Patent Application
20080006235
The invention relates generally to control of a fuel injection system equipped with a fuel injection solenoid. In particular, the invention relates to an apparatus and method for controlling a fuel pump solenoid by detecting a closure of the solenoid.
In engine systems, the accurate detection of fuel pump solenoid closure or closing point detection for each of the cylinders of the engine is desirable to provide accurate timing of injection of fuel and control of combustion, emissions, smooth regulation of engine speed, and fuel efficiency. Prior art system and methods for detecting fuel injection solenoid closure involve detecting the change of slope of the solenoid current within a window, as the current is reduced from peak pull-in current, to hold current. However, prior art systems are susceptible to variations due to cable lengths between engine control unit and the engine cylinder solenoids, aging, ambient conditions, fuel pumps, and engine speed. Moreover, the location and the characteristic signature of such a detection scheme may vary as a function of pump, impedance of electronics/cabling/solenoid, power supply voltage, engine speed, hydraulic fuel pressure, cam location, aging, and ambient conditions. These variations reduce the reliability and accuracy of conventional methods of closing point detection.
Detecting solenoid closure in a in a fuel injection system is desirable for determining any required compensation in timing of fuel injection. The timing of fuel injection is a critical variable to control in order to achieve operational goals such as combustion quality, fuel efficiency and reduced emissions. Any inaccuracies in closing point detection impact these important goals. An appropriate algorithm with consistent detection and accuracy helps achieve these goals.
Therefore, there is a need for an improved method of detection of closing point and minimizing the susceptibility of closing point detection due to variations that change the characteristics of current waveform. The present invention provides an improved method of closing point detection by regulating the slope from peak current to the time when the closing point is detected.
These and other problems are solved by one of the following features: (i) the signature characteristic of the current being controlled by regulating current slope in response to capacitive charge circuit; (ii) the closing point detection time window used for closing point detection being a function of power supply voltage, and (iii) the time window being adaptively varied to account for drifts, pump and solenoid variations, ambient condition changes, and aging.
In one aspect, the invention relates to a system for detecting a closing of a solenoid. The system includes a capacitive charge circuit electrically coupled to the solenoid. The capacitive charge circuit conditions the current flow through the solenoid to increase the current in response to the closing of the solenoid.
In another aspect, the invention relates to a method for detecting a closing of a solenoid. The method comprises conditioning the current through the solenoid such that the current increases in response to the closing of the solenoid.
In another aspect, the invention relates to a system for detecting a closing point of a fuel injection pump solenoid associated with a cylinder of a locomotive diesel engine. The system includes a solenoid driver circuit for providing a current and a voltage to a winding of the solenoid. The solenoid has a low side driver and a high side driver. The system also includes a sensor for detecting the current through the winding of the solenoid. A capacitive charge circuit is included for modifying the voltage and the current as a function of a capacitive charge value. The system also includes a processor configured with computer instructions for determining the capacitive charge value. The processor is electrically coupled to the capacitive charge circuit and provides the capacitive charge to the capacitive charge circuit. The system also includes a detector for detecting a change in a slope of the current as a function of time. The change of slope changes from a decrease in current over time to an increase in current over time in response to the closing of the solenoid.
In yet another aspect, the invention relates to a system for detecting a closing of a solenoid that includes a capacitive charge circuit electrically coupled to the solenoid. The capacitive charge circuit conditions a current through the solenoid in response to a current slope parameter. The current slope parameter defines a change in the current after removal of an electrical charge to the solenoid.
In another aspect, the invention relates to a method for detecting a closing of a solenoid. The method includes detecting a current through the solenoid and determining a current slope characteristic. The current slope characteristic is a function of the current and time. The method also includes conditioning an electrical characteristic of the solenoid such that the conditioning is in response to a current slope parameter. The current slope parameter defines a change in the current after removal of an electrical charge to the solenoid.
In yet another aspect, the invention relates to a system for detecting a closing of a solenoid. The system includes a current detector electrically coupled to the solenoid for detecting a current through the solenoid. A processor determines a current slope characteristic wherein the current slope characteristic is a function of the detected current and time. The system also includes a capacitive charge circuit electrically coupled to the solenoid for conditioning the current through the solenoid. The conditioning provides a change in the current slope characteristic after removal of an electrical charge responsive to a current slope parameter.
In another aspect, the invention relates to a method for detecting the closing point of a fuel injection pump solenoid associated with a cylinder of a locomotive diesel engine. The method includes specifying a detection time period. The method also includes conditioning an electrical characteristic applied to the solenoid by applying a capacitive charge characteristic to the solenoid. The conditioning changes the current as a function of time. The method also includes detecting a current through the solenoid and determining a current slope. The current slope is the incremental change in the current in respect to time. The method further includes detecting a change in the current slope from a negative current slope to a positive current slope and generating a signal in response to detecting the change in the slope.
In yet another aspect, the invention relates to a method for detecting a closing of a solenoid that includes determining a closing point window for detection of the closing of the solenoid as a function of solenoid power supply voltage.
In another aspect, the invention relates to a system for detecting a closing point of a fuel injection pump solenoid associated with a cylinder of a locomotive diesel engine. The system includes a solenoid driver circuit providing a current and a voltage to a winding of the solenoid. A sensor detects the current through the winding of the solenoid. The system also includes a processor that is configured with computer instructions that specify a closing point window as a function of solenoid power supply voltage. The processor also detects a change in the current as a function of time during the closing point window.
In yet another aspect, the invention relates to a method for operating a fuel injection system of a locomotive diesel engine that includes specifying a closing point detection value for a cylinder of the diesel engine in response to a failure to detect the closing of the solenoid. When a speed and a timing of the diesel engine at the time of specifying is comparable to a speed and timing of the diesel engine of the previously detected closing point detection values, the specified closing point detection value is a function of an exponentially weighted average of previously detected closing point detection values. When the speed and the timing of the diesel engine at the time of the specifying varies from a speed and a timing of the diesel engine associated with a previously detected closing of the solenoid, the specified closing point detection value is a function of an average of generally contemporaneously determined closing point values.
In another aspect, the invention relates to a system for operating a fuel injection system of a locomotive diesel engine. The system includes a solenoid driver circuit that provides a current and a voltage to a winding of the solenoid. A sensor detects the current through the winding of the solenoid. A memory stores the detected closing point detection values and a speed and a timing associated with the detected closing point values. The system also includes a processor that is configured with computer instructions. The computer instructions specify a closing point detection value for a cylinder of the diesel engine in response to a failure to detect the closing of the solenoid. The specified closing point detection value is a function of an exponentially weighted average of previously detected closing point detection values stored in the memory when a speed and a timing of the diesel engine at the time of specifying is comparable to a speed and a timing of the diesel engine of the previously detected closing point detection values. The specified closing point detection value is a function of an average of generally contemporaneously determined closing point values as stored in the memory when the speed and the timing of the diesel engine at the time of the specifying varies a predetermined amount from a speed and a timing of the diesel engine associated with a previously detected closing of the solenoid.
Other aspects and features of the present invention will be in part apparent and in part pointed out hereinafter when the following description is read in conjunction with the accompanying drawings.
Corresponding reference characters and designations generally indicate corresponding parts throughout the drawings.
The present invention proposes a method and system for controlling a fuel injection system equipped with a fuel injection solenoid. More specifically, the invention provides a method and system for reducing fuel injection timing errors associated with a fuel pump solenoid by detecting a change in the closing point timing of the solenoid over a period of time. The time between the application of a voltage to the solenoid and the instant at which the solenoid actually starts moving is critical and is the Closing Point Detection (CPD) timing.
Referring to
Processor 102 provides control signals to solenoid driver circuit 104 including a high side command signal 106 indicating an on/off for the high side powering of the solenoid, a low side command signal 108 indicating an on/off for the low side powering of the solenoid, and a capacitive charge circuit command signal 110 indicating and on/off for the capacitive charge circuit. Processor 102 is also electrically coupled to a linear voltage digital to analog converter (DAC) 118. Processor 102 provides a linear voltage signal 126 to the linear voltage DAC 118 that in turn provides the analog charge value signal 120 for each channel or for each cylinder to solenoid driver circuit 104.
Solenoid driver circuit 104 provides the capacitive charge to solenoid 112 in response to the analog charge value signal 120. The high side current measurement circuit of the solenoid driver circuit 104 provides a high side current measurement output 122 to an analog to digital converter 116. The low side current measurement circuit of the solenoid driver circuit 104 provides a low side current measurement output 124 to the analog to digital converter 116. The analog to digital converter 116 provides the measured data to processor 102. Processor 102 and analog to digital converter 116 coordinate the determination of the channel or cylinder associated with the measurements and determinations.
Referring to
At time to, current 202 is I0 that may in some embodiment be an un-energized state of zero current. In other embodiments, current I0 may be an amount greater than zero. At time t0, processor 102 initiates the solenoid activation sequence which includes generating an on condition in both high side control signal 106 and low side control signal 108 as indicated in timing diagrams 240 and 220, respectively. In response, solenoid driver circuit 104 applied high side powering and low side powering to solenoid 112 and the current increases from I0 at time t0 to I1 at time t0. At this point, solenoid driver circuit 104 measures the currents in solenoid 112 and provides these measurements to ADC 116 that in turn provides them to processor 102. Processor 102 compares the current 202 to a predetermined phase 1 maximum current level and when current 202 reaches the predetermined phase 1 maximum current level provides an off low side driver signal 108 to solenoid driver circuit 104. Phase 1 is shown as 204 that is the time interval between t0 and t4. Upon turning off the low side powering, current 202 decreases from I1 to I2 at time t2. Processor 102 provides an on/off control signal 108 to the low side driver such that current 202 is maintained between a maximum value of I1 and a minimum value of I2, averaging at a predetermined pull-in current value of I3.
Processor continues to pulse the low side driver for a defined period of time and then at time t3 turns off the low side driver as indicated by timing signal 220 at time t3. Upon turning off the pulsing of the low side driver, current 202 through solenoid 112 decreasing as the stored energy dissipates to ground. Processor 102 initiates an on control signal 110 for the capacitive charge circuit at time t4. Time t4 being a predefined period of time following t3, which is referred to as a blanking window 218. Current 202 continues to decrease as a function of the capacitive charge circuit and other electrical characteristics of solenoid 112 and solenoid driver circuit 104. This decrease is illustrated as the decrease in current 202 at time t4 of I2 to the value of I4 at time t5. The slope of the current line as a function of time from I2 at t4 to I4 at t5 being a function of the capacitive charge circuit and the provided analog charge value 120. Processor 102 determines the desired capacitive charge value 120 to regulate, control, or condition solenoid current 202 during this phase 2 (indicated by 206). Processor 102 regulates the slope of current 202 as a function of time between t4 and t5 such that the slope is within an upper bound 210 and a lower bound 212.
In graph 200, time t5 indicates the point in time when the core or actuator of solenoid 112 closes. Upon closing, current 202 increases due to the change in the electrical characteristics of solenoid 112 as a function of solenoid driver circuit 104 and specifically as a function of the capacitive charge circuit. Processor 102 continues to measure current 202 based on measurements of solenoid driver circuit 104 and feedback on the high side current 122 and low side current 124 to ADC 116. When processor 102 detects an increase of current 202 from a previous value (shown as I4) to a value I5, the processor determines that solenoid 112 has closed. The determination of the closing of solenoid 112 may, in one embodiment, be based on a change of the current as a function of time from a negative slope to a positive slope as indicated by the slope of line 220 from I2 at t4 to I4 at t5 and the slope of line 214 from I4 at t5 to I5 at t6. As illustrated, the processor requires a time period t6-t5 to determine the change in current 202 over time is indicative of a closing of solenoid 112.
Upon detection of the closing of solenoid 112, processor 102 provides an off signal via the capacitive charge control signal 110 to the solenoid driver circuit 104. The solenoid driver circuit 104 removes powering of the capacitive charge circuit as indicated at time t6 on capacitive charge signal 260. Processor 102 provides time t6 or the determined time t5 in fuel timing signal 132 to a fuel injection controller 134. Upon removal of the capacitive charge powering of solenoid 112, current 202 decreases from current I5 at time t6 to current I6 at time t7. When processor 102 determines that current 202 has reached predetermined current level I6, processor 102 initiates an on signal via the low side control signal 110 so that low side driver provides powering to solenoid 112. Processor 102 once again pulses low side driver between an on state and an off state such as to maintain current 202 at between a maximum value of I7 and a minimum value of I6, with an average current of I8. I8 is referred to as the holding current and this period of time is referred to as the current hold phase or phase 3 (208). After a predetermined period of time, processor 102 terminates the pulsing of the low side driver powering and the high side powering as indicated at time t9 as shown in graphs 220 and 240. In response to the termination of both the high side powering and the low side powering, current 202 decreases from value I7 at time t9 to I0 at time t10. This is referred to as phase 4210.
When solenoid 112 is closed, fuel is provided by the fuel injection system to the associated cylinder of the engine. This fuel-passing period begins with the actual opening of solenoid 112 and ends when the solenoid opens (not shown) which is some point in time after t9.
In operation, one aspect of the invention is a solenoid closing point detection (CPD) apparatus and method wherein CPD is time or degrees from the time of solenoid actuation to the time when solenoid 112 is closed. The solenoid closure results in a characteristic slope change in the current 202. The computer instructions 130 of the processor 102 determines when and how to condition the current slope change. The accuracy of the detection can be validated using a measurement system for solenoid movement and comparing it to the value computed in software.
As discussed above, solenoid current 202 has four phases as a function of time. Phase 1(204) is the pull-in current phase. At the beginning of this phase, the high side driver is turned on and kept on until the beginning of phase 4 (injection end). During phase 1 (204), the low side driver provides low side powering until a specified pull-in current level is attained I1. After this, the pull-in current is regulated by pulsing the low side driver to average current value of I3. The solenoid current reaches peak current in a specified time t1 that may be 850 microseconds (μsec). The initial current slope during pull in phase can vary as a function of cable length, solenoid, and power supply voltage. The pull-in phase overcomes the damping and/or friction of the stationary solenoid 112. At the end of pull-in phase, low side driver is turned off.
Phase 2 (206) is the closing point detection phase during which a capacitive charge circuitry is enabled thereby providing capacitive charging powering. The window between the time low side driver is turned off t3, and time t4 when capacitive charge circuitry is enabled is called blanking window 218. The capacitive charge value 120 used for the charge circuitry is based on feedback calculation from previous firing of solenoid 112. The closing point of solenoid 112 is detected when the decaying current slope goes from a steady negative to a positive value, as long as this occurs within a specified “CPD window.” The CPD window may be restricted to a predetermined period of time such as 600-700 μsec to avoid saturation of solenoid, current 202 dipping to or below the hold current minimum I6 during this phase. At the instant the closing of solenoid 112 is detected or when the CPD window or predefined time period expires, the capacitive charge circuitry is turned off and phase 2 ends when the current reaches the minimum hold current level I6. As a result of solenoid closure, pressure increases in high-pressure fuel line causing a fuel injector needle to lift.
At the end of pull-in current phase, the low side solenoid driver is turned off, and then the current starts decaying down, at a rate dictated by impedance of the circuit (this varies as a function of solenoid, cable length, ambient conditions, power supply voltage). The capacitive charge circuit is enabled by a charge on/off digital signal provided by the microcontroller for each channel that can reduce/increase the slope at which current decays during the CPD phase. The amount the slope is changed for a particular solenoid depends on the analog charge value provided by the microcontroller through a linear voltage digital to analog converter. Without this feature, it is possible that the decay is so rapid where the hold current is reached before solenoid closure occurred. The capacitive charge circuit is controlled by processor 102 such that the slope of current 202 as a function of time can be varied within reasonable bounds 210 and 212. In the “linear” voltage mode, the transistor is “partially” open, thus resulting in slowing of the solenoid driver turn off. Processor 102 executes instructions 130 that provide a closed loop algorithm to manipulate the capacitive charge circuit to regulate the current slope to a specific value, with dead zone on slope error, and saturation on charge value. The slope estimation from the current measurement is done by various methods for differentiation.
Phase 3 (208) is the current hold phase. During phase 3, solenoid 112 is held in the closed position by holding current 202 at a specified hold current level averaging I8. Current 202 is held at the hold current level I8 by turning low side driver on and off. During phase 3 while solenoid 112 is held in the closed position, fuel is delivered to the cylinder.
Phase 4 (210) is end of injection phase. During phase 4 (210), current 202 in solenoid 112 decreases from I7 to I0 or to zero causing solenoid 112 to return to the open position. Low side and high side drivers are turned off to initiate the beginning of phase 4.
Referring now to
In another comparison,
One embodiment of a solenoid driver circuit for CPD detection is illustrated in
Thus, in operation, the solenoid driver 600 as illustrated in
Mode 4, the On Mode, is used during the positive current slope of the pull-in and the hold phase. Both high side driver 602 and low side driver 610 are ON. Current 202 flows from power source 606 through solenoid 112 and low side driver 610 to ground 624. Current 202 increases rapidly during this mode.
Referring now to
Referring to
In another embodiment of the invention, processor 102 determines a closing point detection (CPD) slope signature by regulating current 202 such that the current slope changes from a negative to a positive slope value in response to a closing of solenoid 112. Processor 102 detects the solenoid closing using an algorithm that places conditions on both di/dt and d2i/dt2 to be positive to determine the closing point. As such, any noise issues by dependence on d2i/dt2 alone are mitigated by the distinctive signature of CPD (as a result of regulated slope with capacitive charge circuitry), and conditioning CPD on di/dt itself. This is an improvement to operating detecting the closing point as a function of d2i/dt2>0). Such a detection scheme is susceptible to noise on current signal, which makes it difficult for the closing to be detected from any noise. Additionally, the present invention is an improvement over simply determining a threshold as a function of d2i/dt2, i.e., declare CPD only if d2i/dt2>threshold. In this case, such a determination is difficult to reliably avoid incorrect detections and at the same time, avoid missing detections.
In another embodiment of the invention, the closing point window 216 may be adaptively determined as a function of real-time measurements and operating characteristics or parameters of the fuel injection system or engine. In this invention, CPD window start is also a function of solenoid power supply voltage and optionally, cylinder position. This multi-dimensional function accounts for known variations of CPD with engine speed, cam position and cylinder position (including effect of solenoid cable lengths to various cylinders and bank). This function is optimized by performing a design of experiments on locomotives with variations of speed and timing for various cylinders. It is expected that other variations such as pump-to-pump and locomotive-to-locomotive still keep CPD within the CPD window. This invention recommends a process for testing with these variations and ensuring that CPD is detected within the CPD window. Beyond these, this invention employs an optional scheme for adaptively varying the CPD window, to ensure more reliable CPD. In this embodiment, the CPD window is calibrated for each locomotive cylinder periodically, either in real time during operation, or at scheduled service intervals. In one embodiment, a CPD window function is initialized per a default function of rpm, timing, solenoid voltage and other known variations. CPD statistics such as a mean and a standard deviation are collected as a function of rpm, timing, etc., for each cylinder, during normal locomotive operation, or per a special test during manufacturing or maintenance service. The CPD window function for each cylinder is adjusted within constraints, such that CPD statistics fall within the CPD window, preferably in the center. The newly computed CPD window function is used for CPD window determination. The CPD window is recalibrated from time t0 time, to account for aging, and locomotive component changes.
This is an improvement over prior art system where the closing point detection window 216 start time, relative to the start of solenoid activation, is a static function of engine speed and injection timing command. Such prior art restricts CPD window 216 to a value typically around 500-700 μsec based on predetermined solenoid saturation measurements.
In another aspect of the invention, a closing point detection (CPD) value from one injection cycle is used for timing compensation for the next injection cycle. Given the variation in operating conditions, variations between CPD are relatively small between two consecutive cycles, this leads to sufficiently accurate timing for emissions and fuel efficiency considerations. However, despite all the features mentioned above, CPD is not detected in certain cases, particularly, when Closing Point is not reached before end of injection. This happens for negative duration angle commands, which may be needed for lightly loaded conditions, or in response to reference speed or load changes. If such conditions persist for a while, CPD may not be detected for a while. Under these conditions where CPD is not detected, a default CPD value is needed for use in timing compensation. Typically, the last detected CPD value from previous cycles is used as default value.
In contrast to prior art system, one embodiment of the current invention determines a closing point deflection default value as a function of the operating engine speed and timing at the time of the failure of detecting the closing point. When the same speed and timing conditions exist, the CPD value is used as an exponentially weighted average of the previously detected CPD values. One such embodiment may be an exponentially weighted average of a sample size of 15 previously detected CPD values. When the conditions of speed and timing have changed significantly since last detected CPD value, a default CPD for the given cylinder is based on the most recent CPD average for the particular operating speed and timing condition.
As can now be appreciated, the systems and methods herein described provide substantial advantages over the prior art. Such advantages include improved accuracy and reliability in detecting the closing of the solenoid. Significantly, the system and method herein described may also be adapted for use with existing fuel injection systems and solenoids.
When introducing elements of the present invention or preferred embodiments thereof, the articles “a”, “an”, “the”, and “said” are intended to mean that there are one or more of the elements. The terms “comprising”, “including”, and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.
Those skilled in the art will note that the order of execution or performance of the methods illustrated and described herein is not essential, unless otherwise specified. That is, it is contemplated that aspects or steps of the methods may be performed in any order, unless otherwise specified, and that the methods may include more or less aspects or steps than those disclosed herein.
In view of the above, it will be seen that several objects of the invention are achieved and other advantageous results attained.
As various changes could be made in the above exemplary constructions and methods without departing from the scope of the invention, it is intended that all matter contained in the above description or shown in the accompanying drawings shall be interpreted as illustrative and not in a limiting sense. It is further to be understood that the steps described herein are not to be construed as necessarily requiring their performance in the particular order discussed or illustrated. It is also to be understood that additional or alternative steps may be employed with the present invention.
| Number | Date | Country | |
|---|---|---|---|
| 60506555 | Sep 2003 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10892852 | Jul 2004 | US |
| Child | 11860297 | Sep 2007 | US |
