6
Automatic Transmissions: Control and Calibration

6.1 Introduction

As discussed in the previous chapter, automatic transmissions (AT) that use planetary gear trains (PGT) realize multiple gear ratios by alternately engaging and disengaging hydraulically actuated clutches according to the transmission clutch table. A gear shift in the transmission involves the engagement of an oncoming clutch and the release of an off‐going clutch. Two technical issues naturally arise: (a) when should transmission shifts be initiated, (b) how should the clutches involved in a shift be controlled so that the shift is made quickly and smoothly? The primary functions of transmission control systems are therefore to address these two issues.

Clutches in automatic transmissions of early types were both hydraulically actuated and controlled [1]. A gear shift was initiated by a governor that was driven by the transmission output shaft [1,2]. The governor assembly is actually a centrifugal mechanism that has weights rotating with the governor and swinging about a pivot attached to the governor. As the governor rotates, the weights will swing outward due to the centrifugal effect. The position of the swinging weights depends on the governor rotational speed which is in turn linearly related to the transmission output speed. As the weight swings outward, it switches on a pilot hydraulic valve to trigger a shift corresponding to a specific vehicle speed. Once a shift is initiated, the responsiveness and smoothness depend on the hydraulic control circuit that controls the apply pressures of the involved clutches. Hydraulically controlled automatic transmissions had to be exclusively applied in the automotive industry right up to the 1970s.

Automatic transmissions with partial electronic control date back to the early 1970s [3]. In these early types, only shift point control, or shift schedule, were controlled electronically. In a transmission controlled fully electronically, a microcomputer is used to control both the shift schedule and the shift processes through hydraulic solenoids. Automatic transmissions fully controlled by electronic systems were developed by the automotive industry from the mid 1980s and were well received by the automotive market due to the enhanced vehicle fuel economy and drivability. After 1990, all newly developed automatic transmissions, typically with four speeds and lock‐up torque converters, were controlled fully electronically. In the control systems of these transmissions, two or three shift solenoids are used as switches in the hydraulic circuits to realize transmission shift scheduling, and one pressure control solenoid is used to control the line pressure and clutch apply pressure [4–13]. This control system layout and its variations were extensively applied for automatic transmissions until the turn of the century, when current generation automatic transmissions were developed and started to enter production in the industry. In comparison with the previous generation, current generation ATs generally have more speeds – mostly with six and some with eight speeds [14,15] – and more advanced control technologies that provide independent control for clutch pressure and line pressure via pressure control solenoids (PCS) or variable force solenoids (VFS). In general, the control system of today’s automatic transmissions possess the following functions:

• Shift schedule control: The transmission shift schedule is directly related to vehicle acceleration performance, fuel economy, and drivability. The transmission control unit (TCU), or the powertrain control module (PCM), which controls the whole power train system, makes the shift decisions with the input signals on the vehicle speed and the engine throttle opening, and other supplemental input signals on vehicle operation status, for the optimization of performance, fuel economy, and drivability. The shift schedule is established through intensive calibration and can be designed with adaptability and self‐learning features to fit the driving style of individual drivers. It can also be designed with different modes, such as “normal”, “economy”, and “sport”, to fit driver priorities and road conditions.
• Shift process control: Transmission shift response and smoothness depend on the accurate torque control for the oncoming and off‐going clutches. Once the TCU sends out the command signal to make a shift, the apply pressure for the oncoming clutch must be ramped up, and release pressure in the off‐going clutch must be ramped down, according to the designed control strategies validated by calibration. Technically, shift process control is the most challenging task in the development of a transmission control system. For transmissions with six or more speeds, there are already a large number of direct shifts – i.e. clutch to clutch shifts – based on transmission design. Each of these direct shifts corresponds to a specific vehicle operation condition and engine operation status and warrants separate clutch torque control and calibration in order to achieve optimized shift quality in the overall drive range.
• Torque converter locking control: The status of the torque converter concerns powertrain efficiency and transmission shift smoothness. An open converter functions as a fluid couple and is conducive to smooth shifts, but at the expense of efficiency which can be up to 3% fuel economy loss. To take full advantage of torque converter characteristics without sacrificing fuel economy, a torque converter locking schedule must be designed and calibrated as an important portion of the overall transmission control system. When the torque converter is locked or released, the lock clutch apply pressure or release pressure is ramped up or down using a pressure control solenoid to minimize powertrain transients.
• Engine torque control during shift: This is usually achieved by spark retarding in a selected number of engine cylinders. As discussed in the previous chapter, spark retarding during a transmission shift decreases the engine torque to lower the oncoming clutch torque required to complete the shift, thus enhancing shift response and smoothness. Obviously, spark retarding can only be made possible technically in vehicles that are equipped with electronically controlled automatic transmissions.
• System diagnosis and failure mode management: In case of component malfunctions, the control system will provide warning signals on the instrument panel and enable vehicle operation under safe conditions in the failure management mode. The vehicle can then be driven to a service station for diagnosis and repair. In addition, control and calibration variables may be modified for further improvements of transmission quality or may even need to be corrected after the vehicle model is already in the market. Therefore, the software of the transmission control system, including control codes and data, should have the capability of being updated if necessary.

The development of transmission control system is always focused on the fulfillment and improvement of the functions just described. This chapter naturally concentrates on the hardware and software technologies applied in the transmission control systems. Following this introduction section, Section 6.2 provides the functional descriptions for the hardware components, including pump, clutches and pistons, accumulators, sensors, various valves, and solenoids, in the transmission control systems. Section 6.3 presents the transmission control system configurations and the related design guidelines. Examples based on production transmissions of the last and current generations will be used to demonstrate the operation logic and functions of the control systems. The section will also provide an introduction to the development of control systems for automatic transmissions for the next generation. Section 6.4 will present concurrent transmission control technologies commonly applied in the automotive industry. The content of this section concentrates on the accurate control of clutch apply pressure during gearshifts. It also details the strategy for torque converter lock and release control. The transmission control system comes with a large number of variables and component characteristics that must be calibrated for the optimization of the control functions. The identification of control variables to be calibrated and calibration of transmission control system will be the topic of Section 6.5 at the end of this chapter.

6.2 Components and Hydraulic Circuits for Transmission Control

The control system of automatic transmissions is typically a synergetic combination of mechanical, hydraulic, and electronic subsystems or components. The mechanical aspects of automatic transmissions have been discussed in detail in Chapter 5. To better understand the functions of transmission control systems, it is necessary to have a general knowledge of the characteristics of individual components and hydraulic circuits. The following text provides the description on the functions of key hydraulic and electronic components, as applied in the transmission control system.

System ATF supply and line pressure control circuit: Driven by the engine, the hydraulic pump generates the ATF flow and pressure in the hydraulic circuit of the transmission control system. Gear pumps and vane type pumps have both been applied in automatic transmissions. The pump consumes power in proportion to its capacity and a fixed volume pump may pump out more ATF than necessary under a given transmission operation, resulting in fuel economy loss. Variable capacity pumps supply an ATF amount to the control system in proportion to the engine RPM to meet system requirements with improved pump efficiency. The design guidelines of hydraulic pumps can be found in open literature, such as SAE publications [2]. Typically, the hydraulic pump forms the system pressure supply circuit with a pressure regulator valve, a pressure limit valve, and a pressure control solenoid or variable force solenoid (VFS) for the supply and control of line pressure by regulating the AFT flow volume entering the main circuit, as shown in Figure 6.1. This circuit design and its variations are generally applied in the control systems of the last generation and current automatic transmissions [4,6,7], such as the GM four‐speed Hydra‐Matic 4 T80‐E [16] and the GM six‐speed AT Hydra‐Matic 6 T40/45 [17]. It can be predicted also that the control system for future automatic transmissions with full direct clutch pressure control will also be based on designs similar to the line pressure control circuit shown in Figure 6.1. Since the current generation ATs do not use accumulators in the control system, the torque signal port to accumulator control valve is eliminated in the PSC in Figure 6.1.

Figure 6.1 System ATF supply and line pressure control circuit.

Pressure regulator valve: The basic structure of this type of valve is illustrated in Figure 6.1. These valves can be designed with in‐port and out‐port variations. A pressure regulator valve regulates the ATF amount entering the main circuit, or pressure in equivalence, through the combination of valve spring and torque signal pressure, which is also termed throttle signal pressure. The sump port of the valve has a variable opening to let pumped ATF bleed through to the pump suction circuit. To boost the line pressure, the torque signal pressure controlled by a pressure control solenoid acts with the valve spring downward against the upward line pressure force on the valve body, diminishing the sump port opening and letting in more ATF. In the circuit shown in Figure 6.1, there is a reverse ATF port in the pressure regulator valve. When the transmission is shifted into the reverse gear, the ATF pressure in the reverse circuit is routed to the reverse port, pushing the valve body down with the valve spring to boost the line pressure that is needed to generate a large reaction torque in the reverse clutch. This also has the effect of filling the hydraulic system circuit quickly to make it ready for shift operations. Note that the pressure supply circuit shown in Figure 6.1 provides stable line pressure to the control system circuit with minimized pressure pulsation caused by the pump.

Pressure limit valve: As shown in Figure 6.1, this valve sets a limit on the line pressure in a very simple structure. The line pressure acts on the valve body against the valve spring. When the line pressure exceeds the limit allowed by the valve spring, the valve body will be pushed by the line pressure to open the exhaust port, thus limiting the ATF pressure passing through the valve. The ATF pressure after the limit valve is then routed to the pressure control solenoid, as shown in Figure 6.1, and to the shift solenoids that actuate transmission shifts. This pressure is therefore termed the actuator feed pressure.

Pressure control solenoid: This is the key valve that electronically controls the ATF pressure with high precision. The structure of a typical pressure control solenoid (PCS) is illustrated by a section view in Figure 6.2. There are two portions – an electronic portion and a hydraulic portion – in a pressure control solenoid. The electronic portion consists of coil assembly, armature, push rod, and spring housed inside the cylindrical frame. The hydraulic portion consists of valve core, spring, valve sleeve, and valve shell. There are three ports in the hydraulic portion on the valve shell: actuator feed fluid in‐port from the pressure limit valve, as shown in Figure 6.1, torque signal pressure out‐port, and exhaust port. In addition, the actuator feed fluid can also be exhausted from the variable bleed orifice between the valve sleeve and the armature. As can be observed in Figure 6.2, the function of the pressure control solenoid is to control the actuator feed fluid and convert it to the torque signal fluid pressure with high precision. The torque signal pressure is then routed to the pressure regulator valve and accumulator control valves as the pilot pressure for the control of line pressure and clutch apply pressure during shifts. Apparently, the pressure control solenoid is crucial for transmission shift feel since it is the primary, if not the only, component for the real time control of clutch apply pressure.

Figure 6.2 Section view of a pressure control valve.

Pressure control solenoids (PCS), also called variable force solenoids (VFS) or variable bleed solenoids (VBS), work on the principle of pulse width modulation (PWM). In general, the operation of the pressure control solenoid can be explained by duty cycle and frequency. A duty cycle is the percent of time that electric current flows through the coil of the solenoid in a cycle whose time duration is defined by the frequency. In other words, a duty cycle is the percent of time in a cycle the TCU sends electric current through the coil. A pressure control solenoid works at a given specified frequency. For example, GM transmission Hydra‐Matic 4 T80‐E uses a frequency of 614 Hz for the pressure control solenoid [16]. A 20% duty cycle means that electric current flows in the solenoid coil for 0.2 × (1/614) s (approximately 0.0003 s) in every cycle that lasts for 1/614 s (approximately 0.0016 s), as shown in Figure 6.2. The current in the solenoid coil generates a magnetic field that fills the center of the electronic portion in which the armature is located. The magnetic force, the force of the push rod spring, the hydraulic pressure force on the end of the valve core, and the spring force on the valve core interact with each other to vary the armature and valve core positions for the accurate control of the torque signal pressure.

The pressure control solenoid shown in Figure 6.2 is actually a variable bleed valve (VBS) and can be either “normally high” or “normally low”. A normally high pressure control solenoid regulates torque signal pressure in reverse proportion to the current in the coil. The line pressure control solenoid in Figure 6.1 is normally low. It controls the line pressure by bleeding more or less ATF from the valve as controlled by the TCU according to the engine load and vehicle operation conditions. The line pressure needed to secure transmission control functionality varies with the engine load which is reflected by the engine throttle opening. As engine throttle increases, the line pressure is also increased by bleeding less ATF from the pressure control valve through the decrease of duty cycles. By contrary, a normally high pressure control solenoid regulates the out‐port pressure in direct proportion to the current in the coil; that is, to increase the out‐port pressure, the TCU increases the duty cycles of the coil. Note that the PCS or VFS is energized by duty cycles through PWM by TCU, but the ATF pressure controlled is calibrated against the current through the solenoid coil. When applied in the transmission control system, the TCU monitors the current in the electric circuit from the solenoid to ground and uses this as the feedback to adjust the duty cycles for accurate control of torque signal pressure.

Shift solenoid: This is actually an electronically controlled switch valve in the transmission hydraulic circuit and only has two states: On or Off. A combination of On and Off states of shift solenoids defines a particular gear position. For example, two shift solenoids can be used to define uniquely the four gears in a four‐speed AT through the four combinations of On and Off states. A sequential shift can then be triggered by the change of status of one of the two shift solenoids. The section view of a normally open shift solenoid is illustrated in Figure 6.3. The line pressure under limit shown in Figure 6.1 is routed to the shift solenoid via an orifice and acts on the check ball. When de‐energized by the TCU, the combined force of the ATF pressure and the spring moves the plunger away from the metering ball, allowing the solenoid ATF from the pressure limit valve shown in Figure 6.1 to unseat the metering ball to be exhausted. When it is energized, current in the coil generates a magnetic field, which in turn generates a magnetic force on the plunger and moves it to seat the metering ball, blocking the solenoid fluid from the exhaust circuit. The solenoid AFT pressure is then used to control the positons of shift valves, which then route ATF pressure for clutch applications in different gears. Note that a pressure solenoid control valve can also function as a switch solenoid with On or Off states, in addition to its pressure control capability.

Figure 6.3 Shift solenoid and shift valve circuit.

Shift valve: These valves are used in the hydraulic system circuit for the transmission control unit to route the ATF flow in various operations, including park, neutral, fixed gear operations, and shifts. Typically, one shift valve is used for the shift between two neighboring gears. For example, GM four‐speed AT Hydra‐Matic 4 T80‐E uses three shift valves: 1–2 shift valve, 2–3 shift valve, and 3–4 shift valve in the hydraulic system [16]. A shift valve has multiple ports and two positions that are controlled by the solenoid ATF pressure from a shift solenoid. At each position, a shift valve will either block ATF from entering the clutch apply circuit or let ATF pressure through it to become clutch apply pressure in the clutch apply circuit. Note that the manual valve connected to the shift stick can also be considered as a shift valve manually controlled by the driver.

Clutch: Gear ratios of an automatic transmission are realized by applying different clutches according to the transmission clutch table, as discussed in Chapter 5. Three types of clutches – multiple disk clutch, band clutch, and one‐way clutch – are used in automatic transmissions. The one‐way clutch is self‐actuated and does not need to be controlled. Current generation ATs only use multiple disk clutches due to their compactness and characteristics in apply and release. Some six‐speed ATs of the current generation still use a one‐way clutch to facilitate the 1–2 shift as explained in Chapter 5. Multiple disk clutches can be used for both coupling and reaction. The basic structure of multiple disk clutch is shown in Figure 6.4. When applied, the two rotating components – the drum and the hub, – are coupled together by the friction generated between the friction disks splined to the hub and the steel plates splined to the drum. To apply the clutch, the clutch apply ATF pressure from a shift valve enters the piston chamber to fill the piston cavity before pressure is built up in the piston. The ATF then pushes the piston against the return spring and then the apply plate. The friction disks and the steel plates are then clamped against each other, and friction is generated on each of the contact faces between the friction disks and steel plates. As the apply pressure is ramped up via clutch pressure control solenoid, the clutch torque will be increased to the level for the full engagement of the clutch.

Figure 6.4 Structure of a multiple disk clutch and apply process.

There are several important variables or characteristics associated with the clutch apply process. As shown in Figure 6.4, clutch apply ATF will fill up the clutch piston cavity for some time while the clutch pressure is built up, and the piston will not move until the pressure build‐up is high enough to overcome the return spring force and push the apply plate afterwards. The clutch hardware designed this way in order to smooth out the harshness of the clutch apply. However, the initial clutch apply characteristics adversely affect shift timing and must be calibrated for the accurate control of clutch pressure ramp‐up profile. To eliminate the effects of the clutch apply side cavity, the return spring side can be designed with a counter chamber or compensator chamber [8]. Low pressure ATF fills up the compensator chamber and the piston apply side cavity when the clutch is open. This improves the accuracy of clutch apply timing and clutch pressure as shifts are initiated. As shown in Section 2.4 of Chapter 2, if disk wear is assumed to be uniform, the torque capacity of a multiple disk is given by:

(6.1)

where n is the number of friction disks, D and d are the friction disk outer and inner diameters, μ is the friction coefficient, A is the clutch piston area, F_s is the return spring force when the clutch is released, and k and x_p are the spring stiffness and piston displacement respectively. Clutch apply pressure is denoted by p_c, which is controlled by pressure control solenoid during shifts through clutch pressure control circuits.

Clutch pressure control circuit with accumulator: In the control system of the previous generation of automatic transmissions [16], hydraulic accumulators were usually used to dampen clutch apply harshness in a circuit shown in Figure 6.5. ATF from the line pressure control circuit shown in Figure 6.1 is routed through an orifice to fill up the accumulator back side (spring side) and to the accumulator control valve counter side (left side as shown in Figure 6.5), when the transmission is in Park. The accumulator spring is stretched when the related clutch is not applied. Generally, the spring side of all accumulators is filled up by accumulator back pressure ATF in Park and stays this way in the whole range to be ready for clutch apply and release. During clutch apply, ATF from the line pressure control circuit shown in Figure 6.1 is routed to the accumulator valve and is directed by shift solenoids and shift valves to enter the clutch apply circuit. Meanwhile, torque signal pressure from the pressure control solenoid acts on the accumulator control valve against the accumulator back pressure, causing the accumulator piston to displace back and forth so as to control the clutch apply pressure. For the clutch pressure control circuit shown in Figure 6.5, the clutch apply pressure p_c is related to the accumulator back pressure p_ab as:

(6.2)

where F_sa is the spring force that balances the accumulator back pressure before clutch apply is initiated, and x_a and k_a are the accumulator piston displacement and spring stiffness respectively. The accumulator piston back area A_ab is designed to be smaller than the area A_ac that receives the clutch apply pressure. During shift operations, the torque signal pressure from the pressure control solenoid acts against the accumulator back pressure to vary the opening in the accumulator control valve. This leads to the control of accumulator back pressure and thus the clutch apply pressure momentarily according to Eq. (6.2) during shifts to satisfactory accuracy. It is noted that p_c is lower than the line pressure during the clutch apply process. The circuit design shown in Figure 6.5 had been applied almost exclusively for clutch apply pressure control in the previous generation automatic transmissions. In general, a particular accumulator is attached to each clutch, and line pressure is routed by shift valves to the accumulator and apply circuit for the related clutch, as shown in Figure 6.5.

Figure 6.5 Clutch pressure control circuit with accumulator.

Clutch pressure control circuit with independent PCS: In typical current generation six‐speed automatic transmissions, no accumulator is used in the hydraulic circuit in the control system. This reduces the number of hardware components, thus lowering the transmission overall size and weight, as well as simplifying the control system hydraulic circuit. Unlike previous generation ATs, which use, as illustrated in Figure 6.1 and Figure 6.5, only one PCS for the control of both line pressure and clutch apply pressure, current generation ATs generally use a PCS to control the apply and release of each clutch in the clutch pressure control circuit shown in Figure 6.6, and a separate PCS to control the system line pressure in a circuit similar to that shown in Figure 6.1.

Figure 6.6 Clutch pressure control circuit with independent PCS.

The advantage of the clutch pressure control circuit shown in Figure 6.6 is obvious in comparison with the old version shown in Figure 6.5. Firstly, it eliminates the accumulator for each of the clutches in the transmission. For a transmission that has five clutches, eliminating five accumulators would mean significant reduction in overall transmission dimension and weight. More importantly, the new circuit allows independent clutch pressure control during shift operations. This is crucial for smooth shifts in current generation ATs that feature multiple speeds and clutch to clutch shifts. During shift operations, the ATF pressures in the oncoming clutch and in the off‐going clutch are controlled by two separate solenoids. This leads to more accurate clutch pressure control and improved shift smoothness and response. In addition, the independent control of line pressure and clutch pressure control minimizes system disturbance and interference between clutches, which is also conducive to accurate clutch pressure control.

In general, each clutch pressure regulator valve corresponds to a particular clutch in automatic transmission with independent clutch pressure control. In addition to pressure regulation functionality, a pressure regulator valve also functions as a switch with On or Off status to route ATF flow. As shown in Figure 6.6, the clutch pressure regulator valve has multiple ports, including an in‐port for line pressure, out‐port for clutch apply pressure, exhaust port and pressure switch signal port. Torque signal pressure controlled by the respective clutch PCS is routed via an orifice to one side of the valve and acts against the valve spring force. When the PCS is energized to be at On status, the torque signal force overcomes the spring force and displaces the valve body toward the spring side, creating an opening between the valve shoulder and the line pressure in‐port, as shown in Figure 6.6. Controlled by the PCS duty cycle, the torque signal force counteracts the spring force to move the valve body back and forth, varying the opening, or “restriction”, for the line pressure to enter the clutch apply circuit.

When the PCS is at Off status, i.e. not energized, there is no torque signal force acting on the valve and the spring force moves the valve body toward the torque signal side and blocks the line pressure from entering the clutch apply circuit. Line pressure is then routed through the valve to become the pressure switch signal which indicates the status of the clutch pressure regulator valve. Switch signals from clutch regulator valves are also used for fault diagnosis and fail mode management function of the transmission control system. To release the clutch, the clutch PCS pressure will be reduced firstly so that the spring force will move the valve body to the right and block the entrance for the line pressure. As the valve body moves to the right, it also creates an opening to the exhaust circuit for the clutch apply circuit. The exhaust opening is controlled afterwards by the PCS to control the clutch release pressure.

As shown in Figure 6.6, the clutch apply pressure is usually fed back to the pressure regulator valve directly via an orifice. If necessary, the clutch pressure control circuit shown in Figure 6.6 can be augmented by a feedback loop on the clutch apply pressure via a clutch pressure boost valve as shown in Figure 6.7. In this setup, the clutch PCS pressure acts on an area differential in the boost valve against the spring to vary the exit opening of the clutch pressure feedback ATF, which is routed via an orifice to the spring side of the clutch pressure regulator valve in the circuit shown in Figure 6.6. As the clutch PCS pressure is controlled to reach a designated level, it will move the boost valve further to the right and let the feedback pressure exhaust.

Figure 6.7 Clutch apply pressure circuit with boost valve.

Clutch compensator feed circuit: When a clutch is designed with a compensator chamber on the return spring side, ATF under a low pressure from the compensator feed valve fills the compensator and also the clutch apply side cavity so that the clutch is ready for apply, with minimal effects of the initial apply characteristics. As shown in Figure 6.8, the compensator feed ATF pressure depends on the spring force acting against the orificed compensator feed pressure feedback and is designed at a low value that is not sufficient to overcome the clutch return spring so that no clutch drag is created in the clutch. If the compensator feed pressure p_cp exceeds the level allowed by the spring, the valve body will be moved to the exhaust position. In addition to minimizing unwanted clutch initial apply attributes, clutch compensator pressure also increases the responsiveness of clutch actuation during shift operations. Note that the clutch compensator feed circuit is independent of other hydraulic circuits in the control system and its status is kept the same in the whole drive range.

Figure 6.8 Clutch compensator feed circuit.

Torque converter circuit: The torque converter clutch (TCC) apply and release is controlled by the circuit shown in Figure 6.9 or its variations in various automatic transmissions [8,16]. The TCC control valve is basically a position valve controlled by the ATF pressure from the TCC pressure solenoid. When the TCC PCS is not energized, the spring force keeps the TCC control valve in the release position for the feed ATF from the line pressure regulator valve shown in Figure 6.1 to enter the torque converter and leave it to cool. The TCC apply port is blocked by the valve land at the release position. Meanwhile, the TCC regulator valve is actuated by its spring and is kept at the position that blocks the line pressure from flowing through. Therefore, the TCC regulator valve does not have any effect on the converter feed circuit when the TCC PCS is not energized. In some transmissions, such as the GM Hydra‐Matic 6 T40/45, the TCC regulator valve also has a shift solenoid pressure port. Shift solenoid pressure is routed to the TCC regulator valve and acts on the shuttle, only in first gear, to close the TCC PCS pressure port, providing a redundant condition that the torque converter will not be locked in first gear. In the whole drive range, the shift solenoid pressure is not routed to the TCC regulator valve, which therefore does not have any effect on the converter feed circuit if the TCC PCS is not energized.

Figure 6.9 Torque converter clutch pressure control circuit.

When the TCC PCS is energized, the TCC PCS pressure acts against the spring to move the TC control valve to the apply position, and meanwhile to move the TCC regulator valve to the pressure regulating position, as shown in Figure 6.9. In the pressure regulating position, line pressure ATF flows through the TCC regulator valve to enter the TCC apply pressure regulation circuit. During the torque converter locking up process, the TCC PCS provides the pilot pressure to the TCC regulator valve to act against the spring and the TC apply pressure feedback, as illustrated by Figure 6.9. The TCU varies the duty cycles of the TCC PCS to control the TCC PCS pressure, which in turn controls the valve opening for the line pressure ATF to enter the TCC apply regulation circuit. The TCC apply ATF then flows through the TCC control valve to the apply side or feed side of the torque converter pressure plate to lock up the clutch.

To release the torque converter clutch, the TCU firstly decreases the TCC PCS pressure to a level enough for the spring force to move the TCC control valve and the TCC regulator valve both to the release position. The TCC apply ATF circuit is then linked with the cooler feed port of the TCC control valve, and the converter feed ATF port is linked to the TC release circuit. ATF is then released from between the pressure plate and the cover of the torque converter to the exhaust circuit, as discussed in Chapter 4.

Speed sensors: Hall‐effect type speed sensors are used in transmission control systems to measure the angular velocities of the transmission input and output. A speed sensor is positioned to face a toothed wheel (called a reluctor wheel) that is mounted on, and turns with, the shaft. As the reluctor wheel turns, the speed sensor produces an electric signal at a frequency correlated to the number of teeth and the rotational speed of the shaft. Typically, there are three speed sensors in the control system of an automatic transmission: one on the transmission input, another on the transmission output, and the third on the engine output or the impeller of the torque converter. The TCU receives signals from these three speed sensors via the respective circuit and processes these signals for various control operations. Note that speed sensors are the most important sensors in the transmission control systems; they provide real time data for the TCU to make decisions for shift schedule control, shift point control, and torque converter lock and release control, as discussed in Section 6.4.

ATF temperature sensor: The ATF temperature sensor is basically a thermistor that changes resistance in reverse proportion to temperature changes. A reference voltage is supplied to the circuit of the temperature sensor. The TCU measures the voltage change in the circuit as a measure of the ATF temperature and it can use this information to make adjustments to the shift schedule control, shift point control, and the torque converter lock and release control.

Shift range sensor or switches: This sensor provides information on the driver’s selection of drive range. For automatic transmissions with manual shift option, this sensor also sends a signal to the TCU on driver’s intention to initiate a shift. For transmissions that use a shift switch instead of a shift stick, the shift switch provides the TCU with similar signals to those of the shift range sensor.

In addition to the sensors mentioned, the transmission control system also shares signals via control area network (CAN) from numerous sensors that are mounted separately for other vehicle systems, mainly the engine control system. The following is a list of these sensors and how signals from these sensors are used in the transmission control system.

Engine throttle opening sensor: The engine operation status is determined by the engine throttle opening and the engine RPM. The engine throttle position is one of the two main variables – vehicle speed and engine throttle position – which determine the shift schedule. Since the transmission control strategy is torque based, the signal from the engine throttle sensor is used by the TCU for almost every aspect of transmission control, including shift point control, shift process control, and torque converter lock and release control, as will be discussed in Section 6.3.

Other engine related sensors include engine coolant temperature sensor, crankshaft position sensor, and manifold pressure sensor. These sensors complement the engine speed and throttle sensors for accurate determination of engine operating status and provide engine related data to the TCU for the control of transmission shift schedule and shift processes.

The operation of the air conditioning system affects the engine load and net transmission input torque, especially for passenger vehicles equipped with small engines. An air conditioner switch provides the On or Off status of the air conditioning system to the TCU, which calculates the net transmission input torque and adjusts the line pressure and clutch pressure accordingly. Note too that the TCU interacts via CAN with other vehicle operations or systems, such as the ABS system, the vehicle stability program, and cruise control operation, for decision making in the transmission shift schedule.

Transmission control unit: An example TCU layout is shown in Figure 6.10. As the core of the transmission control system, the TCU receives and processes signals from various sensors described above, performs calculations with the software that implements control strategies for shift schedule and shift processes, and sends commands to shift solenoids and pressure control solenoids for the accurate control of line pressure, shift point, clutch pressures during shifts, and pressure for the torque converter lock and release clutch. The TCU of current generation ATs may use 32‐bit or 64‐bit microprocessors (CPU). For example, the Delphi TCM8 transmission controller uses a 32‐bit, 80 MHz microprocessor with 1.5 MB flash memory and 56 kB RAM. Transmission controllers, such as the Delphi TCM8, must possess the high‐speed processing capability required for real time transmission control and sufficient memory for the storage of the control software and database. A TCU with a 32‐bit CPU is usually sufficient for handling all function requirements for ATs with up to six speeds. Typically, transmission controllers are equipped with built‐in input–output devices or drives for signal reception and control command delivery. The Delphi TCM8 features a configurable pulse width modulation (PWM) and pressure control solenoid drive with current feedback.

Figure 6.10 Configuration of transmission control unit (TCU).

Stored in the TCU as EEPROM, the transmission control software consists of two parts: programs and database. The programs include the CPU operating system, I/O interface, and driver functions, software code for transmission control functions, as well as diagnostics and failure mode functions. The database contains data in several groups: (a) attributes of powertrain subsystems, such as engine maps and torque converter characteristics; (b) transmission calibration data, such as line pressures and clutch pressures under various operation conditions, clutch torque profiles, and clutch friction coefficient look‐up table; and (c) shift schedules for different transmission operation modes. Note that transmission TCUs are capable of software and database upgrading.

6.3 System Circuit Configurations for Transmission Control

The system hydraulic circuitry for the control of automatic transmissions is the integration of the components and sub‐circuits discussed in the previous section. In general, the sub‐circuits for line pressure control, the torque converter lock, and the release control and clutch pressure control are similar in the control systems of various ATs. As discussed in Section 6.2, the status of certain sub‐circuits, such as the accumulator back pressure circuit and the clutch compensator feed circuit (if applied), remains unchanged in all drive ranges. The main difference lies in the ATF routing circuits that are controlled by the shift solenoids via the shift valves. The following guidelines should be useful in the design and analysis of the hydraulic circuitry for automatic transmission control systems.

• The transmission gear position and shift logic are uniquely defined by the combinations of On and Off status of shift solenoids. In ATs with four or five speeds, two or three shift solenoids are sufficient to define all the transmission operation states, including fixed gear operations and shift operations. For automatic transmissions with all clutches independently controlled by pressure control solenoids (PCS), the number of PCSs is the same as the number of clutches, and the transmission operation state is naturally defined by the On or Off status of the related PCSs. A clutch to clutch shift would then correspond to the changes of status of the two PCSs that control the two concerned clutches respectively. If a one‐way is used in a transmission, then a shift solenoid can be used, in addition to the PCSs that control all the other clutches, to define the transmission status and route the clutch apply ATF.
• The system hydraulic circuit should be designed to fulfill control functionality with minimized complexity. This is not only cost advantageous but also reduces unwanted attributes in the hydraulic circuit that worsen hydraulic pressure control response and accuracy.
• For automatic transmissions with clutches independently controlled by PCSs, the clutch pressure control circuits are laid out similarly to that shown in Figure 6.6, and if necessary a clutch apply pressure feedback circuit shown in Figure 6.7 can be applied for a particular clutch.
• Clutch compensator circuits are designed for clutches involved in shifts that are frequently made and most sensitive to the driver in response and harshness. As mentioned previously, compensator feed circuits are separate from the rest of the hydraulic system.

This section will briefly present the system designs for the hydraulic circuitry for the control of the previous generation of ATs, with the GM Hydra‐Matic 4 T80‐E as the example for illustrative purposes. The section will then focus on the control system configurations of the current generation ATs as exemplified by the GM Hydra‐Matic 6 T40/45. The transmission control strategies to be covered in Section 6.4 will be discussed in reference to the current generation ATs. Finally, the section will provide an introduction to the control systems of automatic transmissions currently under development.

6.3.1 System Hydraulic Circuitry for the Previous Generation of ATs

As discussed in Section 5.2, the GM four‐speed AT, Hydra‐Matic 4 T80‐E, uses five multiple disk clutches, two band clutches, and three one‐way clutches to realize clutch to one‐way clutch shift for all sequential shifts. For readers’ convenience, the stick diagram with the clutch table and the status of the shift solenoids is shown in Figure 6.11. Clutch C₅ is only applied in third gear for engine braking during coasting. Two solenoids, A and B, are used in the transmission control system to define the gear position and to initiate shifts. The park position (P), reverse gear (R), neutral position (N), and first gear share the same solenoid status. The manual shift valve routes the ATF flow in these positions respectively. In the drive range, a gear position is defined respectively by a combination of “on” and “off” of the two solenoids. Because all sequential shifts are clutch to one‐way clutches, each sequential shift is triggered by the change of status of one solenoid. A skip downshift would be triggered by the change of statuses of both solenoids, as can be observed in the clutch table.

Figure 6.11 Stick diagram, clutch, and shift solenoids table for GM Hydra‐Matic 4 T80‐E.

Except for the coasting clutch C₅ and the band clutch B₂ that is applied only in P, R, N, and first gear, the apply circuit for the other five hydraulically actuated clutches features an accumulator respectively. Each of the five accumulators is used in the clutch apply pressure control circuit for the shift process involving the related clutch, in a layout shown in Figure 6.5. Two accumulator control valves are used for the control of clutch apply and release pressure for all sequential and skip shifts. For a sequential shift, one shift solenoid will change its status to control the position of a shift valve in Figure 6.5 and route the line pressure to the accumulator apply side to close the clutch apply circuit. In a skip shift, the statuses of both solenoids will change to reposition the shift valves and route the line pressure to the two accumulators for the two clutches involved in the shift.

The system configuration of the hydraulic circuitry for the GM four‐speed AT, Hydra‐Matic 4 T80‐E is illustrated in Figure 6.12. The sub‐circuits for line pressure control, TC clutch apply and release control, and clutch pressure control have been discussed previously and are drawn as blocks in the system configuration. In addition to the manual valve, the two shift solenoids control the positions of the three shift valves, routing the line pressure to the respective clutch apply circuit for fixed gear and shifting operations.

Figure 6.12 Configuration of hydraulic circuitry for the previous generation of ATs.

As can be observed from Figure 6.12, there is only one PCS, namely the line PCS, which controls both the line pressure and the clutch apply pressure during shifts. When a sequential shift is to be initiated, the TCU signals one of the two solenoids, either A or B, and flips its status. Then the shift solenoid pressure acts against the spring of a related shift valve and repositions it. This will route the line pressure through the shift valve and connect it to the intended clutch pressure control circuit. The TCU then sends electric current to the line PCS by pulse width modulation (PWM) and controls the torque signal pressure. The torque signal pressure acts as a pilot pressure to regulate the clutch apply pressure in the corresponding clutch pressure control circuit. For a skip downshift, the TCU flips the statuses of both solenoids A and B, which reposition the shift valves. The ATF in the off‐going clutch circuit is then connected to the exhaust circuit, while the oncoming clutch apply circuit is connected to the line pressure, both by the repositioned shift valves. The torque signal pressure from the line PCS then rapidly ramps up the oncoming clutch apply pressure to complete the downshift.

The shortcomings of the hydraulic circuitry shown in Figure 6.12 are obvious. Firstly, the clutch apply and release pressures are not controlled by PCS independently, making it difficult to accurately control apply pressure and shift timing. Secondly, since the line PCS controls both the line pressure and the clutch apply pressure during a shift, the pressure of other clutches that need to be applied during the shift may be affected. This may even result in slippage in these clutches if the line pressure fluctuates too much during the shift in the control of the oncoming clutch apply pressure. To overcome these shortcoming, engineers had adopted two approaches in the design and control of the previous generation ATs. On the design side, one‐way clutches are widely used in the transmission to realize clutch to one‐way clutch sequential shifts. This is typified by the GM Hydra‐Matic 4 T80‐E shown in Figure 6.11, which uses three one‐way clutches and has all sequential shifts clutch to one‐way clutch. On the control side, hydraulic accumulators are extensively used in the control systems for the clutch apply circuits. The adoption of these two approaches greatly enhances the shift smoothness of the previous generation ATs with four or five speeds, but of course at the expense of hardware cost and overall transmission weight and dimensions.

6.3.2 System Hydraulic Circuitry for ATs with Independent Clutch Pressure Control

ATs of the current generation in mass production typically have six speed and feature clutch to clutch sequential shifts, except for the 1–2 shift in some designs. There are two apparent differences in the clutch control circuits between the previous and current generation ATs, as can be observed in Figures 6.5 and 6.6. In the ATs of the current generation, accumulators are eliminated in the clutch pressure control circuit, and a PCS designated to a particular clutch controls the pressure regulator valve position to control the clutch apply pressure independent of other sub‐circuits. This setup overcomes the shortcomings of the hydraulic circuitry of the previous generation ATs, with substantial reduction in overall transmission dimension and weight. The system hydraulic circuitry of ATs with independent clutch pressure control is illustrated in general form in Figure 6.13.

Figure 6.13 Hydraulic circuitry for ATs with independent clutch pressure control.

In the system hydraulic circuitry shown in Figure 6.13, the line pressure control circuit and the torque converter clutch control circuit are illustrated from Figures 6.1 and 6.9, as discussed in Section 6.2. The compensator feed circuit shown in Figure 6.8 is designed for clutches with compensator pistons, and the status of this circuit remains unchanged in the whole transmission operation range. The clutch pressure control circuit can be designed in the setup shown in Figure 6.6 or Figure 6.7, depending on the pressure control requirements for the specific clutch. The manual shift valve has five positions, Park, Reverse, Neutral, Drive, and Manual, which correspond to different ports, and serves as a switch to route the ATF under line pressure, respectively. In some designs, a position sensor provides the signal indicating the shift level position for the TCU to control the statuses of the clutch solenoids accordingly, eliminating the need for the manual shift valve to route line pressure ATF.

The hydraulic circuit design illustrated in Figure 6.13 and its variations can be applied in all ATs with clutch to clutch shifts and independent clutch pressure control, such as the Ford six‐speed RWD AT discussed in Section 5.3 and shown in Figure 5.11. In the Ford six‐speed RWD AT, there are five multiple disk clutches and no one‐way clutch, with all shifts clutch to clutch. Therefore, there will be five clutch pressure control solenoids (PCS) and five clutch pressure control circuits. Each clutch PCS controls independently one of the five pressure control circuits designated for a particular clutch. In addition to controlling clutch pressure, each clutch PCS also possesses the On or Off status. Therefore, all transmission operation statues, fixed gear operations or shift operations, are uniquely defined by the combinations of the On or Off status of the five clutch pressure control solenoids.

As discussed in Chapter 5, many current generation automatic transmissions with six or eight speeds – such as the Ford six‐speed FWD AT and the Lexus eight‐speed RWD AT shown in Figures 5.10 and 5.12 respectively – still use a one‐way clutch in first gear as the reaction clutch and the 1–2 shift is therefore clutch to one‐way clutch. In these transmissions, it is not necessary to use a PCS per each of the clutches. A clutch pressure control circuit can be shared by the clutch applied in first gear for engine braking and another clutch that is applied in higher gears, similar to the design shown in Figure 6.12, by using a shift solenoid for ATF routing and a PCS for clutch apply pressure control. The pressure control circuits for all other clutches are laid out as shown in Figure 6.13.

In the GM six‐speed FWD AT, Hydra‐Matic 6 T40/45, one‐way clutch F serves as the reaction clutch in the first gear of the drive range and clutch D serves as the reaction clutch in the first gear with engine braking capability, as shown in Figure 6.14. When the vehicle is launched in first gear, clutch D is applied as the reaction clutch, and is then released after launch before the 1–2 shift is initiated. This design makes the 1–2 shift in the drive range clutch to one‐way clutch, with engine braking capability at low vehicle speed in first gear. As shown in the clutch and solenoid status table, the control system uses four PCSs and one shift solenoid (SS). The shift solenoid status is On in P, R, N, and D1 EB (engine braking), and is Off in the whole drive range, flipping only once for the status change from D1 EB to D1. The P‐R, R‐N, and N‐D changes are triggered by flipping the status of one PCS respectively, and 1–2 shift is triggered by the flipping of the status of the clutch C PCS. All other sequential shifts are triggered by flipping the status of two PCSs. It is also interesting to note that skip downshifts to first gear from 3rd and 4th gears are triggered by flipping one PCS.

Figure 6.14 Clutch and solenoid status table for GM Hydra‐Matic FWD six‐speed AT.

The hydraulic circuit for the control system of the GM Hydra‐Matic 6 T40/45, shown in Figure 6.15, is based on the configuration shown in Figure 6.13 with minor modifications. Clutch A, applied in the fourth, fifth, and sixth gears, and Clutch D, applied in the reverse and first gear (with engine braking), share the same PCS (PCS A), with a shift solenoid to control the shift valve position for line pressure ATF routing. The shift valve has two out‐ports for clutch apply pressure, one for clutch A and another for clutch D. It also provides passages for the line pressure ATF in the drive range for other purposes, such as the line pressure input for the torque converter clutch pressure control circuit shown in Figure 6.9. The shift solenoid is energized as On in P, R, N, and D1 EB and the shift valve is positioned by the SS pressure acting against the valve spring. At this position, clutch D remains applied by the pressure regulated by the clutch A and D pressure control circuit. Meanwhile, the manual valve provides additional routing for the line pressure ATF. When the manual valve is at P, PCS A is energized to be On and all other PCSs are Off. Line pressure ATF is routed to the clutch A and D pressure control circuit to apply clutch D and to fill the compensator feed circuit. When the manual valve is switched to R from P, it routes the line pressure through the shift valve passage to enter the clutch B pressure control circuit, and PCS B is energized to be On and controls clutch B apply. When the manual valve is switched from R to N after a short time of operation in R, the status of the hydraulic circuit returns to that of P, with PCS B flipping back to Off. In a similar fashion, when the manual valve is switched to D, it routes the line pressure ATF to the in‐ports of the pressure control circuits of clutches B, C, and E, making them ready for apply as required by the shift schedule. At this time, only PCS E flips from Off to On to control clutch E apply via its pressure control circuit. The transmission will then operate in D1 EB (first gear with engine braking) for a short time until both the SS and PCS A flip from On to Off. Then, the SS pressure is exhausted and the shift valve is repositioned by the valve spring. The shift valve remains in this position in all gears in the drive range and serves only as a line pressure ATF router. Note that the transfer from D1 EB to D1 is completed by only releasing clutch D; the one‐way clutch F will automatically take over the role as the reaction clutch in first gear from clutch D. Once the transmission operates in the drive range from D1 to D6, the gear positions and shifts are then defined by the statuses or status flips of the four PCSs, as illustrated in the clutch and solenoid status table in Figure 6.14.

Figure 6.15 Hydraulic circuit for GM Hydra‐Matic FWD six‐speed AT.

6.3.3 System Hydraulic Circuitry for ATs with Direct Clutch Pressure Control

The hydraulic circuits of automatic transmissions with direct clutch apply pressure control are further simplified from the hydraulic circuits of the current generation ATs as shown in Figure 6.15. As the industry trend, next generation ATs with eight or more speeds will feature direct clutch pressure control technology. In this new transmission control technology, clutch apply and release pressures are directly controlled by large volume pressure control solenoids (PCS), also called variable force solenoids (VFS) or variable bleed solenoid (VBS), without the pressure control circuits shown in Figures 6.6 and 6.7. Meanwhile, the manual valve, the shift valve, and the shift solenoid are also eliminated from the hydraulic system. The shift lever or shift nub only provides signals to the TCU on the positions indicating P, R, N, and the drive range D. Each clutch in the transmission corresponds to a specific PCS or VFS. Therefore, the apply pressure of the oncoming clutch and the release pressure of the off‐going clutch are independently controlled by the respective VFS in all shifts. This results in enhanced accuracy in clutch pressure control and shift timing control, leading to optimized shift smoothness and shift response. In addition, the elimination of complex hydraulic circuits is conducive to overall transmission weight and cost reduction. Note that the hydraulic circuits for line pressure control and torque converter clutch control in ATs with direct clutch pressure control are the same as, or similar to, the circuits shown in Figures 6.1 and 6.9. If allowed by package space, clutches with dominating effects on shift smoothness and response are designed with compensators. The compensator feed circuit shown in Figure 6.8 is designed to fill the clutch piston compensator to minimize clutch initial apply attributes and to enhance shift response. The compensator feed circuit remains the same status in the whole drive range. The hydraulic system configuration implementing direct clutch pressure control is illustrated in Figure 6.16. This configuration or its variations can be used for the hydraulic circuit design for the control systems of next generation automatic transmissions. As an example, it can be well fitted to the control system of eight‐speed ATs [14,15], such as the ZF RWD eight‐speed AT that is shown in Figure 5.14 and analysed in Sections 5.3 and 5.4. For the ZF RWD eight‐speed AT, the five clutches, A, B, C, D, and E are respectively and directly controlled by PCS A, PCS B, PCS C, PCS D, and PCS E. In any direct shifts, sequential or skip shifts, one PCS controls the apply pressure of the oncoming clutch, and another PCS controls the release pressure of the off‐going clutch.

Figure 6.16 Hydraulic circuit for ATs with direct clutch pressure control.

6.4 Transmission Control Strategy

As described in Section 6.1, the control system of an automatic transmission must possess five basic functions: (1) shift schedule control, (2) torque converter locking control, (3) engine torque control during shifts, (4) shift process control, and (5) system diagnosis and failure mode management. These functions are actuated by the transmission controller, based on the signals from relevant sensors and are executed according to pre‐designed and calibrated strategies or algorithms. The previous sections have provided a general description of the hardware components and hydraulic circuits for the transmission control systems. This section highlights the strategy and techniques for the implementation of the control system functions.

6.4.1 Transmission shift schedule

The transmission controller makes shift decisions according to the shift schedule based on two primary inputs – the transmission output speed and the engine throttle position – and other supplementary inputs such as ATF temperature and brake pedal depression. It is the shift schedule that defines the operating status of the vehicle power train system under any road condition. Therefore, the shift schedule is crucial to the vehicle characteristics such as fuel economy, dynamics performance, drivability, and pollutant emission levels. As discussed in Chapter 5, the fuel economy and acceleration performance of an automatic transmission vehicle with a specific shift schedule can be simulated by computer over various driving ranges, with the engine output data, the transmission data, and other vehicle data provided. These simulations can provide a model‐based validation for the initial shift schedule in terms of fuel economy and performance. This initial shift schedule can then be used in test vehicles that undergo the intensive powertrain calibration process. The finalized shift schedule will be optimized in the calibration process as a well‐balanced trade‐off between fuel economy and dynamic performance. In some production transmissions [6,8,10], the shift schedules are designed with several modes, such as Economy, Normal, and Sport, which are chosen by the driver to meet different priorities and preferences.

An example shift schedule for a five‐speed AT is shown in Figure 6.17, where the horizontal axis represents the vehicle speed and the vertical axis the engine throttle position. The solid lines are the thresholds for upshifts and dotted lines are for downshifts. As can be observed from Figure 6.17, some of the important attributes in an AT shift schedule are as follows:

• The current vehicle operation status is defined in Figure 6.17 by a point whose horizontal coordinate is the vehicle speed and vertical coordinate is the engine throttle position. Upshifts between two adjacent gears are made at higher vehicle speed than downshifts. This means that there is a buffer zone between upshift and downshift. This buffer zone is designed for any two adjacent gears to keep the transmission from shifting back and forth along the shift threshold line. Without the buffer zone, a slight decrease of the vehicle speed right after an upshift is made may cause the status point to cross back over the downshift threshold line and trigger the downshift, resulting in frequent up/downshifts and poor drivability.
• The same upshift is made at higher vehicle speed at wider engine throttle opening, as indicated by the inclined upshift threshold lines in Figure 6.17. This design reflects the desire of the driver for acceleration performance. Deeper gas pedal depression will keep the vehicle running in the current gear for a longer time to achieve better acceleration. On the other hand, lighter throttle openings correspond to earlier upshifts and reach the high gear operation more quickly, resulting in better fuel economy as compared with aggressive gas pedal depressions.
• Each shift threshold line consists of several straight line segments and can be defined by the coordinates, i.e. the vehicle speed and the throttle position, at the intersection points of these segments. Therefore, the whole shift schedule is defined by these coordinates of the intersection points along the respective shift threshold lines. Adjustments or modifications on the shift schedule can be done by simply changing the coordinates of these intersection points.
• A shift is defined by a point on a shift threshold line. A number of points on each shift line are chosen to represent the corresponding shift conditions. These points include the points of intersection of the straight line segments mentioned above and may also include intermediate points between the intersections, usually chosen at nearly even steps of engine throttle openings. For example, ten points might be chosen on the 1–2 shift line in Figure 6.17, six of them are at the intersections of the straight line segments and four are between the intersections for a nearly even throttle position interpolation. Theoretically, each one of such points constitutes a different shift and needs to be calibrated separately against the corresponding vehicle operation condition. Therefore, the number of shifts can become very large for an AT with multiple gear ratios.
• A transmission shift schedule works interactively with the torque converter clutch operation schedule. The apply and release of torque converter clutch closely follows the transmission shift schedule. As discussed later separately, the converter clutch operation schedule has two important effects on the transmission performance: shift smoothness and fuel economy.
• The shift schedule defined by threshold lines in Figure 6.17 is used for normal operation conditions. The shift thresholds are adjusted on a real time basis to better fit vehicle operation conditions, such as operation in low temperatures or when towing of a trailer. During a shift process, the controller will decide whether to proceed or discontinue while the driver depresses the brake.

Figure 6.17 Example shift schedule for a five‐speed automatic transmission.

Some vehicles are equipped with ATs that feature driver selectable shift modes, usually termed Economy, Normal and Power. Each of these modes is designed to fit an intended priority of the driver. For example, Figure 6.18 shows the shift schedule under the Normal mode and the Power mode respectively for a Toyota passenger car four‐speed AT [8]. The Normal mode implements a balanced transmission shift schedule with the torque converter locked near the coupling point with the torque ratio approximately equal to 1.1. The Power mode is achieved by a power oriented shift schedule with an open torque converter for the full advantage of torque multiplication. It can be observed from Figure 6.18 that the shift lines of the Power mode are shifted to the right as compared with those of the Normal mode, which keep the vehicle accelerating in lower gears for a longer time to reach the target vehicle speed. The Economy mode combines a fuel efficient shift schedule with a schedule for torque converter lock‐up that is optimized for fuel economy and is shown in Figure 6.19. This schedule locks up the torque converter when the torque ratio becomes less than 1.3. According to the dyno test and actual vehicle test data obtained by Toyota engineers [8,9], the Economy mode achieves about 3% fuel economy improvement over the Normal mode, and the Power mode achieves a 3.6% improvement in 0–100 km/h acceleration performance over the Normal mode under WOT test conditions. These data support the necessity of having different shift modes to be selected by the driver for preferred priorities since the improvements in fuel economy and performance are not insignificant.

Figure 6.18 Shift schedules for Normal and Power modes.

Figure 6.19 Torque converter clutch lock‐up schedule optimized for fuel economy.

6.4.2 Torque Converter Lock Control

The objective of torque control lock control is mainly to achieve an optimized trade‐off between efficiency and shift smoothness. As a torque multiplier and a fluid couple, an open torque converter functions as the vehicle launcher and as a damper for powertrain harshness during transmission shifts. However, there is a power loss in the operation of an open torque converter. This loss is more significant in the low speed range which corresponds to low torque converter speed ratios and may translate into about 3% fuel mileage loss. According to the model simulations and experiments conducted by Toyota researchers [8,9], vehicle fuel economy is improved in the whole speed range by locking up the torque converter, especially at low vehicle speed. In general, an earlier converter lock‐up scheme, as shown in Figure 6.19, results in better fuel economy. In most ATs, torque converters start to be locked in second gear operation above some threshold vehicle speed and are locked afterwards in fixed gear operations. However, a locked torque converter loses its function as a fluid couple to dampen the transients and harshness during shift operations. In order to optimize shift quality, the torque converter is therefore released during shift operations for it to function as the harshness damper. Once a shift is completed, the torque converter will be locked again to enhance the powertrain efficiency. Note that torque converter locking or unlocking is a dynamic process by itself that may cause unwanted powertrain transient behavior and needs to be controlled for optimized driver and passenger feel. In summary, torque converter lock control concerns two technical issues: Lock‐release schedule and lock‐release operation control.

6.4.3 Lock‐Release Schedule

The torque converter stays open in first gear operation and thus the lock‐release schedule applies to transmission operations above second gear. Similar to the transmission shift schedule, the torque converter lock‐release schedule makes the lock‐release decisions based the vehicle speed and the engine throttle opening, as shown in Figure 6.19. A buffer zone is designed between the locking and unlocking threshold lines for each gear to keep the converter from locking and unlocking frequently. As mentioned previously, earlier locking‐up results in better fuel economy and later locking‐up corresponds to better acceleration performance. If fuel economy is the target, then the converter is locked at lower speed in each gear above second gear, as designed in the lock‐release schedule. However, locking up the torque converter too early, i.e. at aggressively low vehicle speed, results in the loss of torque multiplication and may even excite powertrain vibration. This happens if the torque converter is locked too early in all gears, according to the model simulation and experiment as conducted by Toyota researchers [8,9]. Therefore, the threshold lines in the lock‐release schedule must be designed for each gear such that the converter torque multiplication capability is used as much as possible, and the lock‐up point is above the lowest speed that will avoid exciting powertrain vibration. This lowest speed can be determined initially by model simulation and validated by experimental means or test vehicle calibration.

The lock‐release schedule shown in Figure 6.19 only determines the torque converter status during fixed gear operations. As mentioned previously, the torque converter is unlocked synchronously during shift operations, especially in low gears, since its functionality as a fluid couple is needed to dampen shift harshness. Since torque converter locking or unlocking is a process by itself, the timing of converter locking or unlocking control is crucial to transmission shift smoothness. The On or Off status of the torque converter during shifts follows the transmission shift schedule shown in Figure 6.17 or Figure 6.18. In addition, the timing of converter unlocking control depends on the operation conditions and type of the shift when it is initiated by the transmission control unit (TCU). In general, the control of converter unlocking during shifts is timed based on the following conditions:

• Threshold line that triggers the shift: this will identify both the type of shift – upshift, or downshift – and the gear positions before and after the intended shift.
• Power on or power off: this signal is from the accelerator pedal sensor.
• Current converter clutch status: this is determined by the converter clutch control solenoid status.
• Downshift triggered by deep depression of accelerator pedal: this is also power‐on downshift.

The synchronization of torque converter clutch locking or release control during shifts is illustrated in Figure 6.20. A shift is signaled by the flip of a solenoid status, from On to Off, and the torque converter clutch status during shift is commanded also by the On or Off status of the related solenoid. The case for upshifts is shown on the left side in Figure 6.20. The shift is triggered by the flip from On to Off as indicated by the shift signal. The torque converter clutch signal On or Off corresponds to the locked or unlocked position. When a power‐on upshift is to be made when the torque converter is locked in the current gear, releasing the converter clutch during the torque phase will give rise to engine flare because the mechanical coupling between the engine and the transmission input becomes momentarily a fluid coupling. The engine speed must be brought down in an upshift and engine flare should certainly be avoided since it elongates shift time and causes other undesirable shift attributes. On the other hand, if the torque converter clutch stays locked way into the inertia phase, shift harshness or even shift shocks may result due to the lack of converter damping effect. Therefore, it is critical to control the converter clutch release timing for the full advantage of fluid couple damping effect during the inertia phase. As shown in Figure 6.20, there is a time delay ΔT between the initiation of the upshift and the flip of the converter clutch signal. This time delay is designed to time the release of the torque converter clutch right at the transition from the torque phase to the inertia phase. Note that the value of time delay ΔT can be calibrated for all upshifts that involve torque converter clutch unlocking. After the upshift is completed, the transmission control unit (TCU) will command the torque converter clutch status from Off to On so as to lock up the torque converter in the high gear as shown in Figure 6.20, which is deemed appropriate by the schedule in Figure 6.19.

Figure 6.20 Torque converter clutch lock‐up schedule optimized for fuel economy.

The order of torque phase and inertia phase in power‐on downshifts are just reversed as compared with power‐on upshifts. In a power‐on downshift, the off‐going clutch must be controlled to slip as soon as the downshift is initiated, so the engine speed will be brought up to the target value of the low gear to be downshifted. The torque converter clutch is usually locked at the time when a power‐on downshift is commanded by the TCU, as indicated on the right side of Figure 6.20. As controlled by the TCU, the TC clutch signal flips from On to Off before the shift signal flips from On to Off by a time ΔT. The value of ΔT can be preset using the transmission shift schedule as shown Figure 6.17 and can be validated through test vehicle calibration. For power‐on downshift triggered by deep accelerator pedal depression, the TC clutch signal can be controlled to flip synchronously as the downshift is triggered. Since the converter clutch is released as soon as the power‐on downshift is initiated, the engine speed will be synchronized to the low gear quickly to complete the shift. At the end of the downshift process, the status of the torque converter clutch is recovered to the status prior to the shift as deemed appropriate by the schedule shown in Figure 6.19, as illustrated on the right side of Figure 6.20. Note that the control logic for power‐off upshifts is similar to that used for power‐on downshifts for the avoidance of powertrain harshness.

6.4.4 Lock‐Release Operation

The torque converter clutch pressure control circuit was shown in Figure 6.9. As mentioned previously, the TC clutch apply or release pressure is controlled by the TC clutch pressure control solenoid using pulse width modulation, i.e. via the duty cycle percentage. The control strategies may differ in specifics for different transmissions but they are all aimed at operation smoothness of the torque converter clutch during locking and release. As an example [17], the control of converter clutch apply and release processes is illustrated in Figure 6.21 for the GM Hydra‐Matic six‐speed AT shown in Figure 6.14. The apply pressure ramp‐up profile is from point A to point H, and the release pressure ramp‐down profile is from point I to point L respectively. It is noted that the electric current in the TC clutch pressure control solenoid follows the same profile due to its proportionality with the pressure.

Figure 6.21 Pressure ramping for torque converter clutch control.

When the vehicle is being driven in second or higher gear with an open torque converter and the transmission control unit (TCU) decides to lock up the torque converter according to the lock‐up schedule similar to that shown in Figure 6.19, the status of the torque converter clutch control solenoid immediately flips from Off to On, as designed in the system control circuit shown in Figure 6.15. As detailed in the Technician’s Guide for the GM Hydra‐Matic six‐speed AT [17], the torque converter clutch locking‐up process is implemented in three steps:

1. The TCU ramps the TC clutch PCS pressure from point A to point B and holds it until point C, through the circuit shown in Figure 6.9. The pressure at point B is sufficient to position the TCC regulator valve for the line pressure ATF to enter the regulated TC apply circuit and to position the TCC control valve to allow the converter release ATF to exhaust. To avoid possible harshness caused by overly high initial apply pressure, the TCU lowers the TC clutch PCS pressure from point C to point D. The TCC PCS pressure at D is controlled to be still high enough to maintain the positions of the TCC regulator valve and the TCC control valve, but at this point, the converter clutch apply pressure from the TCC regulator valve is not yet high enough to apply the converter clutch.
2. The TCU increases the TCC PCS pressure from point E to point F in a small jump and further ramps it to point G following a designed slope. In this process, the regulated TC apply pressure from the TCC regulator valve enters the TCC apply circuit and becomes high enough to start applying the converter clutch. The converter slip rate – i.e. the speed difference between the impeller and the turbine – will decrease gradually as the TCC PCS pressure increases. At point G, the converter slip rate approaches zero.
3. To eliminate the residual slip in the torque converter and to secure the lock‐up, the TCU further increases the TCC PCS pressure in a jump to point H. At this point, the torque converter becomes a mechanical couple that links the engine and the transmission.

When the vehicle is being driven with the torque converter locked‐up and the transmission control unit (TCU) decides to release the torque converter in a shift event or according to the lock‐up schedule, the status of the torque converter clutch control solenoid flips from On to Off. In a shift event, the time interval ΔT between the shift signal and the clutch signal depends on the shift type, as discussed previously and illustrated in Figure 6.20. The control of the release process is implemented following the release pressure ramp‐down profile from point I to point L, as shown in Figure 6.21. In a shift event, the torque converter clutch needs to be released quickly in order to take full advantage of the damping effect of the converter fluid couple functionality. This is indicated by the steepness of the release pressure ramp‐down profile from point I to point L. For the GM Hydra‐Matic six‐speed AT [17], the torque converter release process is controlled in two steps:

1. At the beginning of the release process, the TCU decreases the TCC PCS pressure by cutting down the current sent to the TC clutch control solenoid in the circuit shown in Figure 6.9 from point I to point J. The reduced TCC PCS pressure at point J allows the repositioning of the TCC control valve by the resultant pressure of ATFs from other circuits and diminishes the opening in the TCC control valve for the regulated TC apply ATF to enter the converter clutch apply circuit. As a result, the ATF pressure on the torque converter clutch apply side – i.e. the apply side of the converter pressure plate – is reduced and at some point the contact surface between the pressure plate and the converter cover will reach the threshold of slippage. This slippage threshold point can be calibrated per vehicle operation condition and the slip can be detected by the speed sensors as soon as it starts.
2. As the release proceeds, the TCU further decreases the TC clutch PCS pressure from point J to point K following the calibrated slope by decreasing the current sent to the TCC pressure control solenoid to zero. At point L, the residual electromagnetic effect of the TC clutch PCS is zero and the release ATF enters the space between the pressure plate and the converter cover from the release opening of the TC control valve as shown in Figure 6.9. The torque converter is now fully open and provides the torque multiplication and damping functionalities as designed.

As mentioned previously, the torque converter locking and release processes are controlled in different ways for different transmissions. The control strategy presented above can be considered as typical and can be modified for general applications. In addition, the profiles shown in Figure 6.21 can also be trimmed for different transmissions and are validated in the calibration process of test vehicles.

6.4.5 Engine Torque Control During Shifts

As discussed in Section 5.5, the oncoming clutch torque drags on the transmission and decreases the transmission output torque during the torque phase of upshifts. During the inertia phase, the transmission output torque largely depends on the oncoming clutch torque and may have a large overshoot if the oncoming clutch torque is too large. The output torque drop forms the so‐called torque hole that bottoms at the transition from torque phase to inertia phase, while the torque overshoot peaks toward the end of the inertia phase, as shown in Figure 5.22. Both the torque hole and the torque overshoot must be minimized for shift smoothness. Typical output torque variation patterns are shown in Figure 6.22 for clutch to one‐way clutch upshifts without engine torque retarding and with engine torque retarding. As shown in Figure 6.22, engine torque reduction by spark retarding has a significant effect on the minimization of the output torque overshoot. For the best effect, engine torque reduction should be started as soon as the shift transfers to the inertia phase. If the engine torque reduction starts when the torque phase is yet to be finished, it will deepen the torque hole, resulting in unpleasant shift feeling. Therefore, the TCU must interact with the engine controller via the control area network to time the spark retarding accurately on a real time basis for power‐on upshifts.

Figure 6.22 Engine torque reduction by spark retarding during clutch to one‐way clutch upshifts.

The effect of the engine torque reduction on shift quality has been analysed in detail in Section 5.5. In an upshift, the engine speed must be brought down quickly to be synchronized with the value corresponding to the high gear. If the upshift is clutch to one‐way clutch, such as the 1–2 shift in the Ford FWD six‐speed AT or the GM Hydra‐Matic six‐speed AT, the torque in the off‐going clutch is zero after the inertia phase starts. Therefore, the magnitude of the engine angular deceleration in the inertia phase is only proportional to the oncoming clutch torque for a given engine output torque or transmission input torque, as shown in Eq. 5.94. The oncoming clutch torque T_C must be ramped up to a certain value in order to decelerate the transmission input speed or engine speed. A steep ramp‐up profile for the oncoming clutch torque T_C results in rapid shift response and shortens shift time. But, on the other hand, larger oncoming torque magnitudes give rise to higher output torque overshoot and shift harshness, as analysed in Section 5.5. This contradiction is well addressed by reducing the engine torque, i.e. the transmission input torque, in the inertia phase as shown in Figure 6.22. As observed in Eq. 5.94, by reducing the engine torque via spark retarding, the transmission input torque is reduced proportionally, and therefore a smaller oncoming clutch torque is able to achieve the engine deceleration rate required to complete the shift in good time. When the engine speed is brought down near the target speed of the high gear toward the end of the inertia phase, spark retarding is cancelled by the engine controller and engine torque recovers to the normal level. The oncoming clutch torque is further ramped up in one step to secure the engagement of the oncoming clutch. If spark retarding is controlled in good time and the oncoming clutch torque ramp‐up profile during inertia phase is controlled properly, it is possible to achieve near perfect upshifts, as shown in Figure 6.22 with minimized output torque overshoot.

As shown in Figure 6.23, the effect of engine torque reduction in clutch to clutch upshifts is similar to that for clutch to one‐way clutch upshifts discussed earlier. In this case, the off‐going clutch is a regular clutch whose torque capacity depends on the hydraulic pressure in the clutch piston chamber. In the torque phase, the hydraulic pressure in the off‐going clutch piston chamber is controlled by the related hydraulic circuit to decrease rapidly for the off‐going clutch to reach the slip threshold, at which point the off‐going clutch torque capacity is equal to the off‐going clutch torque determined by the system dynamic status, as shown in Figure 6.23. As the inertia phase starts, both the oncoming clutch torque and the residual torque in the off‐going clutch act against the engine torque to decelerate the engine speed or transmission input speed, as observed in Eq. 5.94. The engine torque reduction starts after the torque phase immediately by timing the spark retarding accurately. Owing to the reduced engine torque, the oncoming clutch torque is controlled to follow a lowered profile for the minimization of the output torque overshoot, while still being high enough to act against the engine torque for the deceleration of the engine speed, as observed in Eq. (5.94). During the inertia phase, the TCU processes the data from the speed sensors on real time and notifies the engine controller to cancel spark retarding for engine torque recovery once it judges that the upshift is near completion. The oncoming clutch torque is then ramped up further in one step to securely engage the oncoming clutch. With the engine torque reduction controlled in good time and the oncoming clutch torque profile ramped up accurately, it is possible to achieve power‐on clutch to clutch upshifts with minimized output torque overshoot similar to that for clutch to one‐way clutch upshifts, as shown in Figure 6.22.

Figure 6.23 Engine torque reduction by spark retarding during clutch to clutch upshifts.

6.4.6 Shift Process Control

This is the key function of the transmission control system. Transmission shift response and smoothness primarily depend on the off‐going clutch torque and the oncoming clutch torque, which are controlled by the clutch pressure control circuits illustrated in Figures 6.6 and 6.7. During shift operations, the hydraulic pressures in the piston chambers of the off‐going clutch and the oncoming clutch are respectively controlled by these circuits using variable force solenoids (VFS). In addition, engine torque reduction and torque converter unlocking are introduced during shifts to enhance shift response and smoothness, as already discussed in detail. This subsection focuses on the techniques and strategies for the ramping‐up and ramping‐down of the ongoing and off‐going clutch pressures. ATs currently in production or under development typically feature the following technical highlights for shift process control:

• The apply or release pressure of a clutch is independently controlled by a VFS in a circuit specific to the related clutch, as shown in Figures 6.6, 6.7, 6.15, and 6.16. This results in better accuracy and response in clutch apply pressure control.
• Torque based control is used for both the torque phase and the inertia phase during shifts. When a shift is initiated, the engine torque is estimated using the engine map and modified on a real time basis using an array of sensor‐provided data, such as temperature, engine manifold pressure, and air conditioner status,The net transmission input torque is obtained by factoring in further the torque converter characteristics.
• Clutch torque profiles for various shifts optimized in transmission calibration are stored in the transmission control unit (TCU) as data. These torque profiles can be converted to clutch pressure profiles for real time shift control using a database on clutch friction coefficients.
• The attributes of clutch behavior during the initial stage of piston stroking are handled effectively by hardware design improvements and by pre‐calibrated control software techniques. Counter piston design and compensator chamber design can be applied to eliminate the time delay caused by clutch piston chamber fill‐up and minimize the effects of follow‐up piston stroking transients. On the software side, a database can be established on the initial piston characteristics through calibration and is used to modify or compensate clutch pressures results on a real time basis [13].
• Open loop control is usually used for the torque phase, and closed loop is used for the inertia phase for power‐on upshift. The feedback signals are mainly the speeds of the engine, transmission input, and output. Fine adjustments of clutch pressures are implemented based on the feedback, keeping shift performance consistent between vehicles and during extended service life.

Typical torque profiles during power‐on clutch to clutch upshifts are shown in Figure 6.23. The pressure profiles of the two involved clutches follow the same pattern as the torque profiles but in a scaled proportion, as shown in Figure 6.24. In Section 5.4, the clutch torque magnitudes were determined for each gear in terms of the transmission input torque. The maximum torque magnitude required for a clutch when the transmission operates in a fixed gear depends on the maximum transmission input torque, which in turn can be determined based on the engine torque map, torque converter characteristics, and transmission shift schedule, such as that shown in Figure 6.17. The maximum torque required of a clutch is used in clutch design in terms of sizing and the selection of the number of friction disks. For a clutch with given design parameters, the clutch torque only depends on the clutch apply pressure. These clutch design parameters are optimized such that the pressure required for full clutch engagement in each gear does not vary significantly between the various clutches in the transmission. Therefore, the system line pressure that is controlled according to vehicle operation conditions will fit all clutches for the control of clutch apply and release pressure profiles. When applied, the hydraulic pressure in the clutch piston chamber is equal to, or close to, the system line pressure. For a factor of safety, the system line pressure is thus controlled at a value somewhat higher than the pressure required to fully engage the clutches during fixed ratio operations. The pressure profiles of the off‐going clutch and the oncoming clutch in a power‐on upshift are shown in Figure 6.24, where the pressure unit is in BA. Usually, the clutch apply pressure in automatic transmission control systems does not exceed 10 BA or 1.0 MPa. The pressure profiles in Figure 6.24 generally refer to power‐on upshifts under heavy engine loads with large to full engine throttle opening.

Figure 6.24 Clutch pressure profiles during clutch to clutch upshifts.

6.4.7 Initial Clutch Pressure Profiles

As already mentioned previously, the pressure profiles for clutch torque control during shifts must be validated and finalized in the calibration process of test vehicles. However, there must be an initial set of clutch pressure profiles for each of the shift events to be implemented by the transmission so that the calibration vehicle can be test driven. For transmissions with multiple gears, the number of shift events can be overwhelming. For example, the ZF eight‐speed RWD AT analysed in Chapter 5 has 36 direct clutch to clutch shifts, and each of these direct shifts can be made under different operation conditions as dictated by the shift schedule, so the total number of shift events will then be in the hundreds. The initial clutch pressure profiles, if selected appropriately, not only significantly shorten calibration time, but also lead to better calibration results in terms of shift response and smoothness. Model simulation, as discussed in Chapter 5, is the main tool for the selection of these initial clutch pressure profiles. Firstly, the torque values of the applied clutches under static condition in each gear can be calculated using the clutch torque table of the transmission for each shift event according to the shift schedule. Secondly, these clutch torque values are converted to the clutch pressure, with a safety factor on the clutch torque capacity, using the clutch torque capacity formula as shown by Eq. 6.1. Thirdly, pressure profiles that follow the patterns illustrated in Figure 6.24 are used as input control variables for the vehicle system model shown in Figure 5.26 to simulate the transmission shift process for the assessment of shift quality. The pressure profiles that are optimized by model simulation can then be used in the test vehicles in the calibration process for validation and finalization. Note that the pressure profiles in Figure 6.24 need to be converted to current signals sent to the related variable force solenoids during transmission shifts, as shown in Figure 6.25.

Figure 6.25 Clutch piston initial stroke attributes.

6.4.8 Initial Piston Stroke Attributes

For clutches that are not designed with counter pistons, there will always be a time delay for the clutch pressure to respond to the control signal at the beginning of the shift, as shown in Figure 6.25. This is because the clutch pressure can only build up after the piston chamber is filled up with ATF. After the piston chamber is filled up, the piston will move against the return spring and the wave disk (if present), to eliminate the backlash in the friction disks. The clutch pressure will then follow the control signal, as shown by the solid line and dotted line respectively in Figure 6.25. It is critical to quantify the initial piston stroke attributes to minimize the time delay caused by them and to control the clutch pressure accurately as soon as possible after the shift starts. Although fill‐up and initial stroke attributes are difficult to model analytically, both of them can be calibrated in a laboratory setup or in a test vehicle. In the calibration setup, a pressure sensor is installed in a location close to the piston chamber and measures the pressure build‐up process in response to the current signal sent to the related VFS. In the control system of some ATs, databases are established for piston fill‐up attributes and initial stroke attributes based on the calibration data [12,13]. During transmission shifts, VFS control signals are interpolated in real time from the database by the transmission control unit (TCU) using an array of sensor‐provided inputs, such as clutch speed, ATF temperature, and system line pressure. By using this technique, it is possible to correlate the clutch response in the initial stage of the shift to the control signal, even though the correlation is not in linear proportion. More importantly, it is therefore technically possible to control the duration of the initial piston attributes so that the clutch pressure can be controlled accurately and in good time.

6.4.9 Feedback Shift Control

Feedback control is only used for the inertia phase during shifts since the transmission ratio is not changed in the torque phase. When a shift is deemed necessary by the TCU, it will also figure out the desired ratio change rate and the clutch pressure profile based on the vehicle operation condition. Pressure control signals are sent to the VFS for the off‐going and oncoming clutches respectively. This will cause the change of dynamic status of the powertrain system immediately. This change is monitored by the TCU via various sensors, such as the engine speed sensor, and the transmission input and output speed sensors. As the oncoming clutch pressure ramps up and the off‐going clutch ramps down, the off‐going clutch will start to slip at some point, which is detected by the sensors as soon as it happens. Closed loop control is then used for the control of the oncoming clutch pressures, with the off‐going clutch pressure brought to zero as quickly as possible. In the inertia phase, the transmission input and output speeds and the engine speed are measured by speed sensors on a real time basis at designed sampling time intervals. The TCU calculates the ratio change rate based on the transmission input and output speeds and compares it to the desired ratio change rate, as shown in Figure 6.26. Clutch pressure is then refined upon the base pressure profile and controlled by the VFS to achieve the desired ratio change rate. In addition to the desired ratio change rate, other target variables such as the engine speed and the vehicle acceleration can also be used as the control reference for the feedback control in similar configurations to those shown in Figure 6.26[8,13,14]. Generally, feedback control for transmission shifts possesses characteristics and benefits described in the following:

• Shift quality is further improved through fine adjustments of clutch pressure profiles in real time control.
• Variations in properties and attributes always exist for production vehicles of the same model on both component and system levels, leading to transmission shift quality variations from vehicle to vehicle. Feedback shift control minimizes the effects of these variations on shift quality and achieves better consistency in transmission shift quality and driver feel among production vehicles of the same model.
• The transmission properties and attributes on both component and system levels change as the vehicle ages, leading to a worsening in shift performance. Feedback shift control minimizes this worsening and achieves consistency in shift performance over the product service life.
• Feedback shift control is implemented by control software at no hardware cost increase.

Figure 6.26 Control loops for torque phase and inertia phase.

The effects of feedback control for transmission shifts are illustrated in Figure 6.27, where the control reference is the engine speed or transmission input speed. During a shift, a target engine speed profile in the shift inertia phase is pre‐designed and the oncoming clutch pressure is controlled so that the engine speed will follow the target profile. The feedback control starts at a point after the transfer from the torque phase to the inertia phase since the feedback signals can only be obtained after the system responds to the initial inertia phase control. In comparison with shift control without feedback, the overshoot in the transmission output torque during the inertia phase can be reduced for shift quality enhancements.

Figure 6.27 Shift control in inertia phase with feedback on engine speed.

6.4.10 Torque Based Shift Control

The advantage of torque based shift control is that it provides accurate clutch torque control in real time during shifts in accordance with the transmission input torque when shifts are initiated. This leads to shift quality optimization based on the powertrain operation status. There are three technical issues that need to be addressed for the implementation of torque based shift control:

• accurate estimation of transmission input torque
• determination of the torque profiles of the off‐going and oncoming clutches during shifts
• accurate conversion of the clutch torque profiles to the clutch pressure profiles

The estimation of transmission input torque is straightforward and is mainly based on the engine map, engine operation condition, status of auxiliary systems, and torque converter characteristics. Firstly, the nominal engine torque value is interpolated from the engine torque map in terms of the engine throttle opening, RPM, and intake air pressure. This nominal torque is then modified based on signal inputs from related sensors, such as temperature, atmospheric condition, air/fuel mix ratio, and any other inputs that may affect engine torque output. The net torque applied by the engine to the torque converter impeller is obtained by subtracting the loads of the accessary systems, such as air conditioner and power steering pump, from the modified nominal torque. The transmission input torque is then determined by considering the mass moment of inertia of the flywheel–impeller assembly and the torque converter characteristics, using Eqs (5.122–5.125).

Shift operations are completed by controlling the off‐going and oncoming clutch torque profiles with the objective of achieving a smooth transfer of the transmission output torque from the current gear to the target gear. As detailed in Chapter 5 and previous sections of this chapter, the variation of the transmission output torque during transmission shifts is critical for shift quality. This is because the transmission output torque is linearly related to the vehicle longitudinal acceleration or the G value to which the driver or passenger is most sensitive. Various techniques can be applied to determine the base torque profiles, which are then modified by model simulation detailed in Chapter 5 and finalized through the calibration process. For example, Honda engineers proposed a so‐called “G design” method for the determination of clutch torque profiles [12,13]. A pre‐selected G value that is validated to be conducive to shift quality is designed for each shift. Based on the transmission input torque that is determined by the steps described previously, the clutch torque profiles are then converted from the target G values, as illustrated in Figure 6.28. Note that the selection of G value in the G design method is equivalent to the minimization of torque hole and overshoot of the transmission output torque because of the proportionality between the longitudinal acceleration and the output torque. Using model simulation detailed in Chapter 5, it is possible to determine the clutch torque profiles that minimize the torque hole and overshoot for all shifts with the transmission input torque as a given variable.

Figure 6.28 Torque based shift control logic.

The clutch torque profiles obtained in this way, as illustrated in Fig. 6.28, must be converted to the corresponding clutch pressure profiles in the shift control processes. This is the most difficult of the three technical issues mentioned previously. There are two main difficulties here: the correlation of the hydraulic pressure in clutch piston chamber and clutch torque; and the correlation of clutch pressure and the current signal sent by the TCU to the VFS in the clutch apply circuit. The first difficulty is mostly overcome if the friction coefficient of the clutch disks can be estimated with accuracy on a real time basis. As shown in Eq. 6.1, for a given clutch, its torque only depends on the pressure in the piston chamber and the friction coefficient. To convert the clutch torque into the control variable, i.e. the clutch pressure, there must be an effective method of estimating the friction coefficient under real time condition. Each clutch is designed with a given nominal friction coefficient for the friction disks but the friction coefficient variates around the nominal value with respect to clutch temperature and slippage. Therefore, it is necessary to establish a database of the variation of friction coefficient under various clutch temperatures and slippage deemed possible by vehicle operation conditions. Such databases have been successfully applied in transmission control systems, such as the control system of the Honda five‐speed clutch to clutch AT [13]. The second difficulty can be solved by improving the accuracy of the variable force solenoid and design optimization of the clutch apply circuit. In automatic transmissions currently in production or under development, each clutch is assigned a VFS in the hydraulic circuit so that the clutch pressure is controlled directly with minimum interference from other clutches or actuators. Accurate correlation between pressure and current signals (in terms of PWM) can therefore be achieved to control the clutch torque during shifts.

In summary, torque based transmission shift control can be implemented in steps that are illustrated in Figure 6.29. When a shift is commanded by the TCU, the nominal engine torque is firstly interpolated from the engine map in terms of the engine speed and throttle opening and is then modified using sensor‐provided data such as intake air pressure, atmospheric condition, and fuel/air ratio. The torque on the converter impeller is then obtained by subtracting the loads of auxiliary systems from the modified engine torque. The net transmission torque is the result after considering the torque converter dynamics and characteristics. Knowing the transmission torque, various techniques can then be used to determine the clutch torque for shift control. The G design method is one of these techniques. The clutch torque can also be determined through model simulation with designed transmission output torque patterns. The clutch apply pressure that produces the clutch torque is then interpolated from the friction coefficient database using sensor provided data on ATF temperature, clutch clamping force, and slip rate. Afterwards, the clutch apply pressure is converted to VFS signal by the TCU which controls the clutch torque via the related circuit. In the initial piston stroke stage, the control signal is interpolated from the piston attributes database to minimize the delay of clutch actuation. As the transmission system responds to the VFS signal during a shift, as shown in Figure 6.29, open loop control and feedback control are then used for the control of the torque phase and inertia phase during the shift, as illustrated in Figure 6.26.

Figure 6.29 Torque based shift control logic.

6.4.11 System Diagnosis and Failure Mode Management

This function of the transmission control system is product specific and is realized by a combination of hardware and software. For example, pressure switches can be used in clutch pressure control circuits, as shown in Figures 6.15 and 6.16, to detect the pressure build‐up. The status of these switches will tell whether or not the circuit is working properly. In case of malfunctions, the pressure switch will detect abnormal pressure values and provide relevant signals to the TCU, which then processes these signals to identify the root cause of the malfunctions. Generally, ATs today have the following features in system diagnosis and failure mode management:

• Control software and database can be updated after the sale of the vehicle. The OEM or transmission system suppliers develop and improve transmission control technologies continuously, the software or the database in existing vehicles can be replenished with updated versions at no cost to the vehicle owner. This allows existing vehicles to benefit from newly developed technologies and, more importantly, provides an easy fix for software issues in the transmission control system.
• Wherever possible, the TCU will send out a warning signal and display it on the instrument panel when malfunctions occur or an abnormal behavior, at component or system level, is detected. This urges the driver to have the vehicle inspected and serviced accordingly in the dealership or repair shop.
• With safety guaranteed, the control system allows the driver to drive the vehicle to a repair shop in case of a malfunction under the failure mode management scheme. This is usually realized by the default setup of the shift solenoids and hydraulic circuits in the transmission control system. For example, the transmission can be set in first gear in case of malfunction to allow the driver to drive the vehicle at low speed to the nearest repair shop.
• Transmission control systems can be checked on a regular basis after specified mileage intervals using specific inspection tools. This allows prognosis of potential malfunctions caused by component aging or function degradation by replacing problematic components ahead of their gross failures.

6.5 Calibration of Transmission Control System

There are many variables in the transmission control system, some of which are related to software that implements control strategies or algorithms, and others are related to hardware such as valves, clutches, and sub‐circuits. These variables affect the transmission system response interactively. Moreover, the attributes or characteristics of control hardware components vary with respect to the system operation condition. Since each transmission shift corresponds to a specific system operation status, transmission control calibration needs to be conducted per each shift event for the optimization of shift quality in all vehicle operation ranges. Due to the large number of shift events in ATs with multiple gear ratios, transmission control calibration is a time consuming and expensive process in terms of labor cost. The calibration process is by nature highly experimental and is based on trial and error techniques. Engineer experiences and effective management are critical for shortening the powertrain calibration process of new vehicle models. Since the transmission works together with the engine and torque converter in the vehicle powertrain, transmission control calibration results directly depend on accurate and reliable data on engine output and converter characteristics which are obtained by engine and torque converter calibrations.

Transmission control calibration can be conducted on both component level and system level, in a laboratory environment or on test vehicles. At the component level, the main objectives are to obtain accurate data on the performance and attributes of key components in the transmission control system, such as clutch pistons, friction elements, and hydraulic circuits. Component level calibration can be mostly conducted in a laboratory setup that emulates conditions that the component is subject to when applied in the transmission system. Shift process control, including torque converter clutch control and engine spark retarding, is mainly calibrated at system level on test vehicles. Note that shift control can also be conducted at the initial calibration stage under laboratory environment, as shown in Figure 6.30. In some respects, the laboratory setup has advantages in emulating extreme transmission operation conditions and off‐season weather patterns. In addition to the various sensors that are to be used in the control system of production vehicles, sensors used only for calibration purposes are installed at key positions in the setup shown in Figure 6.30 for a laboratory test or in the transmission of the test vehicles. For example, torque sensors, which are not used in production vehicles, may be used for transmission calibration to measure the torque values at key locations, such as the input and output shafts. An array of data is collected and processed by calibration hardware and software tools from sensors on various variables:

• Pressure values and profiles at key locations in the hydraulic circuit of the transmission control system. This indicates the response of the clutch apply or release pressure to control signals and is the main task in transmission calibration.
• Speeds of the engine, torque converter turbine, and transmission output. Additional speed sensors may be used in calibration to measure clutch slippage in shift operations. An accelerometer can be used in calibration vehicles to measure vehicle longitudinal acceleration in real time.
• ATF and environment temperatures.
• Engine intake pressure, throttle position, air/fuel ratio, etc.
• Torque on transmission output shaft or haft shaft.

Figure 6.30 Transmission control system testing set‐up.

6.5.1 Component Level Calibration

This can be conducted in a laboratory setup for the component concerned, and the main purpose is to obtain accurate quantitative data on the performance specifications and attributes when the component is applied in the transmission control system. As discussed previously, these data are critical for the correlation between control signals and clutch pressure profiles in the torque based shift control for transmission shifts. The calibrations on clutch piston attributes, clutch disk friction coefficient, and response of hydraulic circuits for line pressure control and clutch apply pressure control are most important on the component level, as described in the following.

Clutch piston attributes: As illustrated in Figure 6.25, for clutches that are not designed with counter pistons, there is a time delay for the piston chamber pressure to respond to the control signal. This time delay can be minimized by minimizing the piston chamber cavity in hardware design. For a given clutch, the time delay can be effectively handled if quantitative and reliable data on the attributes of clutch initial actuation are available. These data can be obtained by calibrating the clutch piston pressure response to the control signal (i.e. VFS current) in a laboratory setup as shown in Figure 6.30 or in test vehicles. In such calibrations, a pressure sensor is installed at a location close to the piston chamber to measure the pressure build‐up when control signals are sent to the related VFS, while other sensors record the variables concerned with transmission operation status, such as temperature, clutch slip, and line pressure. A database is established using the calibration data on the clutch piston attributes during filling‐up and initial stroke. This allows accurate timing for the actuation and pressure ramping‐up of the oncoming clutch during shifts, as discussed previously.

Clutch disk friction coefficient: The friction coefficient of clutch disks depends on clutch temperature, clamping force, and clutch slippage. Accurate estimation of the friction coefficient is necessary for accurate clutch pressure control for the target clutch torque values during shifts. The friction coefficient variations of multiple disk clutches can be quantified under laboratory conditions that emulate transmission operations. In the experimental setup, torque sensors are installed on the shafts on both sides of the multiple disk clutch. The torque values are converted to the friction coefficients based on the clutch design parameters using Eq. 6.1. This leads to the establishment of the friction coefficient database shown in Figure 6.29.

Line pressure and clutch pressure circuits: The circuits for line pressure control and clutch pressure control are shown in Figures 6.1 and 6.5–6.7 respectively in various design variations. Each of these circuits is a subsystem consisting of valves, orifices, and hydraulic passages. The target pressure, whether it is line pressure or clutch piston chamber pressure, is controlled by the VFS in the related circuit. How the target pressure responds to the VFS control signal depends on the circuit attributes. It is therefore important to know the accurate correlation between the target pressure and the VFS control signal. This correlation can be calibrated either on test vehicles or in a laboratory setup. Pressure sensors are installed in the related circuits to measure the pressure values under real time conditions or conditions emulating real time transmission operation. The calibrated correlation can then be used to control the target line pressure or clutch pressure via the respective variable force solenoids.

6.5.2 System Level Calibration

The ultimate target of transmission calibration is to optimize the shift schedule and the quality of all shifts in the schedule that covers the whole vehicle operation range. Therefore, system level calibration concerns two mutually related issues: the calibration of the shift schedule and the calibration of all shifts in the schedule. As discussed in Chapter 5, model simulation can be used to predict the performance and fuel economy of an automatic vehicle with a given shift schedule. The simulation results and experiences on existing vehicles are used to establish an initial shift schedule for the optimized trade‐off on fuel economy and performance. This initial shift schedule is then used for test vehicles in the calibration process and will be further adjusted to achieve the optimized trade‐off on fuel economy, performance, and drivability. As for the control of the shifts in the shift schedule, model simulation can also be used to obtain initial clutch torque profiles that serve as the baseline shift control for test vehicles in the calibration process.

System level calibration can also be conducted in both the laboratory setup shown in Figure 6.30 and on test vehicles. Of course, laboratory calibration can only provide preliminary and complementary results on transmission shift control quality. The shift schedule and shift control for production vehicles have to be finalized through a painstaking calibration process on the test vehicles. There are several technical issues to be addressed in the vehicle calibration process, as described in the following.

Selection of shift events: As discussed previously, there are dozens of direct shifts, which only involve an off‐going clutch and an oncoming clutch, in a multi‐ratio automatic transmission. Each of these shifts can be made at different engine throttle opening and vehicle speeds along the particular shift threshold line in Figure 6.17. A number of points are selected on the shift threshold line and shift control calibration is conducted per each shift defined by one of these points. For example, if 10 points are chosen on the 1–2 upshift threshold line in Figure 6.17, there will be 10 1–2 upshift events that need to be separately calibrated. Calibration data on control variables, such as clutch pressure, line pressure, engine spark retard timing, and VFS signal current, are recorded for each of these shift events in the database of the transmission control system. If a shift is to be made between the selected points, the shift control variables are then determined by interpolation from the database. Counting all upshifts and downshifts as well as kick‐down power‐on downshifts, the total number of shift events to be calibrated for the shift schedule in Figure 6.17 will be in the hundreds. Note that the selection of shift points on the shift threshold line must reflect the most frequently used vehicle operation patterns. The shift threshold itself, as defined by the shift events, is a calibration target and is refined in the calibration process for the optimization of vehicle shift quality and drivability.

Converter clutch control calibration: The torque converter works together with the engine and transmission in the vehicle powertrain and its operation status is determined by system dynamics. The torque converter clutch control therefore needs to be calibrated on test vehicles. The converter clutch is applied or released according to the schedule shown in Figure 6.19. The calibration of converter clutch control concerns two issues: clutch release timing ΔT during shifts, as shown in Figure 6.20; and clutch apply and release process control, as shown in Figure 6.21. As discussed previously, the torque converter clutch release timing ΔT shown in Figure 6.20 directly affects the shift smoothness of upshifts and downshifts. This timing, as a delay after upshift initiation or as an amount of time ahead of downshift initiation, needs to be calibrated in each shift for the optimized damping effects of the torque converter on the shift process. This timing amount is just the difference between the points of time at which shift signal and converter release signal are sent by the TCU to the respective VFS. The process of converter clutch apply and release is controlled by the circuit shown in Figure 6.9. Pulse width modulation is used to control the converter clutch pressure during apply or release in a profile illustrated in Figure 6.21. The duty cycles in each step during apply and release are to be calibrated following the profile shown in Figure 6.21 for optimized smoothness when the converter clutch is released at the beginning of a transmission shift or applied after a shift in fixed gear operation.

Engine spark retarding timing calibration: As discussed in Chapter 5 and early in this chapter, transmission input torque reduction reduces the oncoming clutch torque required to complete the shift and lowers the transmission output torque overshoot. The transmission input torque reduction should be timed to start right after the inertia phase starts in upshifts. To control this timing, the engine controller should retard the spark to stop firing the selected cylinders at the exact crankshaft rotational position upon receiving the command from the TCU via control area network (CAN). The amount and timing of engine output torque reduction depend on the response to the engine controller and the engine characteristics. The key to the perfect engine spark retard timing is for the engine controller to send out spark retard commands in good time, so the engine responds with the reduced output torque as soon as the inertia phase starts. Through test vehicle calibration, engine spark retard timing can be determined for all shifts to minimize the torque hole and torque overshoot for the optimized shift quality.

In summary, calibration of transmission control is a lengthy trial and error practice where experience and existing data on similar products are critical in enhancing the effectiveness and shortening the process. This section only provides a guideline on control variables to be calibrated and the basic techniques used in transmission calibration. Publications are hard to find in the public domain in transmission control and calibration areas. Transmission control and calibration engineers are often internally trained at work by OEMs or suppliers through technical sessions or by technical manuals that do not circulate outside.

References

1. 1 Gott, P.G.: Changing Gears: The Development of the Automotive, Society of Automotive Engineers, 1991, ISBN 1‐56091‐099‐2.
2. 2 ASE Transmission/Axle/Driveline Forum Committee, Design Practices: Passenger Car Automatic Transmissions, Third Edition, AE‐18, SAE Publication, 1994, ISBN 1‐56091‐506‐4.
3. 3 Schwab, M.: Electronically‐Controlled Transmission Systems – Current Position and Future Developments, SAE Paper No. 901156.
4. 4 Taga, Y., Nakamura, K., Ito, H., and Taniguchi, T.: Toyota Computer Controlled Four‐Speed Automatic Transmission, SAE Paper # 820740.
5. 5 Wilfinger, E. and Thompson, J.: Borg‐Warner Australia Model 85 Automatic Transmission, SAE Paper No. 880480.
6. 6 Shinohara, M, Shibayama, T., Ohtsuka, K., Nawata, K, Ishii, S., and Yoshizumi, H.: Nissan Electronically Controlled Four‐speed Automatic Transmission, SAE Paper # 890530.
7. 7 Kondo, T., Iwatsuki, K., Taga, Y., Tanguchi, T., and Taniguchi, T.: Toyota “ECT‐I”: a New Automatic Transmission with Intelligent Electronic Control System, SAE Paper # 900550.
8. 8 Taniguchi, H. and Ando, Y.: Analysis of a New Automatic Transmission Control System for Lexus LS400, SAE Paper # 910639.
9. 9 Narita, Y.: Improving Automatic Transmission Shift Quality by Feedback Control with a Turbine Speed Sensor, SAE Paper # 911938.
10. 10 Hojo, Y., Iwatsuki, K., Oba, H., and Ishikawa, K.: Toyota Five‐speed Automatic Transmission with Application of Modern Control Theory, SAE Paper # 920610.
11. 11 Taga, Y., Nakamura, K., Ito, H., ad Taniguchi, T.: Toyota Computer Controlled Four‐Speed Automatic Transmission, SAE Paper # 820740.
12. 12 Ohashi, T., Asatsuke, S., Moriya, H. and Nobie, T.: Honda’s 4‐speed All clutch to Clutch Automatic Transmission, SAE Paper # 980819.
13. 13 Wakamatsu, H., Ohashi, T., Asatsuke, S., and Saitou, Y.: Honda’s 5‐speed All clutch to Clutch Automatic Transmission, SAE Paper # 2002‐01‐0932.
14. 14 Konda, M., Hasegawa, Y., Takanami, Y., Arai, K., Tanaka, M., Kinoshita, M., Ootsuki., T., Yamaguchi, T., and Fukatsu, A.: Toyota AA80E 8‐speed Automatic Transmission with Novel Powertrain Control System, SAE Paper # 2007‐01‐1311.
15. 15 Lee, S., Zhang, Y., Jung, D. and Lee, B.: 2014, A Systematic Approach for Dynamic Analysis of Vehicles with 8 or More Speed Automotive, ASME Journal of Dynamic Systems, Measurement, and Control, Vol. 136, No. 5, 051008(1–11), 2014.
16. 16 General Motors Corporation, Hydra‐Matic 4T80‐E, Technician’s Guide, Second Edition, Copyright 1994 Powertrain Division.
17. 17 General Motors Corporation, Hydra‐Matic 6T40/45, Technician’s Guide, Copyright 2008 Powertrain Group.

Problem

1. The stick diagram of the Ford 10‐speed RWD automatic transmission for pickup trucks is as shown. Construct the framework hydraulic circuit similar to Figure 6.16 for the transmission control system.