GM pioneered the automatic transmission with their Hydra-Matic back in 1939, a four speed unit with a fluid coupling (no torque converter). It was rather complex, but it worked well enough, and soon proved itself in automobiles as well as military armored vehicles, tanks and other applications during the war. After the war, the GMC M-135 2.5 ton 6×6 truck proved its capabilities in that application, with a two-speed auxiliary transmission for a total of eight speeds.
But it wasn’t until 1953 that GM began to offer the H-M in civilian trucks, starting with the basic four-speed unit in 1/2 – 1 ton light trucks. Later that year, GM expanded the availability of H-M in both the medium and heavy duty lines: the 8-speed Hydra-Matic, which was a 4-speed unit with a two-speed reduction unit to yield the eight speeds in 4-5 ton trucks—essentially the same unit used in the M-135 trucks—and the much more complex Twin Hydra-Matic, which was essentially two 4-speed H-M units in one case yielding 7 speeds, backed by a three auxiliary transmission for a total of 21 speeds, 13 of them non-duplicated.
The Twin Hydra-Matic turned out to be overly complex, expensive and unreliable, and were withdrawn within just a couple of years, with GM offering to replace them with conventional transmissions. Sound familiar?
This ad from 1954 references the basic 4-speed unit for ½, ¾ and 1 ton units as well as the 8-speed unit for 4-5 ton trucks, although no real specifics.
This is from a brochure for the Twin H-M for sale on eBay, and decided to buy it, as I couldn’t make out the print on the ad for it.
The Twin H-M came standard on the DFM660-47 version of the “cannonball” GMC HD COE unit. It was teamed only with the 150 hp DD 4-71, which sounds a bit modest in power, but 150 hp in the early-mid 50s was common and typical. Trucks back then rarely exceeded 45-55 mph in the pre-interstate era, except for certain big West Coast units.
This gives some basic information on the unit, including its gear ratios.
Here’s a cross section. Note that this is a composite of two scans, and the shaft appears as being offset between the two units.
This cross section shows how the two 4-speed units interact to create 7 speeds.
Here’s a table of all of the available 21 speeds. Note that a number of the duplicate.
This refers to the column below that shows all of the possible gear steps with the unit utilizing the three speed auxiliary, in “creeper” and “low” range as well as direct.
Here’s that table. It shows various rpm on the bottom, and the respective range of road speeds in each gear.
I couldn’t find much detailed in on the 8-speed H-M, but essentially it worked similarly, with two ranges and four speeds.
By 1955, a revised Twin H-M appeared with a lighter 2-speed reduction unit, which was deemed adequate for most applications. The three-speed reduction unit was still available too.
The Twin H-M proved to be unreliable. Most of them were later replaced by manual transmissions without cost to the owners.
In 1957, wide ratio range two–speed rear axles were teamed with Hydra–Matic transmissions in medium duty 4x2s, replacing the unreliable 2–speed reduction boxes. Meanwhile the 7-speed Twin Hydra-Matic transmission was discontinued because of poor reliability and low sales.
It was replaced by the Allison Torqmatic (called Powermatic by Chevrolet) 6–speed automatic transmission, which could of course also be teamed with a two-speed rear axle.
The Allisons included torque converter lockup clutches and hydraulic retarders were available, providing superior performance and reliability. They were made in several version covering a wide range of applications from medium duty trucks to giant oil drilling rigs. These were widely used for many years. I drove a big Ford Super Duty dump truck and some others with this transmission. It banged out the gears with a bit of a nudge in the back each time.
It’s difficult finding more information on the 8-speed and Twin Hydra-Matics, as they were only built briefly, and not in large numbers. But clearly there were limits to what could be done with the H-M, which continued to be built in several evolutionary steps for passenger cars through at least 1964.
Related: CC 1953 GMC Pickup – First Year For Automatics In Light Trucks
Cadillac doubled the number of speeds in its manual trans in the early 20’s (IIRC) by adding a second gear in the rear differential. That didn’t take, either.
I wonder about the marginal benefit and longevity of this century’s rush to 6-8-10 speed trans for gas cars, particularly with turbo-ed DOHC engines, but I guess CAFE and emissions conquer all else. In my limited experience with them, waiting for a downshift to accelerate even a little was irritating.
I’ve not experienced a modern 10-speed unit, but the 9-speed automatic in my Mercedes works superbly and transparently. Having more gears to choose from means that an appropriate “next gear” is always close by, and shifts are almost undetectable. I was a skeptic too, but there is a tangible benefit to having more cogs.
Following Mercedes tradition, it normally starts out in second gear, but in sport mode it will start out in a super short first gear for blistering off the line acceleration. Conversely, it has a super tall ninth gear for relaxed, long-legged highway cruising.
Ralph L, that Cad 2-speed was a bit earlier. (Yes, for a casual conversation I’m too lazy to look it up and confirm exact details lol) Roughly, Cadillac and Austin got into a legal wrangling over the 2-speed’s design. Austin prevailed, Cadillac paid royalties for the ones they had sold, then promptly dropped the design. Or something close to that.
It was a neat piece, two gear sets of different ratios running in constant mesh, the desired set “dogged in” as needed.
I wonder if the Allison transmission used in trucks do the same sound as the ones used for school buses?
There are many different Allison transmissions. When Allison was sold off by GM, GM retained the rights to use the Allison name and transmissions in their light trucks. The school bus probably has an earlier Allison Trans. Current transmissions started coming out around 1991. I would be leery of any of the earlier trans, parts can be difficult to get. Their transmission is the trans to have in refuse and construction. Allison has a test track at their Indianapolis plant. The hill they have is ridiculously steep. What they have you do is drive a loaded dump truck up the hill, stop half way up and then you finish the climb to the top. They have you let off the brakes before any throttle application. As the truck starts to roll back you nail the throttle and you feel that trans grab and pull you over the top. They have trucks with manual trans if you want to try the same test. The other test is driving thru a “wash” of pea gravel. First you drive a stick truck into the pea gravel, go a far as you can before you get stuck. Then you back a truck with the Allison in it into the pea gravel in front of the stuck dump truck. Then you chain the stuck truck to your Allison truck and your Allison pulls itself and the stuck truck out of the pea gravel. The trucks are all loaded equally so no cheating.
Stéphane Dumas, I would intuitively speculate that you’re listening to a 500 Series Allison. Truck, bus, basically the same sound. With a truck possibly the engine cooling fan might compete for the bus’s “drum” sound.
On a side note I’ll predict the starter motor’s early demise. lol
xr7, that’s interesting about Indy. I’d like to take the standard vs automatic challenge.
To keep things fair I’d like to add the deceleration-on-greased-ice-patch test. lol
It’s wicked when an automatic loses its “sense” when even a quick loss of traction “confuses” it.
That’s an AT-540 (or similar 500-variant) you hear on that GMC, as described and commentated here. The AT-540 came in for 1970, so much later than the Hydra-Matic subjects of this post.
Appreciate the perky pixie. She reminds me of a gal I knew many years ago. What about engine braking on these transmissions? Remember the whine and “crispy” shifting on my father’s ’47 Olds. Seeing a similar Olds a a car show last year refreshed my memory. I figure the engine braking on these is pretty much non existent, unlike the lock up torque converters of today.
A car with VVT has virtually no engine braking. I first encountered it on my 2015 Kia Rio and my Golf is the same. It is especially noticeable with a manual transmission car.
Dad taught me the importance of gearing down before a stop. He claimed it increased brake life, which it did. Problem is it increased clutch wear and I’d much rather replace brakes than a clutch.
Dad also attempted to teach me how to drive. He was a terrible driver and I failed my first driving test. I insisted I take driver’s training and I would pay for it myself. He was very put out over that but I still use that training today.
It did, however, take some time to get used to stopping while still in fourth gear and shifting to first at the stoplight.
That must be the programing of those cars, my pickup with VVT and a manual trans provides great engine braking, no different than other vehicles I’ve had w/o VVT.
Engine braking is a bad idea under most conditions, as it defeats the proportioning feature of the braking system which heavily favors the front brakes, leading to loss of braking control and less positive stops. I did witness an inexperienced motorcycle rider doing just that and the rear wheel lost traction resulting in a nasty slide and fall.
You don’t need a lock up converter for engine braking. The way a lot of lock up converters have been programed in the past is that they only lock up in the top gear, so it doesn’t play a factor when you downshift for engine braking. Yes some of the newest transmissions with umpteen gears and multiple OD ratios do lock up in lower gears but that is a relatively new thing.
The young lady above appears to be pointing out the girdlespring on the upend of the grammies.
…The young lady above appears to be pointing out the girdlespring on the upend of the grammies…
“My Girdle is kill…OH, NEVER MIND, I forgot this isn’t a turbo-encabulator.”
She’s simultaneously tying that in with pointing out Torqmatic’s retarder. The single most efficient device ever conceived for converting mechanical energy into brutal heat. lol
Great piece, Paul, and thank you for it.
I have always been really interested in Hydramatic because they were just before my time. Added to that they were mostly in higher end cars like Oldsmobile, Cadillac and Pontiac. For this reason there couldn’t have been a lot around in Canada, as in the Hydramatic era a Canadian only had about 60% of the buying power of an American.
The Hydramatic seems like such a Rube Goldberg device. It was very over-engineered and with its complexity, it’s amazing GM made it work so reliably. I have only ever driven one and that was a short jaunt in a customer’s car to try to diagnose a problem not caused by Hydramatic.
I prefer all-manual, but my current Jag has manual override on 6speed auto. The only other auto I’ve owned, bought for its excellent engine and suspension modifications, was a BMW e9 Coupe, transmission by GMH, basically Hydramatic, always reliable. BMW original equipment, so highly respected for acceleration and appropriate ratios for cornering. The Fordomatic 2 speed on one of my girlfriends’ dad’s treasured Customline was a rather sluggish impediment to performance (not that a young lad would take that wallowing wonder past 5 tenths)
Thanks Paul. As much as I thought I knew about medium and heavy trucks of this era this was completely under my radar. Fascinating info, thanks for the good work
It seems that both the 8 speed and Twin Hydramatics were made to judge the acceptance of automatic transmissions in heavy duty trucks. Both transmissions made the maximum use of existing components. Though not terribly successful, it appears there was enough interest to justify introducing a proper heavy duty automatic transmission in the Allison. I believe the program was transferred from the Hydramatic Division to the Allison Division due to Allison’s production capacity and experience with their Cross-Drive military transmissions.
When I started my career I was working with an old-timer that had experience with the 8 speed Hydramatic. Seems our company had some in the fleet in mid-50’s gasoline powered (Olds V-8 I believe) GMC tractors. The guy said they were very smooth and nice to drive loaded, but shifted abruptly and were jerky when empty. So much so the trucks went through U-joints and driveshafts constantly. Today our fleet’s medium and heavy trucks are near 100% Allison equipped.
There was a guy that used to regularly attend Southern California A.T.H.S. shows that had a fully restored DFM660-47 tractor. At some point the transmission had been replaced with a manual box (still had the Hydramatic emblems though!). The truck owner had a Twin-Hydramatic transmission he displayed with the truck.
Bob B’s speculation that Twin Hydra-matic was a taste test is probably spot on.
However, the efforts put in to refine Twin HM details, such as improved PTO operation, makes it seem that GM had hopes for Twin Hydra-matic to live on even when expecting Torqmatic’s birth.
As to the gear count, beyond take-off, there isn’t a dime’s worth of difference between Creeper-Direct and Low-Direct ratios. Effectively this is about a 9-speed box.
L-D has the auxiliary in reduction until about 25mph, while C-D has the auxiliary in reduction until about 14 mph before ratio duplication begins. Other than the added reduction for take-off in C-D, and of course the different power-flow paths utilized there isn’t a significant difference in gear ranges. Above about 15 mph even plain old Direct is offering basically the same gearing.
Not available with Dual Hydra-matic is the ideal “gear splitting” used to keep an engine in its sweet spot when pulling hard. Because… with a load so heavy that it really needed all that shifting below 10 mph, recovering between the 5/6/7 shifts would be like climbing back up after falling off of a cliff. lol
Basically Dual Hydra-matic wasn’t offering anything in road gearing that a broad torque-curve engine couldn’t live without.
Sure, there are lots of potential gear combinations but the punchline is that other than at initial take-off they are mostly redundant. Not even a “lazy” overdrive gear for the ride home. This is probably due to the “let’s just run what we have” attempt at reutilizing existing Hydra-matic components.
In comparison, a purpose engineered transmission, or transmission and 2-speed axle combination, will have progressive steps in gearing with little or no redundancy; each gear combination having its unique work range.
What’s underscored here is Hydra-matics most significant shortcoming, namely, the lack of a torque converter.
As Torquematic and later Allison transmissions clearly demonstrated, so many cogs just aren’t necessary when utilizing a well matched torque converter.
CC effect – Recently I hunted for an OD for Paul’s pickup. During the hunt, at a “transmission museum” I was looking over a batch of relics, including HM and Torqmatic.
Thanx Paul, Ive never heard of these much less driven one they might have been easier than a twin stick manual but other than that theres only really 13 gears, I owned a hydramatic car and that wasnt too bad the car was on its 3rd engine and I replaced the worn beyond recovery diff head but the trans lasted a couple of years before it wouldnt retain oil, my car was the last model the hydramatic was used by either GMH or GM UK they both switched to the two speed powerglide for 1965.
An eye-opener. Never heard of it before this.
Jim Dandy’s point about the many redundant gears shows up at least one weakness of the thing, especially as it likely cost a good bit more than a manual for no useful gain over, say, a 9-speed self-shifter. (I don’t think driver effort/fatigue was even a consideration in those times!) Certainly not enough gain to spend the money in a hugely cost-driven sector, one would think.
I do wonder why it was unreliable, though. The Hydramatic seemed to be very reliable indeed: possibly the mechanisms to connect one of the 4-speeds to the other was what was prone to failure.
Great post, Dr N.
Fascinating stuff, and all new to me. It seems like GM enjoyed a great burst of success with the original H-M and spent the next 20 years refining it and (as with the trucks) trying to extend the concept. But with the exception of the Chevrolet Powerglide, GM seems to have spent the first fifteen years or more after WWII so convinced of its engineering superiority in automatic transmissions that it didn’t make a single one other than the H-M that was really much good.
By 1956-57 when both Ford and Chrysler had introduced “modern” automatics that coupled a torque converter to a 3 speed gearset, GM spent nearly a decade bringing out one odd/bad design after another – TurboGlide, the 1958 Flight-Pitch (or was it Triple Turbine?) Dynaflow and the Roto-Hydra-Matic. Until the THM became common in the mid 60s GM offered one compromised system after another. Other than the durable (but complex and expensive) final H-Ms of 1956-63, you either got a 2 speed that would last or a 3 speed that would not. Not that buyers seemed to care that much, though.
This chapter of GM trying to make the original H-M do something it really wasn’t designed to do was one I was not familiar with, but it sort of falls into GM’s “We are brilliant and everyone else is an idiot” mindset that put out so many vehicles that were failures in either tech terms or in market success.
Ironically, for all of their teething pains with automotive transmissions, GM hit home runs with their industrial boxes. As with their cars, GM’s heavy transmissions followed two different theories of operation. One series heavily dependant on torque converter operation. The other series more dependant on clutched gearing. Both were excellent boxes and the initial designs lived on for decades.
Others made early attempts at industrial powershift transmissions, none saw the success GM did.