We’re going to take a different approach for this week’s article. Instead of presenting another history, we’ve decided to give you a look at the way we approach the research for these articles, and tackle a challenging comment posed by one of our readers: did inventor Oscar Banker design the 1937-1939 Oldsmobile/Buick Automatic Safety Transmission, the predecessor of Hydra-Matic?
SOURCES, FACTS, AND OTHER SLIPPERY SLOPES
Periodically, we get comments or e-mails from readers taking issues with our facts. These are sometimes polite, occasionally quite snarky, accusing us of malign agendas of various kinds. (This fortunately doesn’t happen that often, as it’s bad for morale.)
We generally take those comments very seriously. If we have an agenda, it’s that we hate being wrong. That’s not to say we don’t make mistakes, because we do, for any number of reasons. Sometimes our notes get garbled. Sometimes our sources were incorrect about something we didn’t realize. (Just because something is in print doesn’t mean it’s right.) Sometimes, we took a guess and got it wrong.
If we make a mistake, we have a strong interest in fixing it. Our goal is for these articles to be canonical, something to which people can return as a reliable reference. We can’t cover every detail, and we recognize that we’ll never be able to match the detailed technical knowledge of experts in a specific vehicle or marque. (That’s why we won’t try to tell you how to troubleshoot mechanical problems, or how to authenticate or restore some collector car; we’re really not qualified for that.) However, we want the facts we do present to be as correct as we can make them.
When we get a comment that challenges our facts, the first thing we do is refer back to our sources. A lot of mistakes are just that — we labeled something a ’53 when it was actually a ’54, we dropped a digit, etc. (Embarrassingly, we often have these details right in our original notes, but make an error in transcription.)
The next thing we do is consider whether the commentator is taking issues with our facts or our conclusions. We make no pretense of pure, Platonic objectivity. We take it for granted that some readers will look at the facts and draw their own conclusions; we do that all the time. We do sometimes reevaluate our theses, especially if they were based on incorrect assumptions, but we aren’t necessarily going to edit the text to accommodate differences of opinion.
If neither of those possibilities covers it, then we consider that the commentator may have information our sources didn’t cover. If that happens, we usually ask the commentator for their sources, so we can evaluate them. To our great frustration, those requests are usually ignored, which leaves us to do our own investigation.
THE MYSTERY OF OSCAR BANKER
As a case in point, a reader recently commented that our article on the Hydra-Matic omitted Oscar Banker, whom the commentator claimed was the inventor of the Hydra-Matic’s semi-automatic predecessor, the Automatic Safety Transmission.
We reexamined our various sources for that article, and found that they made no mention of Banker. That isn’t necessarily conclusive; even official histories often mention only key executives. If you read a lot of automotive histories, you might think that Harley Earl and Bill Mitchell personally designed every GM car made from around 1930 to 1980. Even the heads of the various design studios are seldom mentioned, let alone the individual stylists and modelers who created the actual designs. If people left the company during the development process, their names often vanish from the official account in thoroughly Orwellian fashion.
A web search found mention of Oscar Banker in a variety of websites related to Armenian culture. Oscar H. Banker, 1895-1979, born Asatour Sarafian, was an Armenian immigrant who lived in the Chicago area. Several of those references claim that he invented the first “practical” automatic transmission in the mid-thirties. Some suggest that he worked with GM engineers on the Automatic Safety Transmission, but that GM made various changes to the design of which Banker did not approve.
We should make clear that our interest is in the second claim — the first we’re inclined to dismiss out of hand. Identifying the “first” of anything, in a technological sense, is a difficult and futile endeavor; many things are tried decades before they become workable or popular. It’s hard to say definitively that a single individual invented the first “X,” because there’s no guarantee that someone’s mad uncle Oslo didn’t devise the same principle in his barn back in 1902, with a patent to prove it. (Your author’s father claims to have had the idea for the minivan years before Chrysler’s T-115 Voyager/Caravan, but we’re not holding our breath waiting for royalty checks.) Many engineers were working on automatic transmissions of various sorts well before World War 1. Some of them even worked, if not very well, which could also be said of the troublesome AST.
We found little useful material on Banker on the web. It appears that he had a son, Oscar H. Banker, Jr., and that he and his wife eventually settled in Bay Village, Ohio, a suburb of Cleveland, where Banker died in January 1979 at the age of 83. A discussion group archive had what appears to be a passage from some uncredited source, describing Banker demonstrating his prototype transmission to GM research chief Charles Kettering (presumably in the mid-1930s, although it’s not specified). The blog of author Robert Hull included a reminiscence of meeting Banker, but contained few solid facts.
We found that Banker wrote a paper for the SAE in early 1946 entitled “Ten years’ experience with automatic transmissions in the bus field,” to which we unfortunately don’t have access, and that he received the Automotive Hall of Fame’s Distinguished Service Citation in 1969, an award also given to notable automotive engineers, designers, and executives like Ed Cole, L.L. “Tex” Colbert, Charlie Sorenson, and Brooks Stevens. There was mention of Banker in Ralph Nader’s 1965 book Unsafe at Any Speed: The Designed-in Dangers of the American Automobile.
We discovered that Bob Hull published Banker’s memoir, Dreams and wars of an American inventor: (an immigrant’s romance), in 1983, apparently released under Hull’s own imprint, but it’s long out of print, and neither the Los Angeles city nor county library system has a copy. A number of out-of-state libraries have copies only in non-circulating reference collections, which doesn’t help.
All of this was intriguing, but frustratingly vague. The Automotive Hall of Fame and SAE references suggest that Banker was an engineer of some note, but we couldn’t find any detailed record of his work. We still had one other resource to tap, however: the records of the U.S. Patent and Trademark Office.
A patent search confirmed that Banker was indeed a prolific inventor: he was granted a lengthy list of patents between 1919 and 1976, in a variety of fields. His transmission patents range from automatic clutches to spur-gear automatic gearboxes and an early torque converter automatic transmissions. The aforementioned 1946 SAE paper indicates that at least some of these designs later saw use in buses and perhaps trucks, although we were unable to find any details.
The patent that most gave us pause was U.S. Patent No. 2,262,747, applied in September 1936 and issued in 1941, simply entitled “Automatic Transmission.” This was not Banker’s first or only automatic design, but it was the first that bore any significant to the GM designs of this period.
The abstract of this patent suggested some remarkable similarities to the Automatic Safety Transmission. Banker’s transmission was a four-speed planetary transmission that shifted automatically from first to third to fourth, just like the AST. The abstract further described the strategy of the gear ratios: that it was intended to be used with a low numerical axle ratio, so that first through third would provide ratios similar to a conventional three-speed gearbox, with fourth providing overall gearing comparable to overdrive. GM applied the same strategy to both the AST and the four-speed Hydra-Matic. A glance at the illustrations revealed that Banker’s design used planetary gearsets for its forward speeds, but spur gears for reverse, again in common with the AST. Was this, in fact, the basis of the AST?
OSCAR BANKER VS. EARL THOMPSON
Before jumping to any conclusions, we took a closer look at both Banker’s patents and the patent records of Earl Thompson and Oliver Kelley, who led GM’s Transmission Development Group.
On closer examination, Banker’s 1936 design revealed some significant differences from the production AST, as well as its similarities. First, Banker’s design did not use brake bands to hold the annulus of each planetary gearset in place. Instead, it used one-way clutches to hold the ring gears stationary in first and third, a feature GM adopted for the second-generation, dual-coupling Hydra-Matic in 1956. Second, and perhaps more significant, Banker’s transmission was not hydraulically operated; the clutch engagements that executed each shift were controlled by a system of centrifugal weights. Both the AST and Hydra-Matic (and many subsequent automatics) used hydraulic servos to control their bands and clutches.
We found a series of patents in Earl Thompson’s name for semi-automatic gearboxes, including some “preselector” transmissions, designed before he joined GM. In 1932, there was a transmission design combining a set of spur gears and an automatically controlled planetary overdrive. By 1934, there was another design using two planetary gearsets in series, one manually controlled by the driver, the other controlled automatically by a hydraulic governor. Thompson patented several variations on this design, filing the application for the last in 1937.
Thompson’s dual-planetary designs bear a pronounced resemblance to the engineering diagrams we’ve seen of the Automatic Safety Transmission. The major difference is that they had two separate driving ranges, one shifting automatically between first and second (like Low in the AST), the other shifting between third and fourth; the patent doesn’t describe a means of automatically shifting between those ranges. Those patents are followed by two more filed by Oliver Kelley, for fully automatic transmissions with fluid clutches; the latter is close to the production Hydra-Matic.
What can we determine from all these patents? First, none of Banker’s patents were assigned to General Motors, which they probably would have been if he had been a GM employee; it was assigned instead to a Chicago company called New Products Corporation. That doesn’t mean that those designs were never licensed to GM (which wouldn’t be indicated in the patent records), but it makes it less likely that he worked directly with GM on these designs. Furthermore, the patents that were assigned to GM by Thompson and Kelley suggest a fairly straightforward evolution of concepts that Thompson’s group was working on between 1931 and 1939. If Thompson and his team were aware of Banker’s designs, or if they licensed them, it doesn’t seem to have brought any dramatic changes in their engineering approach.
What about the 1-3-4 shift pattern? There are no references to it in Thompson’s patents. Reader Dan Rolfe recently referred us to Philip G. Gott’s 1991 book Changing Gears: The Development of the Automotive Transmission, which says that the 1-3-4 pattern was a design accident, not part of the original specification. The original design called for starting in Low (which engaged first gear) and then shifting manually to High, as described in the patent applications. Testing determined that if you started in High, there was not enough hydraulic pressure to engage servos for the rear band and clutch, which effectively put the transmission in first gear at low speeds. While Oscar Banker developed that pattern deliberately, it appears that GM stumbled onto it as a happy accident. It’s possible that Banker’s patent application prevented GM from patenting the 1-3-4 shift, but that doesn’t necessarily mean that one was derived from the other.
(Why not an automatic shift from first to second to third? The mechanical complexity of the 2-3 shift was a major problem; Thompson’s patent abstracts mention it in some detail and it remained a sore point even with the fully automatic Hydra-Matic. A shift from first to third was mechanically much simpler and skipping second was a lot more practical than starting in third.)
The unattributed anecdote we found about Banker’s encounter with Boss Kettering does not indicate that GM purchased or even seriously examined Banker’s design. We’re reluctant to put too much stock in unattributed quotes, but if that account is true, it would suggest that Thompson’s design was a parallel development, with some similar features, developed independently to address a common set of engineering problems. (Gott’s book does indicate that GM subsequently licensed another of Banker’s designs for use in buses, but as far as we’re aware, the coach transmissions were developed separately from the AST and Hydra-Matic.)
OSCAR BANKER AND RALPH NADER
Oscar Banker is mentioned in Ralph Nader’s Unsafe at Any Speed, again in connection with GM transmissions. Nader mentions that during an SAE presentation on the new Roto Hydra-Matic in early 1961, Banker questioned why it retained the NDLR shift pattern of the earlier Hydra-Matics (and early Powerglide and Dynaflow transmissions). Afterward, Banker wrote a letter to GM president Jack Gordon, asserting that the layout was hazardous, because it made it too easy to accidentally shift into reverse while in motion.
Thanks in part to pressure from Banker and other engineers, the federal General Accounting Office subsequently pushed for standardization of the PRNDL shift pattern for federal fleet vehicles. GM finally acquiesced, applying the PRNDL pattern to all its transmissions by the 1966 model year.
We suspect that account may have been the source of Bob Hull’s claims that Banker clashed with GM engineers over the design of the Hydra-Matic. We’re not sure, however, whether those criticisms arose from firsthand experience with the Transmission Development Group, or simply Banker’s observations of a common and popular design. GM’s patents were a matter of public record, and by 1940, shop manuals and other technical information were widely available. Banker was clearly a talented engineer who worked extensively with transmission design, so it wouldn’t be unlikely for him to have strong opinions about the design, whether he was directly involved with its development or not.
MAKING SENSE OF IT ALL
All of this is an example of what’s called inductive reasoning. We can make some reasonably logical speculation based on what facts we have, but we don’t have enough information to judge whether or not all our assumptions are correct. In short, the best we can give you is our best guess.
So, did Oscar Banker invent the Automatic Safety Transmission? Our guess is probably not. He did patent a similar design around the same time; he may well have pitched it to GM. However, we haven’t seen any compelling evidence that he worked directly with GM engineers beyond that. The patent evidence, meanwhile, indicates that the AST could have been simply an evolution of Earl Thompson’s earlier designs.
With so little to go, though, we can’t make any authoritative statement, and if we find more information, we may change our tune. Maybe we’ll stumble on a pile of internal memoranda talking about Banker’s contributions, or some other compelling evidence for a different version of events. If so, c’est la guerre.
We wish we had more detailed and reliable information on Banker. Even if his design did not become the AST, it’s clear that he was a talented engineer and a prolific inventor: a patent search on “Oscar H. Banker” returned 17 pages of results. We find it frustrating that we’re not able to provide a clearer picture of his life and accomplishments.
Lacking the full picture, though we’ll go with the most reasonable explanation we can muster, based on the facts we do have. Which is, in turn, how we approach the research for each of these articles.
# # #
NOTES ON SOURCES
Our scant online sources on Oscar Banker included Automotive Hall of Fame: Distinguished Service Citation list, http://www.automotivehalloffame. org/ honors/ index.php?type=distinguished& cmd=name, accessed 3 July 2010; Demirdjian Enterprises, Inc., “Armenia: Contributions” (11 January 2009, http://www.jdemirdjian. com/ Pages/ Contributions.htm, accessed 3 July 2010); Philip G. Gott, Changing Gears: The Development of the Automotive Transmission (SAE Historical Series) (Warrendale, PA: Society of American Engineers, 1991); Bob Hull, “Hull of Fame” (6 October 2002, http://bhull.blogspot. com/ 2002_10_06_archive.html, accessed 3 July 2010); Stepan Karadian, “Stamps of Armenia Worldwide” (n.d., Cilicia Armenia, http://www.cilicia. com/ armo_stamps_ worldwide.html, accessed 2 July 2010); “Med-E-Jet inoculation gun, United States, 1980,” (n.d., The Science Museum, http://www.sciencemuseum. org.uk/ objects/ public_health_and_hygiene/1981-1398.aspx, accessed 3 July 2010); Stepan Partamian, “Yes, We Have” (date unknown, http://www.yeswehave. org/ book.html, accessed 3 July 2010); “rod,” “My big fat Greek riot,” (10 December 2008, http://www.rhythmism. com/ forum/ archive/ index.php/t-75468.html, accessed 3 July 2010); Social Security Death Index records for Oscar H. Banker, WorldVitalRecords.com, http://www.worldvitalrecords. com/ zsearch.aspx? ix=ssdiall& qt=l&zfn=Oscar& zln=Banker, accessed 12 July 2010; Society of American Engineers, Technical Papers, http://papers.sae.org/ 460099/, accessed 3 July 2010; and E.K. Zsigmond, MD, DSc, “Jet Anesthesia and Jet Local Anesthesia for the 21 Century,” Journal of the National Medical Association Vol. 94, No. 11 (November 2002), pp. 1004-1006. We also consulted Chapter 2 of Ralph Nader, Unsafe at Any Speed: The Designed-in Dangers of the American Automobile (New York: Grossman Publishers, 1965). Special thanks to reader Dan Rolfe for pointing us toward the Gott book in an email on 17 July 2010.
Our principal original sources on the Automatic Safety Transmission were “Almost Automatic,” Special Interest Autos #20 (January-February 1974), pp. 24-27; Helen Jones Earley and James R. Walkinshaw, Setting the Pace: Oldsmobile’s First 100 Years (Lansing, MI: Oldsmobile Division of General Motors Corporation, 1996); and Terry B. Dunham, Lawrence R. Gustin, “Appendix Five: The Buick Shifts for Itself,” The Buick: A Complete History (An Automobile Quarterly Magnificent Marque Book) (Kurtztown, PA: Automobile Quarterly, 1980), pp. 382-389.
We also consulted the following patents:
- Oscar H. Banker, “Change Speed Transmission,” United States Patent No. 1,985,884, applied 14 December 1932, issued 1 January 1935.
- Oscar H. Banker, “Change Speed Transmission Mechanism,” United States Patent No. 2,198,072, applied 31 May 1934, issued 23 April 1940.
- Oscar H. Banker, “Change Speed Planetary Transmission,” United States Patent No. 2,077,387, applied 16 July 1934, renewed 22 March 1935, issued 20 April 1937.
- Oscar H. Banker, “Automatic Transmission,” United States Patent 2,199,095, applied 18 October 1934, issued 30 April 1940.
- Oscar H. Banker, “Automatic Transmission,” United States Patent No. 2,262,747, applied 18 September 1936, issued 18 November 1941, reissue applied 31 January 1942, reissued 18 May 1943.
- Oscar H. Banker, “Power Operated Inoculator,” United States Patent No. 3,292,622, applied 21 September 1964, issued 20 December 1966.
- Oscar H. Banker, “Gun Type Inoculator,” United States Patent No. 3,518,990, applied 2 May 1968, issued 7 July 1970.
- Oliver K. Kelley, “Combination Fluid Turbo Clutch and Variable Speed Gearing,” United States Patent No. 2,176,138, applied 5 February 1937, issued 17 October 1939.
- Oliver K. Kelley, “Fluid Flywheel Gearing Arrangement,” United States Patent No. 2,211,233, applied 10 April 1939, issued 13 August 1940.
- Earl A. Thompson, “Change Speed Gearing with Automatic Overdrive,” United States Patent No. 2,285,760, applied 6 March 1933, issued 9 June 1942.
- Earl A. Thompson, “Change-Speed Transmission and Control,” United States Patent No. 2,195,605, applied 8 October 1934, issued 2 April 1940.
- Earl A. Thompson, “Change-Speed Mechanism and Control,” United States Patent No. 2,193,304, applied 16 October 1935, issued 12 March 1940.
- Earl A. Thompson, “Clutch and Gearing Control,” United States Patent No. 2,362,418, applied 15 March 1937, divided 14 February 1940, issued 7 November 1944.
- Hydra-Matic History: The First Automatic Transmission
- Dynaflow, Turboglide, Roto Hydramatic, and Other Early GM Automatics