Before the computer age, the automatic transmission was the most complex thing on an automobile. For internal combustion engine vehicles, some form of variable gearing is necessary, so having a reliable link between powerplant and wheels is vital.
Needless to say, complex things don’t always work the way they should. Sometimes, engineering incompetence is the culprit. Sometimes it’s penny-pinching.
The modern automatic transmission got its start in 1934, when REO—a Lansing, Michigan, automobile and truck manufacturer—developed the Self-Shifter. A clutch pedal was required for shifting into forward or reverse, but not for shifting between forward gears. This was a “nearly-there” effort about which automotive historians generally forget. REO, which Ransom E. Olds started after leaving Oldsmobile, spent a fortune developing this transmission, which did enjoy some success and reasonable reliability. One downside was that while the transmission did indeed shift itself, it always happened at the same set speed. Not great for economy or performance.
General Motors had its own false start with the Automatic Safety Transmission, introduced by Buick in late 1937, the year after REO quit the car business and began concentrating on trucks. It was also a “nearly-there” effort. Reverse could only be had by stopping and using a clutch, otherwise no clutch pedal was needed. The transmission shifted between gears while moving forward. The rather-expensive ($59) option was dropped mid-1938 and considered a failure. Bear in mind, cost of a new Buick started at $945. Today that’s like adding a $1000 option on a $16,000 car.
A winner finally arrived in 1940 when General Motors introduced the Hydra-matic on Oldsmobiles, giving car buyers a “true” automatic transmission. “No clutch to press! No gears to shift!” This transmission had a 15-year gestation period under Cadillac engineering, from the late 1920s until 1934, when funding for it was dropped—no doubt due to the Depression. Oldsmobile General Manager Charles McCuen lobbied hard to have it brought to Olds, and his persistence paid off.
The four-speed Hydra-matic offered a huge advantage in that it could shift at varying speeds based upon car speed and throttle position, just like today’s step-gear ratio automatics. The optional cost was $57. Cadillac was “allowed” to adopt the unit as an option for 1941. By the time WWII began and car production ceased in February 1942, more than 200,000 Hydra-matic units had been manufactured and sold.
The invention and utilization of the automatic transmission certainly changed the industry, but manufacturers didn’t always get it right. Like we said, complex things don’t always work out the way they should. Here’s a dirty half-dozen that should have stayed on the drawing board.
GM’s Roto-Hydramatic 375 “Slim-Jim”
This transmission, used in 1961–64 Oldsmobiles and Pontiacs, was a dead-end technology very loosely based on some of the technology employed, starting in 1956, to improve the original Hydra-matic. The intent was to reduce production costs, reduce the size and weight of the unit, and improve smoothness of operation. (Smoothness was “in.” Efficiency, not so much).
The transmission ended up being one of the most unreliable ever produced. Failures occurred with depressing regularity. One problem was with a fluid coupling that filled and emptied in order to shift; it proved problematical and complex.
Chevrolet’s TurboGlide and the similar Buick FlightPitch
In this era, not only was each General Motors division responsible for its own engine designs, but—with the exception of Cadillac, Oldsmobile, and Pontiac—each also developed its own automatic transmissions. Still, there was much cross-pollination in GM engineering, and in this case it resulted in a two-fer of abysmal proportions.
The virtually identical technology behind these two distinct transmissions, which were used in 1957–61 Chevrolets and 1958–59 Buicks, was intended to make for ultra-smooth progress by using torque converters that were made up of multiple turbines, each connected to its own planetary gear-trains. Motion was achieved more by “shunting” than “shifting,” as one turbine dropped out of play and another picked up the slack. Nominally, there were three ratios forward, with a variable pitch stator on breakaway adding a little extra torque.
The complex nature of these beasts proved to be so bad for business that Buick pulled the plug after 24 months. Chevrolet soldiered on and tried to improve their beast—unsuccessfully.
For decades, mail-order parts vendor JC Whitney made a small fortune selling Powerglide automatic transplant kits for cars affected by a TurboGlide or FlightPitch—requiring a stop at a salvage yard for a used Powerglide two-speed automatic, of course. Say no more.
General Motors Turbohydramatic 200
Let’s keep this simple. Developing a small three-speed Simpson-gear-train automatic for the Chevrolet Vega and six-cylinder Nova compacts certainly improved the performance of these cars over the two-speed Powerglides. So in that regard, mission accomplished.
But putting the same transmissions (which featured robust pressed-metal planetary gears) behind V-8 engines in reasonably heavy down-sized/full-sized cars—all 1976–80 General Motors products in the U.S. except Cadillac—created reliability issues in spades.
Putting these behind V-8 diesel engines made for double-jeopardy.
These rather blatant copies of the Chevy Powerglide two-speed automatic were admittedly reliable and decent units, but that’s where the good news ended.
Building a cheap-o alternative to save a few coins was a rather cynical ploy, although 1959 Ford buyers could still opt for the much heavier Fordomatic three-speed units if desired… until 1960, when the all-new compact Ford Falcon came out.
Sure, the Falcon was a huge best-seller and genuine success story. But a tiny, 144-cubic-inch 85-horsepower six-cylinder engine combined with only two speeds forward—and a wide ratio between them—meant the little Falcon Six would face-plant again and again. Can you spell S-L-O-W? Falcons got a three-speed automatic until 1964, so thankfully (wisely), no Mustangs were ever built with a Fordomatic two-speed automatic.
Pontiac fans will cringe, but one of the engineers responsible for this disaster was none other than John Z. DeLorean, who worked as lead engineer on Packard’s automatic transmission team.
Originally, Packard had more or less copied the very inefficient Buick Dynaflow, which relied upon a torque-converter as the primary means of varying speed and load between the engine and the road. Packard added a lock-up torque converter to vastly improve highway fuel economy, and like Buick, it had a manually-selected “emergency low” planetary gear train (which also allowed reverse) from 1949–54.
For 1955–56, the Ultramatic was changed so that it automatically shifted from low to high, then “shifted” again as the torque converter clutch locked up. The problem was the development wasn’t really up to snuff; and using Packard luxury-car buyers in beta-testing really harmed the company’s reputation.
Another issue was the extremely powerful and large all-new Packard V-8 engines had too much torque for the transmission. Finally, even Packard engineers admitted that two-speed automatic were literally going to be passe’ by 1958, so the huge corporate expense and effort put towards developing Packard’s own automatic never paid off. It was a money pit for car owners and stockholders.
The Twin-Ultramatic was one of several things that helped kill off the automaker, which offered gussied-up Studebaker-based pretender cars for 1957–58 before finally expiring.
Oh dear, I know I’m going to get letters from Poncho fans, but here goes anyway.
DeLorean, who joined GM after Packard, opted to join the near-moribund Pontiac Division. As chief engineer, he began to breathe life into the organization, culminating in his extremely successful GTO in 1964.
But when GM’s brass wanted to foist a version of the Corvair onto Pontiac in the late 1950s, DeLorean balked—and rightly so. The Corporate 14th floor brainiacs pushed the matter, insisting that Pontiac had been “scheduled” to receive Corvair-based transaxles starting in the fall of 1960—GM’s way of saying “Pontiac-ize the Corvair.”
DeLorean and his team had other ideas. Buick and Oldsmobile were collaborating on a small compact with an aluminum V-8 for 1961 introduction, and Pontiac signed on to use the same body from Fisher instead of the Corvair body shell. But the Buick-based alloy V-8 was too expensive, so with no budget for a new engine, engineers essentially took the Pontiac 389 V-8 and sawed it in half, giving the 1961–63 Tempest a massive slant-four engine, which did a terrific job of emulating a paint-shaker under the hood.
Meanwhile, only 1 percent of buyers opted for the more expensive Buick alloy V-8.
In anticipation—by more than 15 years—of the Porsche 928, DeLorean and his team took the Corvair-based gearboxes and put them in the Tempest, culminating in a highly advanced front engine, rear transaxle set-up with all-independent suspension.
The problem was, the Corvair transaxle was developed for a rear-engine car, not a front-engine car, and the power flow literally meandered back and forth within, which sucked power (and fuel). Worse, it was only a two-speed automatic.
Even more problematic was the 1963 Tempest’s optional Pontiac V-8 of 326 cubic inches. The torque-filled Pontiac V-8 overwhelmed the little Tempestorque automatics, causing reliability issues. Sometimes the flexible driveshaft, connecting engine to transaxle, would snap.
The following year, Pontiac’s Tempest (and subsequent GTO) had a fully-conventional drivetrain layout with front engine and transmission and a live rear axle robust enough for that 389 V-8 and tri-carb setup option.
So there you have it—six lackluster transmissions to slow the march, or at least the shift, of automotive progress. Will future generations remember Ford’s PowerShift or Honda’s “glass” automatics in the same way? Only time will tell.