, while Cadillac specified 7.5 to 1 compression and 88 RON premium.
These engines soon became the model for the U.S. industry, but Motor Trend, kicking off a decades-long tradition of being wrong about nearly everything, called them “premature,” remarking:
[T]he manufacturers were forced into a more expensive type of engine construction and a noisier type of operation (overhead valves) by the anticipation of compression ratios thought possible a number of years from now. Actually, an improved design “L” head engine, which breathes adequately at compression ratios up to 7½ and 8:1, would have been cheaper and would have easily taken full advantage of fuels available.
This was ridiculous: It wasn’t like the new Oldsmobile and Cadillac V-8s didn’t offer ample benefits even with existing fuels — these engines were the envy of the world for quite a few years, for good reason. Besides that, the conundrum of high compression was that the oil industry wasn’t going to make the investments needed to produce higher-octane commercial gasolines so long as engines capable of taking advantage of such fuels remained strictly experimental. By introducing production engines that were ready for higher compression ratios, GM was trying to show oil refiners that there would be a market and thus spur development. Gulf Research & Development Co. engineer C.J. Livingstone called the new engines a “challenge to [the] oil industry.”
Oldsmobile and GM Research Laboratories went a step further than that, constructing a number of very high-compression Rocket engines. These were not offered to the general public — which was just as well, since finding fuel for them would have been difficult unless you were next to an airport — but Jack Wolfram said several were “sold to the fuel companies for fuel development.” Those engines were supplied with three different sets of cylinder heads that gave compression ratios of 8 to 1, 10 to 1, and 12 to 1, allowing for easy comparisons.
Wolfram’s 1949 SAE presentation on the production Rocket engine included a surprising amount of detail about the performance of the 12.0 to 1 compression version. The high-compression Rocket was very similar to the production engine, but had a shorter stroke (3.25 inches rather than 3.4375 inches), reducing its displacement to 287.5 cid (4,706 cc) (which had actually been the original intended displacement of the production engine). Other changes were modest: thicker piston heads, copper-lead bearing material, and a revised ignition system with a higher-output ignition coil and distributor and spark plug insulator changes to prevent cross-firing and flash-over. Despite its reduced displacement, the high-compression Rocket was about 10 percent more powerful than the production engine, with 134 net hp and 260 lb-ft of torque.
Specific fuel consumption of the high-compression engine was a further 24 percent better than the production Rocket. For demonstration purposes, the high-compression engine was installed in a 1949 Oldsmobile 88 (the “1948” in the above graph is a typographical error) with its axle ratio changed from 3.23 to 2.95. With the taller axle ratio, there was no performance gain, but there was a big improvement in fuel economy: The high-compression engine returned 3.75 mpg better mileage than the standard 88 at 50 mph. Wolfram was vague about the high-compression engine’s octane requirements, but the test program used iso-octane with 2 cc/gallon of tetraethyl lead additive, implying something in excess of 100 RON.
Whether Oldsmobile would actually have replaced the production Rocket with a smaller, higher-compression version is an interesting question. Wolfram indicated that they intended to “eventually,” and emphasized that it would involve only minor tooling changes. If suitable fuels had been available early on, I think they might have, but the octane ratings of premium gasoline increased more slowly than auto engineers of the ’40s seemed to have expected: The average octane rating of premium went from 88.2 RON in mid-1949 to 90.8 RON by fall 1952 and 91.5 RON by October 1953, a gradual progression. In 1948, Socony-Vacuum Oil Co. research director W.M. Holaday declared flatly that marketing a third grade of super-premium gasoline was “not feasible” because of the huge investment involved; he was wrong, but it took until 1956 before oil companies were willing to try it even on a tentative basis.
By the mid-1950s, automakers’ interest in better fuel economy (or at least the sincerity of their expressed interest) seemed to drop off. Average U.S. gasoline prices, which had been rising in the late ’40s, stabilized in the ’50s; when adjusted for inflation, U.S. gasoline actually got cheaper over time until the early ’70s. Steady increases in premium fuel octane ratings allowed gradually higher compression ratios — by 1958, the average for new cars was up to 9.5 to 1 — but not enough to produce the kind of dramatic improvements in efficiency Kettering had predicted, especially with most new cars getting bigger, heavier, and more powerful.
Despite the proliferation of high-compression engines, a lot of Americans also remained very reluctant to pay extra for premium gas. Through the ’50s, less than one-third (around 31–32 percent) of retail gasoline sales were premium grades, even though by 1958 premium fuel was recommended for nearly half (49 percent) of all new domestic cars. The greater efficiencies of high-compression gasoline engines COULD reduce fuel consumption — through the 1960s, some automakers, including Oldsmobile, offered high-compression two-barrel economy engines — but the idea of paying more per gallon of gas in hopes of using less of it was a difficult equation for many buyers.
Also, I think that Kettering and other engineers of the ’40s may have been overly optimistic when it came to the octane needs of ultra-high-compression engines. Even when super-premium fuels with 100+ RON ratings became more generally available, compression ratios of domestic production engines very seldom went over 11 to 1. A meticulously tuned experimental engine with 12 to 1 or 12.5 to 1 compression, run on carefully blended experimental fuels of known octane rating, might get by okay on 99 or 100 RON fuel. In the real world, the actual octane of pump gasoline varied quite a bit from region to region and even tank to tank (the advertised octane rating was supposed to be a minimum guarantee, but that wasn’t always true in practice), and automakers had to make allowances for carbon buildup and poor tune.
So, the small-displacement, ultra-high-compression Rocket probably wouldn’t have ever been very practical. It was eventually forgotten, becoming another of the array of experimental Oldsmobile engines that never made it to production, like the later W-43 32-valve version of the third-generation Rocket 455 — an Olds engine from some parallel timeline where familiar events happened a little differently.
Related Reading
1949 Cadillac Series 62 Coupe de Ville – From Image Leader To Standard-Bearer For The U.S. Industry (by me)
1948 and 1949 Cadillac Series 62 Sedans: One Great Design, Two Very Different Faces (by Aaron65)
Curbside Classic: 1949 Oldsmobile 88 – Ghost Of The Future, Legend of the Past (by Bellinghamster)
Fascinating article and analysis.
Lately, I’m thankful that all of the vehicles in our fleet run on regular gas rather than premium or diesel. Current pricing at our closest Costco:
Regular — $3.559
Premium — $4.559
Diesel — $4.959
That price for regular is almost exactly the same as in 1950 and right about at the historic average, inflation adjusted. And modern cars are about twice as efficient. Yet one reads constantly about “high gas prices” and people trading in their cars for a more efficient one or an EV. I’ve seen this cycle so many times before…
Hudson also joined the octanefest. In ’49, the engineers jacked up their big L-head six to 9.3:1. CEO Barit invited the press to examine and drive the car, fueled up with 100 octane aviation fuel. The reporters asked Barit when he thought 100 octane would be available for cars, and he said “Probably not for another 20 years.” “Well then, why are we here?” “Proving a point, sir, proving a point.”
These engines turned my FIL into an ‘Oldsmobile Man’. After he got out of the Army in ’53 until the late ’70’s an Olds was his car of choice. He purchased other brands, but the family car had to be Oldsmobile!
“… kicking off a decades-long tradition of being wrong about nearly everything,…”
Dayum. Now you made coffee come out my nose.
I find it more than a bit odd that the higher cost for higher octane gas wasn’t given more weight in these considerations.
A minor modern point of comparison: Stephanie’s TSX has a “premium recommended” engine (2.4 L K series four), but it’s not “required”, unlike some engines. I’ve run some informal tests which suggest that it gets 10% better mileage on premium. So if premium is priced about 10% more than regular, it’s a wash, and sometimes I’ll put that in, but regular is the default. And it seems the pricing premium for premium has gone up, like up to 28% in the Costco prices given in the first comment here. That’s a pretty stiff increase.
When my Golf was new, I experimented with using 87 vs 94 octane fuel. The car ran smoother with the higher octane fuel and got 10% better fuel economy.
The extra cost was a wash until recently. As Paul stated, premium fuel is now 25% more expensive than regular. My car has a performance tune, meaning premium fuel is required. Since only buy one tank of gas per month, the extra $10 a tank is not a big burden.
The regular-premium price spread most certainly widened, I’m pretty sure it was only 40 cents difference here 20 years ago, where now at most stations it’s as much as $1.20 more than regular which is most certainly not worth the economy tradeoff.
Funny enough I know a few people in CA and even though their fuel prices are high, the regular-premium price spread isn’t nearly that wide, maybe like 50 cents, so where regular 87 is ludicrously expensive from my perspective, premium is actually almost the same as it is here in IL
Whatever they use to raise octane numbers now is probably way more expensive than the lead brain poison they used to use – that they called Ethyl for obvious reasons.
I miss MBTE, Ethanol blends the corn lobby have stuck us with for years last 20 years are godawful.
Ethanol blended into gasoline today has an octane rating of about 120, so that helps. More expensive? Not with subsidy to corn growers. MTBE polluted ground water, like lead, from leaky underground tanks at gas stations.
The main high octane gasoline component is alkylate, an expensive synthetic gasoline that made high performance aircraft engines possible during WWII, up to 115/145 aviation octane rating with lots of additional lead.
I think that in the early postwar period (i.e., 1945 to 1950), engineers figured that the future price of gasoline, much less premium gas, was still too hard to predict, since there were shifts taking place in how fuels were made. Of the 99 RON/87 MON gasoline Kettering’s engineers used for the experimental six, Kettering remarked, “It is understood that such a fuel could be made and sold in large quantities at a comparatively small premium.” The oil industry didn’t necessarily see it that way, because for them, there was a huge cost difference between “making higher octane fuel available as one grade in certain markets” and “effecting a large general increase in octane numbers.”
The price premium for premium-grade gasoline seems to be highly variable. Here, gas is currently well over $5 a gallon, but premium is barely 10 percent more than regular. When I went by it the other night, the station down the street (the one that’s had tanks and Sambars and such parked out front) was at I think $5.459 or $5.499 for regular, while premium was $5.999. Both are pretty high for U.S. gas, but at that level, the extra 50 cents or so doesn’t seem like a big deal comparatively.
And I’m waiting for someone to comment that the reason the Studebaker V8 was so heavy is because they were anticipating the higher compression ratios to come. That just doesn’t fly.
Speaking of weight, you said the Olds V8 was 35 lbs heavier than the Cadillac? I did not realize that and I’m a bit surprised. But then they both were punched out to about 7 liters so I assume they were very close in physical size too.
Give it time. Or comment with a link on the Hemmings article on the Studebaker V8.
Studie engine was so heavy because they didn’t have enough engineers to fully analyze everything and a lot of eyeball estimating on the conservative side created a chunky beast.
The only thing to possibly support the argument is that they didn’t design the engine to allow for much of an overbore. Either the people who established the bore centers etc. were incompetent or they were planning on higher octane gasoline. Probably beyond their capability.
The weight difference between the Cadillac and Oldsmobile V-8s were probably close to 70 lb. Cadillac listed the dry weight of the new V-8 as 699 lb including flywheel and manual clutch. Oldsmobile put their V-8 at 745 lb with flywheel. That’s a difference of 46 lb, but it’s actually more than that because Oldsmobile didn’t offer the V-8 with manual transmission in 1949, so the Cadillac weight included a 22½ lb clutch assembly the Olds didn’t have.
Unfortunately, while Cadillac published a detailed weight breakdown, Oldsmobile didn’t, at least not that I’ve been able to find (although they did release exterior dimensions, which I don’t have for the Cadillac). I don’t think they’re terribly different in physical size, but I’m pretty sure the Olds block is somewhat heavier, and the Olds is probably heavier in other areas in a “few pounds here, few pounds there” kind of way.
The Olds engine was used in drag racing far more than the Cadillac was, but that might be due to fewer junkyard reclaims.
It could be argued that wouldn’t matter much because how many Chrysler hemi’s were produced yet they were plentiful enough for drag racing.
The Cadillac was used in boats in the 50s by the marine outfitter Crusader but it had a reputation for spinning bearings during high speed runs although it worked quite well in normal operation and I’m sure the Cadillac name helped appeal to customers.
For the third time in a row I typed a relatively short comment on an article here, only for the page to reload on its own and wipe it out. Nice site but hopelessly unusable.
Odd. That’s not ever happened to me.
or me………………
It appears to be dependent upon the browser and OS used. I don’t see it as much on the laptop but using an iPhone with Firefox the aggressive popups are constantly refocusing/sliding the text content around, blocking the text and usually reloading the entire page every few minutes.
I have a similar issue with my Samsung S23 and Firefox mobile. All is fine with Chrome.
Yeah I had this issue with Safari on my iPhone, also seemed to be more prevalent when I had a few other tabs open.
One of my aunts had ’55 Oldsmobile Super 88 hydramatic. Among its many attributes was that it would get 20 mpg at 70 and 18 mpg at 80, using premium fuel. Even considering it would have had an optimistic speedometer/odometer as was the common deceptive practice at the time, this was quite impressive and a frequent topic of conversation.
I would nominate this model as the best car in the world, at that time, for the money.
My grandmother had a ’50 Olds 88, mostly driven (and worn out before I could get to it) by an uncle and his sons. Their take was that it could be a handful to drive on a slippery surface, what with the mandatory 1st gear start of the contemporary Hydra-Matic, skinny bias-ply tires, and forward weight bias. Other than that, they felt the coil-and-link rear suspension setup Olds used could set up a disconcerting side to side swaying at higher speeds, which may have been why Olds went back to Hotchkiss drive for several years, I dunno. It didn’t keep them from being Olds loyalists for decades, however.
Most owners of non-daily drivers -especially classic car owners- use premium gasoline here, best known and readily available are Shell V-Power (RON 98) and BP Ultimate 98 (ditto). Apart from the octane number, there’s the bioethanol-factor. Both Shell and BP claim their premium fuels are ethanol-free. Hooray!
The price difference can be ignored, given the low yearly mileage of classic cars. And owners of exclusive, high-end/high-performance cars don’t give a damn about the price difference either.
Our 94 octane fuel was ethanol free until recently. The powers that be legislated the end of ethanol free fuel two years ago. I saw the fuel consumption of my Golf increase by 5%.
I could get away with 91 for my car but 94 is only a few cents a litre more.
Bioethanol is highly hygroscopic. Crap fuel for non-daily drivers/classic cars, in other words.
1. The Olds engine was heavy in part due to the huge rear overhang–the cast-iron block extended rearward, surrounding the flywheel/torque converter area. This is visible in the second-to-last photo in the article. Chrysler did the same thing with the Hemi for the early years, but redesigned the block to eliminate it later. I don’t know about Cadillac.
2. “Phillips 66” and “Union 76” were both referencing their octane ratings in the early years.
3. The “W-43” is not DOHC. It is 4-valve per cylinder, but using pushrods and a single cam. Olds did build a DOHC experimental engine, but it wasn’t the W-43. There’s a poor photo of the DOHC engine towards the end of this article:
https://www.hagerty.com/media/automotive-history/oldsmobiles-w-43-v-8-engine-was-killer-32-valve-prototype/
Oops, thanks. Fixed now.
Regarding the block, I assume the “overhang” you mean is the upper half of the flywheel housing, which was integral with the block. The attached photo shows what it looks like from underneath on a bare block.
Cadillac did the same thing, for the same reasons: They wanted to minimize flywheel displacement, since GM Research Labs had found that that was a big contributor to engine roughness. Withrow and Fry wrote in 1944, “the crankshaft itself is able to follow the more rapid pressure development [of combustion] rather closely, whereas the flywheel displacement lags behind the pressure development by several crankshaft degrees. Consequently, the inertial forces of the flywheel cause it to be set into oscillatory vibration.” Stiffening the flywheel housing reduced that displacement.
When the military, and then the airlines, transitioned from piston engines to jets, very high octane alkylate gasoline refining capacity was available for automotive fuels. Chevron 100 RON became available in 1958. Im sure there were similar products in other parts of the country about the same time. Compression ratios above 10:1 definitely needed it. In the 1960’s we had Chevron Custom Supreme, 104 RON at third pump. Richfield Boron was the other option for those of us on the West Coast.
Even at 10.5:1, our ’65 327 (relatively big bore, short stroke) pinged on normal premium and really liked the expensive stuff.
Even with wedge combustion chambers and good quench, the larger the bore, and a singe spark plug, detonation was a problem, as bores kept getting larger. 12:1 was only going to be achievable with diesel-style direct injection.
I don’t know if we should read anything into this at all, but the 288-in^3 displacement of the stillborn Kaiser V8 was very close to that of the proposed Oldsmobile high-compression V8.
Let me check something…
“You women have heard of jalopies
You’ve heard the noise they make
But let me introduce my new Kettering ’88
Yes, it’s straight, just won’t wait
Everybody likes my Kettering ’88
Baby, we’ll ride in style
Movin’ all along
V-8 motor and this modern design
Black convertible top and the gals don’t mind
Sportin’ with me, ridin’ all around town for joy
Blow your horn, baby, blow
[Saxophone Solo]
Step in my Kettering and don’t be late
Baby, we’re pullin’ out about a half-past eight
Goin’ ’round the corner and get a beer
Everybody in my car is gonna take a little nip
Move on out, oozin’ and cruisin’ along”
Yup, just as I suspected.. Calling it the “Rocket” was a MUCH better decision than “Kettering”.