U.S. patent application number 09/819229 was filed with the patent office on 2001-11-15 for method for measuring component of a gaseous emission.
Invention is credited to DeAmicis, Fabio, Patrick, Ronald S..
Application Number | 20010040104 09/819229 |
Document ID | / |
Family ID | 25416015 |
Filed Date | 2001-11-15 |
United States Patent
Application |
20010040104 |
Kind Code |
A1 |
Patrick, Ronald S. ; et
al. |
November 15, 2001 |
Method for measuring component of a gaseous emission
Abstract
A modified universal exhaust gas oxygen sensor, referred to
herein as a CEGA sensor, is provided which can be used to measure
the concentration of a variety of components of a gaseous fuel
emission including CO, CO.sub.2, O.sub.2, H.sub.2, and H.sub.2O.
The CEGA sensor-employs at least one additional electrode on a
ceramic substrate which possess a different catalytic activity
relative to the electrodes that normally found on a UEGO sensor.
The ceramic substrate may be made of any suitable ceramic and is
preferably made of zirconia. The difference in catalytic activity
between the additional electrode(s) and the electrodes native to
the UEGO sensor create an oxygen gradient which enables a measure
of combustion completeness to be calculated. In combination with an
air/fuel ratio measured by the sensor, the concentrations of
different components in the emission can be calculated. Several
methods, devices and systems which can be used with various types
of ceramic sensors including a CEGA sensor in order to improve
their performance are also provided.
Inventors: |
Patrick, Ronald S.;
(Mountain View, CA) ; DeAmicis, Fabio; (Los Altos,
CA) |
Correspondence
Address: |
WILSON SONSINI GOODRICH & ROSATI
650 PAGE MILL ROAD
PALO ALTO
CA
943041050
|
Family ID: |
25416015 |
Appl. No.: |
09/819229 |
Filed: |
March 27, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09819229 |
Mar 27, 2001 |
|
|
|
08902552 |
Jul 29, 1997 |
|
|
|
6254750 |
|
|
|
|
Current U.S.
Class: |
205/784.5 ;
204/432 |
Current CPC
Class: |
G01N 27/419
20130101 |
Class at
Publication: |
205/784.5 ;
204/432 |
International
Class: |
G01N 027/407 |
Claims
What is claimed is:
1. In a universal exhaust gas oxygen sensor (UEGO) for measuring
properties of a gaseous emission comprising at least one oxygen
pumping cell and a sensing cell in contact with a detection cavity,
the pumping cell including a ceramic in gas communication with the
detection cavity, a first electrode in contact with a first side of
the ceramic inside the detection cavity, and a second electrode in
contact with a second side of the ceramic outside the
detection-cavity, a first voltage potential externally applied
between the first and second electrodes for pumping oxygen across
the ceramic into and out of the detection cavity, the first voltage
potential being controlled by a second voltage potential across a
third and fourth electrode of the sensing cell, an air/fuel ratio
measurement of the gaseous emission being obtainable from the
current passing between the first and second electrodes, the
improvement comprising: a fifth electrode which has a different
catalytic activity than the first electrode positioned inside the
detection cavity in contact with the pumping cell ceramic, a third
voltage potential being externally applied between the fifth
electrode and either the second electrode or a sixth electrode
located on the second side of the pumping cell ceramic, the third
voltage potential controlled by a fourth voltage formed between the
first and fifth electrodes, a measure of combustion completeness
being obtainable from a difference between the first and second
voltage potentials.
2. A sensor according to claim 1 wherein the sensor includes a
sixth electrode, the third voltage potential being externally
applied between the fifth and sixth electrodes.
3. A sensor according to claim 1 wherein the ceramic is
zirconia.
4. A sensor according to claim 2 wherein the sensor includes a
sixth electrode, the third voltage potential being externally
applied between the fifth and sixth electrodes.
5. A sensor according to claim 1 wherein measurements of the
air/fuel ratio and combustion completeness enable concentrations of
CO, CO.sub.2, O.sub.2, H.sub.2, and H.sub.2O in the emission to be
determined.
6. A ceramic sensor for detecting components of a gaseous emission
comprising: a detection cavity; a diffusion passage across which
the gaseous emission enters the detection cavity; an oxygen pumping
cell defining a portion of the detection cavity formed of a ceramic
substrate and having a first electrode in contact with a first side
of the ceramic substrate within the detection cavity and a second
electrode in contact with a second side of the ceramic substrate
outside the detection cavity, an application of a first voltage
potential between the first and second electrodes causing oxygen to
be pumped into or out of the detection cavity across the ceramic
substrate to maintain an oxygen concentration in the detection
cavity at a target value; a sensing cell defining a portion of the
detection cavity formed of a ceramic substrate, the sensing cell
including a third electrode within the detection cavity, a fourth
electrode outside the detection cavity, a second voltage potential
being formed between the third and fourth electrodes due to a
difference in oxygen concentration across the third and fourth
electrodes, the second voltage potential controlling the first
voltage potential applied between the first and second electrodes,
an air/fuel ratio measurement of the gaseous emission being
obtainable from current passing between the first and second
electrodes, and a fifth electrode which has a different catalytic
activity than the first electrode positioned inside the detection
cavity in contact with the pumping cell ceramic, a third voltage
potential being externally applied between the fifth electrode and
either the second electrode or a sixth electrode located on the
second side of the pumping cell ceramic, the third voltage
potential controlled by a fourth voltage formed between the first
and fifth electrodes, a measure of combustion completeness being
obtainable from a difference between the first and second voltage
potentials.
7. A sensor according to claim 6 wherein the sensor includes a
sixth electrode, the current flowing across the fourth and sixth
electrodes.
8. A sensor according to claim 6 wherein the ceramic is
zirconia.
9. A sensor according to claim 8 wherein the sensor includes a
sixth electrode, the current flowing across the fourth and sixth
electrodes.
10. A sensor according to claim 6 wherein measurements of the
air/fuel ratio and combustion completeness enable concentrations of
CO, CO.sub.2, O.sub.2, H.sub.2, and H.sub.2O in the emission to be
determined.
11. A method for measuring concentrations of components of a
gaseous emission comprising: measuring an air/fuel ratio using a
ceramic sensor; measuring combustion completeness using the ceramic
sensor; determining concentrations of components of a gaseous
emission based on the measured air/fuel ratio and measured
combustion completeness.
12. A method according to claim 11 wherein the ceramic is
zirconia.
13. A method according to claim 11 wherein the components are
selected from the group consisting of CO, CO.sub.2, O.sub.2,
H.sub.2, and H.sub.2O.
14. A method according to claim 13 wherein the ceramic is
zirconia.
15. A method according to claim 11 wherein the air/fuel ratio is
measured based on a current between a first electrode attached to
the ceramic of the oxygen pumping cell and a second electrode
attached to the ceramic of the oxygen pumping c ell and combustion
completeness is measured based on a current between a fifth
electrode attached to the ceramic of the oxygen pumping cell and a
second or sixth electrode attached to the ceramic of the oxygen
pumping cell.
16. A method according to claim 15 wherein the ceramic is
zirconia.
17. A universal exhaust gas oxygen sensor modified to enable
simultaneous detection of an air/fuel ratio and a measure of
combustion completeness.
18. A universal exhaust gas oxygen sensor modified to enable
measurement of CO, CO.sub.2, O.sub.2, H.sub.2, and H.sub.2O
component of a combustion emission.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to sensors for use in
detecting gaseous components and more particularly to ceramic
sensors for use in analyzing combustion emission components.
BACKGROUND OF THE INVENTION
[0002] A variety of sensors have been developed for detecting
different gaseous combustion emission components. Examples of the
different gaseous components which these sensors can detect
include, but are not limited to oxygen (O.sub.2), carbon monoxide
(CO), carbon dioxide (CO.sub.2), hydrocarbons (HC), and nitrogen
oxides (NO.sub.x). These sensors can be used in a variety of
devices including, for example, automotive engines, diesel engines,
gas turbine engines, jet engines, power plants, furnaces, and
barbeques. Many of these gaseous components are hazardous.
[0003] Information derived from these sensors can be used for a
variety of purposes. Data from the sensors can be used for feedback
control of different aspects of a device which is producing a
gaseous emission. Alternatively, these sensors can simply be used
to monitor the content of the emission. For example, these sensors
can be used as a component of an on-board, OEM emissions control
system for an automotive engine or as an off-board emissions
measuring device used for inspection and maintenance, for example
as a tool for an automotive mechanic.
[0004] A need exists for sensors which can detect a wide array of
gaseous components. For example, a need exists for a sensor which
can determine the concentrations of oxygen, carbon monoxide, carbon
dioxide, hydrocarbons, and nitrogen oxides in a sample. The sensors
should have a high signal-to-noise ratio and thus be able to
accurately determine the concentrations of various components of a
gaseous sample. The sensors should be simple, reliable, and
inexpensive to manufacture. These and other objectives are provided
by the sensors, devices, and methods of the present invention.
SUMMARY OF THE INVENTION
[0005] The present invention relates to a modified universal
exhaust gas oxygen sensor, referred to herein as a CEGA sensor,
which can be used to measure the concentration of a variety of
components of a gaseous emission including CO, CO.sub.2, O.sub.2,
H.sub.2, and H.sub.2O. The CEGA sensor employs at least one
additional electrode on a ceramic substrate which possess a
different catalytic activity relative to the electrodes that are
normally found on a UEGO sensor. The ceramic substrate may be made
of any suitable ceramic and is preferably made of zirconia.
[0006] The difference in catalytic activity between the additional
electrode(s) and the electrodes native to the UEGO sensor create an
oxygen gradient which enables a measure of combustion completeness
to be calculated. Combustion completeness is a parameter
quantifying the degree to which the gaseous emissions of combustion
are in chemical equilibrium. In combination with an air/fuel ratio
measured by the sensor, the concentrations of different components
in the emission can be calculated.
[0007] A method is also provided for measuring concentrations of
components of a gaseous emission by measuring an air/fuel ratio
using a ceramic sensor, measuring combustion completeness using the
ceramic sensor and determining concentrations of components of a
gaseous emission based on the measured air/fuel ratio and measured
combustion completeness. The CEGA sensor of the present invention
enable these functions of the method to be performed by a single
sensor.
[0008] In one regard, the CEGA sensor is an improved universal
exhaust gas oxygen sensor (UEGO) for measuring properties of a
gaseous emission which includes at least one oxygen pumping cell
and a sensing cell in contact with a detection cavity, the sensing
cell including a ceramic in gas communication inside the detection
cavity, a first electrode in contact with ceramic positioned inside
the detection cavity, and a second electrode in contact with the
other side of the ceramic, a first voltage potential externally
applied between the first and second electrodes for pumping oxygen
across the ceramic into and out of the detection cavity, the first
voltage potential controlled by a second voltage potential formed
across a third and fourth electrode of the sensing cell, an
air/fuel ratio measurement of the gaseous emission being obtainable
from the current passing between the first and second electrodes,
the improvement comprising the addition of a fifth electrode which
has a different catalytic activity than the first electrode
positioned inside the detection cavity in contact with the pumping
cell ceramic, a third voltage potential externally applied between
the fifth electrode and either the second electrode or a sixth
electrode located on the same side of the pumping cell ceramic as
the second electrode, the third voltage potential controlled by a
fourth voltage potential formed between the first and fifth
electrodes, a measure of combustion completeness being obtainable
from the current passing between the fifth and either the second or
sixth electrodes.
[0009] In one particular embodiment of a CEGA sensor, the sensor
includes
[0010] a detection cavity;
[0011] a diffusion passage across which the gaseous emission enters
the detection cavity; an oxygen pumping cell defining a portion of
the detection cavity formed of a ceramic substrate and a first
electrode in the detection cavity and a second electrode outside
the detection cavity for pumping oxygen into and out of the
detection cavity across the ceramic substrate to maintain a target
oxygen level concentration in the detection cavity, an air/fuel
ratio measurement of the gaseous emission being obtainable from
current passing between the first and second electrodes; and
[0012] a sensing cell defining a portion of the detection cavity
formed of a ceramic substrate, the sensing cell including
[0013] a third electrode within the detection cavity,
[0014] a fourth electrode outside the detection cavity, a first
voltage potential being formed between the third and fourth
electrodes due to a difference in oxygen concentration across the
third and fourth electrodes, and
[0015] a fifth electrode in contact with the ceramic within the
detection cavity which has a different catalytic activity than the
first electrode, a second voltage potential being formed between
the fifth electrode and either the second electrode or a sixth
electrode due to a difference in oxygen concentration across the
fifth and either second or sixth electrodes, a measure of
combustion completeness being obtainable from a current passing
between the fifth and either second or sixth electrodes.
[0016] The present invention also relates to several methods,
devices and systems which can be used with various types of ceramic
sensors including the CEGA sensor of the present invention in order
to improve their performance.
[0017] In one regard, the invention relates to a method for
calibrating a ceramic sensor which, as one of its functions,
determines an air/fuel ratio. This method can be used in
combination with any sensor which calculates an air/fuel ratio
including, but not limited to UEGO, NO.sub.x and CEGA sensors.
[0018] According to the method, a ceramic sensor is operated at a
constant, known air/fuel ratio. While being operated at a constant,
known air/fuel ratio, the pumping current (I.sub.pm) of the sensor
is measured. A basic relationship which correlates the air/fuel
ratio to the pumping current for the family of sensors to which the
specific ceramic sensor belongs is then used to calibrate the
sensor by comparing the measured pumping current (I.sub.pm) to the
expected pumping current from the basic relationship for that
air/fuel ratio (I.sub.p). A transformation between the measured
pumping current (I.sub.pm) and the current that the basic
relationship gives for a known air/fuel ratio is created. During
subsequent sensor usage, this transformation is used to modify the
measured pumping current to create a value which is used with the
basic relationship to obtain an air/fuel ratio that is accurate for
the specific sensor.
[0019] In one particular embodiment, the method for calibrating a
ceramic sensor which, as one of its functions, determines an
air/fuel ratio includes the steps of:
[0020] operating the ceramic sensor at a constant, known air/fuel
ratio;
[0021] measuring a pumping current of the sensor;
[0022] comparing the measured pumping current to an expected
pumping current for the constant, known air/fuel ratio; and
[0023] calibrating the sensor using a basic relationship which
provides the expected pumping current for the air/fuel ratio at
which the ceramic sensor was operated.
[0024] The present invention also relates to a software algorithm
which can be incorporated into a system in which the sensor is used
which compares I.sub.pm versus I.sub.p for one or more air/fuel
ratios and produces a look-up table for I.sub.pm versus air/fuel
ratio which can be used during the operation of the sensor.
[0025] The present invention also relates to a semiconductor memory
device which can be used in combination with or incorporated into a
ceramic sensor, the memory device including logic and data for
performing a variety of functions. For example, the memory device
can include logic for calibrating the sensor as well as memory for
calibration data for the sensor. The memory device can also include
logic and memory for storing usage information regarding the
sensor. The memory device can also include logic which monitors and
controls the operation of the sensor. The memory device can also
include logic for detecting when the sensor is being used or has
been used beyond its recommended limits, e.g., temperature, time,
voltage, etc. The memory device can also include a mechanism for
warning the user of the improper use or overuse.
[0026] A method is also provided for correcting for temperature
transients by measuring the temperature of the sensor; and
correcting an output of the sensor based on the measured
temperature. The system for operating the sensor can also include
logic for adjusting the sensors output based on a determination of
the sensor's temperature.
[0027] The present invention also relates to a method for reducing
noise from leakage current from the sensor's heater by taking
measurements when the heater is off or after the effects of the
leakage current have reached steady-state, most preferably just
prior to turning the heater off.
[0028] The present invention also relates to a method for reducing
noise due to coupling between the heater wires and sensing
element's wires by taking sensor measurements before transitions in
the heater's voltage occur.
[0029] The present invention also relates to a method for reducing
noise due to the use of a sensor impedance measuring method for
determining a sensor's temperature by taking measurements just
before the impedance measuring event.
[0030] The present invention also relates to logic for performing
any of the above methods for avoiding noise by controlling when
sensor measurements are taken. The present invention also relates
to logic for determining whether the heater duty cycle is low or
high and for selecting the measurement times based on the duty
cycle.
[0031] The present invention also relates to a method for reducing
noise due to a regulated voltage-type heater in a ceramic emission
sensor system by measuring the noise due to the regulated
voltage-type heater at and subtracting the noise from the sensor
signals in order to compensate for this source of noise.
[0032] The present invention also relates to a method for improving
the accuracy of measuring oxygen-containing species in a gaseous
emission in multiple cavity sensors.
[0033] According to one embodiment, the method is performed by
applying a gaseous emission to the sensor; measuring a pumping
current in a first cavity of the sensor which has a functional
relationship to an air/fuel ratio of the gaseous emission;
measuring a pumping current in a second cavity of the sensor which
has a functional relationship to an amount of oxygen-containing
species in the gaseous emission and the air/fuel ratio of the
gaseous emission; and using a combination of the measured pumping
currents of the first and second cavities to measure an amount of
oxygen-containing species in the gaseous emission. This method can
be incorporated into the sensor by incorporating logic and data for
performing the method into the sensor.
[0034] The present invention also relates to a method for field
calibrating sensors using gaseous emissions. According to one
embodiment, the method of field calibration is performed by
applying a gaseous emission having a known amount of
oxygen-containing species to the sensor; measuring a pumping
current in a first cavity of the sensor which has a functional
relationship to an air/fuel ratio of a model gas; measuring a
pumping current in a second cavity of the sensor which has a
functional relationship to the amount of oxygen-containing species
in the gaseous emission and the air/fuel ratio of the gaseous
emission; and using a combination of the measured pumping currents
of the first and second cavities and the known amount of
oxygen-containing species in the gaseous emission to calibrate the
sensor. This method can be incorporated into the sensor by
incorporating logic and data for performing the method into the
sensor.
[0035] The present invention also relates to a method for
minimizing the effect of rapid emission composition transients on
the accuracy of multi-cavity exhaust sensors. According to the
method, the effect of rapid emission composition transients on the
accuracy of a multi-cavity exhaust sensor is minimized by measuring
the sensor values; detecting for an occurrence of a rapid emission
composition transient; discontinuing usage of the measured sensor
values when the rapid emission composition transient is detected;
detecting for a subsidence in the rapid emission composition
transient; and resuming usage of the measured sensor values when
subsidence of the rapid emission composition transient is
detected.
BRIEF DESCRIPTION OF THE DRAWINGS
[0036] FIG. 1 illustrates an embodiment of an UEGO sensor.
[0037] FIG. 2 provides a graph illustrating that the concentrations
of CO, CO.sub.2, O.sub.2 and H.sub.2 in an engine's exhaust are a
function of the normalized air/fuel ratio.
[0038] FIG. 3 illustrates an embodiment of a NO.sub.x sensor.
[0039] FIG. 4 illustrates an embodiment of a CEGA sensor with two
additional electrodes relative to a UEGO sensor.
[0040] FIG. 5 illustrates an embodiment of a CEGA sensor with one
additional electrode relative to a UEGO sensor.
[0041] FIG. 6 illustrates the basic relationship between the
pumping current (I.sub.p) in a ceramic sensor and the air/fuel
ratio.
[0042] FIG. 7 illustrates an embodiment of a memory device coupled
to a sensor in a system according to the present invention.
[0043] FIG. 8 illustrates an alternate embodiment where the memory
device is built into the sensor.
[0044] FIG. 9 is a graph plotting current from a ceramic sensor
(I.sub.p) as a function of temperature.
[0045] FIGS. 10A-10I illustrate a series of measuring timing
patterns for enhancing the signal-to-noise ratio of the sensor.
[0046] FIG. 10A illustrates a noiseless signal (I.sub.p2) in the
sense that extraneous noise has been eliminated.
[0047] FIG. 10B illustrates the duty cycling of voltage to the
heater.
[0048] FIG. 10C illustrates how the signal illustrated in FIG. 10A
is modified as a result of the leakage current.
[0049] FIG. 10C illustrates how the signal illustrated in FIG. 10A
is modified as a result noise generated due to coupling between the
heater wires and sensing element's wires.
[0050] FIG. 10D illustrates the noise effects associated with the
signal illustrated in FIG. 10B.
[0051] FIG. 10E illustrates the noise effect associated with the
timing of a sensor impedance measurement.
[0052] FIG. 10F illustrates a low heater duty cycle.
[0053] FIG. 10G illustrates a signal with noise generated from a
heater operating with a low heater duty cycle.
[0054] FIG. 10H illustrates a high heater duty cycle.
[0055] FIG. 10I illustrates a signal with noise generated from a
heater operating with a high heater duty cycle.
[0056] FIG. 11 illustrates a NO.sub.x sensor with a regulated
voltage heater controller in a first configuration.
[0057] FIG. 12 illustrates a NO.sub.x sensor with a regulated
voltage heater controller in a second configuration.
[0058] FIG. 13 illustrates signals from a dual-cavity ceramic
exhaust sensor during engine load transients.
[0059] FIG. 14 is a plot of V.sub.s.
[0060] FIG. 15 illustrates a software flowchart for the
technique.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
[0061] The present invention relates to a modified universal
exhaust gas oxygen (UEGO) sensor which can be used to measure the
concentration of a variety of components of a gaseous fuel emission
including CO, CO.sub.2, O.sub.2, H.sub.2, and H.sub.2O. The
modified UEGO sensor, referred to herein as a CEGA sensor, has an
analogous structure to a UEGO sensor but employs at least one
additional electrode on the ceramic substrate which possess a
different catalytic activity relative to the electrodes that
normally found on a UEGO sensor. The ceramic substrate may be made
of any suitable ceramic and is preferably made of zirconia.
[0062] The difference in the catalytic activity between the one or
more additional electrodes in the CEGA sensor and the electrodes
native to a UEGO sensor causes an O.sub.2 gradient to be formed
when the emission is not in chemical equilibrium due to an excess
of either O.sub.2 consuming or O.sub.2 generating reactions
occurring in the vicinity of the electrode with a higher catalytic
activity. By monitoring the size of the O.sub.2 gradient, a measure
of combustion completeness can be calculated. The CEGA sensor, like
a UEGO sensor, is able to measure an air/fuel ratio. By comparing
the combustion completeness and air/fuel ratio measurements, the
concentrations of different components in the emission can be
calculated.
[0063] The present invention also relates to several devices,
methods and systems which can be used with various types of ceramic
sensors including the CEGA sensor of the present invention in order
to improve their performance.
[0064] 1. EGO, UEGO & NO.sub.x Type Ceramic Sensors
[0065] A variety of different ceramic gas sensors have been
developed for detecting different products in combustion emission
components. These ceramic sensors can include a variety of ceramic
substrates including, for example, zirconia (ZrO.sub.2) and
titania. One type of ceramic gas sensor that has been developed is
the exhaust gas oxygen sensor (EGO). These sensors are used to
maximize the efficiency of a catalytic converter which receives the
emissions. For maximum converter efficiency, it must be fed with
emissions from combustion processes operating at a stoichiometric
balance between air and fuel. EGO sensors can only detect whether
an engine is running rich or lean of the stoichiometric point. The
voltage potential (V.sub.s) generated by an EGO sensor can be
expressed according to the following equation:
V.sub.s=(RT/4F)ln(Po.sub.21/Po.sub.22) (I)
[0066] where:
[0067] R is the gas constant;
[0068] T is the temperature;
[0069] F is the Faraday constant;
[0070] Po.sub.21 is the partial pressure of oxygen on side 1 of the
sensing cell exposed to the combustion emissions; and
[0071] Po.sub.22 is the partial pressure of oxygen on side 2 of the
sensing cell exposed to a reservoir of oxygen molecules of a known
concentration.
[0072] Equation I has a very sharp transition value at the
stoichiometric point and hence can be used to identify this point.
The sensor can also be used to measure oxygen concentrations of the
emission in the vicinity of the stoichiometric point. While EGO
sensors can detect whether an engine is operating at, above or
below the stoichiometric point, the sensor cannot detect the actual
air/fuel ratio of the emission.
[0073] A second type of ceramic gas sensor that has been developed
is the universal exhaust gas oxygen sensor (UEGO). Unlike the EGO,
UEGO sensors are "linear" oxygen sensors in the sense that they can
detect the actual air/fuel ratio of the emission.
[0074] FIG. 1 illustrates an embodiment of an UEGO sensor. As
illustrated, the sensor includes three ceramic cells 14, 18, 33 and
a porous diffusion passage 20 into which emissions 23 enter the
sensor. Two of the ceramic cells 14, 33 are used for pumping oxygen
while the third cell 18 is used as a sensing cell.
[0075] All three ceramic cells have platinum electrodes 22 attached
to each side of the ceramic substrate. When a voltage potential is
placed across the pumping cell's electrodes and the cell's
temperature is above 300.degree. C., oxygen is pumped through the
ceramic from one side 26 (or 27) of a cell to another side 27 (or
26) of the cell via the chemical reaction:
O.sub.2+4e.sup.-2O.sup.2- (II)
[0076] The oxygen ions 31 generated during pumping pass through the
ceramic while the electrons travel through external circuitry 29
connecting the electrodes. The amount of oxygen that is pumped
through the cell is detected by the external circuitry by measuring
the amount of current generated, four electrons equaling a pumped
oxygen molecule.
[0077] Sensing cell 18 is an EGO sensor which generates a voltage
potential (V.sub.s) across electrodes 33, 35 when a difference in
oxygen concentration is present across its surface. The voltage
potential (V.sub.s) is used to drive external circuitry 29 whose
purpose is to pump oxygen into and out of the detection cavity 28
so as to maintain a stoichiometric exhaust composition therein.
[0078] The diffusion passage 20 allows combustion constituents to
flow into and out of a detection cavity 28 located between one of
the pumping cells 14 and the sensing cell 18. Overtime, the
composition of components on either side of the diffusion passage
would be equal if not for the pumping of oxygen into and out of the
detection cavity 28 by the pumping cell 14 and the combustion
reactions that occur inside the detection cavity 28. By measuring
the amount of oxygen pumped via the external current flow (I.sub.p)
into or out of the detection cavity 28 to maintain a stoichiometric
exhaust composition, the exhaust's air/fuel ratio is
determined.
[0079] When the exhaust mixture contains excess oxygen relative to
a stoichiometric air/fuel ratio, commonly referred to as lean of
stoichiometric, oxygen from the exhaust diffuses through the
passage into the detection cavity 28. In this case, the pumping
cell 14 removes oxygen from the detection cavity 28 to create a
stoichiometric exhaust composition.
[0080] When the exhaust mixture contains less oxygen relative to a
stoichiometric air/fuel ratio, commonly referred to as rich of
stoichiometric, CO and H.sub.2 from the exhaust diffuse through the
passage into the detection cavity 28. In this instance, the pumping
cell 14 pumps oxygen into the detection cavity 28 which reacts with
CO and H.sub.2 to create a stoichiometric exhaust composition.
[0081] The sensing cell 18 has one side 30 exposed to the pumping
cell 14 and another side 32 exposed to a reference cell 33 which
has a constant oxygen concentration. In some embodiments, the
reference cell is passive and is the environment outside the
sensor. Alternatively, as illustrated in FIG. 1, the reference 33
cell can be an active type reference cell whose O.sub.2
concentration is held constant via a small pumping current.
[0082] Also shown in FIG. 1 are pumping cell control electronics
35. The positive side 34 of an op-amp 36 is held at approximately
0.45 V (the V.sub.s target) and depending on whether the sensing
cells voltage is >0.45 V (i.e., a rich mixture in the detecting
cavity) or <0.45 V (i.e., a lean mixture in the detecting
cavity) the op-amp 36 either supplies electrons (for rich mixtures)
or removes electrons (for lean mixtures) from the reference cell
33. This results in O.sub.2 either being supplied to the detection
cavity for rich mixtures or removed from the detection cavity for
lean mixtures.
[0083] The amount of oxygen pumped by the pumping cell 14 to
balance the diffusion of O.sub.2, CO, and H.sub.2 into the
detection cavity 28 is a function of the following parameters:
[0084] a) the diffusivity Of O.sub.2 through the diffusion
passage;
[0085] b) the diffusivity of CO through the diffusion passage;
[0086] c) the diffusivity of H.sub.2 through the diffusion passage;
and
[0087] d) the concentration of O.sub.2, CO, and H.sub.2 in the
exhaust which is a function of the fuel's chemical composition (H:C
and O:C ratios) and the air/fuel ratio, as illustrated in FIG.
2.
[0088] The diffusion coefficients are principally set by the
structure of the diffusion passage. These coefficients can be
determined by calibration instruments and are provided with the
sensor. The fuel's composition (H:C and O:C ratios) can be input by
the user. With this background information, the air/fuel ratio can
be calculated.
[0089] A third type of ceramic sensor that has been developed is a
NO.sub.x sensor. A NO.sub.x sensor is essentially an UEGO sensor
with an additional diffusion passage and cavity. See U.S. Pat. No.
5,145,566. In addition to detecting the air/fuel ratio as does the
UEGO sensor, the additional diffusion passage and cavity of the
NO.sub.x sensor enables the detection of other emission components
such as CO, CO.sub.2, H.sub.2O and NO.sub.x.
[0090] An embodiment of a NO.sub.x sensor is illustrated in FIG. 3.
As illustrated in FIG. 3, the sensor 40 integrates a UEGO sensor
with a second cavity 48.
[0091] The UEGO portion of the sensor 40, defined by first
diffusion layer 20 and detection cavity 28, acts to keep the oxygen
concentration of the detection cavity 28 at a low and constant
amount by pumping oxygen into and out of the detection cavity 28.
Combustion gases enter the detection chamber 28 via the first
diffusion layer 20 as in a UEGO sensor. By monitoring the pumping
current of the detection cavity (I.sub.p1), oxygen concentrations
and air/fuel ratios are measured.
[0092] Gases within the detection cavity 28 pass through a second
diffusion layer 51 where they enter the dissociation cavity 48. In
the dissociation cavity 48, a constant pumping potential is
maintained which selectively strips oxygen from NO.sub.x molecules
producing a dissociation cavity current I.sub.p2. The dissociation
cavity current is proportional to the NO.sub.x concentration in the
exhaust.
[0093] It is noted that although the NO.sub.x sensor is described
with regard to the embodiment illustrated in FIG. 3, several other
configurations of NO.sub.x known and/or possible and are intended
to fall within the scope of this invention.
[0094] 2. CEGA Sensors
[0095] A modified universal exhaust gas oxygen (UEGO) sensor which
can be used to measure the concentration of a variety of components
of a gaseous fuel emission including CO, CO.sub.2, O.sub.2,
H.sub.2, and H.sub.2O has also been developed. The modified UEGO
sensor, referred to herein as a CEGA sensor, has an analogous
structure to a UEGO sensor but employs one or more additional
electrodes on the ceramic substrate which possess a different
catalytic activity relative to the electrodes that are normally
found on the ceramic substrate of a UEGO sensor.
[0096] The difference in the catalytic activity between the one or
more additional electrodes in the CEGA sensor and the electrodes
native to a UEGO sensor causes an O.sub.2 gradient to be formed
when the emission is not in chemical equilibrium due to an excess
of either O.sub.2 consuming or O.sub.2 generating reactions
occurring in the vicinity of the electrode with a higher catalytic
activity. By monitoring the size of the O.sub.2 gradient, a measure
of combustion completeness can be calculated. The CEGA sensor, like
a UEGO sensor, is able to measure an air/fuel ratio. By using the
combustion completeness and air/fuel ratio measurements, the
concentrations of different components in the emission can be
calculated.
[0097] CEGA sensors provide advantages over existing ceramic
sensors. For example, CEGA sensors are less complex to build and
have a greater speed of response than the sensors described in U.S.
Pat. No. 5,145,566 for measuring exhaust components including CO,
CO.sub.2, and H.sub.2O.
[0098] Two embodiments of a CEGA sensor are illustrated in FIGS. 4
and 5. As illustrated, the CEGA sensor is essentially a wide-range
UEGO sensor such as the sensor illustrated in FIG. 1 where one or
more electrodes have been added. For example, the CEGA sensor
illustrated in FIG. 4 has two additional electrodes 80 and 82. The
CEGA sensor illustrated in FIG. 5 has one additional electrode
80.
[0099] With regard to FIGS. 4 and 5, the first electrode 80 is
positioned in the detection cavity 28 of the sensor on the same
substrate as the pumping cell electrode 26. First electrode 80 has
a different (more or less) catalytic activity than the pumping cell
electrode 26. With regard to FIG. 4, the CEGA sensor also includes
a second electrode 82 positioned outside of the detection cavity
28.
[0100] With regard to the CEGA sensor illustrated in FIG. 5, this
CEGA sensor differs from the CEGA sensor illustrated in FIG. 4 in
that it does not include electrode 82. Instead another electrode
which is used in the basic UEGO sensor, for example electrode 22,
can perform the function otherwise provided by electrode 82.
[0101] The CEGA sensor pumps oxygen into and out of the detection
cavity 28 to maintain a constant sensing cell voltage (V.sub.s) 81
independent of the stoichiometry of the combustion device's exhaust
23 which enters the detection cavity 28 through diffusion aperture
20.
[0102] Electrodes 80 and 26 have different levels of catalytic
activity. If the gas entering the detection cavity 28 is not in
chemical equilibrium, these different levels of catalytic activity
create a difference in the oxygen concentration between electrodes
80 and 26 which can be used to determine the degree of combustion
completeness of the combustion device's exhaust.
[0103] The voltage potential between electrodes 26 and 80 created
by the difference in oxygen concentrations can be used to drive
external electronics whose purpose is to pump oxygen to or away
from electrode 80 in order to reduce the voltage potential between
electrodes 26 and 80. The polarity and amount of oxygen pumped is a
function of the degree of combustion completeness of the exhaust
gas entering the sensor and can be measured as a function of an
oxygen pumping current (I.sub.pe)
[0104] The chemical composition of the emission can be determined
from the detected air/fuel ratio and degree of combustion
completeness as described below. First the total pumping current
(I.sub.p+I.sub.pe) is measured and used to determine the air/fuel
ratio. Determination of the air/fuel ratio can be performed
according as is known in the art with regard to UEGO sensors.
[0105] At any given air/fuel ratio, the percentage of component X
can be represented by the equation:
Xi=Xi,afr+Gi,afr.times.CC (III)
[0106] where: Xi is the mole fraction of exhaust component i;
[0107] Xi,afr is the mole fraction of component i realized in a
combustion device with low combustion completeness;
[0108] Gi,afr is the difference between the mole fraction of
component i realized in a combustion device with high combustion
completeness and Xi,afr; and
[0109] CC is the degree of combustion completeness, 0 representing
0% combustion completeness. and 1 representing 100% combustion
completeness.
[0110] As illustrated in FIG. 2, the chemical composition is a
function of the air/fuel ratio. However, different compositions can
exhibit the same air/fuel ratio. The fact that the air/fuel ratio
does not uniquely determine the chemical composition of the
emission has previously made it impossible to use an air/fuel ratio
sensor as a chemical composition measuring device.
[0111] In a CEGA sensor, the degree of combustion completeness is
determined in addition to the air/fuel ratio. By determining both
the air/fuel ratio and degree of combustion completeness, this
combination of measurements can be used to uniquely determine the
chemical composition.
[0112] With regard to the CEGA sensor illustrated in FIG. 4, oxygen
pumping current (I.sub.pe) is measured across electrodes 80 and 82
and can be used to determine combustion completeness (CC) via the
equation:
CC=CC.sub.MAX-G.sub.CC.times.I.sub.pe (IV)
[0113] where: G.sub.CC and CC.sub.MAX are determined by
experimentation.
[0114] With regard to the CEGA sensor illustrated in FIG. 5,
pumping current (I.sub.pe) measured across electrodes 80 and 22 and
can be used to determine combustion completeness (CC) via the
equation:
CC=I.sub.pe/I.sub.p (V)
[0115] CEGA sensors provide several significant advantages over
prior art air/fuel ratio sensors such as UEGO sensors. For example,
CEGA sensors can be used to determine the chemical composition of
an exhaust from a combustion device as well as the condition of an
upstream catalytic converter, the condition of the catalytic
converter being determined by placing the CEGA sensor downstream of
the converter. If the catalytic converter is working effectively to
bring the exhaust into chemical equilibrium, the potential between
the electrodes 26 and 80 in the CEGA sensor will be small. In this
regard, the CEGA sensor functions much like a second catalytic
converter. Accordingly, if the potential between the electrodes 26
and 80 is small, this indicates that the first catalytic converter
is effectively bringing the emissions into chemical equilibrium. By
contrast, if the potential between the electrodes 26 and 80 is
large, this indicates that the first catalytic converter is
operating at less than 100% effectiveness. As a result, the
effectiveness of the catalytic converter can be measured as a
function of the I.sub.pe.
[0116] Automotive manufacturers currently use two exhaust sensors
to determine the condition of the catalytic converter, one upstream
and one downstream of the catalytic converter. The stoichiometry of
the engine is varied to give a sinusoidal signal to the sensor
upstream of the catalyst. If the sensor downstream of the catalyst
sees the same sinusoidal signal then the catalytic converter is not
effective. The advantage of the described invention is that it
removes the necessity of the exhaust sensor normally positioned
upstream of the catalytic converter.
[0117] The present invention also relates to several methods for
operating ceramic sensors as well as devices which can be used in
combination with ceramic sensors which improve the performance of
the sensors.
[0118] 3. Method for Calibrating Air/Fuel Ratio Sensors
[0119] One embodiment of the present invention relates to a method
for calibrating a ceramic sensor which, as one of its functions,
determines an air/fuel ratio. This method can be used in
combination with any sensor which calculates an air/fuel ratio
including, but not limited to UEGO, NO.sub.x, and CEGA sensors.
[0120] In this embodiment, the accuracy of a ceramic sensor which
determines an air/fuel ratio is improved by using a basic
relationship between a pumping current (I.sub.p) and an air/fuel
ratio and modifying the use of that relationship to calibrate a
specific sensor. FIG. 6 illustrates the basic relationship between
the pumping current (I.sub.p) in a ceramic sensor and the air/fuel
ratio.
[0121] According to the method, a ceramic sensor is operated at a
constant, known air/fuel ratio. While being operated at a constant,
known air/fuel ratio, the pumping current (I.sub.pm) of the sensor
is measured. A basic relationship which correlates the air/fuel
ratio to the pumping current, such as FIG. 6, is then used to
calibrate the sensor by comparing the measured pumping current
(I.sub.pm) to the expected pumping current for that air/fuel ratio
(I.sub.p). A transformation between the measured pumping current
(I.sub.pm) and the current that the basic relationship gives for a
known air/fuel ratio is created. During subsequent sensor usage,
this transformation is used to modify the measured pumping current
to create a value which is used with the basic relationship to
obtain an air/fuel ratio that is accurate for the specific
sensor.
[0122] A software algorithm can be used to compare I.sub.pm versus
I.sub.paf for one or more air/fuel ratios and produce a look-up
table for I.sub.pm versus air/fuel ratio which can be used during
the operation of the sensor.
[0123] This method has the advantage of being computationally
simple, thus enabling the calibration to be performed quickly.
Because the algorithm for performing this method is simple, a small
amount of memory is needed to store the algorithm. In addition, the
algorithm does not require detailed knowledge of the
characteristics of the sensor.
[0124] 4. Memory Device For Ceramic Sensors
[0125] This embodiment of the invention relates to a semiconductor
memory device which can be used in combination with or incorporated
into a ceramic sensor. The memory device can be used in a system
which includes the sensor to control the sensor, calibrate the
sensor, and/or monitor the sensor's usage and performance.
[0126] FIG. 7 illustrates an embodiment of a memory device coupled
to a sensor in a system according to the present invention. As
illustrated, the memory device 90 is designed to be attached to the
gaseous component analyzer 94. The sensor is also attached to the
analyzer.
[0127] FIG. 8 illustrates an alternate embodiment where the memory
device 90 is built into the sensor 92. In this embodiment, the
memory device 90 and sensor 92 have a single connector 98 which
attaches to the analyzer 94 via connector 99.
[0128] The memory device can include logic for performing a variety
of functions. For example, the memory device can include logic for
calibrating the sensor. In addition, the memory device can include
a look-up table for use with the logic to calibrate the sensor. By
using a memory device in combination with a sensor, automated
calibration of gas sensors can be performed.
[0129] For standardized quality control methodologies, it may be
required that the user calibrate the sensor. Calibration is
generally performed at a central location, after which the sensors
are distributed to sites where they are used. The memory device of
the present invention can include logic to store field calibration
information which can be transferred from site to site as the
sensor is used. By using the memory device in this manner,
opportunity for calibration information loss or mistakes in its use
are significantly reduced.
[0130] The memory device can also include logic and memory for
storing usage information regarding the sensor. For example, the
usage memory can be used to record the number of hours that the
sensor has been used. This would allow a sensor manufacturer to
prorate warranty settlements based on actual recorded sensor usage.
The usage memory can also be used to record the conditions under
which the sensor has been used. This information would allow a
sensor manufacturer to see the conditions under which the sensor
was used and to use this information for sensor development and/or
marketing and for warranty issues should the user operate the
sensor outside of its recommended limits.
[0131] The memory device can also include logic which monitors and
controls the operation of the sensor. For example, some sensors can
be damaged if they are heated too rapidly. The memory device can
function to control how fast the sensor is heated.
[0132] The memory device can also include logic for detecting when
the sensor is being used or has been used beyond its recommended
limits, e.g., temperature, time, voltage, etc. The memory device
can also include a mechanism for warning the user of the improper
use or overuse.
[0133] 5. Mechanism and Method for Compensating for Thermal Load
Transients
[0134] During thermal load transients, ceramic sensors can
experience periods where the sensor is not at the desired
temperature. These temperature errors can reduce the accuracy of
the sensor. Even with the use of a closed-loop temperature control
system, significant errors in temperature can often occur. In order
to accommodate for these errors in temperature, logic is provided
for use with ceramic sensors which compensates for errors that can
occur due to temperature transients that are not corrected by a
closed-loop control system. By correcting for these errors in
temperature, the accuracy of measurements from ceramic sensors is
improved.
[0135] FIG. 9 is a graph plotting current (I.sub.p) from a ceramic
sensor (y axis) as a function of sensor impedence. The sensor is
being exposed to a constant quantity of combustion emissions
components and, as such, the current (I.sub.p) should be constant.
The temperature of a sensor can be determined based on the
impedence of the sensor (x axis). Impedence is preferably measured
near the time that the sensor's output signal(s) is measured.
[0136] According to one embodiment, the temperature of the sensor
is measured based on the sensor's impedence. The sensor's output is
then corrected based on the detected temperature if the sensor's
temperature is found to deviate from the desired temperature,
according to the following equation:
I.sub.corrected=I.sub.measured-I.sub.correction=I.sub.measured-G.times.BR.-
times.(R.sub.actual-R.sub.target) (VI)
[0137] where:
[0138] I.sub.correction is the correction to the pumping or
dissociation currents measured in a ceramic exhaust gas sensor;
[0139] G is the gain of the sensor at the particular operating
point;
[0140] BR is the slope of the I.sub.p versus sensor impedence error
curve in the vicinity of the target impedence (R.sub.target) (BR is
normalized to the average sensor gain);
[0141] R.sub.actual is the actual sensor temperature (given here in
terms of an impedence); and
[0142] R.sub.target is the target sensor temperature (given here in
terms of an impedence).
[0143] In an alternative embodiment, the gain of the sensor is
modified by the actual sensor temperature according to the
following equation:
G.sub.corrected=G.times.(1+G.times.MR.times.(R.sub.actual-R.sub.target)
(VI)
[0144] where:
[0145] MR is the sensitivity of the gain to a sensor temperature
error (MR is normalized to the average sensor gain).
[0146] The present invention also relates to a sensor which can
include logic for receiving the resistance data from the sensor and
adjusting the sensor's output based on the above method.
[0147] 6. Mechanism and Method for Timing Sampling to Increase
Signal-to-Noise Ratio
[0148] A need exists to maximize the signal-to-noise ratio of
sensors. FIG. 10A illustrates a noiseless signal (I.sub.p2) in the
sense that extraneous noise has been eliminated. Applicants have
determined the existence of several sources of noise which mask
this signal and have designed a timing pattern for taking sensor
measurements which significantly reduces the effects of these
sources of noise. By incorporating this timing pattern for taking
measurements (i.e., sampling), the signal-to-noise ratio of a
ceramic sensor was improved by a factor of 20.
[0149] Applicants have detected the presence of leakage current
from the heater of a ceramic sensor to its sensing element. FIG.
10B illustrates the cycling of voltage to the heater. As
illustrated, the heater transitions 101 from an off state 105 to an
on state 103 and transitions 107 from the on state 103 to the off
state 105. FIG. 10C illustrates how the signal 111 illustrated in
FIG. 10A is modified as a result of the leakage current. As can be
seen in FIG. 10C, the leakage current causes the signal amplitude
to increase 115 relative to signal 111 when the heater is on 103
and also introduces a noise effect 113 where the signal amplitude
changes over time during an on or off state. This leakage current
has been found to vary from sensor to sensor and change with sensor
temperature.
[0150] To minimize the effect of heater leakage noise, measurements
are preferably taken from the sensor when the heater is off or
after the effects of the leakage current have reached steady state,
most preferably just prior to turning the heater off. These time
periods are illustrated in FIG. 10C as 119 and 121
respectively.
[0151] Applicants have also detected noise generated due to
coupling between the heater wires and sensing element's wires. To
minimize the effect of noise generated by coupling between the
heater wires and the sensing element's wires, measurements are
preferably taken just before transitions in the heaters voltage
occurs. These transitions are illustrated in FIG. 10B as 101 and
107 and their noise effects are illustrated in FIG. 10D as 117.
[0152] Another source of noise that has been detected is leakage
current due to the use of an impedence method for measuring a
sensor's temperature. FIG. 10E illustrates the timing of the
impedence method where the noise effect is illustrated as 109. To
minimize the effect of leakage current due to the impedence method,
measurements are taken just before the impedence method is to
occur.
[0153] A method is provided according to the present invention for
taking sensor measurements based on a timing pattern which is
designed to avoid the effects of these different sources of noise.
This method can involve consideration of the heater's duty cycle on
the timing pattern. For example, sensor measurements can be timed
to not coincide with the delivery of current to the heater.
[0154] As illustrated in FIGS. 10F and 10G, at low heater duty
cycles, noise 117 dominates the period when the heater is on 103.
It is therefore preferred that sampling be done when the heater is
off when the heater duty cycle is low.
[0155] As illustrated in FIGS. 10H and 101, at high heater duty
cycles, noise 117 dominates the period 105 when the heater is off.
It is therefore preferred that sampling be done when the heater is
on when the heater duty cycle is high. In a preferred embodiment,
the system in which the sensor is used includes logic for
determining whether the heater duty cycle is low or high and for
selecting the sampling times based on the duty cycle.
[0156] 7. Method and System Involving Use of Regulated Voltage-type
Heater with Ceramic Sensor
[0157] Pulse-width modulated (PWM) heater controllers have
traditionally been used with ceramic sensors. These heater controls
serve to cycle the heater between on and off modes and thus include
on-to-off and off-to-on transitions. This type of controller
presents the operational disadvantages of having a long duty cycle
at low voltages and exhibiting coupling between the heater wires
and the sensing element wires during the on-to-off and off-to-on
transitions of the heater.
[0158] According to this embodiment of the invention, a regulated
voltage-type heater may be used with a ceramic sensor. Regulated
voltage-type heaters have not used with ceramic sensors due to the
greater complexity of their design and the observance of current
leakage from the continuous heater voltage which affects the
pumping and dissociation current readings. In this embodiment, the
contribution to the pumping and/or dissociation current by the
heater is measured at regular intervals and subtracted out in order
to compensate for this source of noise. This enables more highly
accurate measurements by allowing for more time for sample
averaging as compared to PWM heater controllers. In addition,
faster sensor start-up is enabled because the controller does not
have to wait until the heater effects have decayed to steady-state
values.
[0159] FIGS. 11 and 12 illustrate a NO.sub.x sensor with a
regulated heater voltage 125. The temperature of the sensor can be
measured by one of a variety of techniques including, for example,
sensor impedance method, heater resistance, and thermocouple.
Measurement of the temperature of the sensor is then used to modify
the regulated voltage 125. The circuit alternatives between the two
configurations, defined by the position of switch 111.
[0160] Configuration 1 is shown in FIG. 11. As illustrated in this
figure, the dissociation voltage V.sub.s2 113 is applied across
electrodes 117 and 119 causing the dissociation of
oxygen-containing species in the dissociation cavity. In this
configuration, the measured current Ip2 is due to the dissociation
and leakage 121 of current from the heater 123.
[0161] Configuration 2 is shown in FIG. 12. As illustrated, the
dissociation voltage V.sub.s2 113 is not applied across electrodes
117 and 119. In this configuration, the measured current I.sub.p2'
is due only to leakage 121 from the heater 123.
[0162] The difference between the measured currents I.sub.p2' and
I.sub.p2 (I.sub.p2'-I.sub.p2) is independent of the effect of the
heater and is used to determine the concentrations of components in
the exhaust 23.
[0163] 8. Method and Logic for Calibrating Sensors
[0164] For a two-cavity sensor, the basic relationship between the
measured pumping current I.sub.p2 57 and the amount(s) of
oxygen-containing species in the exhaust 23 is:
I.sub.p2=current due to pumping out O.sub.2 from the second cavity
(where the dissociation occurs) 48 that enters the second cavity
from the first cavity 28 (the cavity prior to the cavity where the
dissociation occurs) via the second diffusion layer 51+current due
to pumping out O.sub.2 from the second cavity 48 that is created in
second cavity from dissociating oxygen-containing species (ex.
NO.sub.x, CO.sub.2, H.sub.2O). [VIII]
[0165] It is standard practice to control the oxygen content of the
first cavity to a constant amount and thus fix the first term in
Equation VIII to a constant quantity C.
[0166] It is also standard practice to have conditions (ex.
temperature, oxygen concentration, electrode composition) in the
first cavity not conducive to the increase or reduction of the
amount(s) of oxygen-containing species to be measured.
[0167] It is also standard practice to have conditions in the
second cavity conditions (ex. temperature, oxygen concentration,
electrode composition) conductive to the dissociation of the
oxygen-containing species to be measured.
[0168] It is standard practice to choose the level of the
dissociation voltage V.sub.s2 55 of the second cavity to select
those oxygen-containing species to dissociate. At a low pumping
voltage, primarily one species will dissociate. As the pumping
voltage is increased, additional species will dissociate
contributing to a greater I.sub.p2. If a series of pumping voltages
are used, the relative contribution of the different dissociated
species, and hence the relative amounts of the dissociating species
in the exhaust can be resolved.
[0169] The reasons for these standard practices is to make the
first term in Equation VIII a constant and the second term a
function of the oxygen-containing dissociating species, i.e.:
I.sub.p2=C+K.times.Xi
or:
Xi=(1/K).times.(I.sub.p2-C)
or:
Xi=H.times.(I.sub.p2-C)
or:
Xi=H.times.I.sub.p2-A [IX]
[0170] where:
[0171] I.sub.p2=pumping current in the cavity where the
dissociation occurs; and,
[0172] C=a constant; and,
[0173] K=is a constant set by the propensity of the sensor to
dissociate species i; and,
[0174] Xi=the concentration of dissociating oxygen-containing
species i in the exhaust; and
[0175] H=1/K; and,
[0176] A=C/K.
[0177] Equation IX represents a prior art method for measuring the
amount(s) of oxygen-containing species in an exhaust. It assumes
that just one oxygen-containing species is dissociating. If
multiple oxygen-containing species are dissociating then Xi in
Equation IX will have to be replaced by .SIGMA.Xi and different
dissociation voltages V.sub.s2 will have to be used to resolve the
contributions of the different dissociating species to
I.sub.p2.
[0178] One shortcoming of Equation IX is that it is difficult if
not impossible to control without error the amount of oxygen
entering the second cavity. Therefore, due to imperfections in
sensor design and control, C will vary with exhaust
composition.
[0179] A second shortcoming of Equation IX is that it is difficult
if not impossible to avoid reactions that increase or decrease the
amounts(s) of oxygen-containing species prior-to their arrival in
the-second cavity (where they are measured). These reactions result
in errors in measurements of the species. The magnitude of these
errors can vary with exhaust composition because the rates of these
reactions can vary with exhaust composition.
[0180] A third shortcoming of Equation IX is that it is difficult
if not impossible to avoid the reactions of some oxygen freed by
dissociation in the second cavity with molecules such as CO and
HCs. These reactions result in errors in the measurements of the
species. The magnitude of the errors can vary with exhaust
composition because the amounts of the interfering species can vary
with exhaust composition.
[0181] The present invention relates to a method for mitigating the
effects of the above shortcomings on sensor accuracy by using a
knowledge of the composition of the exhaust. The present invention
also relates to sensors and sensor systems which incorporate logic
for performing the method.
[0182] As can be seen from the graph illustrated in FIG. 2, the
composition (i.e. amounts of CO, O.sub.2, HC) of the exhaust of
combustion devices is strongly influenced by the air-fuel ratio of
the combustion device (ex. engine) producing the exhaust. The
air-fuel ratio of the combustion device is a function of the
pumping current I.sub.p1 53 of the first cavity via prior art.
[0183] Using these relationships, the following equation has been
derived:
Xi=H(I.sub.p1).times.(I.sub.p2-C(I.sub.p1)) [X]
[0184] where:
[0185] H and/or C are not constants but rather functional
relationships of the pumping current I.sub.p1 of the first cavity.
These relationships can be determined experimentally.
[0186] The pumping current I.sub.p1 used in Equation X is
preferably normalized to its value in a common and stable gaseous
environment (ex. I.sub.p1 in air). Accordingly, Equation X can be
recast as:
Xi=H(I.sub.p1/I.sub.p1n).times.(I.sub.p2-C(I.sub.p1/I.sub.p1n))
[XI]
[0187] The advantage of the construct of Equation XI is that the
functional relationships between H or C and I.sub.p1 do not degrade
with sensor degradation. This is because the relationship between
I.sub.p1 and air-fuel ratio will scale with the relationship
between I.sub.p1n and the common gaseous environment.
[0188] Using Equations X and XI, methods have been developed for
determining the amounts of each oxygen-containing species (ex.
NO.sub.x, CO.sub.2, H.sub.2O) in a sample of combustion exhaust
using any sensor that dissociates each species and uses the
quantity of oxygen molecules produced by the dissociation as an
indication of the amount of each species in the exhaust. One type
of sensor which may be used in this method are two-cavity sensors,
such as the one illustrated in FIG. 3. However, the method is not
intended to be limited to two-cavity sensors. Rather, the method
can be applied to a single-cavity, or three-cavity, or any other
construct of sensor that uses dissociation to determine the amounts
of oxygen-containing species.
[0189] According to one embodiment, the method is performed by
applying a gaseous emission to the sensor; measuring a pumping
current in a first cavity of the sensor which has a functional
relationship to an air/fuel ratio of the gaseous emission;
measuring a pumping current in a second cavity of the sensor which
has a functional relationship to an amount of oxygen-containing
species in the gaseous emission and the air/fuel ratio of the
gaseous emission; and using a combination of the measured pumping
currents of the first and second cavities to measure an amount of
oxygen-containing species in the gaseous emission.
[0190] One advantage of the present method is a decrease in errors
caused by variations in exhaust composition and sensor degradation,
thus improving the sensor's accuracy. This method allows accurate
calibration of the sensor using a simple combination of model
gases. The use of model gases simplifies sensor calibration which
makes the calibration less costly and quicker to perform, while
retaining the accuracy of a calibration using actual combustion
exhaust. This method permits sensor field calibration in a variety
of environments which broadens the range of applications for the
sensor.
[0191] Sensors are not conventionally calibrated in actual
combustion exhaust. Instead, "model gases" of simplified
compositions are generally used. A model gas is made by blending
gases from tanks of specific molecules. For example, a simple model
gas composition of NO.sub.x, O.sub.2, CO, and N.sub.2 might be used
to calibrate a sensor to measure NO.sub.x. The problem with such
calibration procedures is that it ignores the possible effects on
the accuracy of the sensor of species that are in the exhaust but
are not in the model gases. Such absent species, when in the
exhaust, may either consume dissociated oxygen or cause more oxygen
to be dissociated. The result being that a model gas calibration
may not result in a calibration that is accurate in actual
exhaust.
[0192] Since Equations IX, X, and XI are not limited to either
model gases or actual exhaust, a relationship exists between the
functions H and C determined in a model gas calibration and those
determined in an exhaust gas calibration. For example, in one
embodiment functions H and C determined in a model gas calibration
are used to accurately calculate the amount of species i in an
actual exhaust according to the equation:
Xi=G.times.H.times.(I.sub.p2-M.times.C) [XII]
[0193] where:
[0194] G and M are corrections to H and C when going from measuring
species i in a model gas to measuring species i in an exhaust;
and
[0195] H and C are determined using model gases.
[0196] G, H, M, and C are preferably expressed as functions of
I.sub.p1/I.sub.p1n. G and M are preferably determined once for a
given construction of sensor and type of fuel combusted. If species
i is to be measured in a model gas, G and M are set to 1. The
advantage of the method of Equation XII is that it allows simple
model gas calibrations to give exhaust gas calibration
accuracy.
[0197] In a further embodiment of the present invention, the
relationship expressed. in Equation XII is used to enable field
calibration of sensors. In general, it is desirable that
instrumentation be able to be simply recalibrated while in use.
Such calibrations are called "field calibrations". Field
calibrations are typically two-point calibrations: one point at
zero amount of the species to be measured and one point at a value
greater than the maximum amount of the species to be measured. In
order to perform field calibrations, Equation XII is modified to
give:
Xi=G.times.H.times.SPAN.times.(I.sub.p2-M.times.(C-ZERO))
[0198] where:
SPAN=Xi/(G.times.H.times.(I.sub.p2-M.times.(C-ZERO)))
[0199] at the upper calibration point; and
ZERO=C-I.sub.p2/M when Xi=0. [XIII]
[0200] According to Equation XIII, the result of field calibration
is the determination of the quantities SPAN and ZERO to be used in
Equation XIII. The advantage of this method is that field
calibration can be performed in either model gases or actual
exhaust. If field calibration is performed in model gases, G and M
are set to 1 in the equations for SPAN and ZERO. If field
calibration is performed in exhaust, G and M assume their values as
determined.
[0201] According to one embodiment, the method of field calibration
is performed by applying a gaseous emission having a known amount
of oxygen-containing species to the sensor; measuring a pumping
current in a first cavity of the sensor which has a functional
relationship to an air/fuel ratio of a model gas; measuring a
pumping current in a second cavity of the sensor which has a
functional relationship to the amount of oxygen-containing species
in the gaseous emission and the air/fuel ratio of the gaseous
emission; and using a combination of the measured pumping currents
of the first and second cavities and the known amount of
oxygen-containing species in the gaseous emission to calibrate the
sensor.
[0202] 9. Method and Logic for Compensating for Rapid Emission
Composition Transients
[0203] This embodiment provides a method for minimizing the effect
of rapid emission composition transients on the accuracy of
multiple cavity exhaust sensors while minimizing the associated
additional costs.
[0204] FIG. 13 illustrates signals from a two cavity exhaust sensor
during engine emission composition transients. The sharp transients
in measured NO.sub.x are not real but rather are caused by the
delay in the first cavity's ability to control the level of oxygen
in that cavity. Should the level of oxygen become greater or less
than its target value, the perceived concentration of NO.sub.x, as
measured in the second cavity, becomes greater or less than it
actually is.
[0205] The method of this embodiment involves controlling when
NO.sub.x is and is not measured such that NO.sub.x is measured when
the measurement is accurate and is not measured when the
measurement is inaccurate.
[0206] As shown in FIGS. 3 and 14, V.sub.s has a very abrupt change
in value with air/fuel ratio and oxygen concentration in the
vicinity of the stoichiometric point. Multiple cavity sensors such
as two-cavity sensors are typically operated on or near the
stoichiometric point and are used to control the concentration of
oxygen in the first cavity to a target value. During rapid emission
composition transients, the concentration of oxygen in the first
cavity will go off-target and this will be reflected in an
off-target V.sub.s value. This embodiment uses V.sub.s information
to control the compensation logic.
[0207] FIG. 15 illustrates a software flowchart for the technique.
The software flowchart contains four conditional loops: threshold
loop, peak detect loop, timer loop, and recovery loop.
[0208] The threshold loop determines whether or not to compensate
the measured NO.sub.x value based on a difference between the value
of V.sub.s and the target value, shown in the flowchart to be 0.45
V. When V.sub.s diverges from the target value by more than a
predetermined value, shown in the flowchart as .beta., then the
peak detect loop is entered. In the peak detect loop, the
previously determined value of NO.sub.x is used instead of
subsequent samples. Within the peak detect loop, a maximum
deviation of V.sub.s is monitored for. When a maximum deviation of
V.sub.s relative to the target value is reached, a timer is started
for a predetermined time period, shown as y, and the timer loop is
entered.
[0209] For the predetermined time period, the previously determined
value for NO.sub.x is used until either the predetermined time
period expires or the difference between V.sub.s and the target
value increases. When the time period expires, the recovery loop is
entered. Meanwhile, if the difference between V.sub.s and the
target value is found to be increasing, the threshold loop is
entered.
[0210] In the recovery loop, the sampled NO.sub.x value is used
until V.sub.s deviates from the target value less than .beta. or
V.sub.s begins to diverge from its target value. When either of
these events occur, the threshold loop is entered.
[0211] While the present invention is disclosed by reference to the
preferred embodiments and examples detailed above, it is to be
understood that these examples are intended in an illustrative
rather than limiting sense, as it is contemplated that
modifications will readily occur to those skilled in the art, which
modifications will be within the spirit of the invention and the
scope of the appended claims.
* * * * *