EP1437501A1

EP1437501A1 - Lambda sensor diagnosis

Info

Publication number: EP1437501A1
Application number: EP20030000494
Authority: EP
Inventors: Mats Karlflo
Original assignee: Ford Global Technologies LLC
Current assignee: Ford Global Technologies LLC
Priority date: 2003-01-13
Filing date: 2003-01-13
Publication date: 2004-07-14
Anticipated expiration: 2023-01-13
Also published as: DE60312298D1; EP1437501B1; DE60312298T2

Abstract

Method of checking for malfunction in a lambdasensor arranged downstream of a catalyst in a vehicle, the method comprising the steps of causing the catalyst to become oxygen loaded or oxygen unloaded, measuring a first time delay until the oxygen loaded or oxygen unloaded condition is sensed by the lambdasond, causing the catalyst to become oxygen unloaded or oxygen loaded, measuring a second time delay until the oxygen unloaded or oxygen unloaded condition is sensed by the lambdasensor, and comparing the conditions and the measured time delays to verify the state of the lambdasensor.

Description

TECHNICAL AREA

The present invention relates to a method of checking a lambdasond arranged behind a catalyst in a vehicle.

BACKGROUND OF THE INVENTION

The development of close coupled catalyst systems (CCC), i.e. catalysts coupled closer to the exhaust manifold, and consequently closer to the engine in conventional vehicles, has lead to that both lambdasonds and catalysts have aged due to high temperatures and thereby deteriorated in a higher extent.
Another reason for deteriorated lambdasonds probably depends on the small start up monolith in the catalyst, leading to a too small catalyst surface to catch up the toxic exhausts before hitting the rear lambdasond.
Particularly, it has recently been apparent that it can be difficult to separate the ageing of catalysts to the ageing of the rear lambdasonds, which will be described in greater detail in the following.
Conventional vehicles with internal combustion engines (ICE) are provided with an exhaust gas filter system, generally comprising a front lambdasond, a catalyst and a rear lambdasond, which are arranged in connection to the exhaust manifold and the exhaust system. Usually, the system also comprises an engine control unit (ECU) or the like, controlling the driveability, emissions, diagnoses, etc. The expression "lambdasond" is synonymous with lambdasensor or oxygensensor, i.e. a sensor, sensing e.g. the oxygen pressure or oxygen concentration of air, exhausts or other gaseous mediums.
In most vehicles the front lambdasond is presently arranged into or in front of the catalyst. The front lambdasond measures the oxygen concentration of the exhausts, which exhausts reaches the front lambdasond and subsequently enters the catalyst. Since the λ-values of the exhausts varies considerably during the different operations of the ICE a sufficient excess or deficit of oxygen is required for optimised exhaust cleaning. Thus, the catalyst is required for storing oxygen and providing the deficit of oxygen for minimising the emissions.
The catalyst is generally composed of two or three bricks, which are manufactured of ceramics and various metals. The catalyst contains a large number of channels, which are coated with a very thin layer of so-called noble metals, e.g. platinum and rhodium. When the catalyst is sufficiently heated, the area of the coated channels reacts with the harmful matter of the through-flowing exhausts. The catalyst can be loaded with oxygen or unloaded of oxygen in connection with said catalytic reactions. A catalyst being completely loaded with oxygen is "set oxidised", and in opposite, a catalyst being completely unloaded of oxygen is "set reduced".
The capacity of catalysts can consequently be estimated as the oxygen storage capacity, OSC, implying the greatest amount of oxygen as can be loaded into the catalyst, and the greatest amount of oxygen which can be unloaded of the catalyst, respectively. These amounts are preferably equally great. The OSC is mainly dependent on 1) the area of said coated channels, 2) the actual working temperature of the catalyst, and 3) the ageing of the catalyst.
For exemplifying the impact of the working temperature: the catalyst works with substantially 50 % efficiency in 350 degrees Celsius, and with substantially 100 % efficiency in 450 degrees Celsius.
The catalyst mainly ages due to high working temperatures and the harmful, toxic matter in the exhausts as mentioned above. The ageing of the catalyst is temporarily or permanently. As a rule, ageing due to high temperatures is permanent but ageing due to the harmful, toxic matter in the exhausts, e.g. sulphur compounds, is temporarily since these can be burnt away in the coarse of time. Consequently, a decreased OSC often implies a deteriorated catalytic efficiency since said required excess or deficit of oxygen for the catalytic reactions are not always available.
The rear lambdasond is generally arranged between the rear bricks of the catalyst, where it measures the oxygen concentration of the partly or completely converted exhausts during the different operations of the ICE. However, the rear lambdasond can be arranged after the catalyst as well.
During combustion the actual Fuel/Air-value of the exhausts is generally referred to as the "λ-value", already mentioned above, describing how the actual Fuel/Air-ratio is related to the ideal Fuel/Air-ratio. The ideal Fuel/Air-ratio is approximately 14,7 kg air/kg fuel. Consequently, the combustion takes place under lean burn, rich burn and ideal conditions in which λ>1, λ<1 and substantially λ=1, respectively. Thus, in this case "a condition" is the F/A-ratio of the exhausts.
For clarifying, a lean burn pulse is a lean burn amount of the exhausts flowing from the engine through the exhaust manifold, the catalyst and the remaining part of the exhaust system. The expression "pulse" refers to a breakthrough of the exhausts having a λ-value not being substantially 1 (one), which is sensed by the rear lambdasond. Thus, the exhausts are not catalytically cleaned at the position of the rear lambdasond. The corresponding is of course valid for a rich burn pulse. Moreover, the lean or rich burn amount is a volume, a weight, a mass flow, a volume flow, a mole amount, etc of the exhausts.
Typical ageing effects of the rear lambdasonds are slow and asymmetrical lambdasonds. "Slow" relates to longer periods of the alternation between the lean and the rich burn conditions, which is illustrated in Fig. 1a. The slow symmetrical lambdasond can cause longer response delays, which in turn may cause a slow fuel regulation leading to a deteriorated catalytic efficiency. An asymmetric lambdasond doesn't operate at λ=1, which is illustrated in Fig. 1b. The asymmetric lambdasond generally leads to incorrect fuel-to-air-mixtures of the combustion and thereby to a deteriorated catalytic efficiency as well.
Thus, both slow symmetrical and asymmetrical lambdasonds lead to a decreased catalytic efficiency and consequently to increased emissions.
In even greater detail, the response delay is the time delay between one of said conditions is measured by the front lambdasond until the rear lambdasond senses said condition. Further, the response delay can differ between a transition from a lean burn condition to a rich burn condition, and the opposite. Then the rear lambdasond is asymmetrical as described and operates accordingly, which is seen in Fig. 1b.
During operation of the ICE, the front lambdasond measures the oxygen concentration of the exhausts flowing through the exhaust system. At the position of the front lambdasond, the λ-value of the exhausts usually differs considerably from one (1). Then the catalyst reduces or oxidises the exhausts for controlling the emissions and thereby the λ-value of the exhausts to a value being close to one (1). When the rear lambdasond measures a lean or a rich burn pulse there is not sufficient excess or deficit of oxygen for said catalytic reactions to take place. Thus, the amount of the lean or rich burn pulse is too great. However, an eventual remaining area of the catalyst can reduce or oxidise the remaining part of said exhausts. Advantageously, the λ-value of the exhausts is one (1) at the outlet of the catalyst. Thus, the catalyst works as an exhaust gas filter.
As mentioned above, it has recently been apparent that it can be difficult to separate the ageing of the catalyst to the ageing of the rear lambdasond, leading to that the malfunction indication lamp (MIL) doesn't always indicating a malfunction, or even indicates the wrong malfunction. In the present day situation, the malfunction indication lamp (MIL) merely indicates the front lambdasond diagnosis and the catalyst diagnosis. The rear lambdasond is verified by means of levels and variability and is purely "electrical", i.e. the system checks that the signal is not constant and shows reasonable voltage values.
The diagnosis of the catalyst for an oxidised condition of the catalyst is illustrated in Fig. 2. The solid square wave-formed curve shows the λ₁-value sensed by the front lambdasond. The dashed curve shows the λ₂-value sensed by the rear lambdasond. The appearance of the flattened, falling λ₂-curve depends on the oxidising, catalytic effect of the catalyst in the course of time. Thus, the response delay is clearly seen, i.e. the time delay until the rear lambdasond senses substantially the same condition as the front lambdasond. The corresponding is of course valid at a transition to a reduced condition of the catalyst, which partially can be seen as well.
In the graph, a measure of the OSC is shown for the oxidised condition of the catalyst as well. If the rear lambdasond senses a longer response delay due to its ageing as mentioned above, the system also comprehends that the catalyst obtains a greater OSC-value than the real value which is apparent in Fig. 2. Consequently, the MIL-lamp doesn't indicate an aged catalyst having a deteriorated OSC in all cases.
In WO 98/38415 an engine control system analyses the electrical signals of the upstream and downstream gas sensors, which signals are used by the engine control unit for computing a numerical value for the λ-values at both the upstream and downstream locations in the exhaust system. The cited document also presents a lean breakthrough, being sensed by the rear lambdasond, indicating that the oxygen storage capacity of the catalyst has been exceeded and defines the point where the λ-value of the downstream sensor switches from less than one (1) to greater than one (1). Correspondingly, a rich breakthrough can be detected when the λ-value switches from a value greater than one (1) to a value less than one (1). Otherwise, the cited document reveals a method for monitoring the performance of the catalyst.

SUMMARY OF THE INVENTION

It is therefore an object of the present invention to provide a method for checking the rear lambdasond so that the engine control module, ECU of conventional internal combustion engine vehicles can distinguish ageing rear lambdasonds to ageing catalysts. By means of the inventive method the ECU can also distinguish a correct rear lambdasond to a malfunctioning catalyst, or vice versa.
It is also an object of the present invention to provide a method, ensuring that the MIL-lamp always indicates a correct malfunction when necessary.
It is further an object to provide rear lambdasond diagnosis in conventional vehicles for checking an eventual malfunction of the rear lambdasond.
These objects are achieved in accordance with the present invention by the method for checking a rear lambdasond as claimed in claim 1.
This object is accomplished by means of a first data set being measured by a first sensor at the start of one of the conditions and a second data set being measured by the lambdasond arranged behind the catalyst when it senses said condition. The first and the second data sets are oxygen concentrations or oxygen pressures. Preferably, the measured first and second data sets are standardised due to different driving cases.
By means of the invention a time delay is measured between the time when one of said conditions is sensed by the first sensor and the time when said condition is sensed by the lambdasond arranged behind the catalyst. Advantageously, the first sensor is constituted by a front lambdasond.
Preferably, the time delay of the oxygen loaded condition and the time delay of the oxygen unloaded condition are compared, of which the difference is reported to an engine control module.
By means of the invention a first mole amount of oxygen for substantially oxygen loading the catalyst and a second mole amount of oxygen for substantially oxygen unloading the catalyst are calculated by the engine control module. Preferably, the mole amount of the oxygen loaded condition and the mole amount of the oxygen unloaded condition are compared, of which the lowest value is reported to the engine control module.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be described in the following by way of example only and with reference to the embodiments illustrated in the drawings, in which:

Fig. 1a: is a graph illustrating a slow lambdasond according prior art,
Fig. 1b: is a graph illustrating an asymmetrical lambdasond according prior art,
Fig. 2: is a graph exemplifying the catalytic diagnosis in the present-day vehicles according prior art,
Fig. 3: is a principal sketch showing a system operating according to the inventive method,
Fig. 4: is a block diagram illustrating the inventive method, and
Fig. 5: is a graph illustrating a delay in form of a time response, obtained by the inventive method.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

In the drawings, reference numeral 10 generally denotes a system for checking an emission component in form of a lambdasond, particularly the rear lambdasond, in accordance with the present invention.
The system 10 comprises an internal combustion engine (ICE) 13 arranged in connection to an engine control unit (ECU) 11, controlling the ICE 13, see Fig. 3. At the exhaust side, the ICE 13 comprises an exhaust manifold (not shown here) and an exhaust system 14. In greater detail, the exhaust system 14 comprises a first sensor 15, a catalyst 16 and a lambdasond 17. The first sensor is preferably constituted of a front lambdasond 15, which is arranged in front of the catalyst 16, i.e. either ahead of the catalyst 16 or in the front part of the catalyst 16 in the exhaust direction. Said lambdasond 17 is hereinafter referred to as the "rear" lambdasond 17, preferably arranged behind the catalyst 16, i.e. either in back of the catalyst 16 or in the back part of the catalyst 16 in the exhaust direction. Further, the system 10 comprises a second sensor 12, which measures the intake air amount flowing into the inlet manifold and subsequently into the combustion in the combustion chamber(s) of the ICE 13 (not shown here). Preferably, the second sensor 12 is a MAF (mass airflow)-sensor or the like, measuring the mass or the volume flow of air of the intake air amount. Naturally, the system 10 is affected by other components such as injectors, throttles etc., which are not mentioned hereinafter.
The ECU 11 controls the combustion of the ICE 13, resulting in lean burn pulses and rich burn pulses of the through-flowing exhaust flow in the exhaust system 14 during the operation of the ICE 13. In greater detail, the Fuel/Air-ratio of the lean burn pulses is lesser than one (1) and the Fuel/Air-ratio of the rich burn pulses is greater than one (1). It is emphasised that the lean and the rich burn pulses preferably occur inherently as a part of the combustion in most of the regular driving cases of a conventional vehicle. However, the lean and rich burn pulses can be particularly generated for performing the inventive method in other embodiments of the invention.
In order to perform the inventive method a lean burn pulse 20 is initially sent through the exhaust system 14, see Fig. 4. Then the front lambdasond 15 senses if the pulse is a lean burn pulse 20 or a rich burn pulse 21. If the front lambdasond 15 senses 22 a lean burn pulse 20, a signal S1 comprising such data is sent from the front lambdasond 15 to the ECU 11. More in detail, the signal S 1 preferably comprises data of the oxygen pressure or the oxygen concentration of the exhausts flowing by the front lambdasond 15 and the time when the oxygen pressure or concentration is sensed as well. However, in other embodiments the signal S 1 may comprise data of the air pressure, the air concentration and the duration of the signal S1, etc. Thus, in order to perform the inventive method the front lambdasond 15 measures a first reference point of the exhausts, which is represented by the signal S1.
The exhausts consequently flow into the catalyst 16 wherein the accompanying lean burn pulse of the exhausts is catalytically cleaned. Under normal operation of the ICE 13, the rear lambdasond 17 senses a λ-value of the exhausts being substantially one (1) when the exhausts is catalytically cleaned successfully. In a preferred embodiment of the invention, the catalyst 16 is set substantially reduced (oxygen loaded) by means of a sufficiently great lean bum pulse. Then the rear lambdasond 17 immediately senses 23 that the λ-value of the exhausts increases above substantially one (1). In greater detail, this is of course due to that the catalyst 16 cannot catalytically clean the remaining part of the lean burn pulse successfully at the position of the rear lambdasond 17.
A signal S2 corresponding to the signal S1 is sent from the rear lambdasond 17 to the ECU 11 when the lean burn pulse is sensed by the rear lambdasond 17. Thus, in order to perform the inventive method the rear lambdasond 17 measures a second reference point of the exhausts, which is represented by the signal S2.
The ECU 11 receives 24 data sets in form of the signals S 1 and S2 and the corresponding times at the position of the front 15 and the rear 17 lambdasond, respectively. For clarifying, in the preferred embodiment the lean burn pulse is firstly sensed 22 by the first lambdasond 15, and said lean burn pulse is subsequently sensed 23 by the rear lambdasond 17 with a response delay Δt1.
In a corresponding way, in order to perform the method of the invention, a rich bum pulse 21 of the exhausts is sent through the exhaust system 14. Advantageously, the front lambdasond 15 senses 22 that the pulse is a rich burn pulse 21, whereupon a signal S3 corresponding to the signal S 1 is sent from the front lambdasond 15 to the ECU 11. When the exhausts flows into the catalyst 16, the accompanying rich burn amount is catalytically cleaned. In the preferred embodiment of the invention, the catalyst 16 is set substantially oxidised (oxygen loaded) by means of a sufficiently great rich burn pulse. Accordingly, the rear lambdasond 17 senses that the λ-value of the exhaust being substantially one (1) when the exhausts is successfully cleaned catalytically. In the corresponding way as in the case of the lean burn pulse, this is of course due to that the catalyst 16 cannot catalytically clean the remaining part of the rich burn pulse successfully at the position of the rear lambdasond 17.
A signal S4 corresponding to the signal S2 is sent from the rear lambdasond 17 to the ECU 11 when the rich burn pulse is sensed by the rear lambdasond 17.
The ECU 11 also receives 24 the data sets in form of the signals S3 and S4 and the corresponding times at the position of the front 15 and the rear 17 lambdasond, respectively. Thus, in the preferred embodiment, the rich burn pulse is firstly sensed 22 by the first lambdasond 15, and said rich burn pulse is subsequently sensed 23 by the rear lambdasond 17 with a response delay Δt2.
Furthermore, in order to perform the inventive method, the ECU 11 preferably calculates 26 the mole amount of oxygen M1 for completely oxygen loading the catalyst 16. Thereby the maximum OSC for a substantially oxidised condition of the catalyst 16 is obtained. The mole amount M1 is calculated by means of the measured signals S1 and S2 constituting the oxygen pressure and concentration of the exhausts, and the corresponding times, and also an estimated air amount of the exhausts flowing through the catalyst 16. In closer detail, the MAF sensor 12 measures the intake air to the inlet manifold and thereby to the combustion of the engine, whereby the air amount of the exhausts flowing through the exhaust system 14 and consequently through the catalyst 16 is estimated by the ECU 11.
The mole amount M2 for completely unloading the catalyst 16 of oxygen is calculated 26 in the corresponding way. Accordingly, the maximum OSC for a substantially reduced condition of the catalyst 16 is obtained as well.
However, the driving cases of conventional vehicles generally varies in time, and the exhaust flow passing by the catalyst 16 varies accordingly, a.o. due to the vehicle velocities, speeds and loads, etc. of the vehicle. Consequently, the time delays Δt1 and Δt2 are preferably standardised Δt1' and Δt2' with regard to the different driving cases so that the time delays Δt1' and Δt2' are comparable with each other. Preferably, the mole amounts M1 and M2 are standardised M1' and M2'accordingly.
In order to perform the inventive method, the time delays Δt1' and Δt2' and the mole amounts M1' and M2' are compared with each other. Obviously, the time delay Δt1' of the lean burn pulse and the time delay Δt2' of the rich burn pulse should preferably show the same magnitude when applying the inventive method, i.e. the lambdasond 17 should of course work in the same way in transitions to lean burn conditions as in transitions to rich burn conditions.
In the same way, the OSC of the catalyst 16 for a lean burn condition and the OSC of the catalyst 16 for a rich burn condition should preferably show the same magnitude, i.e. it should take the same amount of oxygen for loading the catalyst 16 as for unloading the catalyst 16.
If these are not the cases, the rear lambdasond 17 is probably defect and may be aged.
If the magnitude of the time delay Δt1 'and the magnitude of the time delay Δt2' differ, the difference Δt' is reported 25 to the ECU 11, see Figs. 4-5. The difference Δt' can be used for compensating in the ECU 11. Consequently, a signal comprising the compensation may be sent from the ECU 11 back to the actual component, for example the rear lambdasond 17. Moreover, the difference Δt' can also be addressed to the rear lambdasond answer delay in the ECU 11 as a rear lambdasond ageing marker.
Accordingly, if the magnitude of the mole amount M1' and the magnitude of the mole amount M2'differ, the lowest value is chosen 26, constituting a correct value for the real OSC of the catalyst 16, see Figs. 4-5. As mentioned in the state of the art, the consequence of an aged lambdasond may be longer response delays, implying that the ECU 11 reads a larger OSC of the catalyst 16. Consequently, the lowest value of M1' and M2' is used for compensating for the aged rear lambdasond 17 in the ECU 11. A signal comprising the compensation may be sent from the ECU back to the actual component.
However, it is equally preferred to perform the inventive method by firstly sending a lean burn pulse 20 to later be followed by a rich burn pulse 21 as firstly sending a richn pulse 21 to later be followed by a leann pulse 20.
The data sets in form of the signals S1, S2, S3 and S4 is sent to the ECU 11 in the preferred embodiment of the invention. However, in other embodiments of the invention the signals S1, S2, S3 and S4 can be sent to another software (SW) or specific component. For example, the specific components can be the front lambdasond 15 or the rear lambdasond 17.
Moreover, it is preferred to combine the rear lambdasond diagnose of the inventive method with the OSC diagnose of the catalyst 16.
The method of the invention is continuously applied by the ECU 11 during the operation of the ICE for checking and diagnosing the rear lambdasond 17. Preferably, the method is particularly preferred for checking the asymmetry of aged rear lambdasonds 17.
With the expression "oxygen" is intended free oxygen ions, oxygen atoms or oxygen molecules.
The driving cases mentioned above may be driving situations such as idle, low-speed, high-speed, etc. The speed can be constant but is probably varying. For example, the driving cases are the fuel shut-off during engine braking, leading to lean burn combustion and thereby lean burn amounts in the exhausts, or high-speed driving leading to rich burn combustion for cooling the engine, and most likely rich burn amounts in the exhausts.
The invention has been described above and illustrated in the drawings by way of example only and the skilled person will recognise that various modifications may be made without departing from the scope of the invention as defined by the appended claims.

Claims

A method of checking a lambdasond (17) arranged behind a catalyst (16) in a vehicle, the method comprising the steps of:

causing the catalyst (16) to become oxygen loaded or oxygen unloaded,

measuring a first time delay until the oxygen loaded or oxygen unloaded condition is sensed by the lambdasond (17),

causing the catalyst (16) to become oxygen unloaded or oxygen loaded,

measuring a second time delay until the oxygen unloaded or oxygen loaded condition is sensed by the lambdasond (17), and

comparing the conditions and the measured time delays to verify the state of the lambdasond (17).
The method as claimed in claim 1,
characterised in that a first data set (S1; S3) is measured by a first sensor (14) at the start of one of the conditions, and a second data set (S2; S4) is measured by the lambdasond (17) arranged behind the catalyst (16) when it senses said condition.
The method as claimed in claim 2,
characterised in that the first and the second data sets (S1, S2; S3, S4) are oxygen concentrations or oxygen pressures.
The method as claimed in claim 3,
characterised in that the measured first and second data sets (S1, S2; S3, S4) are standardised.
The method as claimed in claim 1,
characterised in that a time delay is measured between the time when one of said conditions is sensed by the first sensor (14) and the time when said condition is sensed by the lambdasond (17) arranged behind the catalyst (16).
The method as claimed in claim 1,
characterised in that the catalyst (16) is caused to become oxygen loaded or oxygen unloaded by means of sending at least one lean burn amount of the exhausts wherein the Fuel/Air-ratio is lesser than one, and one rich burn amount of the exhausts wherein the Fuel/Air-ratio is greater than one, respectively, to the catalyst (16).
The method as claimed in claim 1,
characterised in that a second sensor (12) measures the air flow coming into the engine, being reported to an engine control unit (ECU) (11).
The method as claimed in claim 5,
characterised in that the time delay of the oxygen loaded condition and the time delay of the oxygen unloaded condition are compared, of which the difference is reported to an engine control module (ECU) (11).
The method as claimed in any of the preceding claims,
characterised in that a first mole amount (M1) of oxygen for substantially oxygen loading the catalyst (16) and a second mole amount (M2) of oxygen for substantially oxygen unloading the catalyst (16) are calculated by the engine control module (ECU) (11).
The method as claimed in claim 9,
characterised in that the mole amount (M1) of the oxygen loaded condition and the mole amount (M2) of the oxygen unloaded condition are compared, of which the lowest value is reported to the engine control module (ECU) (11).