JPH11182672A - Lockup control device of torque converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily surely detect the judder of a lockup clutch, during the process of determining differential pressure, by achieving coastal lockup at as small differential pressure as possible. SOLUTION: Lockup during coasting is achieved by means of differential pressure calculated by adding a predetermined differential pressure to the learned value of a slip-starting engagement differential pressure at which the amount of torque converter slip attains a microscopic set value. When the learning of the slip-starting engagement differential pressure is judged as being complete (S751), the judder of a lockup clutch is judged depending on whether or not the stored learning time TGB is greater than a set time TGT (S752), and a process against failures is performed at the occurrence of the judder (S753). When the learning is judged as being incomplete (S751) or, even if it is complete, TGB is judged as being less than TGT (S752), then the judder is recognized from a state under which the number CGE of times that the learning is complete is judged as being less than a predetermined value RTOG (S755), even if the number CGS of times that learning is started is judged (S754) as having reached a predetermined value CGSmax ' and thereby a process for failures is effected (S753).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a lock-up control device for inserting a torque converter inserted into a transmission system of an automatic transmission into a lock-up state in which input and output elements are directly connected, and more particularly to a vehicle. The present invention relates to a coast lock-up control during inertial running (coast running) and a technique for determining an abnormal engagement of a lock-up clutch during the control.

[0002]

2. Description of the Related Art Since a torque converter transmits power between input and output elements via a fluid, it performs a torque fluctuation absorbing function and a torque increasing function, but has poor transmission efficiency. Therefore, under running conditions in which the torque fluctuation absorbing function and the torque increasing function are not required (in the lock-up area), the torque converter is in a lock-up state in which the input and output elements are mechanically directly connected by a lock-up clutch. So-called lock-up type torque converters are widely used today.

[0003] In a vehicle equipped with an automatic transmission having such a lock-up type torque converter, while the vehicle is in an inertia running (coast running) state, a decrease in engine speed is prevented to prevent a fuel cut time ( To increase the fuel injection stop time) and increase fuel efficiency, the torque converter is changed from a converter state in which the input and output elements are not directly connected to a lockup state in which the input and output elements are directly connected (coast lockup state). Is common practice.

Here, in order to enhance the effect of the above fuel consumption measures, the coast lock-up region is generally expanded to the lowest possible rotational speed range. In this case, the lock-up clutch engagement differential pressure is reduced in the coast lock-up state. At the maximum value, when the wheels lock due to sudden deceleration from the coasting state, the lock-up clutch engagement differential pressure is reduced from the maximum value. There is a concern that the release of the up may be delayed and the locked wheel may stop the operation of the engine (generating engine stall).

To solve this problem, it is conceivable to set the lockup clutch engagement differential pressure during coast lockup to a value lower than the maximum value. However, in this case, the lock-up clutch is not completely engaged, and the lock-up clutch generates an oscillating slip called a judder depending on the driving state and running conditions, and the torque converter slip amount is repeatedly fluctuated to make the vehicle uncomfortable. It is feared that a severe vibration is caused.

Therefore, when performing control to lower the lock-up clutch engagement differential pressure below the maximum value during coast lock-up, it is necessary to take measures such as detecting the judder of the lock-up clutch and releasing the coast lock-up. . By the way, as a device for detecting the judder, conventionally, as described in JP-A-7-42768, for example, a judder of a lock-up clutch is determined based on a difference between a rotational fluctuation of a driving wheel and a rotational fluctuation of a non-driving wheel. A device for detection has been proposed.

[0007]

However, in such a conventional judder detection technique, it is necessary to detect both the rotation fluctuation of the driving wheel and the rotation fluctuation of the non-driving wheel. There is a problem that both hardware and software have many problems, mainly due to the need to provide them, which makes the hardware redundant, and the troublesome algorithm for detecting rotation fluctuations complicates the software. Is concerned.

According to the first aspect of the present invention, when the lock-up clutch engagement differential pressure at the time of coast lock-up is set to a value lower than the maximum value, the low lock-up clutch engagement differential pressure is set to a minimum without any trouble. It is an object of the present invention to be able to set a low differential pressure value, and to make sure that judder of a lock-up clutch can be detected without the above-mentioned concern in the process of determining the differential pressure.

According to a second aspect of the present invention, the judder of the lock-up clutch can be detected in real time, and at the same time, the judder of the lock-up clutch can be predicted. The purpose is to be able to achieve the object more reliably.

[0010] A third aspect of the present invention is to easily and surely achieve the object of the first aspect of the invention.

A fourth aspect of the present invention is directed to an invention combining the advantages of the second and third aspects.

A fifth invention according to claim 5 is the first invention,
It is another object of the present invention to further reliably detect the judder of the lock-up clutch in the second and fourth inventions.

A sixth object of the present invention is to prevent the judder detection of the lock-up clutch from being erroneously detected.

[0014]

SUMMARY OF THE INVENTION To achieve these objects, a lock-up control device for a torque converter according to a first aspect of the present invention provides a lock-up control device that responds to a differential pressure applied to a lock-up clutch when the vehicle is coasting. In a lock-up control device for a torque converter in which the input / output elements are directly connected by a clutch to perform a lock-up operation by coasting, the lock-up state at the time of coast lock-up is such that the torque converter slip amount is very small. A coast lockup differential pressure obtained by adding a predetermined differential pressure to a learned value of a slip start lockup clutch engagement differential pressure that is a set slip amount is achieved, and the slip start lockup clutch engagement differential pressure is learned. From the lock-up state during torque conversion And amount of slip determines the abnormal engagement of the lockup clutch from the change degree of the torque converter slip in until reaching the small setting slip and is characterized by being configured to perform abnormality correction.

The lock-up control device for a torque converter according to a second aspect of the present invention is the first aspect, wherein learning of the slip differential lock-up clutch engagement differential pressure for changing the torque converter slip amount from 0 to the minute set slip amount is started. From the above, based on the learning time until the end of learning when the torque converter slip amount reaches the minute set slip amount, it is determined that the lock-up clutch is abnormally engaged when the learning time exceeds the set time. It is characterized by the following.

In the lock-up control device for a torque converter according to the third invention, in the first invention, the start of learning of the slip-up lock-up clutch engagement differential pressure for changing the torque converter slip amount from 0 to the minute set slip amount. When the learning success rate is less than the set rate, the lock-up clutch is used when the learning success rate is less than the set rate, based on the learning success rate, which is the ratio of the cumulative number of learning terminations in which the torque converter slip amount has reached the minute set slip amount with respect to the integrated number of times. Is determined to be abnormally fastened.

According to a fourth aspect of the present invention, there is provided a torque converter lock-up control device according to the first aspect, wherein learning of the slip initiation lock-up clutch engagement differential pressure for causing the torque converter slip amount from 0 to the minute set slip amount is started. From, based on the learning time until the end of learning when the torque converter slip amount reaches the minute set slip amount, it is determined that the lock-up clutch is abnormally engaged when the learning time is equal to or longer than the set time, or Based on a learning success rate represented by a ratio of the cumulative number of times of learning to the cumulative number of times of learning, based on a learning success rate, when the learning success rate is less than a set rate, it is determined that the lock-up clutch is abnormally engaged. It is characterized by having comprised.

According to a fifth aspect of the present invention, there is provided a lock-up control device for a torque converter according to the first, second or fourth aspect of the invention, wherein the average learning time is obtained by averaging the learning time over the past several times. It is characterized in that it is determined that the lock-up clutch is abnormally engaged when the above is reached.

According to a sixth aspect of the present invention, there is provided a lock-up control device for a torque converter according to any one of the first to fifth aspects, wherein the engine speed is less than a set speed, the vehicle speed is less than the set vehicle speed, When at least one condition of whether the oil temperature is out of the set temperature range is satisfied, it is characterized in that the determination of the abnormal engagement of the lock-up clutch is stopped.

[0020]

In the coasting state of the vehicle, the torque converter is brought into a lock-up state in which the input and output elements are directly connected by a lock-up clutch responsive to the lock-up clutch engagement differential pressure. By the way, in the first invention, the coast lockup is determined by adding a predetermined differential pressure to a learned value of the engagement start differential pressure of the lockup clutch for slip starting at which the torque converter slip amount becomes a minute set slip amount. Therefore, the lockup clutch engagement differential pressure is not set to the maximum value during coast lockup, but is set to the minimum differential pressure required to achieve lockup. When the vehicle is locked, the lock-up release is promptly completed, and the problem of the engine stall does not occur.

Further, in the first invention, the torque converter slip amount between the lock-up state and the torque converter slip amount reaches the minute set slip amount during the learning of the slip differential of the lock-up clutch for slip start. An abnormal engagement of the lock-up clutch, such as a judder, is determined from the degree of change, and when the abnormality is determined, an abnormality countermeasure is taken. Therefore, it is possible to determine the abnormal engagement of the lock-up clutch without detecting both the rotation fluctuation of the driving wheel and the rotation fluctuation of the non-driving wheel as in the above-described conventional apparatus, and it is possible to determine whether the non-driving wheel has the rotation sensor. It is possible to solve the problem of the conventional device such that the hardware becomes redundant due to the necessity of providing the software, and also the problem that the software becomes complicated without the need for a troublesome rotation fluctuation detection algorithm.

In the second invention, the torque converter slip amount is reduced from the start of learning of the slip differential lock-up clutch engagement differential pressure for causing the torque converter slip amount to go from 0 to the minute set slip amount. Is determined as abnormal lock-up clutch engagement when the learning time is equal to or longer than the set time based on the learning time up to the end of the learning, so that abnormal lock-up clutch engagement can be detected in real time. At the same time, since the characteristic change of the lock-up mechanism can be detected using the length of the learning time as a scale, it is even possible to predict abnormal engagement of the lock-up clutch,
The object of the first invention can be more reliably achieved.

In the third aspect of the present invention, the torque converter slip amount is smaller than the accumulated number of times of starting the learning of the slip-up lock-up clutch engagement differential pressure for increasing the torque converter slip amount from 0 to the minute set slip amount. Based on the learning success ratio represented by the ratio of the number of times of learning completion that has reached the set slip amount, when the learning success ratio is less than the set ratio, it is determined that the lock-up clutch is abnormally engaged. A timer for measuring time and software are not required, so that the object of the first invention can be achieved more easily than in the second invention, and there is no learning of the engagement start differential pressure of the lock-up clutch for slip start. Even if not performed, it is possible to reliably determine the abnormal engagement of the lock-up clutch.

In the fourth aspect, the torque converter slip amount is reduced from the start of learning of the slip differential lock-up clutch engagement differential pressure for shifting the torque converter slip amount from 0 to the minute set slip amount. Based on the learning time until the end of learning to reach, it is determined that the lock-up clutch is abnormally engaged when this learning time is longer than the set time, or
Based on a learning success rate expressed as a ratio of the integration number of the learning end to the integration number of the learning start, when the learning success rate is less than the set rate, it is determined that the lock-up clutch is abnormally engaged. Thus, the above advantages of the second invention and the third invention are combined, so that it is possible to more accurately and reliably determine the abnormal engagement of the lock-up clutch.

In the fifth invention, when the average learning time obtained by averaging the learning time over the past several times exceeds the set time, it is determined that the lock-up clutch is abnormally engaged. Accordingly, no error is generated in the abnormal engagement determination of the lock-up clutch, and the reliability can be improved.

In the sixth invention, it is determined whether the engine speed is lower than the set speed, the vehicle speed is lower than the set vehicle speed,
When at least one condition of whether the hydraulic oil temperature is out of the set temperature range is satisfied, the abnormal engagement determination of the lock-up clutch is stopped, so that the detection of the abnormal engagement of the lock-up clutch is erroneously detected. Originally, the detection itself is stopped, and erroneous detection of abnormal lock-up clutch engagement can be prevented.

[0027]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows a drive system of a vehicle including a torque converter having a lock-up control device according to an embodiment of the present invention, wherein 1 is an engine as a prime mover, 2 is a torque converter, and 3 is a gear of an automatic transmission. The transmission mechanism 4 is a differential gear device, and 5 is wheels. These are sequentially drive-coupled as shown in the figure to form a drive system of the vehicle.

The output of the engine 1 is determined by an accelerator pedal 14 operated by the driver.
Is a pump impeller 2a as an input element driven by the engine 1, a turbine runner 2b as an output element coupled to an input shaft of the gear transmission mechanism 3, and a lock directly connecting the pump impeller 2a and the turbine runner 2b. A so-called lock-up type torque converter including the up clutch 2c.

The engagement force of the lock-up clutch 2c is determined by the differential pressure of the apply pressure P _A and the release pressure P _R at the front and rear (lockup clutch engagement differential pressure), apply pressure P _A is lower than the release pressure P _R For example, the lock-up clutch 2c is released and does not directly connect between the pump impeller 2a and the turbine runner 2b, so that the torque converter 2 functions in a converter state in which the slip is not limited.
When apply pressure P _A is larger than the differential pressure set value higher than the release pressure P _R, the lock-up clutch 2
c is fastened, the relative rotation between the pump impeller 2a and the turbine runner 2b stops, and the torque converter 2 functions in a lock-up state.

[0030] In this embodiment, the applied pressure in order to perform the predetermined lock-up control P _A and the release pressure P
_The lock-up control system that determines _R has the following configuration.
Lock-up control valve 11, apply pressure P _A and the release pressure P in response to the signal pressure P _S from the lock-up solenoid 13 is duty-controlled by the controller 12
_{In order} to determine _R , the lock-up control valve 11 and the lock-up solenoid 13 are assumed to be well known in FIG. That is, the lock-up solenoid 13 as a source pressure constant pilot pressure P _p, the controller 12
Locking the up command D which generates a signal pressure P _S in accordance with the (duty) from.

The lock-up control valve 11, as well as receiving the signal pressure P _S and the fed-back release pressure P _R in one direction, receives the apply pressure P _A that is the spring force and the feedback of the spring 11a in the other direction, As the signal pressure P _S from the lock-up solenoid 13 is increased with an increase in the lock-up command D (duty),
The lock-up control valve 11 increases the apply pressure P _{A above the} release pressure P _R and increases the pressure difference between them, that is, the lock-up clutch engagement differential pressure P _L = (P _A -P _R ), as shown in FIG. Thus, the lock-up clutch 2c can be engaged, and the torque converter is finally brought into the lock-up state. The lock-up control valve 11
Conversely with a decrease of the lock-up command D (Duty) As is decreasing signal pressure P _S from the lock-up solenoid 13, releasing the lock-up clutch 2c and a release pressure P _R to be higher than the apply pressure P _A Then, the torque converter 2 is set to the converter state.

As shown in FIGS. 1 and 2, a signal from a throttle opening sensor 21 for detecting a throttle opening TVO of the engine 1 and a vehicle speed VS are transmitted to the controller 12, as shown in FIGS.
Impeller rotation sensor 2 for detecting a signal from a vehicle speed sensor 22 for detecting the P, and the rotational speed N _I of the pump impeller 2a
3, a signal from a turbine rotation sensor 24 that detects the rotation speed _NT of the turbine runner 2b, a signal from an idle switch 25 that is turned on when the accelerator pedal 14 is released, and a hydraulic oil temperature F _T. And a signal from the oil temperature sensor 26 to be activated.

The controller 12 executes the control programs shown in FIGS. 4 to 11 based on the input information to perform lock-up control of the torque converter 2 as follows. Fig. 4 is executed only once when the engine is started with the ignition switch turned on.
First, in step 27, the slip start lock-up clutch engagement differential pressure learning completion flag FG is initialized to 0, and in the next step 28, the slip start lock-up clutch engagement differential pressure learning time measurement flag FGM is set.
Is initialized to 0.

FIG. 5 shows a main routine executed by a repetition process by a periodic interruption after the engine is started or by an irregular repetition process.
In steps 1 and 32, the current transmission state of the torque converter 2, that is, the complete lock-up state in which the input / output elements 2a and 2b are directly connected by the engagement of the lock-up clutch 2c, or the converter state in which the direct connection has been released is determined. The determination is made based on the duty D in FIGS.

In step 33, which is selected when the vehicle is in the complete lock-up state, when the region is determined from the throttle opening TVO and the vehicle speed VSP to determine the converter region, lock-up release control is performed as usual, and the control is not performed in the converter region. , The current complete lock-up state is maintained. In step 34, which is selected when it is determined in step 32 that the torque converter is in the converter state, when the region is determined from the throttle opening TVO and the vehicle speed VSP to determine the lock-up region, the engagement of the lock-up clutch 2c is determined. When performing control as usual and determining that it is not in the lockup area,
Maintain the current converter state.

When it is determined in steps 31 and 32 that the torque converter is not in the complete lock-up state and is not in the converter state, that is, when it is determined that the torque converter is in the lock-up control transition period, step 3 is executed.
Steps 3 and 34 are skipped and the progress of the control is continued.

In step 35, it is determined from the vehicle speed VSP and the throttle opening TVO whether or not the coast lockup control needs to be performed based on whether or not the vehicle is in the coast lockup region. The coast lock-up control and the abnormal engagement determination data of the lock-up clutch 2c, which will be described later, are performed. If the coast lock-up control is not necessary, the control proceeds to step 36. An abnormal engagement (judder) determination of the lock-up clutch 2c, which will be described later, and an abnormality process are performed, and the control ends.

The coast lock-up control and the preparation of the lock-up clutch abnormal engagement determination data in step 35 will be described with reference to FIG. 6. This control program is also assumed to be repeatedly processed by a periodic interrupt or an irregular interrupt. In step 41, the current coast lockup operation mode is checked. This operation mode is indicated by “step”, and it is assumed that there are states of step = 0 to 4 described below.

Step = 0 indicates that the vehicle is in a running state other than coast lockup and the torque converter is in a transmission state. Step = 1 indicates that the coast lock-up control has been started in the complete lock-up state and the vehicle has entered the coast running state. step
= 2 indicates that the initial rapid decrease control of the lock-up clutch engagement differential pressure during coast running is in operation. s
Step = 3 indicates that the initial rapid decrease control is completed and the slip differential lock-up clutch engagement differential pressure that is generated by generating a minute target slip is being detected. step
= 4 indicates that the coast lock-up state is achieved by the coast lock-up differential pressure obtained by adding a predetermined differential pressure to the detected slip start lock-up clutch engagement differential pressure.

In step 41, when the operation mode is "step =
0 ", that is, other than coast lockup, in steps 42 and 43, based on the duty D and the signal of the idle switch 25, in the complete lockup state and in the coast running state. Check if it is. When it is determined that the vehicle is in the complete lock-up state and the coasting state, in step 44, the coast lock-up control is started, so that the operation mode is changed from “step = 0” to “st
ep = 1 ”and the control is terminated. If it is determined in step 42 that the vehicle is not in the complete lockup state, or if it is determined in step 43 that the vehicle is not in the coasting state, the control is terminated and the operation mode is changed to“ step = 0 ".

As described above, when the operation mode is “step =
If "1" is set, the control proceeds from step 41 to step 45.
Proceed to. In this step 45, it is checked again whether or not the vehicle is in the coast running state, and if it is determined that the vehicle is in the coast running state, the control proceeds to step 46 and thereafter to perform the coast lock-up control. Thus, step 45
If it is determined that the vehicle is not in the coasting state, the learning time measurement process of the slip start lock-up clutch engagement differential pressure is initialized (interrupted) in step 720, and the coast lock-up control is initialized in steps 47 and 48. , And a process of returning the operation mode to “step = 0”.

The initialization (interruption) of the learning time measurement process in step 720 is performed as shown in FIG. 7. First, in step 721, the learning time _TG is cleared, and in step 722, the learning time measurement flag is set. The FGM is reset to 0, and a learning completion flag FG is set at step 723.
Is reset to 0.

In step 46, which is selected when it is determined that the vehicle is in the coast running state in step 45, it is determined whether the operation mode is "step = 1", that is,
It is determined whether or not the coast lockup control has been started.
If at the start of coast lockup control, step 49
In determining the impeller rotational speed N _I is less than the set rotational speed N _(INH), and, at step 50, if it is determined that the vehicle speed VSP is less than the set vehicle speed V _(INH), i.e. the lock-up clutch engagement When it is determined that the initial pressure drop condition is satisfied, the operation mode is set to “step” in step 51.
= 1 ”to“ step = 2 ”and control is performed in step 5
Advances to 2, the lock-up clutch engagement differential pressure P _L as at the instant t ₁ in Figure 15 from the current value P _{L / U,} the slip start lockup clutch engagement difference below the previously stored pressure P _{L / S (new)} Is rapidly reduced to the initial reduced pressure difference P _{L / O} obtained by adding the set pressure difference ΔP to the pressure difference ΔP.

In step 49, the impeller rotation speed N _I
Is not determined to be less than the set rotation speed N _(INH) ,
Alternatively, if it is not determined in step 50 that the vehicle speed VSP is _lower than the set vehicle speed V _(INH) , steps 51 and 52 are not executed until these conditions are satisfied, and step 4 is executed.
The determination at steps 9 and 50 is repeated until the condition is satisfied.

In the present embodiment, the slip start
Lockup clutch engagement differential pressure P_{L / S (new)}To add to
The constant pressure difference ΔP is, for example, the initial pressure reduction pressure P_{L / O}But Figure 1
5 coast lock-up differential pressure
Small lockup fastening differential pressure) P _{L / U (min)}Slightly higher value
It is preferable to set so that

Thereafter, in step 51, the operation mode is set to "s
Since “step = 2”, the control proceeds to step 53, where it is determined whether the operation mode is “step = 2”, that is, as shown at the instant t _{1 in} FIG. It is determined whether or not the rapid pressure reduction of the pressure P _L has been started. If it has been started, in step 54, the slip rotation speed S generated between the input and output elements of the torque converter 1
_L (S _L = N _I- N _T ) is the set slip speed S _(INH)
By determining whether or not the above is the case, it is determined whether or not abnormally large slip rotation has occurred when the lock-up clutch engagement differential pressure P _{L / U is set} to the initial pressure-reducing differential pressure P _{L / O.}

If an abnormally large slip rotation has not occurred, in step 55, the initial pressure-reducing differential pressure P _L of the lock-up clutch engagement differential pressure P _L at the instant t ₁ in FIG.
_It is determined whether or not the rapid decompression to the _{L / O} has been completed. If not, the control is terminated and the rapid decompression is advanced. Then, when the rapid pressure reduction is completed, the control proceeds to steps 56 and 57, the operation mode is changed from "step = 2" to "step = 3", and the slip differential lock-up clutch engagement differential pressure (P _{L / S ( new)} ) detection control is started.

[0048] Thus, when the slip rotation speed S _L that occurs between input and output elements of the torque converter 1 in step 54 is determined to set it slip rotation speed S _(INH) above, when an abnormal slip occurs In step 58, the lockup clutch engagement differential pressure P _L is increased by a predetermined capacity P _(up1) to eliminate abnormal slip in step 58, and then the operation mode is set to “step = 3” in step 56. ,
In step 57, the detection control of the slip differential lock-up clutch engagement differential pressure (P _{L / S (new)} ) is started.

[0049] Incidentally, the detection control of the slip start lockup clutch engagement differential pressure at step _{57 (P L / S (new} )), the lock-up clutch engagement pressure difference as at the instant t ₁ after the Figure 15 P _L Is gradually reduced by u each time,
For example, the following proportional-integral (PI) control can be used. _{e (t) = S L (} t) -S O (t) ··· (1) u (t) = K p · e (t) + (K I / s) · e (t) ··· ( 2) where S _O : target slip rotation speed _SL : slip rotation speed ( _SL = N _I- N _T ) s: differential operator K _p : proportional gain K _I : integral gain

In step 730 following the detection control of the slip differential lock-up clutch engagement differential pressure (P _{L / S (new)} ) in step 57, the learning time of the slip start lock-up clutch differential pressure is started to be measured. I do. This learning time measurement start process is as shown in FIG. 8. First, the learning time _TG is cleared in step 731, the learning time measurement flag FGM is set to 1 in step 732, and the learning start is started in step 733. the counter C _GS for integrating the number of times causes increment (stepping).

Here, the learning time _TG is a fixed time Δ
According to the program shown in FIG. 9 that is repeatedly executed at intervals of _TG , that is, while it is determined in step 735 that the learning time measuring flag FGM (see step 732 in FIG. 8) is set to 1, By incrementing the learning time T _G (see step 731 in FIG. 8) at 736, the measurement can be performed as shown in FIG.

When the detection control of the engagement start differential pressure (P _{L / S (new)} ) for the slip start lock-up clutch is started in step 57, the operation mode is “step = 3”. the process proceeds to determine where whether the slip rotational speed S _L is abnormal slip by determining whether setting is slip rotation speed S _(INH2) above has occurred. When an abnormal slip occurs, in step 59, after performing processing for the abnormal slip, such as stopping the detection control of the slip differential of the lock-up clutch engagement start pressure (P _{L / S (new)} ), the above steps 720 and 47 are performed. , 48 are executed.

When it is determined in step 58 that no abnormal slip has occurred, it is determined in step 60 whether the operation mode is "step = 3", that is, the slip start lock-up clutch engagement differential pressure detection control is started. It is determined whether or not it has been performed. Operation mode is "step = 3"
If not, the control is terminated as it is. If it is determined that the operation mode is “step = 3”, it is determined that the slip start lock-up clutch engagement differential pressure detection control has been started, and the control proceeds to step 61. Advance.

In step 61, the slip start lock-up clutch engagement differential pressure (P _{L / S (new)} is determined based on whether or not the slip speed S _L between the input and output elements of the torque converter 1 is a small set slip speed. _{) It} is determined whether the learning condition of (1) is satisfied. This learning condition is when the slip rotation speed S _L matches the target slip rotation speed S _O as shown in FIG. 15, or when the slip rotation speed S _L falls below the target slip rotation speed S _{O for a} predetermined time. It can be determined that the learning condition has been satisfied when the learning condition is maintained. When this condition is satisfied, at step 62,
The operation mode is changed from “step = 3” to “step = 4”, and in step 63, the previously stored slip start lock-up clutch engagement differential pressure is replaced with the newly detected slip start capacity P _{L / S (new)} . Update and remember.

At step 64, the newly detected slip start lock-up clutch engagement differential pressure P _{L / S (new)}
The coast lock-up differential pressure P _{L / U (min)} is obtained by adding the predetermined differential pressure α as in the instant t ₂ in FIG. 15, and this is reset as the minimum lock-up engagement differential pressure, and the corresponding duty D is set. The lock-up solenoid 13 shown in FIGS. 1 and 2 is commanded.

As described above, the predetermined pressure difference α is obtained by adding the predetermined pressure difference α to the slip-up lock-up clutch engagement pressure difference P _{L / S (new)} when the torque converter slip rotation speed S _L reaches the minute set value S ₀ . Since coast lockup is achieved with coast lockup differential pressure P _{L / U (min)} , lockup is achieved without setting lock-up clutch engagement differential pressure P _L to maximum value P _{L / U} during coast lock-up. The minimum differential pressure P _{L / U (min)} required for
In sudden braking as is clear from the change state of the lock-up clutch engagement differential pressure P _L and the torque converter slip rotation speed S _L also in FIG when the wheel is locked at the instant t ₃ in Figure 15 in this state, the lock-up The release is completed promptly, and the problem of the engine stall does not occur.

In step 740 following step 64, the above-described learning time measurement end processing is performed as follows. This processing is as shown in FIG. 10 in detail. First, in step 741, it is determined whether a state in which abnormal engagement (judder) of the lock-up clutch can be reliably detected or a state in which erroneous detection is likely to occur. In other words, a value between hydraulic fluid temperature F _T is the allowable lower limit value F _TL and the allowable upper limit F _TU, and the engine speed N _E is not less set rotational speed N _ES or more and the vehicle speed VSP is set When the vehicle speed is equal to or higher than VSP _S, it is determined that abnormal engagement (judder) of the lock-up clutch can be reliably detected. If any one of the conditions is missing, it is determined that there is a possibility of erroneous detection.

If detectable, steps 742-7
At 45, the learning time measurement process is normally terminated as follows. In step 742, the in-learning measurement flag FGM is reset to 0 in accordance with the end, so that step 735 in FIG. 9 does not execute step 736, thereby reducing the learning time _TG to the instant (FIG. 15). The measurement value at the instant t ₂ ) is held. Then, in step 743, the learning time T _GB stored for detecting abnormal engagement (judder) of the lock-up clutch described later is updated to the current learning time T _G (instantaneous t _{2 in} FIG. 15). In step 745, the learning completion counter C _GE is incremented, and the number of learning completion times is integrated.
Then, the learning completion flag FG is set to 1 to indicate that the learning has been completed normally.

If it is determined in step 741 that the abnormal engagement (judder) of the lock-up clutch is erroneously detected, in step 746, the learning start counter _CGS is reset to 0, and
The learning completion counter _CGE is also reset to 0 to interrupt detection of abnormal engagement (judder) of the lock-up clutch as described later.

As described above, after the coast lock-up control and the lock-up clutch abnormal engagement determination data prepared in step 35 of FIG. 5 are prepared, the lock-up clutch abnormal engagement (judder) is performed in step 36 of FIG.
The details of the determination process will be described below with reference to FIG. First, in step 751, as described above with reference to FIG. 10, the learning completion flag FG that is set to 1 when the learning of the slip start lock-up clutch engagement differential pressure (PL _{/ S (new)} ) has been completed normally is 1 or not. Is determined.

If learning has been completed normally, step 7
At 52, the storage updated in step 743 of FIG.
Learning time T_GBIs the set time T_GTDepending on whether or not
Lock-up clutch 2c is abnormally engaged (judder)
Is determined. Where the slip starts
Lock-up clutch engagement differential pressure (P_{L / S (new)}Learning)
Lock-up clutch 2c is abnormally engaged during control (judder
ー) When the lock-up clutch engagement difference occurs
Pressure P_LAnd torque converter slip speed S _LStrange
As is clear from FIG.
Learning is sudden deceleration instant t _ThreeDoes not complete even if it reaches, save learning time
T_GBIs the set time T_GTLock up with the above conditions
If the clutch 2c is abnormally fastened (judder)
You can judge.

If it is determined in step 752 that the connection is abnormal (judder), abnormal connection (judder) is detected in step 753, the abnormal connection is displayed, and the abnormality information is stored in the nonvolatile memory. Then, an abnormality countermeasure process such as prohibiting the engagement of the lock-up clutch 2c is performed.

Note that the learning completion flag F
If it is determined that G is not 1, that is, if learning of the slip-up lock-up clutch engagement differential pressure (P _{L / S (new)} ) is not completed, or even if completed, the stored learning time T _GB If it is determined that the time is less than the set time T _GT , the control proceeds to step 754 and the subsequent steps, where the control is represented by the ratio of the integrated number of learning end times to the integrated number of learning start times of the slip start lock-up clutch engagement differential pressure. Based on the learning success rate, the abnormal engagement (judder) of the lock-up clutch is determined based on whether the learning success rate is less than the set rate.

That is, in step 754, the count value (integrated value of the number of times of learning start) of the learning start counter C _GS incremented in step 733 of FIG. 8 is increased to a predetermined value number C _GSmax necessary for making the above determination. It is determined whether or not it has reached, and if it has not, the control is terminated as it is, and the abnormal engagement (judder) determination is not performed. When the count value of the learning start counter C _GS has reached the predetermined value C _GSmax , in step 755, step 744 of FIG.
In depending on whether incremented by the count value of the learning completion counter C _GE (integrated value of learning completion count) is less than the predetermined value RTOG, determines whether a learning success rate is less than the set ratio, C _GE When it is <RTOG (the learning success rate is less than the set rate), it is determined that the lock-up clutch is abnormally engaged (judder).

If it is determined in step 755 that the engagement is abnormal, in step 756, the learning start counter C _GS and the learning completion counter C _GE are reset to prepare for the next learning. And perform the abnormality countermeasure processing. Step 75
If it is determined in step 5 that the engagement is not abnormal, step 757
, The learning start counter _CGS and the learning completion counter _CGE are each reset to prepare for the next learning.

As described above, according to the algorithm for judging abnormal engagement when the learning start counter C _GS has reached the predetermined value C _GSmax and the learning completion counter C _GE is less than the predetermined value RTOG, the learning is successful. Compared to the case where the ratio is calculated and this is less than the set ratio, it is extremely useful because the calculation can be simplified as compared with the case where it is determined that the engagement is abnormal.

Next, at step 741 in FIG.
_TIs the allowable lower limit F_TLAnd allowable upper limit F _TUIs not a value between
Or the engine speed N_ESet rotation
Number N _ESIt is determined that it is not the above, or the vehicle speed VSP
Is the set vehicle speed VSP_SIf it is determined that it is not more than
explain. Under these conditions, lock-up
It is difficult to reliably detect abnormal fastening (judder)
In some cases, erroneous detection may occur. In this case,
In step 746, the learning start counter C_GSReset to 0
And the learning completion counter C_GEAlso reset to 0
From step 745, the learning completion flag FG is set
Since the process of setting to 1 is not performed, the control in FIG.
The control is terminated via steps 751 and 754.
It will be. As a result, the lock-up shown in FIG.
Medium without judgment of abnormal clutch engagement (judder)
And an abnormal judgment is made in spite of the above conditions.
Can be prevented from being erroneously detected.
Wear.

In the above-described embodiment, FIG.
In step 743 of 0, the stored learning time T _GB is updated to the current learning time _TG to contribute to judder detection performed in step 752 of FIG. 11. In this case, the learning time _TG varies. Erroneous judgment of judder. To solve this problem, as shown in FIGS. 12 and 13, the learning time _TG is averaged over the past several times to obtain an average learning time, and judder determination is performed based on the average learning time. Good.

FIG. 12 is an alternative to FIG.
Corresponds to the one obtained by replacing step 743 in step with step 943 and adding step 947. Step 94
In FIG. 3, the FIF of the stored learning time T _GB as shown in FIG.
Latest learning time T in O (First In First Out) buffer
_G is saved, and the data of the oldest learning time in the FIFO buffer is overwritten and erased. Of course, the number of memories in the FIFO buffer corresponds to the number of learning times used for obtaining the average learning time. Then, in step 947, all the learning time data in the FIFO buffer are cleared and discarded.

FIG. 14 is an alternative to FIG.
Is equivalent to the one obtained by deleting step 757 in step, replacing step 752 with step 952, and adding steps 950 and 951 between steps 751 and 952. In step 950, it is checked whether or not the conditions for obtaining the average learning time have been met, that is, more specifically, whether or not learning has been performed for the number of times necessary for averaging the learning time. If the averaging conditions are not satisfied, the control proceeds to step 754 to make judder judgment based on the success rate of the learning. If the averaging conditions are satisfied, the average learning time T _GBM is calculated in step 951. 952, the average learning time T _GBM is equal to the set time T _GT
The lock-up clutch 2c
It is determined whether or not has caused abnormal fastening (judder).

In this embodiment, the lock-up clutch 2c is abnormally engaged (judder) based on the average learning time T _GBM obtained by averaging the past several learning times.
Is determined, the determination does not become erroneous even if the learning time varies, which is very advantageous because the reliability can be improved.

[Brief description of the drawings]

FIG. 1 is a schematic system diagram showing a drive system of a vehicle including a lock-up control device according to an embodiment of the present invention and a control system thereof.

FIG. 2 is a system diagram showing a lock-up control system of the torque converter according to the embodiment.

FIG. 3 is a diagram showing control characteristics of a lock-up clutch engagement differential pressure in the embodiment.

FIG. 4 is a flowchart showing initialization processing executed by a controller when an ignition switch is turned on in the embodiment.

FIG. 5 is a flowchart showing lockup control executed by a controller in the embodiment.

FIG. 6 is a flowchart showing a process of producing coast lockup control and lockup clutch abnormal engagement determination data executed by a controller in the embodiment.

FIG. 7 is a flowchart showing a control for interrupting and initializing a process of measuring a learning time of a slip start lock-up clutch engagement differential pressure.

FIG. 8 is a flowchart showing a process for starting the measurement of the learning time of the slip differential of the lock-up clutch engagement differential for start.

FIG. 9 is a flowchart showing a process for measuring a learning time of a slip start lock-up clutch engagement differential pressure.

FIG. 10 is a flowchart showing a process for terminating the measurement of the learning time of the engagement start differential pressure of the slip-up lock-up clutch.

FIG. 11 is a flowchart illustrating a process for determining abnormal engagement of a lock-up clutch.

FIG. 12 is a flowchart showing a control program according to another embodiment of the present invention, which is substituted for FIG. 10;

FIG. 13 is an explanatory diagram of a FIFO buffer that stores a learning time of a slip start lock-up clutch engagement differential pressure.

FIG. 14 is a flowchart of a control program instead of FIG. 11, showing another embodiment of the present invention together with FIG.

FIG. 15 illustrates the coast lock-up control of FIG. 6 when the lock-up clutch is normally engaged.
6 is an operation time chart shown together with control during rapid deceleration.

FIG. 16 is an operation time chart showing the coast lock-up control of FIG. 6 together with the control at the time of rapid deceleration when the engagement of the lock-up clutch is abnormal.

[Description of Signs] 1 Engine 2 Torque converter 2a Pump impeller (input element) 2b Turbine runner (output element) 2c Lock-up clutch 3 Gear transmission mechanism 4 Differential gear device 5 Wheel 11 Lock-up control valve 12 Controller 13 Lock-up solenoid 21 Throttle opening sensor 22 Vehicle speed sensor 23 Impeller rotation sensor 24 Turbine rotation sensor 25 Idle switch 26 Oil temperature sensor

Claims (6)

[Claims] In a coasting state of a vehicle, coast lockup is performed by setting a torque converter in a lockup state in which an input / output element is directly connected by a lockup clutch responding to a lockup clutch engagement differential pressure. In the lock-up control device for a torque converter, the lock-up state at the time of coast lock-up includes adding a predetermined differential pressure to a learning value of a slip start lock-up clutch engagement differential pressure at which a torque converter slip amount becomes a minute set slip amount. During the learning of the slip start lock-up clutch engagement differential pressure, the torque from the lock-up state until the torque converter slip amount reaches the minute set slip amount. Change in converter slip amount To determine the abnormal engagement of the lock-up clutch from the engagement, the lock-up control device for a torque converter, characterized by being configured to perform abnormality correction. 2. The method according to claim 1, wherein the torque converter slip amount is reduced from the start of the learning of the slip start lock-up clutch engagement differential pressure for changing the torque converter slip amount from 0 to the small set slip amount. A lock-up control device for a torque converter, wherein the lock-up clutch is determined to be abnormally engaged when the learning time is equal to or longer than a set time, based on a learning time until learning is completed. . 3. The torque converter slip amount according to claim 1, wherein the torque converter slip amount with respect to the cumulative number of times of starting learning of the slip start lock-up clutch engagement differential pressure for increasing the torque converter slip amount from 0 to the minute set slip amount. Based on the learning success rate represented by the ratio of the number of times of completion of learning that has reached the minute set slip amount, when the learning success rate is less than the set rate, it is determined that the lock-up clutch is abnormally engaged. A lock-up control device for a torque converter. 4. The method according to claim 1, wherein the torque converter slip amount is reduced from the start of the learning of the slip start lock-up clutch engagement differential pressure for changing the torque converter slip amount from 0 to the small set slip amount. Based on the learning time up to the end of learning to reach the amount, when this learning time is equal to or longer than the set time, it is determined that the lock-up clutch is abnormally engaged. Based on the learning success rate represented by the ratio of the number of times of integration of the torque converter, wherein when the learning success rate is less than the set rate, it is determined that the lock-up clutch is abnormally engaged. Lock-up control device. 5. The lock-up clutch according to claim 1, wherein the lock-up clutch is abnormally engaged when the average learning time obtained by averaging the learning time over the past several times is equal to or longer than a set time. A lock-up control device for a torque converter. 6. The method according to claim 1, wherein the engine speed is lower than the set speed, the vehicle speed is lower than the set vehicle speed, and the hydraulic oil temperature is out of the set temperature range. A lock-up control device for a torque converter, wherein the determination of abnormal engagement of the lock-up clutch is stopped when at least one condition is satisfied.