GM’s Y-body “senior compacts” — the Buick Special/Skylark, Oldsmobile F-85/Cutlass, and Pontiac Tempest/Le Mans — boasted many interesting engineering features, from their lightweight aluminum V-8 engines to the Tempest’s flexible driveshaft and rear transaxle. One novelty that’s often overlooked is a peculiar feature of the lightweight Dual-Path Turbine Drive two-speed automatic transmission offered on the Buick Special: This transmission had no reverse gear!
Now, I know that some CC readers have driven Specials or Skylarks with this transmission, and you may be thinking, “Aaron, that’s crazy, of course it had reverse.” It did — and it didn’t. Let’s take a look at how this worked.
Buick Dual-Path Turbine Drive
The Dual-Path Turbine Drive is one of the most unusual transmissions GM has ever offered. Used only in the 1961–1963 Buick Special and Skylark, it was a lightweight, light-duty two-speed automatic with a very “loose” torque converter and a split-torque top gear. It was by far the lightest of any contemporary American-made automatic transmission: just 84 lb (38 kg) dry, or 95 lb (43 kg) with fluid — 10 lb less than the three-speed manual transmission used on the Y-body Special and Olds F-85. To put it another way, combined weight of the aluminum V-8 AND the Dual-Path Turbine Drive transmission was less than 450 lb, which was at least 90 lb lighter than a cast iron SBC engine alone!
1961 Buick Special DeLuxe four-door sedan / SMclassiccars.com
One of the ways the Dual-Path transmission achieved such light weight was by using only one planetary gearset. This had two sun gears, but both meshed with the same set of planet pinions — this wasn’t a compound or Ravigneaux gearset like the Chevrolet Powerglide. The gearset was tucked neatly into the hub of the three-element torque converter, with the rear section of the conical aluminum transmission case full of various assorted clutches. Compared to most torque converter automatics, the Dual-Path Turbine Drive torque converter was mounted backwards, with the turbine (driven torus) facing the engine and the impeller (pump) facing toward the rear axle, the opposite of the usual arrangement.
Here’s a Buick cutaway showing the layout:
The torque converter turbine was bolted to the ring gear of the planetary gearset, while the gearset’s planet carrier was fastened to the output shaft. In low gear, the sun gears were held stationary, giving a reduction gear ratio, while in high, the engine drove the sun gears through the converter clutch, giving a split-torque direct drive gear.
Here are schematics showing the power flow in low and high. (The red arrows indicate clutch engagement.)
Although the Dual-Path transmission had two sun gears, it was still functionally a simple planetary gearset: Since both sun gears were meshed with the same planet gears, the sun gears always had to rotate at the same speed and in the same direction, and if one stopped, the other had to stop as well.
Backwards Without Reverse
If you know a little about planetary gears, you may recall that the way you get reverse with a simple planetary gearset is to hold the planet carrier stationary so that the sun gear and ring gear are forced to turn in opposite directions. Since the Dual-Path gearset’s planet carrier was attached to the output shaft, holding the carrier stationary would just prevent the car from moving at all. With no other planetary gears to compound, there was no way for the gearset to reverse the rotation imparted by the torque converter, and hence no reverse gear.
Much like the famous myth of Alexander the Great solving the impossible Gordian Knot by cutting it in two with his sword, the Dual-Path Turbine Drive solved the reverse problem with lateral thinking. Since there was no way for the transmission gears to provide reverse rotation, the transmission instead reversed the rotation of the torque converter!
For this to make sense, we need to first take a step or two back.
Unlike older Buick automatics, which had complex four- and five-element torque converters, the Dual-Path Turbine Drive had a simple three-element converter, consisting of the impeller (converter pump), which was bolted to the engine flex plate; the turbine, which was bolted to the ring gear of the planetary gearset; and the stator, which was mounted on a sleeve shaft connected to an overrunning (one-way) clutch.
In a torque converter, whenever the impeller is turning significantly faster than the turbine, the rotary flow of the oil leaving the turbine to return to the impeller tends to be in the “wrong” direction (if the engine is rotating clockwise, the returning oil has a counterclockwise rotation, or vice versa). The job of the stator, which sits between the turbine and the impeller, is ensure that this oil’s rotary flow is in the correct direction before the oil reenters the impeller, so the moving oil doesn’t oppose the rotation of the engine.
The stator is mounted on a sleeve shaft connected to a one-way clutch that only allows it to rotate in the same direction as the engine and the flex plate. At higher turbine speeds, when the rotary flow of the returning oil is already in the correct direction, the stator just freewheels idly. If the rotary flow of the returning oil is in the “wrong” direction, the oil hits the stator blades and attempts to turn them backwards (against engine rotation). This locks the stator against its one-way clutch, so the oil hitting the stator blades is redirected, reversing its rotary flow to match the direction of engine rotation. Instead of opposing the engine’s rotation, the returning oil now aids the rotation of the impeller, which multiplies engine torque.
This was how the Dual-Path Turbine Drive torque converter worked in the forward gears, but in reverse, the reverse clutch engaged to lock the converter turbine to the transmission case, preventing the turbine from rotating in either direction. At the same time, the other clutches were conditioned so that the one-way stator clutch was disconnected from the case, but was connected via a separate sleeve shaft to the rear sun gear of the planetary gearset.
This will be easier to visualize if you look at the schematic. (The red arrows indicate clutch engagement.)
The converter pump (impeller) was still bolted to the flex plate, so it always rotated at engine speed. Oil propelled by the impeller entered the turbine, which was now locked in place by the reverse clutch. Hitting the blades of the stationary turbine caused the oil to leave the turbine with a “backwards” rotary flow. This returning oil then hit the stator blades, but because the stator clutch was no longer anchored to the case, the stator was free to rotate in either direction, so the oil hitting its blades cause it to spin backwards. Oil leaving the stator blades then reentered the impeller with the “correct” rotary flow, multiplying engine torque. Basically, the turbine was now acting as the stator, while the stator became the turbine.
Because the stator shaft was connected to the rear sun gear shaft by one of the transmission’s many clutches, the stator’s backwards rotation caused both sun gears of the planetary gearset to also rotate backwards. The ring gear, which as I mentioned above was bolted to the turbine, was also held in place while the turbine was locked, becoming the reaction member and driving the planet carrier and the output shaft backwards at reduced speed.
So, there was a reverse speed, but the reverse rotation actually came from the torque converter, not the gears. (The input and output members of the planetary gearset both rotated in the same direction, even in reverse!) It was an ingenious and mechanically elegant solution, and fairly unique. The Chevrolet Turboglide and Buick Flight Pitch Dynaflow did something loosely similar for reverse, but both those transmissions were much more complicated.
Sadly, this clever little transmission was short-lived: It was probably more expensive to build than the two-speed Super Turbine 300 automatic that replaced it, which had much more torque capacity and was far easier to beef up. The last production application of the Dual-Path Turbine Drive was the 1963 Buick Special and Skylark.
1963 Buick Skylark hardtop / RM Sotheby’s
We often think of the Buick Special/Skylark as the most conventional, least technologically ambitious of the three Y-body “senior compacts.” However, it was also a fascinating design with many interesting features. It’s too bad the details weren’t quite sorted and that it cost way too much to produce — it could have been a world-beater.
Further Reading
Curbside Classic: 1962 Buick Special – A Truly Special Buick (by J P Cavanaugh)
Vintage Car Life Comparison: 1962 Pontiac Tempest 4, Buick Special V6, Olds F-85 V8 – Decisions, Decisions (by Paul N)
Dynaflow, Turboglide, Roto Hydra-Matic, and Other Early GM Automatics (at Ate Up With Motor)
Giving Slip the Slip: Lockup Torque Converters and Split Torque Automatic Transmissions (at Ate Up With Motor)
Nice write up Aaron, I could not have said it any better! And I’m the person that sent you the info on these! 🙂
Yes, I’m still very grateful to you for sending me the shop manual section, which I finally realized earlier this year that I should probably scan for posterity and ease of reference.
I’ve always found this transmission to be very compelling due to its small size, low weight and efficiency. This was one of the most remarkable things to come out of GM’s “progressive era” (ca. 1957-1961).
In the couple of reviews I’ve read of cars equipped with this, the reviewers were quite enthusiastic about it. I can’t remember ever driving one or even riding in one equipped with it, although the latter is possible. But I did have quite a bit of driving seat time in a ’62 Cutlass with the Roto-Hydramatic, and I was somewhat appalled by its shift quality and all-round feel. Of course that car wasn’t exactly new by then (5 years old) but still.
I can’t help but wonder if this might have been a suitable transmission for Opel and other off-shore GM divisions to adopt, but then I know the realities of building it in small numbers simply wouldn’t have planned out.
We had a 62 Special wagon from ’66-’84. I was brought home from the hospital in it and I learned how to drive in it. They leak fluid like they own a oil well, but we never had any problems with it. Quite the remarkable trans. And IIRC, they will work in limited capacity with no fluid thru one of the dual paths, but you get no reverse. One of the first uses(I could be wrong) of a u shaped rubber gasket for the pan and a single central attaching bolt. And I don’t think it was mentioned that it’s air cooled, that helps save weight also.
I think that it probably came about specifically because Buick had made a huge investment in expanding and upgrading their transmission plant to build the ill-fated Flight Pitch/Triple Turbine Dynaflow, including equipment for aluminum castings, only to have the new transmission fail and their sales volume tank. The basic design came out of Engineering Staff and, surprisingly, Pontiac (one of the underlying patents is DeLorean’s!), but Buick decided to build it because they were sitting on something upwards of 200,000 units of additional capacity that they were not using.
There’s probably a parallel there with what you deduced about the relationship of the Corvair engine and the Vega aluminum block to the Tonawanda Engine plant: whatever the merits of the designs, the driving force seems to have been “What can we do with this plant space?”
As for Opel, Dual-Path Turbine Drive would have worked much better for them than the awful Model 5 Hydra-Matic, but I have been given to understand that German automotive engineers had a lot of ingrained mistrust of torque converters — ironic given that the torque converter was invented in Germany! — and saw them as fuel-sucking monstrosities that compromised efficiency. ZF was actually sufficiently concerned about that that they developed a lockup clutch for their 3HP transmission as early as 1963, although I don’t think it was used until quite a bit later. The idea of using a torque converter as a continuously variable extra speed rather than just as a glorified fluid clutch would probably have been received with horror, and I’m not sure the split-torque high gear would have been sufficiently mollifying.
Aaron,
I’ve read and enjoyed a lot of your transmission articles online and always wondered if you worked for one of the big 3 in the transmission design department. I’ve never seen anything about that.
cheers
Great article, glad you posted it here. I knew there was a ‘trick’ to how these transmissions went into reverse but didn’t really understand it until I saw your write-up on A.U.W.M.. Sometimes these transmissions were called the ‘skillet’ because of their odd rounded pan that had one central bolt holding it on.
Here’s a picture:
https://charlietranny.com/bsnewpics%20(6).JPG
This is new info to me, and I appreciate your explanation. My amazement at the differing engineering solutions employed by the three Divisions pushing Y body cars out the door continues to increase.
Fascinating transmission, but I haven’t quite got it. I was thinking the stator always has to be the driven element somehow. And why have two sun gears, if they are connected to the same pinion don’t they have to do the same thing regardless?
I might need to print out those flow diagrams and highlight the paths.
It may be easier (and basically correct) to think of it as a two-piece sun gear. The reason it’s separate like that is each gear is physically connected to a different component on opposite sides: The front sun gear is fastened to the driven shaft of the converter/direct drive clutch, the rear sun gear is fastened to a sleeve shaft that connects it to its overrunning clutch. Having them separate is a production/maintenance convenience, and because they’re both in constant mesh with the same set of planet pinions, they have to act as one gear.
Very impressed by your ability to tackle technical topics, while possessing excellent design analysis skills. Quite rare to see folks, that can bring both. While importantly, delivering it in layman’s terms.
Skills, that you and Paul, share.
Thanx Aaron I recently watched a guy dismantle and reassemble one of those Buick trans, One very clever device, the torque converter was totally different to anything Ive messed with.
I was totally unaware of this transmission. To my mind it raises questions of Why? As you explain it, it’s rather ingenious, which makes it all the more of a shame that it was so short-lived.
GM put an amazing amount of unique engineering into these cars, so much of which wasn’t carried forward to any future product. Guess they could afford that sort of thing back then.
Finally! My questions are answered! I now know what transmission my Mom’s 1961 Buick Special had! Hers looked just like the one the picture with the article, except hers was white instead of blue! Thank you for sharing your knowledge with us.
Could one reason the transmission was discontinued be its non-standard shift pattern? Remember, before the birth of NHTSA with the passage of the Motor Vehicle Safety Act in 1966, there were wide variations in the shift pattern for automatic transmissions. GM was notorious for this, especially placing Reverse below Low in the Dual-Path transmission (PNDLR), both in the conventional automatic and in the “Rope Drive” Tempest transaxle. When the Feds mandated that everyone use the now familiar PRNDL shift pattern, the Dual path couldn’t be re-engineered to comply, and was dropped. Granted, the cost of building three (3) different two-speed automatics (Powerglide, ST300 and Dual-Path Turbine Drive) was a larger factor, and the DPTD was cancelled three (3) years before the NHTSA bureaucrats started to throw their weight around, but maybe, GM saw the handwriting on the wall with the Corvair, and cancelled it before the bureaucrats could do so? Comments.
No. Oscar Banker had been complaining at SAE meetings that the GM PNDLR pattern was dangerous (which was arguable), but by the time that became a public issue, the Dual-Path transmission was already long dead. (Furthermore, the Roto Hydra-Matic stuck with the PNDSLR pattern through 1964.)
If Buick was seriously concerned about that, they could have just redesigned the valve body — Chevrolet did with Powerglide. It wouldn’t have required redesigning the transmission or changing how it obtained reverse, it was just a matter of rearranging the relationship of the manual shift valve to the hydraulic passages, and not a big deal.
Also, given that GM offered automatic transmissions well before other automakers did, they had a case for arguing that the PRNDL pattern was the “nonstandard” one: Hydra-Matic, Dynaflow, and Powerglide all used the NDLR (or PNDLR) pattern for years in very large numbers.
The original point of having R below L was to make it easier to shift between low and reverse to rock the drive wheels out of snow or mud. GM first used the PRND pattern on the Buick Flight Pitch Dynaflow/Triple Turbine and Chevrolet Turboglide because those transmission didn’t have any kind of manual low gear, so rocking the car in that way wasn’t an option anyway. Chevrolet realized that having a different shift pattern between Powerglide and Turboglide was a problem, so they redesigned the Powerglide valve body to PRNDL. However, Buick and Detroit Transmission (which made the Hydra-Matic transmissions) would have undoubtedly argued — and probably did to Oscar Banker — that it was more logical, and safer, to stick with the pattern their longtime customers were used to — and that some of those buyers might still want to be able to rock the drive wheels out of mud or snow, as was the original purpose.
Obviously, this was arguable, and industry standardization was really in everybody’s interests in the long run, but GM wasn’t just being capricious about it.
I knew my Mom’s 1961 Buick Special had a different transmission than either a Chevy Powerglide or a Super Turbine 300, now I know which one it was! Thank you for educating me!
When I was still a pre-schooler, my Dad used Mom’s car to pull a small utility trailer, to move an old wooden desk out of my Grandparent’s house, which killed that Dual Path Transmission, requiring a rebuild! To be fair, that desk weighed a ton, but still, I wouldn’t recommend the combination of the 215 V8 and the Dual-Path transmission for trailer towing, that’s for sure!
We had a 62 Special from ’66-’84. IIRC, these would work in limited capacity with no fluid, but no reverse. And it was air cooled, how crazy is that!
Always amazed me how much $$$$ GM spent on unique engineering for their compact cars, how they were all radically different except for the Olds and Buick. No way they made any $$$$ on them except maybe for the end run of the Corvair. Which stuck around long enough unchanged to have paid for itself eventually.
I think this creative engineering and associated costs really stifled the company from improving their products in later years. I mean by the early 80s the A-body and B-body cars were still body on frame, carburator, overhead valve V8, live rear axle with coil springs, same way GM had been making cars since the late 1940s. Part of the reason GM struggled so much in the 80s is instead of incremental improvements over the years they pretty much wadded up the blueprints of how they made cars. Started from scratch all over again with the FWD x car, j car, c body(FWD LeSabre, 88, Deville etc) This lack of improvement/innovation kinda goes back to the early 1960s senior compacts and GM losing their ass on those cars