by Tom Hand
Courtesy of the Walter P. Chrysler Club with our most sincere thanks. First published in 1983. See our TorqueFlite overview / main page, Part II (modifications), and Part III (repairs).
This is part one of a three part paper. The first part explains the individual units of the Torqueflite automatic transmission. The second part deals with modifications and justifications as well as the advantages and disadvantages of altering the Torqueflite shift qualities. Part three give some of the techniques that can be used to troubleshoot Torqueflite irregularities that I've seen most often. I have written this:
This article is not intended to give specifics about every Torqueflite made. If you are interested in more detail than I have given, get the Fundamentals of Automatic Transmission and Transaxles (TM-508 or TM-508a). This will explain just about everything you would want to know about 1966 and later Torqueflite 904s and 727s. As in any repair situation, a person should refer to the appropriate Chrysler service manual for exact details and adjustments of particular units.
The photographs included with this article are of a 1976 727-A that I built for a police car (I spent a lot of time cleaning this unit and at the same time, learning that washing parts in a dishwasher works but tends to rust them). The 904 is different in some regards but I didn't have one available for photographs. Therefore, I will explain construction and operation of both 904s and 727s where the differences are of concern to adjustments and/or assembly procedures.
Many thanks to David Frey, Vicki Milley, Debbie Hoflo, and a very special thank you to Tom Mallot for his photographic help.
Torqueflites first appeared in 1956, as an option in Imperials and 300Bs, with a cast iron body and an aluminum converter housing and tail shaft housing. The gear ratios were the same in the forward gears as they are now:
Reverse was 2.21:1 in the early units and 2.20:1 in all models since 1962. These transmissions were coded A-488s and were standard on 1957 Imperials and other Chrysler models. The 1957 Torqueflite was heralded as “the best automatic in the world” and was used in several European sport-luxury cars (including Monteverdi, Facel Vega, and Jensen).
In 1962, an aluminum case replaced the cast iron one, changing the code name to “A-727;” together with the torque converter, it weighed about 160 pounds. The original 727 was basically unchanged through 1965. This transmission was very different, internally, from its iron-cased predecessor. [Hemi Andersen wrote, “The 904 and 727 are much simpler, yet still tough and reliable, transmissions.”]
Hemi Anderson wrote, “Humbly, I will tell you that I was at least one of the originators of [the 1971 front clutch bushing], if not the only originator. The other modification I suggested at that time was to reduce the input shaft end play down to 0.016 - 0.034 inch.” He elaborated:
In 1968, I took a job as transmission mechanic at Will-Mar Dodge on LaBrea Blvd in Los Angeles. Some time in 1969, the Service Manager told me that an almost new Dodge Polara wagon was to be brought in with transmission trouble. The owner and his family had left for a trip to Sierra National Park; on the way there and back, the transmission had failed and been repaired two or three times. A factory representative was to observe the disassembly of the transmission.
The transmission was behind a 383 two-barrel engine, just a run-of-the-mill Polara wagon. With the transmission on the bench, the first thing was to check the input shaft end play. It felt like a good quarter of an inch. Removing the front pump/reaction shaft support, we saw good reason for all that end play!
The reaction shaft support was destroyed where the front clutch retainer rides, the seal ring lands had eaten into the front clutch retainer and the bushing was destroyed. Both the input shaft and the output shaft were heavily scored at the point where the thrust washer rode between them.
The transmission was completely torn down, all the parts thoroughly cleaned and inspected, and many replaced, including all the seals, clutch discs, clutch plates, and kickdown band. The valve body was completely disassembled and cleaned, cooler lines backflushed, and torque converter replaced.
The representative asked me what I thought had caused the failure. My answer was the input shaft end play was too great; the specifications allow too much clearance, (0.034 to 0.084 inch). My recommendation was 0.016 inches to as little more as possible. This would prevent the front clutch retainer from wobbling on the reaction shaft support, especially with that narrow retainer bushing, causing wear to the front clutch retainer and the resulting front clutch failure. Clearance is controlled by a selective thrust washer on the reaction shaft support, available in 3 thicknesses. The other was that the front clutch retainer bushing was too narrow, and I suggested it be widened.
I had no idea that they would consider upgrading to a wider front clutch retainer bushing, so the end play was my primary suggestion. There could have been other, similar failures, but the timing seems strangely related.
I assembled the transmission under the watchful eyes of the factory representative, and we road tested the car making sure the throttle pressure linkage was properly adjusted.
I moved on from Will-Mar Dodge some time in 1970, going to work for the Automobile Club of Southern California. Later, moving on to the Oxnard/Ventura area, I worked for four dealers over the years and finally, in 1988, opened my own shop in the Ventura area.
Some time around 1998, I had a phone call from a gentleman living in Oxnard who asked me if I was the same “Hemi” who had worked at Will-Mar Dodge in the 1960s. He went on to tell me that he had been the owner of that Dodge Polara and had recently sold it. He went on to say that since I had repaired his transmission in 1969, he had not had any further issues and it was still working perfectly when he sold the car.
This was with a front clutch retainer that still had the narrow bushing, proving that reduced end play plays a greater role in preventing failure than the wider bushing. The wider bushing could tolerate greater end play, so the cure was resolved that way by Chrysler. It is still my opinion that less end play is the way to go in either case.
I have to think that this incident in 1969 had some impact on the upgrading to the wider bushing in the A727 transmission in 1971 and subsequent years. Line mechanics in dealerships don't get any credit, though.
Certain items were altered in the 1965 model year: enlarging of the torque converter splines on some 1965 models, elimination of their flanged output shaft, and changing the shift mechanism from two cables to one rod-and-lever arrangement. These changes have been incorporated on A-904s and 727s since 1966.
Modifications made since 1966 have been small and related to longevity and driver comfort. 1971 saw the wider front clutch bushing that eliminated premature ring and drum wear. I consider this to be the best modification, as far as high performance is concerned.
The automatic transmission is a convenience item for most people. The shift lever is moved to get the car in motion and then left alone until the car is stopped. No clutch pedal to push or shifter to change between three, four, or even five forward gears makes for a minimum of driver effort [editor’s note: by 2014, up to nine forward gears!]. Not only were automatics originally used for general driving comfort, they quickly found a place in competition vehicles.
In the 1960s, Chrysler products dominated various drag strip classes, in both AHRA and NHRA. The majority of the cars were running manual transmissions as they were faster and easier to repair. However, the shock of 4000 to 6000 rpm drops on the starting line created enormous loads on tires and driveline components. If it wasn't for the fact that tires were a lot slicker then, there probably would have been an incredible amount of additional parts breakage.
Chrysler was the first to develop better racing automatics since they already had the most dependable automatic transmissions on the road. It took G.M. until 1964 to develop the famous Turbo-Hydramatic 400. Up to that point in time, only the Powerglide was seriously used in G.M. racing vehicles.
The aluminum torqueflite revealed many more classes that Chrysler could dominate. Evidently, early engineering groups at Powertrain Design did their homework, because the original design has changed little as Chryslers still dominate A through D stock automatic in NHRA classes. With bracket racing coming alive in the 1970s, it has been almost mandatory to have a good automatic in 90% of the classes if you want to be consistent and win races. Torqueflites still fulfill the needs of Chrysler racers.
Automatics have usually been four to six percent less efficient than manual transmissions. Torque converter slippage was a big contributing factor to this inefficiency; in 1978 lock-up clutch-type converters [a Chrysler invention] were added with Torqueflites, to help reduce the gas mileage difference between automatic and manual transmissions. I will not go into much detail on these because they are very similar in design to stock 727s and 904s with the addition of a few control valves and a slightly different torque converter.
As gas mileage demands increased, thereby decreasing engine size, Chrysler changed from 727s to more 904-type transmissions (G.M. created the Turbo 200-type automatic for smaller vehicles, and what a money maker this has been for automatic transmission repair shops!). No real problems were noted by changing to the smaller unit except for 1978 early lock-up shuddering, remedied by delaying the lock-up speed a few mph later that model year.
There has to be something to transfer motion from the engine to the transmission and that item is a Torque Converter. This unit does what its name implies; it converts or “multiplies” engine torque at low speeds and couples the engine to the drive train hydraulically at higher speeds.
The converter consists of three elements, all necessary for proper transmittal of motion. The Impeller and Turbine transfer fluid between each other to drive the Input Shaft of the transmission. The Stator gives proper direction to the fluid flowing between the impeller and turbine. The converter hub has slots that engage the tabs on the inner rotor of the oil pump to create oil pressure and oil flow from the pump while the engine is running.
The overall purpose of the torque converter is to:
Refer to either Chrysler literature or other applicable information for more details about the internal “Vortex, Rotary and Turbulent” flow of fluid if you are interested in more of the converter operation and design. Converter diameters and stall speeds will be discussed in the modification section.
Before going into details of each subassembly in the Torqueflite, refer to the exploded view of a complete 1976-727 transmission. The over-running Clutch is not shown and the Valve Body is actually a 1970 model. There are reasons for this but they are not important to the overall explanation of the internal operations.
From top left you will see the Pan, the Case, and the Tail shaft, or Extension Housing and associated hardware. The top right corner contains all the Servos, Bands, Operating levers and pins, and the Accumulator and its spring.
The next row has the Oil Pump and the Front Clutch assembly shown. Row three displays the Rear Clutch assembly. The fourth row holds both Planetary Gear sets, the Front Clutch hub, Sun Gear and Sun Gear Driving Shell, and the Low-Reverse Drum assembly.
To the far right is the oil filter and “Direct Connection” oil filter extension. This extension is used with the deep oil pan shown in the upper left corner. The last row contains the Output Shaft, Governor Assembly and the Output Shaft Support Bearing. To the right of this is the “Brain” of the transmission; the Valve Body.
I will explain the pictures of the various units in the general order they are displayed and tell what functions the pieces have in the overall transmission operation in the following text.
Shown is a 727 case and associated pieces (the 904 is almost identical except the size and shape of everything is slightly reduced). This case is 727-A which fits all small blocks. The 727 can also be a 727-B which fits the big blocks, or 727-RG which bolts to the six-cylinder blocks. The 904 comes in a 904-A, 904-RG, as well as units for some of the earlier 4-cylinders. Some higher performance 904 units were designated as 998s and 999s in the later 1970s.
The case is in the upper left corner. Below it are the cooler line fittings, the shifter shaft seal, and neutral start switch. The neutral start switch allows the engine to be started when the transmission is in only park or neutral range, and it turns on the reverse lights when needed (1968 and earlier cars had the reverse lights controlled from a switch on the steering column or console shifter).
Next is the case-to-tail shaft gasket and the governor support. The governor support bolts to the case and it has three purposes. It supports the low-reverse drum, the governor-output shaft assembly, and it directs oil to and from the rotating governor weights (The governor works with the throttle pressure circuit to control shift points). Next is the tail shaft housing that contains the parking gear lever (used to lock the governor-output shaft assembly in park) and the output shaft support bearing snap ring. A seal is used at the end of the housing to ride on the drive shaft yoke to keep oil contained in the case.
Below all this is the oil pan (I've shown the Direct Connection deep pan with a drain plug), the fluid source for the hydraulic system (some oil cooling is also done by the pan). Not shown is the filler tube and the dipstick, used both for checking the level and for filling the system.
Shown in this photo are the bands necessary to give second gear (kickdown band), and reverse or manual low (low-reverse band).
In the upper left corner is the low-reverse band assembly consisting of the retaining snap ring, the spring retainer, the large diameter return spring, the piston plug, the piston plug spring, the piston itself, the synthetic rubber piston seal, and the snap ring that holds the piston plug and spring to the piston.
Moving to the top of the photo again, is the low-reverse lever assembly, its adjustment screw and lock nut, the reverse lever shaft and O-ring that the lever assembly pivots on and the rectangular strut that applies the low-reverse band. Right below the band is the anchor that holds the band when the piston pushes the low-reverse lever, which is also held in place by the reverse lever shaft.
An accumulator spring, a cast iron sealing ring, the accumulator piston and one more sealing ring are in the center of the photo. Both 904 and 727 Torqueflites use the accumulator to give a cushioned drive breakaway only, depending on which side of the piston the spring is placed.
In the right half of this picture is the kickdown servo assembly, and the kickdown band. The kickdown servo assembly consists of the retaining snap ring, cast iron seal ring, the aluminum pin guide, the large servo return spring, and a controlled-load servo with its cast iron rings below it.
This particular servo piston is called a controlled load servo; it cushions the three-two downshift. The other type of servo assembly does not use an internal spring-piston assembly, and is usually found in high performance applications where soft shifts would be detrimental. Individual pieces cannot be interchanged between the standard and controlled-load servos.
The lever at the lower right corner comes in different ratios, again depending on engine usage. Usually the higher performance transmissions use a 3.8 to 5.0 ratio lever. The higher numbered levers multiply the apply force more to tighten the band quicker and harder around the front clutch retainer drum.
The 904 uses two different types of low-reverse bands. One type is called a double wrap band, similar to the Hemi kickdown band; it is used with the V8 engines to give greater low-reverse drum holding capability. The other band is a single width band that has less torque holding capacity and is used with six-cylinder engines. The 727’s single width low-reverse band and is used with all engines.
The low-reverse band works with the over-running clutch (not shown) to hold the low-reverse drum during manual low. The 904 has a smaller over0running clutch so it depends on the band to give it sufficient holding power to prevent breakage of the clutch. I'll explain later why the low-reverse drum must be held stationary, but for now bear in mind that the drum needs to be held during manual low, drive breakaway (or low), and reverse. The over-running clutch is like a stator in the torque converter, but its outer race is firmly affixed to the rear of the case whereas the outer race of the converter stator is free to lock-up or spin depending upon engine-vehicle speed difference.
The 904 uses different kickdown bands to hold the front clutch retainer drum during second gear. Most 904s have a single width band, but some 998 and 999 versions of the 904 have a drum with five clutches, and need a wider band to hold it. Some 904s and later 727s use what is called a “flex band” that works the same way as the standard cast band, but will not retain the circular shape when removed from the transmission.
A double wrap band (also called the Hemi or 440-6 pack band) was used with the five clutch retainer drums in the maximum performance 727s. This band could produce much greater holding power than the single width cast or flex band on most other 727s. One might think it wise to use the Hemi-band with the standard 727 drums, but it will not work because of the width difference between the four and five clutch drums. I will explain various tricks that can be done with bands, apply levers and drums in the modification section.
The oil pump provides the hydraulic pressure needed to operate the units in the transmission. It is made up of the pump body, two rotors, and the reaction shaft support. The first rotor has 14 external teeth and the second rotor has 15 internal teeth, The rotor with external teeth is driven by the torque converter hub and it meshes with the other rotor at only a few locations thereby creating a gap. As the rotor spins, fluid is picked up in this gap and is forced out where the teeth are meshing creating pressure to drive the fluid through various passageways of the pump body (The 904 pump is physically smaller and the pump body and reaction shaft support each make up one-half of the pump body assembly).
The reaction shaft holds the stator stationary in the torque converter and it supports the input shaft in its I.D. and the front clutch retainer on its O.D. The front clutch is sealed to the reaction shaft by the two cast iron rings that direct oil from the pump to the inside of the front clutch.
The oil pump body has a bushing to support the converter hub and a seal to prevent fluid leakage around it. The reaction shaft has a bushing in it to support the input shaft (which is part of the rear clutch assembly), oil pump assembly.
Pressure forces the lip seals against their surfaces and pushes the piston away from the clutch hub and applies force to the clutch pack through the Belleville spring-pressure plate assembly.
The rear clutch has to hold a larger torque load than the front clutch so the Belleville washer is used to multiply the apply force of the piston. The Belleville washer also acts as a spring to return the piston when the clutch is not hydraulically functioning. Once the plates and discs are locked in the rear clutch assembly, power can be transmitted through the input shaft, to the locked clutch-pack and to the front annulus gear (see planetary assembly photo). Once the planetaries are utilized, you get gear reduction and torque multiplication; or in other words, first or second gear depending on planetary usage.
Both the front and rear clutch have a check ball in the apply piston, not easily seen in the photos. When the clutch is spinning but not hydraulically operating, residual fluid can be forcing the piston out against the pressure plate (front clutch) or Belleville washer (rear clutch), creating clutch drag. This ball-check valve lets that fluid (that is under pressure from the clutch assembly spinning and throwing the fluid out due to centrifugal force) out of the clutch cylinder or drum. The 904 rear clutch is smaller but similar to the 727; and the front clutch hub-rear clutch retainer is in two separate pieces in the 727 but only one in the 904.
I've grouped three basically separate assemblies in this one photo, both above and below. The planetaries and sun gear in the top row give gear reductions and reversals; the output shaft transfers output torque; and the governor controls the shift points.
The planetary parts include four thrust washers and:
The last three pieces are contained in the low-reverse drum. The front annulus gear is splined to the rear clutch discs and it drives the front planetary. Depending on gear selection, the sun gear-driving shell assembly will be driven by the front or rear clutch packs and will transmit torque to the rear planetary assembly. The thrust washers prevent wear between the rotating members.
The output shaft is shown with its thrust washer on the left; it prevents wear between the input shaft and the output shaft. It is fiber or bronze, depending on the year of the transmission.
The output shaft has four sets of splines cut into it. The splines on the left fit into the front planetary gear, while the second set locks the output shaft to the rear annulus gear and the third set holds the governor-park pawl assembly on the shaft. The splines on the far right are used to connect the drive shaft yoke solidly but still allow it to slide easily on the output shaft. The shaft also has gear teeth cut into it to drive the speedometer gears.
In the lower right portion of the picture is the governor assembly and the output shaft support bearing with their associated snap rings. The bearing supports the loads imposed on the output shaft by the internal mechanisms of the transmission, and maintains the correct end play in the gear train.
The governor works with the throttle pressure rod to control shift points, according to the vehicle speed and engine torque requirements.
Two cast iron sealing rings, the governor body-park pawl assembly, and the governor valve-governor weight assembly all work together to send the governor signal to the valve body to tell it to operate various shift valves, controlling up and down shifts. There is one snap ring used to hold the governor body park pawl assembly secure on the splines on the output shaft.
As in the other units, the 904 planetarys, output shaft, and governor assembly are similar to the 727 pieces, but smaller in size. Depending on the size, both will have three or four planet pinion gears in the assembly. The four pinions will be found in higher performance units.
The valve body can be considered the brain of the transmission. It takes hydraulic signals and fluid pressure to time and control all transmission functions. The springs and valves work together to modulate fluid pressure to time application of the bands and clutches. The channels that resemble endless mazes transfer fluid from valve to valve, or from valve to case passageway.
The 904 and 727 valve bodies are very similar, as are the lockup units. The valve bodies are extremely well designed, and it is hard to make a mistake while assembling them. As long as the springs are not interchanged or the balls mixed up, they always seem to work. Comparing Chrysler valve bodies with Ford units is like comparing 426 Hemis with flathead Fords, no comparison can be found. GM units are pretty slick but not as much fun to work on.
After reviewing all the previous information to get an idea what the various assemblies look like and do, the Torqueflite can now be studied as a complete unit. I'll go through a gear by gear description explaining the active components and how they combine to get the different gears. Because of the skill and generosity of a close friend, I'm able to supply a block diagram of the Torqueflite in the various gears.
Neutral and Park are similar in the Torqueflite. No hydraulic units are operating and, with no clutches or bands applied, the input shaft-front clutch hub assembly and the driving discs in the rear clutch pack are rotating at engine speed. The rear clutch driving discs that are splined to the front annulus gear are free to spin, and the annulus gear is stationary. Therefore, all power flow stops at the rear clutch retainer.
When in Park, something has to lock the output shaft solid to prevent unwanted vehicle motion. This is done by a piece of linkage that connects to the parking gear lever and the valve body manual gear lever. This parking gear lever is pushed into the parking gear-governor assembly by a cam on the parking linkage. The parking gear-governor assembly is splined to the output shaft so when the gear is stopped, the output shaft is stopped.
There is also a hydraulic difference between neutral and park: the “Park” location of the manual valve (inside the valve body) allows fluid to leak from the valve, creating a line pressure drop. This low pressure in park keeps the converter from completely filling and loading the engine unnecessarily.
In neutral, the converter is filled and there are no “controlled leaks,” as in park. This is the reason Torqueflites should always have the level checked only in neutral. If you check the level in park, the level checks higher than it really is.
As mentioned earlier in the text, the valve body takes the hydraulic signals from the governor, throttle pressure rod, and manual shift valve, and combines these with fluid pressure from the oil pump to time and control all transmission functions.
I am not going to describe each valve or spring just yet. I'll incorporate that into the modification section. For now, I'll just point out that the filter and filter extension are at the top, the valve body and associated valves, springs, balls, and end plates are on the left, the separator plate is in the right-center and the transfer plate is at the right.
The springs and valves work together to modulate the fluid pressure to time application of the appropriate bands and clutches. The channels in both valve body and transfer plate that resemble endless mazes, transfer fluid from valve to valve, or valve to case passageway.
Sometimes Chrysler owners complain of delayed motion of the vehicle after dropping it in gear from park. This is due to the converter not being fully charged in park. The best remedy for this is to start the vehicle in park and then immediately shift to neutral. This will fill the converter which gives a quicker vehicle response.
The block diagram (1) has the converter and pump shaded to show they are the only units in operation. In park, the driveline will be locked as explained above.
Drive breakaway (or low gear) is used to move the vehicle from a dead stop or very slow roll.
When the shift lever is placed in drive, the rear clutch drive plates rotate the front annulus gear pinions which then spin the sun gear in a reverse direction. The sun gear then rotates the pinion gears of the rear planetary in the same direction as the engine, which gives a 2.45:1 gear reduction. The rear planet carrier is splined to the low-reverse drum which is held stationary by the over running clutch. The rear annulus gear is locked to the output shaft, so when the rear planets are rotating, the output shaft is spinning.
Reviewing the mechanics of this again will show that the planet pinions of both planetaries transfer the torque from the input shaft to the output shaft. As the input shaft turns 2.45 times, the output shaft turns once.
In drive breakaway, the rear clutch is applied and line pressure is directed to various valves in the valve body. The same oil is also present at the governor and the signal is modulated and returned to the valve body to prepare for the 1-2 shift.
When the selector lever is placed in the low gear position, something else is needed to supplement the existing Drive Breakaway units. Manual low slows the vehicle down when in motion or holds the transmission in low gear longer than the governor and throttle pressure circuit normally allow. Engine braking uses the low-reverse servo to apply the low-reverse band around the drum. When the vehicle is stationary, the over-running clutch will hold the drum, but when the vehicle is already out of low gear, the drum is free wheeling and can only be stopped by the low-reverse band. (Low gear is a combination of the front and rear planetary sets, so the low-reverse drum has to be stationary to hold the rear planet carrier).
It can be seen that manual low is identical to Drive Breakaway with the addition of the low-reverse servo-band assembly and the blocking of the governor signal to the 1-2 shift valve.
The Torqueflite is shifted automatically into second gear by a governor signal pushing the 1-2 shift valve over and allowing line pressure to be directed to the front servo-kickdown band assembly and to the accumulator. The governor signal is created by the rotation of the output shaft, which throws the governor weights outward and sets up the proper pressure (I'll go into more of the pressure relationships in the modification section). Governor pressure is also directed to the 2-3 valve in preparation for that shift.
Since the rear clutch is functioning, and the kickdown band had now been applied, the front clutch drum has now been stopped. The drum is interlocked to the sun gear driving shell with tabs that engage slots in the shell. The rear clutch is still rotating the annulus gear and the sun gear is stopped so no torque is transmitted to the rear planetary. With the sun gear held and the annulus rotating, the planet carrier is spinning at a reduced speed. The planet carrier is splined to the output shaft so it, too, rotates at a reduced speed which is 1 turn for every 1.45 turns of the input shaft.
Hemi Andersen added, “The overrunning clutch holds in all forward drive conditions. For every action there is an equal and opposite reaction, so if the overruning clutch was not holding, the car would not move forward. Since it is shown ‘holding’ in drive breakaway, it would continue holding in second and direct drive as well.”
Manual second is exactly the same mechanically as Drive second, but there is a hydraulic difference. With the lever in second, the vehicle starts in low and shifts to second but line pressure blocks the 2-3 shift valve in the valve body to prevent any 2-3 upshift. The governor pressure will never overcome the line pressure on the 2-3 valve until the transmission is shifted into drive. All the same units are operating when the transmission is downshifted from drive into second. The transmission cannot upshift until the lever is moved back into Drive because the 2-3 shift valve is being blocked by line pressure.
While the transmission is in second gear, there is governor pressure on the 2-3 shift valve, the front clutch retainer is stopped by the kickdown band, and the rear clutch is applied, turning the front planetary and, therefore, the output shaft.
When the governor signal moves the 2-3 shift valve, fluid is applied to the release side of the front servo and the front clutch is applied. With the band released, the front clutch starts spinning. Now torque is applied to the front annulus gear by the rear clutch and torque is also applied to the sun gear driving shell by the front clutch (the driving shell is attached to the sun gear). When the sun gear and the front annulus gear spin at the same speed the planet carrier has to rotate at that speed (gear train rules dictate that when two members of the same gear rotate at the same speed, the whole gear set rotates at that speed).
Remember that this planet carrier is splined to the output shaft so it, too, turns at engine speed. Direct Drive is named for the fact that engine speed is transferred directly in a 1:1 ratio through the transmission.
When the Torqueflite is shifted into reverse a couple of interesting things happen. First, the rear clutch, used in all forward gears, is deactivated. Then the front clutch is applied, which locks on to the front clutch hub-input shaft assembly.
Along with application of the front clutch, fluid is directed to the low-reverse servo, which locks the low-reverse band tightly around the low-reverse drum-rear planet carrier assembly. Engine torque can now be transferred through the transmission.
Looking at this in order, it happens like this: The front clutch is appliedm transferring engine torque through the sun gear driving shell, driving the sun gear in the same direction as the engine. With the low-reverse band holding the rear planet carrier stationary, torque is transmitted only through the planet pinions (via the sun gear), driving them in a reverse direction. The only thing left to spin is the rear annulus gear which is splined to the output shaft.
Using this planetary only gives one turn to the output shaft for every 2.2 turns of the input shaft, again remembering the output rotates the opposite direction of the input shaft.
Hydraulically, there are also a couple of interesting things to note. Governor pressure will not be used in reverse, and a quick line pressure check shows reverse to have about 250 + or - 20 psi (where normal line pressure is 75+ or - 19 psi). It is not immediately obvious why reverse should be so high until the rear clutch and front clutch are compared.
The rear clutch is on in all forward gears. It uses a Belleville spring washer to multiply the piston apply force the necessary amount to transmit engine torque through the rear clutch pack. However, the front clutch has no multiplication factor to depend on when it is applied.
Since direct drive is the only other time the front clutch is used, and the vehicle is already in motion, no extra holding power is needed to sufficiently lock the front clutch drum. But going into reverse, the vehicle is stationary, and much greater torque (2.2 times as much) will be transmitted through the front clutch plates. Therefore, through a couple of hydraulic tricks, line pressure is almost tripled to give the front clutch the necessary holding power.
This is the second part of a three part series devoted to the Chrysler Torqueflite Automatic Transmission. I'll give the reasons that I modify Torqueflites and why the majority of 904s and 727s can benefit from a few well proven alterations. Advantages and disadvantages to modifications will be explained. There are many types of kits made to improve Torqueflite and I'll give my recommendations on the ones I'm familiar with. Prices of kits and recommended usage will he discussed.
Of part one of this series has been read, it will probably be bringing up a few questions as to why a Torqueflite could be so well designed yet still need modifications. Alterations to any original design do not necessarily make that design bad; it just shows that some individuals like things to be different in one way or another.
Prior to learning the Torqueflite’s internal operation and configuration, I was reluctant to tamper with originality. To overcome this reluctance and to learn at the same time, I got a spare valve body from a salvage transmission and followed the Chrysler service manual in disassembly and reassembly procedures. After I determined that the valve body was operational and was a suitable substitute for the one in my car, I performed all my modifications. This was an easy way to eliminate worries in case the valve body functioned incorrectly. The old unit could always be put back into service while the discrepancy was corrected.
I will present information that applies to both the 904 and the 727 (Most front wheel drive vehicles are not used for heavy duty purposes or racing as of yet so not much has been written on them). The information I'll give comes from my past experience and selected other references.
Stock Torqueflites, since the very beginning, have been very reliable. In a vehicle’s lifetime with normal operations, little, if any, maintenance was needed. The trucks and high performance cars were equipped with stronger, harder shifting automatics; the reason being that Chrysler was aware of the intended usage of these vehicles.
The “stronger” transmissions consisted of front and rear planet carriers with more pinions, front and rear clutch assemblies with more discs and plates and sometimes larger, higher friction coefficient kickdown bands. Various hydraulic modifications were made by changing spring rates in the valve body and the kickdown band apply strength was altered by a mechanical leverage change. These changes all worked together to give a higher torque-capacity transmission as well as a stronger, more defined shift.
Even these changes weren't enough in some cases. Muscle car units and fleet service units (taxi, ambulance, police, rental, etc.) needed more torque handling capacity as well as severe service capabilities. These requirements brought the home mechanics and aftermarket manufacturers to some of these same conclusions; design and construction of transmission shift modification kits were needed.
Several manufacturers introduced shift improving kits. These kits, when used as directed, were helping to eliminate a lot of premature failure of severe service transmissions. However, most of these modifications for competition and heavy duty service sometimes made the transmission too severe for everyday use. It became apparent that modifications should meet three important requirements: shift performance, longevity, and driver comfort.
The last sentence of the previous paragraph determines both the pros and cons of Torqueflite modifications. In order to get longevity you must give up a small amount of driver comfort. It is basically a trade off. The Torqueflite is designed in such a way to give relatively slow, soft shifts; and the 2-3 shift is the one I think needs the most improvement. To refresh memories, I'll explain the 2-3 shift function again. When the Torqueflite shifts from 1st to 2nd, the kickdown band locks around the Front Clutch Retainer to stop the sun gear driving shell. To make the 2-3 shift, the kickdown band has to hold until the front clutch is applied and then it has to be quickly released. This timing is very important; too early of a band release causes the engine to “rev up” until the front clutch is applied. Too late of a band release and/or too early of a clutch apply is more common; this gives a “putting on the brakes sensation” which is actually called shift overlap (Since the rear clutch is applied in both 1st and 2nd, and nothing has to be released, kickdown band timing is not as critical in the 1-2 shift). Delay in the kickdown band-release and premature clutch-apply both cause overlap and, therefore, friction material wear.
Advantage number one to Torqueflite modification is elimination of the 2-3 shift overlap which gives a crisp, efficient shift from second to third, prolonging the friction material life. Why is this not factory accomplished? Probably due to the fact that 90% of vehicle owners believe a soft, slow shift is better on the transmission than a firm one, so original equipment units (excluding factory high performance units) are programmed for this shift.
The second advantage is firmer 1-2 and 2-3 shifts. The word firmer is relative; the shifts are not severe (in most cases this depends on kit manufacturer as well as intended usage), they are just more noticeable. Again, these firm, properly timed shifts eliminate slippage of friction material during shifts. A couple of other advantages are: more fluid is supplied to various hydraulic units which sometimes compensates for seal leakage or wear; and certain kits give control of the 3-2 and 2-1 downshifts to the driver (this comes in handy during hilly terrain). Some valve body modification kits give complete control of the shifts to the driver. These are normally referred to as “full manual” or “full race” valve bodies. Obviously, these are not intended for Imperials or anything used for normal transportation.
I've only experienced a couple of things I call disadvantages. The most common complaint after modification is due to improper product selection (I use only modification kits from a company called Transgo in El Monte, California. This will be justified and explained a little later). (Neither Allpar nor the WPC Club endorses any product; the recommendations above are the author’s opinion only.) The wrong kits will cause too severe 1-2 and 2-3 shifts, harsh park or neutral-to-reverse shifts, and some light throttle 2-3 overlap. Improper adjustments after the modifications are performed also causes light throttle 2-3 “thunk.” I use this term to describe the noise you hear when the 2-3 shift is not correctly timed. This is still 2-3 shift overlap (at slow speeds) and minor adjustment usually eliminates this.
Another disadvantage might be the price. Valve body modification kits generally cost from $25 to $65. On the average, when performing the modifications yourself, expect to pay about $55 for a kit, $10 for new fluid and a couple of dollars for a filter. [Note: 1989 prices]
One way to eliminate disadvantages is to select the proper kit. I'll give my recommendations shortly.
This is the point where I'll plug for what I know works. For the same reason we stick with Chrysler products, I stick with Transgo; performance and dependable results. I've experimented with different types of products and they won't be named here. I've had no severe catastrophes; it’s just that the shifts were too firm all the time or they weren't firm enough at the correct time, and sometimes the instructions weren't easy to follow. (Transgo has the exclusive rights to the “Shift Kit” and “Reprogramming Kit” names I'll mention from now on.)
The “shift kit,” or “SK-TF,” is a small version of a reprogramming kit; it can be used in fleet service or everyday vehicles. It eliminates the 2-3 shift overlap, firms up the 1-2 and 2-3 shifts (not severely), provides more fluid to the front clutch circuit, and gives a bypass to prevent excessively high fluid pressure during cold weather starts and higher RPM service.
Transgo manufactures three reprogramming kit for Torqueflites. These actually reprogram the valve body; various springs are changed and fluid passages enlarged or restricted to give the correct shifts at the correct time.
A.“Tow or Go or TF-1” is for the fleet service and everyday vehicles. Additional driver downshift control and all the features of the “SK-TF” are provided. This is a good kit for motor homes, taxis, and police vehicles; good shifts, firm, yet comfortable
B.“Competition or TF-2” is good for full race and extremely heavy campers or motor homes. Shifts are definitely firmer and some driver comfort will be sacrificed. Additional full throttle and down shift control is provided in addition to all the standard Shift Kit features. However, this is still very streetable and reliable (95% of my Torqueflites get these).
C.“Stick shift or TF-3” is a full-race, full-manual controlled “Reprogramming Kit.” It is very similar to the TF-2 except that the driver controls all up and down shifts (It is convertible back to automatic shifting if required). This is good for full-race vehicles or ones where the driver likes to do a lot of shifting.
For people that stick to Chrysler-engineered products, Direct Connection offers two “shift-improver” kits; one for the 904 and one for the 727. These are inexpensive and easy to install. They work well and they, too, are a good choice for everyday as well as some fleet use. You don't get as much driver control and the 2-3 shift is not as good as the Transgo kits provide but they are still a good value for the dollar.
Everything in the previous section on kit variation deals with external modifications, or things that can be done with the transmission still installed. However, if the transmission is removed, a few more key items can be changed to give bullet proof operation.
Along with a “Shift Kit” or “Reprogramming Kit,” revised clutch clearances and band adjustments can be made. I also add a higher friction coefficient kickdown band to every high-performance unit I build. This makes an incredible 1-2 shift and it only costs an additional $30-$35 (The 1-2 shifts are firm enough that I don't recommend this for luxury vehicles). I use Raybestos-Manhatten waffle-cut front clutch discs and Raybestos-Manhatten smooth rear clutch discs and Perfect Circle steel sealing rings. Good overhaul gasket and seal kits are available from many different sources so check around. To finish it off, normally I install the “Direct Connection deep pan and pick-up” package from Chrysler.
By following simple recommendations available from Transgo, a home mechanic can really assemble a neat package. Using these suggestions and products has allowed me to build many street and race Torqueflites with zero failures. I attribute 75% of this to product selection and 25% to very careful attention to assembly details.
To write an article on modifications is rather difficult without mentioning manufacturers’ names. I don't intend to down-grade ones I don't use or aren't familiar with; that’s why I mentioned only the few vendors I did. I know what has worked best for me and I'm sure the same results are possible for anybody.
Should you modify your Torqueflite? If you want it strictly original, no. But if you want added reliability and performance and you have some mechanical skill or a friend that does, set aside a couple of hours and start on your transmission. If it’s already out of the car, read Transgo’s additional recommendations, and decide what you really want your transmission to do for you.
The Torqueflite is relatively straight forward. you'll get a lot of enjoyment with a minimum of problems out of your modified Torqueflite if you are careful. Remember to keep your Chrysler service manual handy and follow whatever kit instructions you finally decide upon. Write in a letter, through the club office, if you have any questions on where to get kits and literature, etc.
TorqueFlite main page • Transmissions page • Identifying and exchanging • Part II (modifications) • Part III (repairs).
More Mopar Car and Truck News
Jeep Wrangler Clockspring Warranty extended • Stratuscaster • Larry and JD Power on FCA quality