1929 Cadillac/LaSalle Synchro-Mesh Transmission – “Shifting Is Made Simple And Clashing Of Gears Is Ended”

 

Before about 1930, most cars used unsynchronized sliding-gear manual transmissions, which were aptly nicknamed “crash boxes” because shifting without clashing gears took real skill. In 1929, however, Cadillac and LaSalle introduced a revolutionary new feature: Silent Synchro-Mesh, which allowed clash-free shifting. Here’s the story of how engineer Earl A. Thompson developed that important innovation and sold it to GM in the 1920s, a rare case of an outside inventor making good in Detroit.

 

Earl Avery Thompson was born in Elgin, Oregon, in 1891, and studied engineering at Caltech and Oregon State University in 1910–1911. He apparently never graduated, perhaps for lack of money, but in those days, two years of college was two years more than most Americans had, and it was enough for him to go on to a successful career as a professional engineer. (Incidentally, Philip Gott’s frequently repeated claim that Thompson roomed with cartoonist Rube Goldberg in college is wrong; Goldberg graduated from UC Berkeley in 1904.)

 

However, it was while Thompson was still at OSU that he first became interested in transmissions and the problem of gear-shifting. Here’s how Thompson himself explained it to the Detroit News in January 1931:

“It was during my school days in Portland, Ore., years ago that I first conceived the idea of a transmission which today is known as synchro-mesh,” said Mr. Thompson.

“The late W. W. Cotton, general attorney of the Union Pacific Railroad, was a neighbor and friend of our family at that time and he frequently allowed me to drive his big [Pierce-Arrow] four-speed automobile. It was a cumbersome thing, the gears were hard to shift, and it seemed that the clutch would spin forever when you did try to shift. Why, at the rate of 10 miles an hour you had to wait 10 seconds before a shift could be made without clashing.

“Naturally, this didn’t appeal to a college youth then any more than now. So I decided to improve the shifting apparatus. I discovered there was an interlocking mechanism between the gear lever and the clutch throw-out which made it impossible to engage the clutch when the gear was in neutral. I made a few changes which remedied this and enabled Mr. Cotton to shift silently at definite car speeds while other motorists were clashing all over town.

“This experiment suggested the possibility of a transmission that would permit silent gear shift at all car speeds. The development of this thought resulted in the invention of the synchro-mesh. This transmission consists basically of a small friction clutch which causes the two spinning members of the gears which are about to be meshed to revolve at the same speed. … Thus shifting is made simple and clashing of gears is ended.”

Thompson didn’t immediately pursue this idea, instead buying a small electrical engineering firm in Eugene and then becoming a consulting engineer in the Portland area, sometimes working with his younger brother Kirk, who had also studied engineering.

 

However, the problem he’d noted wasn’t an isolated one. The four-speed transmission of the Pierce-Arrow may have been particularly stubborn, but the sliding-gear manual transmissions used on most cars of the era were challenging to shift. Gear-changes required meshing gear teeth, so if the driver didn’t perfectly match speeds (which many early transmissions and clutches didn’t exactly facilitate), graunching and grinding were routine. The British automotive writer Michael Sedgwick later remarked that one of the reasons Americans became so partial to engines of enormous displacement was that once you got into high gear, you could mostly stay there, minimizing the usual cacophony of clashing gears.

 

For Americans, the main alternative was the two-speed planetary transmission on the Ford Model T, which was different, but not necessarily better. Thompson and his brother later recalled their miserable experience shifting their Ford over and over again on Oregon’s rough, hilly roads, which left them both thinking that there had to be a better way.

 

Around 1915–1916, the brothers drafted and built a prototype of a synchronized transmission, which they installed in a six-cylinder Buick. This was just a side project for the elder Thompson, who was now running a water company in Jefferson County, but in 1918, he applied for a patent on his transmission design, which was finally granted in November 1922 (as US1435430).

 

Thompson called his invention an “Automatic Gear Shifting Mechanism for Sliding Gear Transmission,” but it was really a semiautomatic preselector transmission. The driver used a small selector lever on the steering wheel to choose the desired gear, which was then automatically engaged when the driver depressed the clutch pedal.

 

Although Thompson wasn’t able to raise the capital to do anything with this design himself, in April 1922, he made the first of three trips to Detroit to pitch his idea to automakers, without any success. Afterward, he and Kirk, who was now working for a Portland Cadillac dealership, bought a new Cadillac 61 and fitted it with the latest prototype of Thompson’s transmission.

 

Thompson then drove the car to Detroit in September 1923 for three months of additional meetings, this time getting some mild encouragement and a few suggestions from Cadillac chief engineer Ernest W. Seaholm. Seaholm later recalled:

The gear box was full of a sort of bewildering assortment of mechanisms that had been whittled out in some local machine shop in his western state. It lacked a professional touch in design, but it worked — and intrigued me.

It wasn’t until sometime in 1924, after Thompson again made the drive from Portland to Detroit, that he got some genuine interest from Seaholm and Cadillac’s managing director. (Seaholm’s account says this was Lawrence P. Fisher, but I think he was misremembering — in 1924, the head of Cadillac was still Herbert Rice, whom Fisher succeeded a year later.)

 

Seaholm recalled:

I introduced him [Thompson] to the Corporation’s New-Devices Committee, recently formed to handle matters of this sort for the Corporation as a whole rather than through our individual companies. They too manifested interest and some agreement was reached pending check-up on patents and other details.

From here, for a time, the going was not smooth. He was referred here and there, but no one was interested in taking on the development. So, one day, he showed up in my office, this time discouraged and ready to pack up and go back home. This shocked me, for it seemed that here we had something of great potential and could not afford to let it slip through our fingers, and we made an agreement for him to stay on until together we could bring out a couple of production prototypes of his transmission.

With this deal finally in hand, Thompson relocated to Detroit to become a Cadillac consultant, taking out several additional patents on his design. According to Seaholm, Cadillac built 10 variants of Thompson’s transmission, which were run in 25 test mules for a total of 1.5 million miles at the GM Proving Grounds in Milford, Michigan.

 

The fussy, complicated preselector system was dropped: Cadillac was most interested in Thompson’s cone clutch synchronizer mechanism, which was placed between second and third gears. Reverse and low were still not synchronized, since Cadillac assumed that “shifts into these gears are usually made when the car is standing still.”

 

Cadillac put the system into production in August 1928 as standard equipment on all 1929 LaSalle and Cadillac models. GM called it Synchro-Mesh. I couldn’t find any indication that the corporation actually registered that term as a trademark, which was just as well — “synchromesh” would soon become a generic term.

 

The early Cadillac/LaSalle synchronizer design was somewhat different from the more familiar baulk-ring synchronizers found in modern manual gearboxes. Here’s an illustration from MoToR showing the synchronizer components:

 

MoToR explained the action like this:

READING from left to right, Fig. 2 [below], shows how the sliding coupling moves from neutral to engage intermediate or second gear, being actuated by the gearshift lever working through the usual shifter shaft and shifter fork. Mainshaft second gear is in constant mesh with its mate on the countershaft and when not in use rotates freely on a bronze bushing splined to the mainshaft. To engage second gear the sliding coupling which is splined to the mainshaft is moved to the left until it meshes with the internal gear which is integral with mainshaft second gear, thus locking second gear to the mainshaft. High gear is engaged in the conventional manner by moving the sliding coupling to the right until it meshes with the internal gear on the clutch shaft.

It is obvious that in any “sliding gear” transmission, silent meshing can be secured only by bringing the meshing teeth to exactly the same speed. In the Cadillac-LaSalle transmission this is done by individual metal-to-metal cone clutches on the mainshaft engaged temporarily to bring the teeth of the high or second internal gear to the speed of the teeth on the sliding coupling splined to the mainshaft.

Reading from left to right, Fig. 3 [above], shows the action of the cone clutch in meshing second gear. The three positions of the clutch mechanism correspond exactly with the three positions of the shifter gear, Fig. 2. Shifter gear and cone clutch are shown separately for clearness.

It should be noted that with plate clutch disengaged and gearshift in neutral, mainshaft second gear, and the countershaft assembly and the plate clutch shaft assembly rotate freely as a single unit, since the mainshaft intermediate gear is free to rotate on its bronze bushing. Therefore when second gear cone clutch is engaged it is able to alter the speed of this whole assembly to the point where mainshaft second gear is rotating at mainshaft speed, thus permitting silent meshing of the shifter gear.

Similarly when high gear cone clutch is engaged it alters the speed of the whole assembly to the point where high internal gear is rotating at mainshaft speed. After the gears are meshed, the cone clutch is disengaged automatically and moves back to neutral.

The actual movement of the control yoke was controlled by a cam-and-roller system that acted on the high and intermediate shift fork through two hydraulic plungers:

 

MoToR explained:

The movement of the roller plunger is hydraulically controlled in order to properly time the period of clutch engagement. It is apparent that time is required to bring the internal gear assembly up to mainshaft speed. Actually the time is so short as to be negligible from the operator’s standpoint, but nevertheless time is required. If the roller plunger were backed only by the spring shown in Fig. 3, the roller would slip back over the flat part of the cam so quickly that the cone clutch would not remain in engagement long enough to bring the assembly to mainshaft speed. To delay this action (to increase the period of cone clutch engagement) the base of the plunger butts against a piston which works in an oil chamber, Fig. 4, and the downward speed of the plunger and piston is limited to the speed at which oil will flow from the oil chamber to the chamber inside the plunger as indicated by the arrows.

This arrangement required an additional relief valve to compensate for the greater thickness of the oil in cold weather.

 

The 1929 Cadillac owner’s manual stressed that Synchro-Mesh required a different shifting technique than drivers of the time were used to:

With the Cadillac Synchro-mesh transmission there is no necessity either for the hesitation in neutral or for the rapid movement of the lever during the later part of the shift. Instead, the movement of the control lever should be one smooth, continuous motion. … The ease and certainty with which a noiseless shift can be made with the new transmission, may tempt some drivers to perform “stunts” for which it is not intended. The synchronizing principle makes it possible for the drive to make use of intermediate speed at any time that it is an advantage to do so. … There is no advantage to be gained, however, in using intermediate at speeds above 30 miles per hour, and any attempt to shift at higher speeds should be regarded as abuse.

Drivers used the Synchro-Mesh gearbox so enthusiastically that Cadillac had to beef it up considerably for 1930, also making some design improvements to the synchronizer mechanism. As Synchro-Mesh and competing synchronized transmissions were more widely adopted throughout the industry, there would be further changes, but Thompson had established sound basic principles on which to build.

 

Thompson became a Cadillac employee in September 1929, but he and his younger brother continued to receive option payments and royalties from GM for a while; Kirk Thompson had used his to buy a Cadillac-LaSalle dealership in Spokane, Washington. In 1930, GM made an attractive offer to buy the patents outright for a lump sum — by some accounts, $1 million, a huge amount of money at the time.

 

General Motors got their money’s worth: Synchro-Mesh was one of the most important developments in automotive engineering. I think it’s fair to say that most Americans still didn’t relish shifting gears manually, but “Silent Synchro-Mesh” made it manageable, and paved the way for the slicker all-synchro gearboxes of later eras, which are sometimes even fun — not an adjective most drivers of the 1910s and 1920s would have ever used.

 

The development of Synchro-Mesh would have been a milestone even if it had been Thompson’s only notable achievement, and it wasn’t. Between 1932 and 1940, Thompson led the development of what became the first commercially successful fully automatic transmission: GM’s monumental Hydra-Matic Drive, for which Thompson would eventually receive the prestigious Elmer A. Sperry Award.

Here’s a 1936 Chevrolet training video, made by The Jam Handy organization, explaining how a three-speed transmission transmits power, with a brief explanation of Synchro-Mesh gear synchronization towards the end:

Related Reading

The World’s First Automatic Transmission – The 1904-1907 Sturtevant Automatic Automobile (by me)
Planetary (Not Interstellar) Overdrive (1934-72) (by Paul N)