Measuring and testing – Instrument proving or calibrating – Volume of flow – speed of flow – volume rate of flow – or mass...
Reexamination Certificate
1998-10-16
2002-04-16
Noland, Thomas P. (Department: 2856)
Measuring and testing
Instrument proving or calibrating
Volume of flow, speed of flow, volume rate of flow, or mass...
C702S100000
Reexamination Certificate
active
06370935
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention concerns the calibration of mass airflow sensors, particularly sensors used with reciprocating combustion engines. More specifically, the present invention relates to a system and method for effecting a calibration on-line or while the engine is in normal operation. The invention further contemplates a self-calibration approach that does not require additional instrumentation or sensors beyond those already present in the engine control system.
Intake mass airflow (MAF) sensors or meters have been widely used in reciprocating combustion engines, particularly with the advent of electronic controls for such engines. One typical application for a MAF sensor is in regulating the air/fuel ratio provided to each engine cylinder. In an application of this type, the engine control module receives signals from the MAF sensor indicative of the mass airflow being drawn into the engine cylinder(s). The ECM can then direct the fuel injector assembly to deliver a particular quantity of fuel into the cylinder(s) to maintain predetermined air/fuel ratios. These ratios can be based upon software algorithms stored within and executed by the ECM or upon a table look-up sequence, also executed by and stored within the ECM.
In another application, MAF sensors are used in exhaust gas recirculation (EGR) control. In order to meet increasingly stringent emissions requirements, most internal combustion engines may include some form of EGR device. In one typical EGR system, a conduit is provided between the cylinder exhaust port and the cylinder inlet port to recycle a certain proportion of exhaust gases back into the inlet air stream. A valve is imposed within the EGR flow path to control the amount of recirculated exhaust gases. While EGR systems have a very positive effect on controlling engine NOx emissions, these systems have a detrimental effect on the engine performance and on other emissions, unless the EGR is well controlled in relation to the intake mass airflow. Thus, the MAF sensor provides data to the ECM that is used in EGR control routines to determine the amount by which the EGR valve is opened, or other mechanisms affecting EGR flow are exercised, to thereby control the recirculation of exhaust gases into the intake airflow.
In most reciprocating engine applications, the MAF sensor is called upon to provide transient airflow information over very large flow ranges between engine idle and full throttle. For example, in a typical diesel using EGR, fresh intake airflow can have a max/min ratio of 20:1 or more.
Like any sensor, whether electrical, electro-mechanical or mechanical, the MAF sensor is subject to output errors. In a typical case, a sensor can have a ±3% error when first manufactured. This error can increase to nearly 20% when the sensor is installed due to uncompensated installation effects.
Thermal MAF sensors are particularly susceptible to problems in maintaining accuracy in their output readings over the engine life. In order to overcome these problems, the MAF sensors can be recalibrated with the engine off-line, or shut down. A variety of systems have been developed for this off-line calibration process. Another problem with recalibration of the sensors is that the calibration is usually dependent upon specific details of the particular engine and intake air system installation. In most cases, the MAF sensor requires an installation specific calibration, which means that the technician performing the calibration must know engine specific data in order to accurately recalibrate the MAF sensor. Most known off-line calibration techniques are disfavored by customers and technicians alike, so that once a sensor is believed to be out of spec the preferred treatment is to replace the unit entirely.
SUMMARY OF THE INVENTION
In order to address the difficulties with prior off-line calibration approaches, the present invention contemplates a system and method for on-line recalibration of the MAF sensor, or more particularly of the sensed MAF values derived from the magnitude of the output of the sensor. Since the present invention contemplates an electronic or software based recalibration, the MAF sensor itself is unchanged so it need not be removed from the engine. The systems and methods of the present invention are implemented at points in the engine operation when the EGR valve can be momentarily closed.
In accordance with aspects of the present invention, a predetermined relationship is maintained between the output voltage of the MAF sensor and a mass airflow value used by the engine control routines. This predetermined relationship is maintained in a memory as a plurality of modifiable values. In one embodiment, these values correspond to constants for a non-linear equation relating MAF voltage to MAF value. In one specific embodiment, this equation takes the form of {dot over (m)}
air
=a(V
MAF
+b)
c
, where the constants a, b and c are the modifiable values stored in memory. With this embodiment, the constants are initially set when the engine is new and remain unchanged until the MAF sensor falls out of calibration. The MAF value {dot over (m)}
air
is used by various engine control routines, such as routines governing emission controls.
In a further feature of this embodiment of the invention, a recalibration buffer is maintained in memory. The recalibration buffer includes associated data pairs of MAF output voltage V
MAF
and an ideal MAF value representative of an MAF value that would be generated by an ideally calibrated MAF sensor. In one embodiment, this ideal MAF value is obtained from a predetermined relationship involving outputs from engine condition sensors other than the MAF sensor. In one specific embodiment, these engine condition sensors include temperature and pressure sensors disposed at the intake manifold downstream from the MAF sensor and the EGR recirculation path. The present invention capitalizes on characteristics of cylinder air charges in which the ideal MAF value based on the condition sensors is substantially equal to the sensed MAF value for a properly calibrated MAF sensor when the EGR valve is closed.
In one preferred embodiment, if the difference between the sensed MAF value and the ideal MAF value exceeds a predetermined threshold, a recalibration procedure is implemented. According to this feature, a regression analysis is performed on the associated data pairs stored in the recalibration buffer. In the illustrated embodiment, this regression analysis generates new constants a, b and c so that the predetermined relationship equation adequately defines the relationship between the MAF voltage and the ideal MAF value in each pair. The results of the regression analysis are evaluated to determine whether the analysis meets certain predetermined quality standards. For example, an r
2
or an RMS evaluation can be used to ascertain whether the actual data pairs are adequately represented by the resulting modified equation.
If the regression quality standards are met, the calculated mass flow values using the new regression constants a, b and c, are compared to the calculated values using the prior regression constants. If the difference between the previous and present values is too great, the MAF sensor is flagged as being too far out of calibration for correction. If the difference is within limits, the new regression constants are stored in memory for subsequent use in calculating {dot over (m)}
Air
for use by the ECM engine control routines. In some cases, a comparison can be directly made between the current and prior regression constants a, b and c themselves to determine the extent of sensor miscalibration.
In accordance with the preferred embodiment, it is anticipated that the process of evaluating the sensed and actual MAF values will occur over several cycles. Consequently, the recalibration buffer will contain several data pairs for use in the regression analysis.
In an alternative embodiment of the invention, the predetermined relationship between MAF output vol
He Chuan
Miller Paul R.
Barnes & Thornburg
Cummins Inc.
Noland Thomas P.
LandOfFree
On-line self-calibration of mass airflow sensors in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with On-line self-calibration of mass airflow sensors in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and On-line self-calibration of mass airflow sensors in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2924133