The unusual driveshaft setup on the 1961 Pontiac Tempest has been called a “rope drive” and described as a giant speedometer cable, but how it really works is far more interesting.
When General Motors unveiled the ’61 Pontiac Tempest in the autumn of 1960, the workbench magazines like Popular Science (Sept. 1960 issue illustration above) were flabbergasted by its unusual drivetrain. Up front, the Tempest Trophy engine was essentially the familiar 389 Pontiac V8 with its left bank removed, creating a 194.5 CID slant four. At the opposite end of the 112-inch wheelbase was a Corvair-based transaxle with swing-axle suspension, highly unorthodox for an American car at the time. And between them, transmitting the power to the rear wheels, was a driveshaft with an obvious bow in the middle that, for many, has seemed to defy understanding ever since.
The Tempest propeller shaft has been described as a “rope drive,” and in side view it does sort of resemble a jump rope with a child holding up either end. But any similarity ends there; it doesn’t function like a rope in any real sense. It’s been called “flexible,” when in fact it’s not (although it is indeed bent). It’s been described as a “giant speedometer cable,” but that’s not accurate either. How the driveshaft works is actually far more interesting.
Courtesy of eBay, here’s a closer look at the stamped-steel torque-tube housing and the driveshaft that runs inside it. The driveshaft is in fact a stiff, solid length of steel, but we won’t call it plain. It was constructed in SAE 8660 nickel-chrome-moly alloy, delicately ground, shot-peened, magnafluxed, and coated with a scratch protectant. For the automatic-transmission cars, the shaft was 87.25 inches long and .650 inches in diameter, while the manual-transmission shaft was .750 inches in diameter and 5.25 inches shorter to make way for a separate clutch shaft inside the bell housing. (Because the driveshaft carried straight engine torque rather than multiplied transmission torque, its diameter could be remarkably small, like a transmission input shaft.)
This carefully prepared shaft was then forced into an arc at installation by the positioning of the engine, transaxle, and torque tube. Almost three inches of bow was installed in the shaft when it was bolted in at the flanges, imparting a uniform stress along its length and forming a radius of around 36.5 feet. That is, if the shaft completed a full circle, its diameter would be nearly 73 feet. The torque tube housing does not support the driveshaft in any way but only maintains the precise alignment between the engine and transaxle, holding the shaft in its curved position. Folks are known to ask how a shaft bent in this manner can still rotate and transmit torque. That might be answered with another question: What else could it do?
While the most obvious purpose of the drivetrain setup was to allow a flatter floor in the passenger cabin, there were other benefits. First, the long, thin driveshaft dampened the significant torque reversals of the slant 4 engine (which in a four-banger are more than 100 percent) and when bent into an arc, the shaft’s critical speed is above the engine’s operating range. Next, with the engine supported at the rear by the torque tube and transaxle, the drivetrain absorbed the engine’s torque reaction and vibration periods. The rubberized front engine mounts could be pillow soft, allowing the Trophy 4 to dance around in the engine compartment, isolated from the passengers in the cabin.
While the bent driveshaft seemed to be tailor-made for the Tempest, in fact it was originally developed for Pontiac’s full-sized cars. The originator of the idea, by all accounts, was the division’s young assistant chief engineer, John Z. DeLorean, and the system was tested on a succession of ’57, ’58, and ’59 full-size GM cars before the Tempest opportunity arose by engineer Bill Collins. While the first-generation 1961-63 Pontiac Tempest had a few problems, the driveshaft wasn’t one of them. Reports are the setup worked just fine. But it’s interesting to note that GM hasn’t used it since.
If you’d like to read more about the ’61 Pontiac Tempest and all its unusual engineering features, the best single source we’ve found is Wick Humble’s excellent, in-depth story in Special Interest Autos #48, Nov-Dec. 1978, which we consulted in preparing this piece.
Fascinating. With more reasonable weight distribution, I expect this car’s axle jacking behavior was better than the Corvair.
My dad told me about seeing Tempests run an endurance race at Virginia International Raceway. There were issues with the transaxle overheating. The “in-the-heat-of-battle” pit stop fix was to knock holes in the trunk floor with a hammer and cold chisel and dump coolers full of ice in. The water from the melting ice cooled the transaxle unit enough for the cars to finish.
Sorry – it wasn’t at VIR, it was at Marlboro, Maryland… https://littleindians.com/racing/1961-marlboro-twelve-hour
Didn’t the Chevy II have a four cylinder engine in 1960?
The “Falcon Fighter” Chevy II was first produced for the ’62 model year.
’37 GM products already flattened the floor, by using a 2-pc. driveshaft; an extra U-joint was located about 2/3 of the way from the engine to differential and the trajectory of the shaft pitched slightly downward past that point, thus eliminating the “tunnel” in the rear floor. This Tempest setup looks like it could be tricky if the unit, which is essentially under spring pressure sandwiched between engine and diff., ever had to be disassembled once in place. I wouldn’t trust a design which, in order to achieve relatively trivial advantages, depended upon such precise fabrication and (esp.) installation tolerances that have to come off a mass-market assembly line.
GM continued the two-piece driveshaft for many years. I once owned a 1965 Buick Electra with it, and the drive tunnel was remarkably smaller than an otherwise similar 1965 Chevrolet Impala. The same design is still used (for different reasons) on large box trucks, where it is easily visible.
Coincidentally, Herb Adams mentioned to me recently that you could save some fuel by going back to a bevel gear rear axle, evolving in the other direction. With the size of trucks these days, there’s no reason to use a hypoid ring and pinion on a truck. Makes sense, but I just never thought about it.
hypoid gear has power loss, ford 9″ diff. the pinion gear is 1.5″ below center of ring gear, chevy 10 or 12 bolt is .75″ below center of ring. theory, move pinon to 0″(zero) off set, would be least power loss, straight spur bevel.
Droopy is not an engineering term I would apply to the 1961-63 torsion bar drive shaft..It is carefully held in a designed radius inside the torque tube. In addition to protection from the elements by the torque tube, it was not coated in shellac but a very sophisticated plastic coating.
I do not think Herb Adams is advocating changing the entire drivetrain for a new gear set.
Kind of surprised that as the shaft spun the arc did not, just like the arc in the aforementioned jump rope.
I owned a 62 transaxle Tempest. The linkage for the 4 speed transmission was the weak link. It was so sloppy that it would alow the side shaft to engage two forward gears at once. Shift into 3 and you would blow the entire trans housing apart. This happened more than once. The handling and weight distribution was very good, but dependability was awful.
My first car was a ’61 Tempest wagon /4 barrel / auto. The automatic had its own problems but it was still a pretty tough car. I actually broke the driveshaft tunnel. But all in all it couldn’t been a better car to learn about mechanics and spend time with my Dad.
Interesting comment about shift linkage. That’s why the Classic Car Club people all covet Buick’s manual-shift transmissions. They’re all one-link control, so even if the pins become worn, you’ll never slam the transmission into two gears at once. You think GM would have smartened-up to the virtues of one of its own companies, but that was not to be. Even Cadillac was prone to linkage failure.
Interesting. A drive shaft that does not whip. GMH here in Oz 20-30 years on made Holdens with a 2 piece tailshaft and when accelerating from low speed you can feel the tailshaft ‘whipping’ and when the rubber mount tears becomes near undriveable.This describes every Commodore from 78-15.
I guess Ralph Nader did not hear about the Pontiac and its swing axle,,, all of which are at least in part cloned from Mercedes of the period. The GMH IRS from around 92 on was too a swing axle,, one that eats tyres and while unusual can too get the dreaded tuck and roll.
GM went through a phase here fascinated with a flat floor, and it’s a fun historical curiosity that the stars aligned so that they were able to commercialize this version of it. They were simultaneously working on 3 versions, the Toronado, the Corvair, and the Tempest. The Toronado took almost a decade to come to market.
Had a Great Aunt that had one of these cars. She only kept it a few years, seems as advanced as the driveline was supposed to be, the rest of the car was constantly falling apart. Quality definitely wasn’t job one in those days…
Pontiac didn’t want the Corvair platform – Pontiac General Manager Bunkie Knudsen thought it was dangerous, and was furious at Ed Cole for backing him into a corner on it. The shaft was produced in-house at Pontiac. However, the spec on the shaft was so tight it made production difficult; the initial reject rate was huge. The stress of the experience is cited as as the reason works manager Buell Starr took his own life. The later 421 SD AFX Tempests from 1963 would inflict revenge on the doubters, using a stock V-8 shaft to set records, win the Daytona Challenge Cup 250 and become the scourge of NHRA experimental stock racing.
My parents had a ‘61 when I was a senior in high school. It was fun car to drive. Not hot off the line, but going around corners at a fast clip roll the outer when under, which then raised the rear as it straightened out. Quite a thrill the first couple of times that was experienced.
Very accurate article. As the transmission/driveline engineer on the Tempest ,it was 24hour job and I did not get an heart attack !! can anyone believe on the first day of production, I was on the line, adding the anti-rattle coupling inside the transmission ??!! The whole project has interesting stories.
Thank you. Very high praise coming from you. It means the world. I’m glad we were able to get it right. It’s not always easy. -McG
Bill you certainly have had a great career! And this is just one of those stories. I did 10 years mostly at Ford and Opel. My stories are nothing like as exciting.
Had one. Horrible car. Lousy mileage, weak engine, constantly overheated and at 20 mph the “flexible” driveshaft snapped and the rear portion fell onto the asphalt. DeLorean must’ve been high on something when he worked on that model.
These shafts were not originally coated with scratch protectant (described by my Automotive Lab professor in College for automotive engineering as being “shellac”). Because of the droop, the stresses in the solid metal drive shaft (at least I THINK they were solid metal) were so significant that the tiniest scratch on the surface of the droopy shaft would become a focal point for stresses and the shaft would invariably break at the precise location of the scratch. This droopy shaft HAD to be inside a protective tube. If it had been left exposed, every stone or grain of sand that hit the shaft while driving could potentially be the one to find a thin spot in the shellac and cause a scratch. So the tube was really a necessity to protect the highly stressed floppy-shaft. The discussion in the article about the critical speed for the shaft was higher than the maximum operating RPM of the engine… that refers to the speed the shaft has to rotate at before it starts whipping around like a big whisk instead of laying droopy and spinning in a nice smooth controlled manner. I never heard what that critical speed might be for the Tempest floppy shaft was. But say the engine rev’d no more than 6250 rpm (a guess, could be a bit lower or higher) and the shaft would remain doopy and controlled at any RPM below, say, 8613 RPM… you had a pretty big safety buffer before you’d have to worry about the shaft not staying in the proper drooped shape as it spins. It was quite an interesting engineering exercise.