DE102007027183A1 - Method for operating gas sensor, for determining concentration of gas components in gas mixture, involves exposing pump electrode to exhaust gas over diffusion barrier - Google Patents

Abstract

The method involves exposing a pump electrode (15) to the exhaust gas over a diffusion barrier (12). The temperature dependency of a variable correcting variable of a pump control circuit, guided by the pump electrode, is determined. The static pressure dependency of the gas sensor (1) is concluded. Independent claims are also included for the following: (1) a circuit arrangement for operating a broadband lambda probe, for determining the gas components in a gas mixture (2) a computer program (3) a computer program product with a program code.

Description

The The invention relates to a method for operating a gas sensor according to the preamble of claim 1 and a circuit arrangement for operating a gas sensor according to the genre of the independent Claim 5.

object The invention also includes a computer program and a computer program product with a program code stored on a machine-readable carrier is stored for carrying out the method.

State of the art

electrochemical Gas sensors in the form of lambda probes are in large numbers used in exhaust systems of internal combustion engines in motor vehicles, for the engine control signals about the exhaust gas composition to be able to provide. That way, the engine can be operated so that the exhaust gases have an optimal composition for after-treatment with in the exhaust system today usually have existing catalysts.

Such a gas sensor, also referred to as a broadband probe, for example, is derived from US Pat DE 101 63 912 A1 out.

The Diffusion barrier for such planar broadband lambda probes is made of zirconium oxide with the addition of a pore-forming agent. Due to an increased yttrium oxide content and a lower Silicon content compared with the adjacent, the electrolyte forming zirconium oxide dampens a reduced sintering activity closing the through the pore former after burning out left behind pores, with silicon during sintering too a liquid grain boundary phase leads to the Sintering accelerates, but yttria increases the sintering activity lowers.

Of the Oxygen transport in the diffusion barrier takes place through the Knudsen diffusion and gas-phase diffusion. The limiting current of a probe with predominantly oxygen transport via gas phase diffusion is pressure independent, because the pressure dependence diffusion, according to the mean free path and thus decrease the diffusion in proportion to the reciprocal of the pressure, is compensated by the higher partial pressure of oxygen. For diffusion barriers with smaller pores, the mean free path length determined by the pore radius and is therefore constant. In this so-called Knudsen diffusion is the oxygen transport thus altogether pressure-dependent, because here the concentration of the oxygen increases without at the same time the diffusion constant decreases and thus the limiting current increases with the pressure.

In reverse as the pressure dependence behaves the temperature dependence of the limiting current. For probes with a predominant oxygen transport via a gas phase diffusion takes the limiting current with the temperature too, while the limiting current in probes with an oxygen transport, predominantly over the Knudsen diffusion takes place, does not depend on the temperature. Therefore, a correlation of the two coefficients can be found determine. Probes with a high pressure dependence of the Limit currents have a low temperature dependence of the limiting current and vice versa.

From the not pre-published DE 10 2006 011 837.5 For example, a method for determining a gas concentration in a measurement gas has become known according to which the static pressure dependence is determined from averaging over a multiplicity of value pairs from gas concentration signals and pressure signals, the pressure being calculated by means of an exhaust gas pressure model. In this case, a so-called compensation parameter of the gas sensor is determined, which in turn is used to determine the gas concentration.

A Signal change due to a pressure variation in the thrust of a for example, the vehicle having the internal combustion engine 200 mbar leads to changes in the pumping current only 10%, so to a relative small change. Besides, you have to be very Many measurements are used because, first, the limiting current with the dilution of the air to be measured due to the variation of the water vapor content of the air changes because second, interference in the transmitter to a variation lead the voltage raw signal for the limiting current and, thirdly, the pressure determined on the basis of a model due to inaccuracies of the print model for various vehicle operating conditions more or less error-prone.

alternative to measure the pressure dependence in the vehicle you can They also firmly specify and, based on this uniform value, the Perform pressure compensation. But there are production-related Variations in the static pressure dependence in such Sensors on. A scatter of static pressure dependence from probe to probe and a temporal change of static Pressure dependence over the lifetime lead to a scattering of the useful signal, so for example to be determined Lambda value, despite pressure compensation.

Disclosure of the invention

Advantages of the invention

The inventive method for operating a Gas sensor with the features of claim 1 and the inventive Circuit arrangement for operating a gas sensor with the features of claim 5 on the other hand allow a Significant reduction in the dispersion of the limiting current signal after one Pressure compensation due to manufacturing variations of the static Pressure dependence in new condition as well as due to aging the static pressure dependence over the lifetime. The determination of the static pressure dependence can over it extended beyond the entire lambda range of the pumping cell with possibly the probe being operated only passively, d. H. the output signal is not evaluated.

Purely In principle, it is also possible in the determination the static pressure dependence used pressure value to compare with the specific pressure due to a print model, whereby a plausibility of the temperature and pressure dependence is possible.

One Another big advantage is that in this process, which aimed at the temperature dependence of the pumping current, accurate determination of the temperature dependence after only a few measured values is possible because the temperature dependence of Pumping current is independent of the water content of the air and because measurements in quick succession have the same water content exhibit.

Brief description of the drawings

embodiments and embodiments of the invention are taken in conjunction with the drawings and the description below.

In show the drawing:

1 a schematic sectional view of a broadband lambda probe;

2 schematically a circuit arrangement for operating the in 1 illustrated broadband lambda probe;

3 two diffusion barriers with different static pressure dependence;

4 the temperature and the pumping current over time at an in 1 presented broadband lambda probe and

5 schematically the temperature dependence a above the value of the static pressure dependence k.

embodiments the invention

An example of an in 1 shown gas sensor 1 , a so-called broadband lambda probe for determining the oxygen content in an exhaust gas of an internal combustion engine (not shown), is a ceramic laminate of individual layers 2 . 5 . 7 . 9 . 14 educated. An electric heater 4 and the associated conductor tracks for the electrical power supply are in an insulating layer 3 embedded. In the shift 5 is a reference air channel 6 provided in which a reference electrode 8th , For example, is arranged from porous platinum material. The reference electrode 8th is connected via a subsequent layer-shaped conductor to a terminal contact of the gas sensor (not shown).

Above the solid electrolyte layer 7 is a structured solid electrolyte layer 9 arranged with a large recess, which centric to a the body of the gas sensor 1 perpendicular to its layers passing through the exhaust gas inlet hole 10 is arranged. Within the recess is released one of the access hole 10 concentric annulus 11 porous material as a diffusion barrier 12 arranged opposite a diffusing measuring gas. The access hole 10 can, as shown, as a blind hole or deviating from the representation as a the body of the gas sensor 1 completely penetrating opening may be formed. In the area of the annulus 11 , which is also referred to as a diffusion chamber, carries the solid electrolyte layer 7 a preferably gas permeable layered Nernst electrode 3 made of porous platinum material.

Over the shift 9 or the porous material 12 is another solid electrolyte layer 14 on their, the annulus and thus the diffusion chamber 11 facing side as well as on their the annulus 11 opposite side each preferably for gas permeable inner and outer pumping electrodes 15 . 16 has at least partially porous platinum material, wherein the electrodes 15 . 16 are shaped so that they are in plan view of the layers of the body of the gas sensor 1 the annulus 11 at least essentially cover. Over the shift 14 is still a gas-permeable protective layer 17 , All solid electrolyte layers 2 . 3 . 5 . 7 . 9 . 14 are for example in the form of films of zirconium dioxide, the yttria is added, made.

It should be noted that the invention is not restricted to a broadband lambda probe shown here, but to all probes which measure the air number over a lambda value of 1 addition, can be used. Thus, for example, instead of the broadband lambda probe shown, it is also possible to provide a lean-type probe, which differs structurally from the one shown in FIG 1 differs characterized in that a reference air channel and a reference electrode arranged therein has. It is also possible to use probes in which the electrodes are arranged on the outside of the solid electrolyte and are covered by corresponding layers, which then form the diffusion barrier.

The broadband probe described above works as follows:
That the access hole 10 having end of the body is disposed in the exhaust stream or in a communicating with the exhaust stream of an internal combustion engine or a heater area, while the other end of the body with a reference gas, usually air, is applied.

About the reference air channel 6 , which at the aforementioned other end of the body 1 has an accessible for the reference gas opening, the reference gas reaches into the in 1 visible end piece of the reference gas channel. About the access hole 10 Exhaust gas reaches the porous material 12 through which the exhaust gas enters the annulus 11 diffuses, which accordingly forms a diffusion chamber.

If the exhaust side of the body 1 by means of electrical resistance heating 4 is heated, can between the reference electrode 8th and the Nernst electrode 13 an electrical voltage to be tapped whose measure of the oxygen partial pressures within the tail of the reference gas channel 6 or within the annulus 11 depends. Here, the effect is exploited that the solid electrolyte layers 2 . 3 . 5 . 7 . 9 . 14 Lead oxygen ions and that the platinum material of the aforementioned electrodes 8th . 13 catalyzes the formation of these oxygen ions. This occurs at the electrodes 8th . 13 a dependent on the respective oxygen partial pressures potential difference, which is also referred to as Nernstspannung. The oxygen partial pressure in the diffusion chamber 11 can be controlled by applying an external voltage with controllable polarity. The corresponding voltage source is connected to non-illustrated contacts with the pumping electrodes 15 . 16 are electrically connected, for example, in the laminate of the body of the gas sensor 1 integrated tracks.

The platinum material of the electrodes 15 . 16 catalyzes the equilibrium reaction of oxygen ions to molecular oxygen, due to the external electrical voltage between the electrodes 15 . 16 an oxygen ion current is generated with voltage and polarity dependent strength and direction. The strength of the pumping current between the pumping electrodes 15 . 16 is tapped as an electrical signal. For example, by detecting voltage and current, the electrical resistance of the circuit passing through the pumping electrodes can be determined. The pumping voltage and thus also between the pumping electrodes 15 . 16 flowing pumping current is controlled by means of a regulator so that the between the reference electrode 8th and the Nernst electrode 13 tapped Nernst voltage always corresponds to a specified setpoint. This is between the pumping electrodes 15 . 16 tapped electric current is a measure of the oxygen content of the exhaust gases relative to the reference gas.

An in 2 illustrated circuit arrangement 20 is via appropriate connections with the reference electrode 8th and the pumping electrodes 15 . 16 connected, wherein the inner pumping electrode 15 within the broadband probe with the Nernst electrode 13 is electrically connected, so that the inner pumping electrode 15 leading connection of the circuit arrangement 20 also an electrical connection with the Nernst electrode 13 having.

As mentioned above, during operation of the broadband probe, the electrical voltage between the Nernst electrode 13 or the associated inner pumping electrode 15 and the reference electrode 8th recorded and compared with a predetermined target voltage. Depending on the nominal-actual value deviation between these two voltages, a current source becomes part of the circuit arrangement 20 is and in one the pumping electrodes 15 . 16 contained pump circuit is electrically arranged, so controlled that the pumping current I _P , that of the outer pumping electrode 16 is supplied, increased or decreased. An output signal generated by the pumping power source is output again as a digital signal after further processing. This signal represents the measured value for lambda.

The diffusion barrier 12 for such a broadband lambda probe is now typically made of zirconia with the addition of a pore-forming agent. Due to the increased yttrium oxide content and the lower silicon content compared to the adjacent area, in particular the adjacent base film 7 and 14 From zirconia, a reduced sintering activity vaporizes the pores left behind by the pore former after burnout during the manufacturing process. Silicon leads to a liquid grain boundary phase during sintering, which accelerates sintering. Yttria reduces sintering activity. In 3a is a thick diffusion barrier and in 3b a thin diffusion barrier is shown. The zirconium oxide ceramic base is denoted by reference numerals 320 designated. In the thick diffusion barrier ( 3a ) exist more strongly sintered areas 330 and a small sintered area 340 whereas in the diffusion barrier shown on the right there is only one strong sintered area 330 exist. To the left 3a Fig. 2 shows schematically the silicon content across the thickness of the diffusion barrier, which is practically zero in the lower-sintered region and different from zero in the more heavily sintered region. In the 3a shown thick diffusion barrier has a lower static pressure dependence, while the in 3b shown thin diffusion barrier has a high static pressure dependence.

Of the Oxygen transport in the diffusion barrier occurs both due to the so-called Knudsen diffusion as well as due to the gas phase diffusion.

The limiting current of a probe with a predominant oxygen transport via gas-phase diffusion is pressure-independent, since the pressure-dependence of the diffusion (the mean free path and thus the diffusion constant decreases in proportion to 1 / p) is compensated by the higher partial pressure of the oxygen. For diffusion barriers with smaller pores, the mean free path is determined by the pore radius and is therefore constant. In this case, a Knudsen diffusion, oxygen transport is pressure-dependent, as the concentration of oxygen increases without the diffusion constant decreasing at the same time. Thus, the limiting current increases with the pressure. Such diffusion barriers with smaller pores (in 3b shown) thus show a high static pressure dependence. In real systems, both transport mechanisms are in different forms side by side.

Conversely, it is the temperature dependence of the limiting current. For probes with predominant oxygen transport via gas-phase diffusion, the limiting current increases with temperature, while the limiting current for probes with predominant oxygen transport through Knudsen diffusion does not depend on the temperature. It is therefore possible to specify a correlation of the two coefficients. Probes with a high pressure dependence of the limiting current have a low temperature dependence of the limiting current and vice versa. This relationship is schematic in 5 represented where the temperature dependence a on the pressure dependence of the limiting current k by a straight line 510 is shown with negative slope.

The not pre-published DE 10 2006 011 837.5 describes a method for determining the gas concentration in a measurement gas with a gas sensor, wherein in the presence of a first operating mode of the internal combustion engine, in which the gas concentration in the measurement gas is known, a gas concentration signal and a pressure signal are detected. The gas concentration signal may, for example, be the limiting flow in the thrust, and the pressure signal may be determined, for example, on the basis of an exhaust gas pressure model. Based on these signals, a compensation parameter k of the gas sensor is determined. This compensation parameter k is taken into account in other operating modes of the internal combustion engine for determining the gas concentration. As a result, the determination of the static pressure dependence is possible.

at The production of such broadband lambda probes now fluctuates the static pressure dependence on. In the compensation of fluctuating pressure in the exhaust tract on the basis of static Pressure dependence in the vehicle can be seen from the state of Technique known method now given only a fixed specified pressure dependence become. A scattering (eg manufacturing dispersion) of the static Pressure dependence and its temporal change over However, the life of the probe lead to an additional Scattering of the useful signal despite consideration of the pressure compensation.

From the not pre-published DE 10 2006 011 837.5 For example, a method for determining a gas concentration in a measurement gas has become known according to which the static pressure dependence is determined from averaging over a multiplicity of value pairs from gas concentration signals and pressure signals, the pressure being calculated by means of an exhaust gas pressure model. In this case, a so-called compensation parameter of the gas sensor is determined, which in turn is used to determine the gas concentration.

The determination of the pressure dependence, as it results from this method, based on signal changes due to the pressure variation in the overrun operation of a vehicle having the internal combustion engine of typically 200 mbar. This leads to changes in the pumping current I _p on the order of only 10%. In addition, many measurements are required because the limiting current changes with the dilution of the atmospheric oxygen to be measured due to the variation of the water vapor content of the air. Moreover, because disturbances of the evaluation electronics lead to a variation of the voltage raw signal for the limiting current and because the modeled pressure is more or less erroneous due to inaccuracies of the print model for different vehicle operating conditions or because the measurement signal output from a pressure sensor is more or less erroneous, the pressure compensation is faulty.

In order to improve the pressure compensation during operation of the vehicle, the static temperature dependency in driving operation is initially determined by operation of the electric heating device 4 at various temperatures, which are measured, for example, by means of the internal resistance of the Nernst cell, determined and concluded therefrom on the static pressure dependence. This happens as in 4 represented, in that, for example, two temperatures by means of the electric heater 4 be set, a first lower temperature T ₁ z. B. 600 ° C, and a second higher temperature T ₂ z. B. 1000 ° C, as shown by the curve 410 , which represents the temperature T over time, schematically in FIG 4 is shown.

The thereby adjusting pumping current I _p , in 4 schematically by the dotted line 420 is detected, and it is a temperature dependence a z. B. determined by the following equation: a = ((I P2 - I P1 ) / I P1 ) / (T 2 - T 1 ).

Because of this temperature dependence a can now be determined by the previously established in the laboratory generally valid relationship between the temperature dependence a and the static pressure dependence k, in 5 by the straight line 510 represented and for example in a characteristic in the circuit arrangement 20 is deposited, be closed in the vehicle during operation on the static pressure dependence k. As a result, the dispersion of the limit current signal after performing the pressure compensation due to manufacturing variations of the static pressure dependence in the new state as well as due to aging of the static pressure dependence over the lifetime of the probe can be reduced.

The Measurement is preferred over the entire lambda range extended the pumping cell, the probe may be passive only is operated, so the output signal is not evaluated.

The above-described determination of the static pressure dependence can also be used in addition to the direct determination of the static pressure dependence from the correlation of the pumping current I _p with an external pressure model, in which case the temperature and the pressure dependence of the limiting current can be made mutually plausible.

From the not pre-published DE 10 2005 056 152.7 Reference is made to a method for calibrating the signal provided by a broadband lambda sensor disposed in the exhaust region of an internal combustion engine wherein the determination of a correction value is made during a predetermined operating condition of the internal combustion engine in which no fuel metering occurs and the engine speed above a threshold. It is provided that the correction value is determined as a function of the temperature of the broadband lambda sensor during the predetermined operating state of the internal combustion engine.

By The method described here can now measure the temperature dependence the pumping current and thus the signal of the gas sensor are improved, because the temperature dependence of the pumping current is independent from the water content of the air. It can therefore be in quick succession lying measurements are carried out and from this one accurate determination of the static temperature dependence the probe are made, which is based on the idea that measurements one after the other have the same water content Have air.

It should be mentioned that the in 4 shown measurements at preferably 600 ° C and 1000 ° C at a known oxygen concentration, ie, for example, in overrun or coasting of the vehicle having the internal combustion engine, by a signal 23 the circuit arrangement 20 is signaled to be made to determine the static temperature dependence. This is preferably controlled by the circuit arrangement 20 with the heater 4 heated in a phase of low mass flow. The expected limit current change of 4% per 100 ° C, ie 16% at 400 ° C temperature difference is sufficient to determine the temperature dependence with very good accuracy. Due to the stored characteristic curve of the functional relationship between the temperature dependence a and the value of the static pressure dependence k, a very precise determination of the static pressure dependence k can take place in this way.

The The method described above can be used, for example, as a computer program on a computing device, in particular a control device of a Internal combustion engine to be implemented and run there. The program code can be stored on a machine-readable carrier, which the controller can read.

The circuit arrangement 20 is also part of the controller or represents the controller itself. In this case, it is assumed that this controller 20 is programmable in a certain range, so that the method described above is implemented can and so far on the controller 20 expires.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - DE 10163912 A1 [0004]
• - DE 102006011837 [0008, 0035, 0037]
• - DE 102005056152 [0044]

Claims (10)

Method for operating a gas sensor, for determining the concentration of gas components in a gas mixture, in particular the concentration of gas components in the exhaust gas of internal combustion engines, with two pumping electrodes forming a pump cell together with a solid electrolyte ( 15 . 16 ), whose one pumping electrode ( 15 ) via a diffusion barrier ( 12 ) is exposed to the exhaust gas, characterized in that one determines the temperature dependence of a variable manipulated variable of an electrode leading via the pumping electrode pumping circuit and from this concludes the static pressure dependence of the gas sensor. A method according to claim 1, characterized in that the variable for the regulation of the pumping circuit variable manipulated variable of the pumping current (I _P ). A method according to claim 1 or 2, characterized in that the gas sensor by means of a heating device ( 4 ) is heated to at least two different temperatures (T ₁ , T ₂ ), the pump currents (I _p1 , I _p2 ) are detected at these temperatures, and a variable (a) characterizing the temperature dependence is determined according to the following equation: a = ((I P2 -I P1 ) / I P1 ) / (T2 - T 1 ), T. 1 <T 2 , Method according to one of the preceding claims, characterized in that the functional relationship between the variable (k) characterizing the static pressure dependence and the temperature dependence (a) is determined empirically and in a characteristic curve ( 510 ) stores. Circuit arrangement for operating a broadband lambda probe ( 1 ) for determining the gas components in a gas mixture, in particular the concentration of gas components in the exhaust gas of internal combustion engines, with two together with a solid electrolyte forming a pump cell pumping electrodes ( 15 . 16 ), a pump electrode ( 15 ) via a diffusion barrier ( 12 ) is exposed to the exhaust gas, characterized by a circuit device which determines the temperature dependence (a) of the variable manipulated variable of the pumping cell and determines therefrom a value representing the static pressure dependence (k). Circuit arrangement according to Claim 5, characterized that representing the static pressure dependence Value (k) by means of a characteristic which determines the relationship between the temperature dependence (a) and the value of the static Pressure dependence (k) represents, is determined. Circuit arrangement according to claim 5 or 6, characterized in that the variable manipulated variable of the pumping current (I _p ). Circuit arrangement according to one of claims 5 to 7, characterized in that the probe heater ( 4 ) is controlled so that the gas sensor can be heated to a desired specific temperature. Computer program that shows all the steps of a procedure according to one of claims 1 to 4 executes when it on a computing device, in particular the control unit an internal combustion engine runs. Computer program product with program code based on a machine-readable carrier is stored for execution of the method according to one of claims 1 to 4, when the Program executed on a computer or a control unit becomes.