WO1990009517A1

WO1990009517A1 - Process for checking the performance of a lambda sensor

Info

Publication number: WO1990009517A1
Application number: PCT/DE1990/000032
Authority: WO
Inventors: Eberhard Schnaibel; Erich Schneider; Martin Stilling
Original assignee: Robert Bosch Gmbh
Priority date: 1989-02-18
Filing date: 1990-01-20
Publication date: 1990-08-23
Also published as: DE3904986A1; KR910700402A

Abstract

A process for checking the performance of a lambda sensor ascertains whether the frequency of the control oscillation in a control circuit which contains the lambda sensor as measurement element lies below a frequency threshold if at the same time the values of the control deviation vary within a predetermined narrow range. If this is the case, an error exists in the control circuit, most probably in the sensor. The process is also suitable for checking the performance of a lambda sensor used for lean control.

Description

Method for recognizing the operational readiness of a lambda probe

The invention relates to methods for recognizing the operational readiness of a lambda probe. There may be a lack of readiness for operation due to numerous deficiencies, for example there may be short circuits or interruptions in the supply lines or the probe may not be sufficiently warm, which is particularly the case shortly after starting an internal combustion engine to which the probe is attached. _4th

State of the art

Recognition of operational readiness is relatively easy if the probe is used to control a lambda value of approximately one. All methods for recognizing the operational readiness of a lambda probe, which are used for regulating to a lambda value of approximately one, have the principle in common that fixed threshold values are determined, upon reaching which readiness for operational readiness is concluded.

REPLACEMENT LEAF If the operational readiness of a lambda sensor that is used for lean control is to be monitored, it is no longer possible to conclude operational readiness or a lack of operational readiness from reaching a fixed lambda threshold value. This is because with lean regulation, load values for lean mixture are permitted within wide limits. If the measured lambda values move within the permissible range, it is difficult to determine whether this is the case due to an error.

The invention is based on the object of specifying a method for recognizing the operational readiness of a lambda probe which also works reliably when it is applied to a lambda probe which is used for lean regulation.

Advantages of the invention

This object is achieved by the method according to claims 1-4. These methods can be carried out individually or in any combination.

In the method according to claim 1, it is checked whether the control oscillation frequency exceeds a minimum value when the control deviations move within a predetermined band. If this is the case, the result of the check indicates that the entire control loop is working properly. So the probe is also ready for operation.

The method according to claim 2 is based on the fact that the probe voltage is processed in two measuring channels with different amplifiers. A check is carried out in each channel to determine whether the measured value indicates a lean or non-lean mixture. The two determined mixture areas are compared. If there is no agreement, this is a sign that there is a probe error. While the methods according to claims 1 and 2 serve to determine the lack of operational readiness of a sufficiently warm lambda probe, the methods according to claims 3 and 4 serve to determine the lack of operational readiness due to the sensor not being operationally warm.

In the process according to Ansp. 3, the probe voltage is compared with a voltage threshold value for lean and / or rich mixture and a ready signal is given when the detected probe voltage reaches one of the voltage threshold values. This sub-method is known from DE 33 19 432 A1 (US 4,528,957). Additional icfo is used in the process according to Ansp. 3, however, checks whether a predetermined period of time has elapsed since the start. If this is the case, a pseudo-ready signal is output if the correct ready signal has not already been issued by fulfilling the aforementioned condition.

In the process according to Ansp. 4, the probe is loaded with a resistance, the value of which is considerably lower than the internal resistance value of the cold probe, but considerably higher than the internal resistance value of the hot probe. The measured voltage is compared with a switch-on threshold voltage and a readiness signal is output if the comparison result is positive. This ensures that readiness for operation is only recognized when the probe is so hot that it delivers measurement results that are sufficiently accurate even for lean control.

The method according to claim 5 can be used with particular advantage in connection with one of the methods according to claims 1-4, but can also be used by itself. According to the method of claim 5, the probe internal voltage is not used directly as a probe voltage for control purposes, but rather a voltage is used which is applied to a Series connection of a load resistor is tapped with an adjustable counter voltage source, which series connection of the probe is connected in parallel. The voltage of the counter voltage source is set in each case to a voltage which corresponds to the respective lambda sol value. This procedure has the advantage that when a probe error occurs, the voltage of the counter-voltage source is used as the probe voltage. Since this assumes a value that corresponds to the respective lambda sol value, there is no control deviation, so that the manipulated value remains unchanged. Even without a method for recognizing the operational readiness of the probe, poor control results due to a probe that has become defective can thus be avoided. However, if a method for recognizing operational readiness is also used, additional measures can be taken with the result of this method.

drawing

1 shows a schematic circuit diagram of a circuit for detecting the probe voltage in two measuring channels;

2 shows a flowchart for explaining a method for detecting errors in a lambda probe that is at operating temperature;

3 shows a flow chart for explaining a method for recognizing the operational readiness of an initially cold lambda probe;

4 shows a schematic circuit diagram of an arrangement with which the operational readiness of a lambda sensor can be recognized; and

5 shows a schematic circuit diagram of an arrangement for measuring a probe voltage. Description of exemplary embodiments

The voltage U_S emitted by a lambda probe 10 is processed in two channels by the circuit according to FIG. 1. The probe voltage star changes at load values by one, specifically from about 200 mV to about 800 mV, with an average value of about 500 mV, which applies to exactly lambda = 1. If, on the other hand, measurement is carried out in the exhaust gas generated by a lean mixture, the voltage fluctuates only slightly in a wide lambda range, eg. B. from about 20 - 50 mV of lambda = 1.4 - 1.1.

In order to raise the low voltage values just mentioned to a level suitable for further processing, a first amplifier 11.1 with a high amplification factor V, in the example V = 75, is present. A voltage ULM is tapped at its output, ie a voltage for lambda values for a lean mixture. For probe voltages ff_S of 200 - 800 mV, this amplifier would deliver voltages that would be far too high for further processing. There is therefore a second amplifier 11.2 with a low gain factor in the example V = 5. A voltage UL1 is tapped at its output, ie a voltage for Lamb values for mixtures of Lambda approximately 1. The second amplifier 11.2 is only directly connected to the probe by a line, which is why the voltage ULI measured by it has a mass offset Um . Since the order of magnitude of the probe voltage that is measured when the mixture is lean is moving, the probe voltage is guided to amplifier 11.1 in two poles. This difference in wiring allows Widersp before determine between measured voltages as _t ste occur particularly in ground fault or short circuit.

Is there a mixture with a lambda value of z. B. 1.1 before, the voltage ULM is in its upper evaluable range, while the voltage UL1 is still below a threshold value, from which it is used for regulation. If the voltage UL1 reaches its usable range, the voltage ULM has left its corresponding range, ahead set there is no probe error. If the voltage UL1 is above a threshold value and the voltage ULM is below a threshold value, these are contradictory results.

Such contradictions can be exploited in a method for recognizing the operational readiness of a lambda sensor. Corresponding steps are included in the method according to FIG. 2, which, however, begins with another investigation.

In the method according to FIG. 2, a method step s1 is reached after starting the method using a mark A. In this it is checked whether the amplitude AMP_RAW of the control deviation remains below an amplitude threshold AMP_RAW_S and at the same time the frequency f of the control oscillation remains below a frequency threshold f_S. If this is the case, an error is displayed in step s2 and the system is switched from regulation to control.

If, on the other hand, the examination in step s1 is negative, an examination follows in a step s3, which was already explained above with reference to FIG. 1. It is namely checked whether the voltage UL1 remains above a lower stomach threshold M_SCHW-U, although the voltage ULM is less than a lean threshold M_SCHW. If this is the case, an error is not necessarily concluded, but it is additionally checked in a step s4 whether the voltage U remains below an upper lean threshold value M_SCHW-0 for more than the duration of a period Z1. This value is z. B. 100 mV, while the lower lean threshold M_SCHW-U is 60 mV. If the condition according to step s4 is also fulfilled, the final step s2 already explained follows. Otherwise, the process returns to mark A.

If the condition in step s3 is not met, the method proceeds to step s5. An extremely lean mixture occurs in overrun mode. If overrun is in front of the probe but does not show an extremely lean mixture, this is a sign of an error. Accordingly, e an error if there is no coasting operation, but the probe still emits a signal for extremely lean mixture. The examination for the first fault takes place in a step s6, which follows step s5 if overrun is present. In step sβ it is examined whether the voltage ULM lies above the lean threshold value M_SCHW for more than a period of time Z2. If this is the case, the final step s2 discussed follows. If this is not the case, the process sequence starts again from the mark A. The second error check, briefly explained in connection with step s5, is carried out in a step s7 which follows the step s5 if there is no overrun operation In step s7 it is examined whether the voltage ULM is below the lean threshold M_SCHW for more than a period Z3. If this is the case, step s4 already explained above follows. If this is not the case, the process starts again from mark A.

The periods Z1, Z2 and Z3 mentioned have a duration v of a few seconds, z. B. of two seconds. The value of the lean threshold M_SCHW is z. B. 5 mV x 75 = 375 At this very low threshold, care must be taken that the probe voltage U_S is not falsified by external voltages. Various methods are available to ensure this, but these are not dealt with in detail here.

In connection with step s1, it is pointed out that, instead of the amplitude of the control deviation, the total stroke of the control deviation can also be monitored and who can be checked whether it lies within a predetermined band should move the specified band, especially the probe voltage. The frequency threshold value strongly depends on dead times, which in the respective application apply. Possible period duration threshold values therefore fluctuate within a wide millisecond range. In practice, however, it is easy to determine which is the minimum frequency at which the control method still works properly. The frequency threshold is set somewhat below this minimum frequency.

The method according to FIG. 3 serves not to switch to control if this is not yet permissible due to the probe temperature. In a step n1 it is checked whether the probe voltage U_S has fallen below a threshold value M_SCHW for a lean mixture or has exceeded a threshold value F_SCHW for a rich mixture. If this is the case, readiness to rule exists. It is then decided in a step n2 whether control should be made on a lean mixture or a mixture with lambda = 1. As a condition for this, it is examined in step n2 whether the engine temperature T_M0T has risen above a temperature threshold value T_M0T_S. If this is the case, regulation to a lean mixture takes place in a step n3. If this is not the case, however, control to lambda = 1 takes place in a step n4.

If the examination in step n1 is negative, step n5 follows, in which a check is made to determine whether a period of time Z_S (T_M0T ) has expired. If this is not the case, ie if none of the conditions according to steps n1 and n5 is met, the method returns to the beginning, ie to step n-1. If, on the other hand, the condition according to step n5 is fulfilled, step n2 already explained follows with one of its subsequent steps.

Another method for examining whether a probe is already sufficiently warm for control operation is now using of the circuit shown in FIG. 4. The circuit has the lambda probe 10 with a load resistor 12 of the resistance value R_L connected in parallel. The probe voltage U_S is tapped at this. The lambda probe 10 is shown with its equivalent circuit diagram, which consists of a probe voltage source 13, with the internal voltage U_SE, and a probe internal resistor 14, with the internal resistance value R_SI. With the values mentioned, the probe voltage U_S results as follows:

U_S = U_SE x R_L / (R_SI + R_L)

This probe voltage is compared in a comparator 15 with a standby threshold voltage U_B, as is output by a reference voltage source 16. As soon as the probe voltage U_S exceeds the standby threshold voltage U, readiness for control is recognized. The standby threshold voltage is set in such a way that control is only released when the probe is so hot that the probe voltage U_S does not change significantly with a constant lambda value if it is heated further.

All methods for recognizing the operational readiness of a probe basically have the disadvantage that a certain period of time inevitably elapses between the occurrence of an error and the detection of the same. During this time, the control deviation changes incorrectly, which also results in an incorrect change in the control manipulated variable. Unwanted changes of larger dimensions can be avoided with a method, as will now be explained on the basis of the arrangement in FIG. 5. This method is advantageously used with one of the methods explained above, but can also be used independently of these.

The lambda probe 10 is also shown with its equivalent circuit diagram in the circuit of FIG. 5. Parallel to the probe there is now a series connection of a load resistor 12, with the resistance value R_L, and a counter voltage source 17, with the adjustable voltage U_G. The probe voltage U_S tapped at the series circuit is determined as follows:

U_S = U_G * + (U_SE - U_G) x R_L / (R_L + R_SI)

If the probe internal resistance value R_SI suddenly increases, be it due to a strong cooling of the probe or by interrupting the probe lines, the probe voltage adjusts to the counter-voltage value U_G. This value is set to a voltage that corresponds to the respective lambda solver. If the probe lines are interrupted, there is no longer any control deviation. In the case of a probe that is no longer operational due to the temperature being too low, there is only a slight control deviation. Accordingly, the control manipulated value no longer changes or only changes very slowly. This missing or, in the worst case, very slow change in the manipulated value means that there are no undesirably large changes in the control within the period of time that elapses between the occurrence of an error and its detection.

Claims

Expectations

1. Method for recognizing the operational readiness of a lambda probe, that is to say that the. The following steps are carried out, especially for lean control:

- detection of the extent of the oscillation of the control deviation,

Comparing the degree of vibration with a low amplitude threshold value,

- detection of the frequency of the control oscillation,

- Comparing the determined frequency with a frequency threshold, and

- Outputting an error signal if the determined extent of vibration remains below the amplitude threshold and at the same time the determined frequency remains below the frequency threshold.

2. Procedure for recognizing the operational readiness of a

I.

Lambda probe, in particular the method according to claim 1, characterized in that the following steps are carried out to identify the operational readiness for lean control: Detecting the slightly amplified probe voltage and determining whether the detected value indicates lean or non-lean mixture,

Detecting the strongly amplified probe voltage and determining whether the detected value indicates lean or non-lean mixture,

- Compare the two determined mixture ranges and

- Outputting an error signal if the determined mixing areas are contradictory.

3. A method for recognizing the operational readiness of a lambda probe, in particular a method according to one of claims 1 or 2, that a d u r c h g e k e n n z e i c h n et that the following steps are carried out to recognize the operational readiness for lean control:

- detection of the probe voltage,

Comparing this voltage with a voltage threshold value for lean or rich mixture,

Determining the period of time which has elapsed since a start time,

Comparing this time period with a time period threshold value and

- Issue a ready signal when

either the detected voltage reaches one of the voltage threshold values,

- or the determined time period exceeds the time period threshold value.

4. A method for recognizing the operational readiness of a lambda probe, in particular a method according to one of claims 1 to 3, characterized in that the following steps are carried out for identifying the operational readiness for lean control: - 15 -

Tapping a measuring voltage at a load resistor at the probe, the value of which is considerably less than the internal resistance value of the cold probe, but is considerably higher than the internal resistance value of the hot probe,

- Compare the measuring voltage with a standby threshold voltage and

- Output of a ready signal when the comparison result is positive.

5. The method according to any one of claims 1 - * ^, characterized in that a voltage is used as the probe voltage, which is tapped at a series connection of a load resistor with an adjustable counter voltage, which series connection of the probe is connected in parallel, the voltage of the counter voltage source is set to a voltage that corresponds to the respective lambda setpoint.